PASTORAL SYSTEMS RESEARCH
IN SUB-SAHARAN AFRICA
PROCEEDINGS OF THE WORKSHOP
HELD AT ILCA, ADDIS ABABA, ETHIOPIA21 TO 24 MARCH, 1983
sponsored by
the International Development Research Centre (IDRC)
and
the International Livestock Centre for Africa (ILCA)
 
AUGUST 1983
INTERNATIONAL LIVESTOCK CENTRE FOR AFRICA
P.O. BOX 5689, ADDIS ABABA, ETHIOPIA

PASTORAL SYSTEMS RESEARCH
IN SUB-SAHARAN AFRICA
PROCEEDINGS OF THE IDRC/ILCA WORKSHOP
HELD AT ILCA, ADDIS ABABA, ETHIOPIA
21 TO 24 MARCH, 1983
AUGUST 1983
INTERNATIONAL LIVESTOCK CENTRE FOR AFRICA
P.O. BOX 5689, ADDIS ABABA, ETHIOPIA
This o 
Key words
PASTORAL SYSTEMS/RESEARCH/AFRICA SOUTH OF SAHARA/ -
REMOTE SENSING/AERIAL SURVEYING/VEGETATION OR PLANT RESOURCES/
LIVESTOCK/PRODUCTIVITY/MARKETING/ANIMAL NUTRITION/PRODUCT ION STRATEGIES/
WEALTH RANKING/HOUSEHOLDS/SURVEYS/TESTING PROCEDURES/PASTORAL
DEVELOPMENT
Abstract
These workshop proceedings give a full report of the papers presented
at the IDRC/ILCA workshop on 'Pastoral Systems Research in sub-Saharan
Africa' held at ILCA's headquarters from March 21 to 24, 1983. Papers
presented in English are summarised in French, and vice versa. All
discussion sessions at the workshop are summarised in both English and
French. These proceedings give an introduction to pastoral systems
research. A majority of papers focus on the survey and diagnostic
stage of pastoral systems research, and include ILCA's experience of
remote sensing techniques and aerial surveys, the survey of vegetation
resources, livestock productivity and animal nutrition, pastoral
production strategies, the importance of wealth effects, household
studies and labour data collection and livestock marketing studies.
Two papers focus on the scope for improvement in pastoral production and
two case studies highlight the experimental design and testing
procedures relevant to pastoral systems research. The final discussions
at the workshop were related to the strengths and weaknesses of ILCA's
approach to pastoral systems research, and how this approach might be
improved .
Preface
This document summarises the proceedings of the workshop on
Pastoral Systems Research in sub-Saharan Africa held at ILCA's
headquarters in Addis Ababa from 21 to 24 March 1983. The workshop
was sponsored by the International Development Research Centre (IDRC),
Ottawa, Canada and by ILCA. It was attended by 15 scientists from
different countries in sub-Saharan Africa, by 20 scientists from
ILCA and by 3 others.
The workshop focused on the techniques and processes in
pastoral systems research with particular reference to ILCA's
experience in sub-Saharan Africa. The 21 presentations were made
by ILCA staff members and the discussion sessions were led by
national representatives from different African countries. At the
end of the workshop a panel of national representatives reviewed
ILCA's progress in pastoral systems research and made suggestions
for future work.
This document contains the written presentations made at
the workshop and summaries of the discussion sessions. Each
presentation appears in its source language and is followed by a
summary in English or French, as appropriate. Summaries of each
discussion session are presented in both English and French.
Acknowledgment is made to the ILCA staff who assisted in
the running of the workshop, in particular Mr Addis Anteneh, the
workshop organiser; and to Senedou Leon and Sophie Gorfineh who typed
these proceedings.
- in -

Contents
Page
General Introduction
Introductory remarks
Workshop background, organisation
and procedure
The development experience*
Towards a framework for pastoral
systems research (PSR)
P J Brumby
Addis Anteneh
S Sandford
C de Haan
7
11
25
Summary/Resume of discussion session 1. 45
The survey and diagnostic phase of PSR
Integration of remote sensing
techniques for resource
evaluation in pastoral
systems research
Low altitude aerial surveys in PSR
P N de Leeuw and
K Milligan
K Milligan and
P N de Leeuw
57
81
Summary/Resume of discussion session 2. 107
Les vegétations et les ressources P Hiernaux
fourrageres dans les systemes
pastoraux
L'evolution du milieu J C Bille
113
133
Summary/Resume of discussion session 3. 147
Livestock productivity and management
Animal nutrition
R T Wilson and
P Semenye
L J Lambourne,
M S Dicko,
P Semenye and
M Butterworth
151
183
N.B. Each presentation made in English is followed by a French
summary, and vice versa.
- v -
Summary/Resume of discussion session 4. 209
Production strategies and pastoral man N Cossins 213
The importance of wealth effects on BE Grandin
pastoral production: A rapid
method for wealth ranking 237
Summary/Resume of discussion session 5. 257
Household studies in PSR B E Grandin and
Solomon Bekure 263
Livestock transactions data collection B E Grandin 277
Household income and expenditure Solomon Bekure
studies 289
Labour data collection B E Grandin 305
Summary/Resume of discussion session 6. 321
Livestock marketing studies Solomon Bekure and
Negussie Tilahun 327
Summary/Resume of discussion session 7. 357
Features and constraints identified in PSR at ILCA
Summary/Resume of discussion session 8. 361
Identifying the scope for improvement in pastoral production
The scope for improvement S Sandford,
F Anderson and
Addis Anteneh 365
Biotechnical options L J Lambourne and
M Butterworth 383
- vi -
Summary/Resume of discussion session 9. 403
Case studies in PSR
Design and testing procedures in R von Kaufmann
livestock systems research: An
agro-pastoral example 407
Summary/Resume of discussion session 10. 431
A proposal for pastoral development in P N de Leeuw and
the Republic of Niger C de Haan 435
Summary/Resume of discussion session 11. 459
Strengths and weaknesses of PSR
Summary/Resume of discussion session 12. 463
Final discussion on PSR
Summary/Resume of discussion session 13. 471
List of Workshop Participants 477
- vn -
i ■■ I i'i
GENERAL INTRODUCTION

Introductory remarks
Peter J. Brumby
Director General, ILCA, Ethiopia
May I warmly welcome you here this morning. This workshop on our re
search in Pastoral Production Systems has been in the making for sev
eral years, plus a bit longer, the bit longer being the additional
time we needed for the hard work in preparing the papers for this
meeting, as well as the time needed to develop the courage to expose
our experience and ideas to you.
Very little has been written on the analysis of the live
stock aspects of farming systems research (FSR) . The 1981 compendium
of Farming Systems Research in Africa contains virtually no references
on the subject. But this situation is changing. In 1982, IDRC
published an excellent monograph on issues and policies influencing
livestock development in Asia and this publication contains several
valuable papers on the analysis of livestock systems. Winrock
International are also increasingly productive in this subject,
particularly in respect of Latin America. We at ILCA have now
published nine major reports on various livestock farming systems in
different parts of Africa. Excellent journals on the subject of FSR
generally are also available; unfortunately, however, they contain
very little livestock material.
Systems analysis is a term, like so many others that will
be used in this workshop, that means different things to different
people. To a computer engineer the meaning of systems analysis is
very clear, but we use the term in a broader and less precise manner.
The phrase 'systems analysis' or 'farming systems research', or the
fancier semantic derivatives that describe a fancier research focus
that extends beyond the farm level, such as 'farming systems per
spectives' and 'farming systems research and development1, provide
little more than an innocuous umbrella under which people of different
disciplines and background can work together, interactively, on the
task of improving livestock and crop output and income.
- I -
Our discussions this week will focus upon rather specific and
technical parts of our approach to livestock systems. The techniques
we use in this work are obviously very important, but it is the
approach we take to livestock and pastoral farming systems that I
would like to see characterizing your views of ILCA's work, rather
than the use of any of the specific research techniques we shall
discuss this week.
The ILCA approach consists of two main items: a character
istic philosophy and a collection of techniques. I shall spend a
few minutes talking about the philosophy that underlies our work; my
colleagues will discuss mainly the techniques used and the results
obtained. In brief, the philosophy is that there are such things as
systems, and that things that operate as systems have to be regarded
rather differently from things that operate in some other manner.
What is a system, are systems important and in what way do
systems differ from non-systems? Let me try to explain by referring
to ILCA as an institution. At ILCA we regard our organisation, ILCA,
as a system. We, the staff, interact and work together for a common
purpose and thereby have the key characteristics of a system. Thinking
unkindly, other organisations may be regarded as a collection of
individuals, and a collection does not constitute a system. With a
collection you can take out a component and it makes little difference,
similarly you can add one without making much difference. But if you
interfere with one component of a system such as ILCA, remove it,
alter it, or add another, you are likely to have important effects on
the whole system, because it is a characteristic of a system that the
components interact. It follows that grey hair and system management
go well together !
There are two other important and inter-linked characters of
a system I want to emphasise. The first is that a system has a bound
ary and that boundary must be clearly established in our description
of the system. The second is that a system is purely a concept. By
this I mean that an adequate description of a specific system will
depend on the purpose of the description and the boundaries that are
perceived for the system; a naturalist defines the same aggregation
- 2 -
of interacting activities in quite a different way to an economist,
a politician or an agriculturalist. Obviously you need a clear notion
of the specific purpose of your description of a system in order to
describe either the system or its boundaries.
You will also appreciate that a description of a system is
highly biased by the training of the person you have asked to describe
it. If you doubt me set loose an agriculturalist, an economist and
a sociologist to describe a specific system such as ILCA. Each will
give an appropriate description that suits their particular purpose,
but none is likely to be appropriate for the purpose of understanding
the ILCA system as a whole. The first task is to think carefully
about what is to be described, and even more importantly about what
does not need to be described - if you try to describe everything you
will never finish and will never produce a description that is of use
to a specific audience.
To decide if a component observed in a system is important,
one has to relate it to the purpose for which the system exists. In
the ILCA system, for example, one finds components such as sociol
ogists, chemists and librarians. It is only when we know the purpose
of ILCA that we can start to decide if describing these specific
components is important. If it is argued that the purpose of ILCA
is to increase milk production, then we can decide on the basis of
the likely effect of our sociologists etc. on milk production whether
they are an important part of our system, or whether they can be
disregarded in our description. If the purpose of ILCA is perceived
as something different than simply increasing milk production, our
description and judgement of the vital component parts must be
different.
A key objective we all share is to improve livestock pro
duction; we also probably share a belief that using a systems approach
is a sensible way to tackle this objective. This accepted, the key
questions we then need to ask. are what systems are to be improved and
what constitutes an improvement.
In many livestock systems of concern to us there is very
little information available and it is tempting to start extensive,
time consuming baseline surveys to try to build the data necessary
- 3 -
for initial modelling and intervention. If this exercise is followed
we frequently find, several years later when modelling and analysis
of the data collected is started, that some key data required have not
been collected. My very important point here is that it is essential
that data collection, data analysis, and modelling go hand in hand.
We must start with a very simple model at the outset to help determine
what information is actually required and not wait until we have
accumulated large lumps of information that are likely to be both too
much and too little.
Having described in adequate terms the systems to be improved
one then has to decide what constitutes an improvement. Improving
milk production per cow may be technically possible in several
different ways, but each way suggested may be financially or organi
zationally impractical. Clearly we need to consider not only increased
outputs but also increased inputs; we need to consider if the inputs
proposed are likely to be available, can they actually be safely used
by our target farmers, and what do they cost? One also quickly
realizes that livestock systems do not produce just one thing, and
that they are not using just one resource. "Improving" a system
that produces several products and uses many resources is a matter
beset with many complications that are frequently overlooked in an
otherwise simple technical approach to livestock improvement.
We must also ask who benefits from any improvement that may
be made. If output is increased does the farmer benefit, or do prices
decrease? In the latter case does the nation as a whole reap the
rewards, or just one small part of the nation?
If the objective is to increase overall food production,
obviously a most important task in Africa, we still need to ask if
fulfilling this objective will feed hungry people. Hungry people are
poor people; if they weren't poor they wouldn't be hungry; if they had
the money to buy food, that food would certainly be produced and
available. Clearly, a food production objective alone is inadequate
and we need to consider equity distribution, cash flows, labour
requirements, multiplier effects and all the rest of the bits and
pieces that are critical in deciding on the value of our improvement.
- 4 -
Farming systems are rarely static, they are usually being
continually changed in many ways by a multiplicity of external forces.
We as systems researchers must make judgements as to where our target
systems may be headed. Research appropriate to improving the pro
ductivity of large farms is frequently inappropriate to improving
farming systems on small farms, and we need to be reasonably sure
that the target system for our research effort has an appreciable
life span.
ILCA's task is twofold : to apply existing knowledge to
improve livestock production, and to undertake research on major gaps
in that knowledge. The technical base for improving livestock pro
duction in Africa has proved to be less extensive than was perceived
when ILCA was established just a few years ago. In consequence, we
have recently been placing greater emphasis in our work at ILCA on
what we loosely call component research. Yet I urge great caution
on your part in regard to the use of the term "component research".
Our primary interest and concern is not in understanding how a
legume grows, or how a cow copes with a protein deficiency, but how
key changes in vital parts of livestock production systems, such as
improving pasture quality and dry-season nutrition, might improve
the productivity and efficiency of the farming system as a whole.
Our main task in many of the systems we work with lies in assessing
the relationships among their key components. Strictly speaking this
is not component research, rather it is process research. Others,
however, may regard it as a development activity. May I suggest
that the distinction between research and development is largely
in the eye of the beholder. In line with our purposive philosophy,
we regard the difference between the two as the reason one gives as
to why the particular work is being done.
May I make a final comment about organising the staff of
any institution to carry out a systems research approach. Many ways
of doing this have been tried and the virtues of multidisciplinarity
are usually loudly emphasised. I want to stress, however, there is
little merit in having many disciplines represented if they are not
required, and that the pursuit of multidisciplinarity as such can be
overdone. In our experience what is required is staff with knowledge
- 5 -
that is highly relevant to the systems you are studying, and this
staff needs to encompass a very wide range of training and experience.
First class generalists backed up by specific selected specialists is
the formula we seek. I am sure you will note this amongst our staff
at ILCA this week.
During the course of the next few days we expect to learn a
great deal from you. We appreciate greatly the time you have given
to share your experience and your wisdom. We also value highly your
willingness to work with us in helping develop our overall programme.
IDRC have made a sizeable cash contribution to the organisation of
this workshop. Their assistance has made it possible to invite more
participants than would otherwise have been the case and it has been
a spur to us to hold this important meeting. We are most grateful to
them. We hope that you will benefit from the discussions of the next
few days, and that you greatly enjoy your stay with us.
- 6 -
Workshop background, organisation and procedure
Addis Anteneh
Workshop Organiser, ILCA, Ethiopia
My sitting here and trying to explain the background to this workshop
arises out of a historical accident. At the time when a definite idea
had been formed about organising this workshop, I happened to be
working with Dr. Poul Sihm, who was then the co-ordinator of arid
zones research at ILCA. He was at the centre of the initial stages
of the preparation of the workshop. Poul Sihm returned to the World
Bank last year, but before we parted he extracted a promise from me
that I would make sure this workshop took place. On his part he
promised to keep a close watch on the workshop's progress and to
participate in the discussions. Unfortunately he is unable to attend
the workshop due to heavy pressure of work, but he has certainly
kept an interest in its organisation - some of our distinguished par
ticipants work in World Bank livestock projects with which he is
associated and were highly recommended by him. To this extent at
least he has kept his side of the bargain. I hope I have succeeded
in keeping mine by helping in the workshop's preparation.
I do not want to burden you with a detailed chronological
account of what has happened since this workshop was first suggested
two years ago. What I would like to give you is a brief pointer to
those events which, in my opinion, have a story to tell about the
dynamics of ILCA's work.
The origins of this workshop are very much related to one of
the first livestock research activities of ILCA - that of monitoring
livestock production systems under induced change. ILCA had then
undertaken to monitor changes taking place in production and the
social and economic impact generated by livestock development projects
in the arid/semi-arid zones of Kenya, Ethiopia and Botswana, and had
entered a commitment to develop and prepare a monitoring guideline.
Monitoring subsequently became a rather unpopular term and activity,
- 7 -
not because it was not necessary but because it was being used and
carried out mostly in the project financing framework, not totally
befitting the character of an international research centre such as
ILCA. Clearly, as you can realize, there could be no such thing as
a monitoring guideline under the circumstances. The move away from
monitoring to livestock production systems research in these areas
entailed the need to clarify conceptual problems as well as questions
of using systems research techniques and methodologies for livestock
which, as the Director General pointed out earlier, have not featured
strongly in the farming systems research approaches now coming into
vogue. And in the meantime, African research and development insti
tutions concerned with livestock as well as ILCA were continually
gaining experience in the field. With all these things taking place,
it was evident that ILCA alone could not produce a pastoral systems
research (PSR) /livestock systems research (LSR) guideline or manual
as a unilateral prescription.
It was thus decided quite early to convene this workshop
to exchange ideas and experiences with selected African leaders in
livestock research and development. The objectives of using this
workshop to provide a forum for discussion of the merits of the
PSR/LSR approach has thus become a primary one. The possible estab
lishment of a network of pastoral/range livestock systems research
in tropical Africa is another important objective which we hope you
will consider. The development and preparation of a set of PSR/LSR
guidelines remains a serious goal. We hope all participants will
consider and give their views on this matter before the closure of
the workshop.
The schedule of presentations and discussions is indeed a
tight one. For the next few days we will have 3£ to 4-hour sessions
each morning and afternoon. We have tried to arrange the sessions to
present a logical flow of ideas. They start with the conceptual
framework and continue with the survey and diagnostic phase, including
the technical, social and economic components. Specific constraints
identified, which follow on the third day, are based on the preceding
phase and are presented in the framework of ILCA's on-going pastoral
systems research programmes. We finish with the identification of
- 8 -
the scope for improvement on the fourth day and will close after the
presentation and discussion of case studies to illustrate in some
detail ILCA's experience in the use of PSR/LSR design and testing
procedures in two areas - Nigeria and Niger.
We have arranged for either ILCA Board Members or senior
members of ILCA's management to chair the different sessions, Their
intimate knowledge of the issues and processes involved in ILCA's
research work or their extensive experience in pastoral systems
and the critical component problems facing these systems' potential
improvements, will be a valuable asset in guiding the discussions.
The agenda submitted to you has gone through several revisions
to take account of the changing picture presented by ILCA's growing
field experience of livestock systems research in its on-going pastoral
programmes and elsewhere. Many of the participants to whom we were
able to send the schedule, together with our invitation to attend the
workshop, are likely to be suprised at the considerable changes intro
duced since then. I hope you will appreciate how much thought and
effort on the part of ILCA's management and staff has gone into the
substantive aspects of the workshop's preparation.
As you will also note from the way the programme is structured
we have gone a little further in our attempt to make participation as
effective as possible. The device we proposed and which was accepted
by most of our colleagues was for a review of specific topics, pre
sented by ILCA staff, by selected discussion leaders from among our
African colleagues. We had hoped to make the papers available some
time before the start of the workshop, but we have not been able to
do so for logistic reasons. We offer our apologies, but are certain
that this particular handicap will be offset by the extensive knowl
edge and experience our lead discussants possess on the subject matter
addressed by the papers. We expect the lead discussants to take the
floor first after the papers have been presented, this to be followed
by a general discussion, except in those cases where there is only
provision for open discussion.
In the final analysis, the outcome of this workshop will have
to be judged more by how much it will succeed in imparting better
knowledge from the experience and the generation of new ideas than by
the logistics which have gone into its organisation and running. My
colleagues in Training and Conferences, Liaison and Travel hope that
your participation will be useful and fruitful and that your days
inside and outside this conference hall will be enjoyable.
Finally, let me be permitted to thank on behalf of ILCA
those individuals and institutions outside ILCA who have been
instrumental in assisting the workshop to take place. We would like
to thank Dr. Barry Nestel for his support and advice both on the
concept and structure of the workshop and in obtaining IDRC support.
Our sincere thanks also go to IDRC who financed a substantial portion
of the cost of the workshop. We are most thankful and appreciative
of Dr. Hubert Zandstra, Associate Director of IDRC's Agriculture
Food and Nutrition Sciences Division and Dr. Bruce Scott, IDRC
Region; 1 Director in Nairobi, without whose continued help and
interest it would have been difficult for many of us to attend.
- 10 -
The development experience
Stephen Sandford
Economist, Livestock Policy Unit, ILCA, Ethiopia
The purpose of this paper is to set our consideration of pastoral
systems research within the framework of the experience of livestock
and pastoral development in Africa during the last half century.
The paper starts with some preliminary background material and then
turns to those elements in past experience which are of particular
relevance to pastoral systems research.
African livestock and pastoral background
I shall start with some statistics on populations (mainly based on
Jahnke, 1982). Pastoral systems occur mainly in arid and semi-arid
zones, which together I call "dry regions". Dry regions occupy about
55 % of tropical Africa's land surface area- and account for 60 % of
its ruminant livestock population (expressed in terms of tropical
livestock units of 250 kg liveweight equivalent) . Not all the live
stock of the dry regions are involved in pastoral systems. Depending
on one's definition of a pastoral system, the proportion of tropical
Africa's total ruminant livestock population involved in pastoral
systems probably lies between 30 % (if one counts only the population
of the arid zone) and 50 % (if one includes up to 2/3 of the livestock
population of the semi-arid zone) . The figure for the human pastoral
population is even more difficult to determine - largely due to
problems of definition - but the true figure probably lies between 15
and 25 million people, representing 6-10 % of tropical Africa's total
rural population.
I now proceed to some data on productivity (drawn largely
from Jasirowski, 1973, and de Montgolfier-Kouevi and Vlavonou, 1981).
Between 1950 and 1970 Africa's human population grew at about 2.5 %
- 75 % of its tsetse-free land surface area.
11 -
per annum but meat and milk output grew at just over 2 %, indicating
declining output per human caput. From 1970 until 1975 the rate of
growth of the human population increased while that of livestock
output decreased, indicating an even faster decline in per caput
output. Effective demand, i.e. demand backed by cash to pay, has been
rising faster than population so that the relation between domestic
output and domestic demand has deteriorated at an even faster rate
than per caput output. The result has been a decline in the export
of livestock and their products and a rise in imports.
Increases in the total output from African ruminants has
more or less matched increases in total ruminant population. In other
words there has been no apparent change in productivity per head of
ruminants at least until the mid 1970s. As far as one can tell these
figures for tropical Africa as a whole are matched by figures for the
dry regions.
I do not want to decry the devoted work of African national
statisticians, and of FAO and ILCA staff who have laboured to produce
chese statistics. However, they would be the first to agree that
these figures are often not well based in reality. The general
picture they convey is probably right but might be wrong.
It is very difficult to find firm evidence concerning
changes in the welfare of the human pastoral population in Africa.
On the one hand there has been an encroachment by non-pastoralists
into previously pastoral areas and there has been some population
growth among pastoralists themselves. These factors suggest an
increase in pressure on resources, with a probable consequent decline
in welfare. On the other hand, mainly due to development of water
supplies, the extent of the dry region effectively accessible to
exploitation by livestock has increased, offsetting some of these
pressures.
Most studies by social anthropologists suggest a decline in
pastoralists' welfare. However a very recent study (Jamal, 1983) of
Somalia suggests that pastoralists, on average, fare significantly
better than crop farmers, mainly because of changes in the ratios
between the prices of livestock and grain which enabled pastoralists
to buy more grain in exchange for each animal sold. But Somalia's
- 12 -
easy access to the oil-rich and meat-hungry markets of the Middle East
may make it a special case. Although in terms of the welfare criteria
traditionally used by pastoralists (e.g. ownership or consumption of
livestock, meat and milk), the welfare, on average, of pastoralists
has declined, this has been more than matched by increased access to
a range of new goods and services. I suspect, therefore, that on
average the material welfare of pastoralists has increased. However,
most studies agree that inequality within pastoral societies has
grown, for a variety of reasons, so that the absolute number of pas
toralists below a given level of material welfare, i.e. the number
of poor people, may have increased, and almost certainly there has
been an increase in both the number and proportion of those who feel
themselves to be under-privileged relative to some norm.
An outline of pastoral development efforts
Pastoral development in the past has consisted of a mixture of general
programmes, e.g. veterinary services, and of special development
projects. It is difficult to summarise effectively the extent of this
mixture. On a financial scale we can note that the total cost of
government livestock development efforts in all ecological zones of
tropical Africa between 1960 and 1975 has been estimated as USjS 600
million (Wissocq, 1978). Funds committed only to those special
livestock projects in dry areas of tropical Africa in which one donor -
the World Bank - was involved between 1965 and 1980 amounted to VS$
600 million and the livestock elements of further mixed crop-livestock
projects in dry areas funded by the Bank amounted to a further US# 200
million (Sandford, 1981) . These figures should be viewed in relation
to a total annual gross value of livestock output in all ecological
regions of tropical Africa of about US$ 6 billion (at 1975 prices) if
one excludes the value of traction and transport services supplied by
livestock, and USj5 10 billion if one includes them (de Montgolfier-
Kouevi and Vlavonou, 1981) .
The nature of, i.e. the type of component involved in, pas
toral development has tended to change over time with fashion and
technology. In the 1920s and 1930s the main emphasis was on veter
inary programmes to fight the three major diseases of the dry areas,
13 -
rinderpest, contagious bovine pleuropneumonia and anthrax. From the
1920s onwards, but primarily from the 1950s, there was a tremendous
surge in the development of water supplies. Over the same period
there was a growth in commercial ranching in those regions with
European settlers. In some cases as early as the 1930s, but more
generally after the Second World War, there were attempts to introduce
controlled grazing schemes in areas used by traditional pastoralists.
Until the mid 1960s these efforts were financed primarily by
domestic resources generated from within the countries of tropical
Africa, with some external supplementation by private capital in the
case of commercial ranching. The pace of development was rather slow
and the main constraint was lack of finance for both staff and other
forms of expenditure.
After about 1965 there was a significant quickening in the
pace of development with the arrival on the scene of both the World
Bank and USAID as important financiers of pastoral development. Since
the mid 1970s the EEC/EDF has also become a financier of pastoral
development on a significant scale. From 1965 to 1980 finance was no
longer the critical constraint but experienced staff (whether local
or foreign) , viable components and suitable government policies and
institutions became the main factors limiting progress.
From 1965 until about 1975 great emphasis was put on the
development of ranching, on the model of the European settlers'
ranches, by individuals, parastatal organisations and cooperatives
of pastoralists. At this time little emphasis was put on general
veterinary services other than the Pan African JP 15 rinderpest
campaign.
From about 1974 onwards there was a progressive disenchant
ment with commercial ranching and a determined search for other
components. There was some return to favour of general veterinary
programmes but these were oriented more than before to problems other
than those of the three major diseases which had been the focus of
attention in earlier periods. There was some emphasis on "stratifi
cation", i.e. specialisation in production by regions and enterprises,
with most attention being paid to intensive feedlots and to fattening
of livestock on peasant smallholdings. Great emphasis was placed on
- 14 -
reforms in land tenure in pastoral areas, and especially on the
association of relatively small groups of people with defined areas
of land. This association was done on a smaller more detailed scale
than the previous general association of one or two tribes with very
large blocks of land. Sometimes this association of land and people
took the title of "group ranches" (which have seldom behaved at all
like commerical ranches) but in recent years the fashion has been
to talk of "pastoral units".
From World War II onwards there has been a fairly continuous
interest by governments in the reforming marketing systems. This has
often involved the introduction of direct trading by government or
of government control over a private marketing system. Sometimes
these marketing systems have been linked with processing facilities,
sometimes not.
The lessons from development efforts in the past
Although livestock development in Africa has often been characterised
as peculiarly disastrous, in fact from 1963 to 1975 (de Montgolvier-
kouevi and Vlavonou, 1981) the growth rate of livestock output was
practically identical with that of crop output. The most recent
figures show that in the last decade (1969/71-1979/81) meat and milk
output grew faster than cereal output (FAO Production Year Books) .
However there has been virtually no change in output
(productivity) per ruminant; and there seems little reason to believe
that there has been any increase in the primary productivity (herbage
yield per unit area) of accessible grazing land. The increase in
total output has been achieved by increases in livestock numbers, and
these increases have been made possible either by grazing a higher
proportion of the existing vegetation on land always accessible, or
by increasing the area of land accessible. The livestock system has
been extended but not intensified.
In extending the system two kinds of improvement have
predominated - the control of rinderpest permitting increases in
cattle numbers, and the development of water supplies permitting
increases in the area of land exploited. In both cases institutional
and administrative as well as technological changes were involved.
- 15 -
Earlier huge losses from rinderpest were initially reduced by an
administrative device - quarantine - albeit at high social cost.
The subsequent development of a safe, easy-to-administer vaccine
made rinderpest control much easier, more reliable and less costly
to all, but the problem was already essentially under control, through
quarantine, long before the tissue culture vaccine was developed.
Improved technology played an earlier role in water development, in
the form of hydrological skills, better borehole technology and
machines that could move large quantities of earth quicker and more
cheaply than could be done by hand. But the use of improved water
technology - developed elsewhere - was in most places only made
possible by organisational changes which permitted economies of scale
in the employment of expertise, the acquisition and use of equipment
and the raising of the necessary finance.
I am afraid that I have to point out here that in these most
successful forms of pastoral improvement nothing remotely resembling
what this workshop will define as pastoral systems research was
employed. Both the technological and the organisational techniques
employed were developed elsewhere and largely for other purposes,
and very little thought or adaptation with respect to local circum
stances were employed.
When we turn from extensive development to the intensifi
cation of production, i.e. the application of new skills and inputs
to raise yields and values, there has been very little success in
tropical Africa on any scale - either in increasing yields per head
of livestock or, more importantly, per hectare of pastoral land. Yet
yields and values have been raised elsewhere - both in pastoral areas
in other countries and in non-pastoral forms of livestock production
in the countries of tropical Africa.
The relative failure to itensify production in pastoral
areas in tropical Africa can be attributed to a number of causes. In
some cases there has been simple neglect. Deliberately or through
lack of interest very little effort has been made in many countries to
intensify production. In some cases this has been caused by a belief
that pastoral production cannot be intensified or that this can only
be done at the expense of ecological degradation.
- 16 -
In many cases government policies towards the livestock
sector in general - or to the pastoral subsector in particular - have
been defective. There may have been wrong policies, or contradictory
policies, or simply an absence of policy. By policy in this case I
mean some general issue, e.g. land tenure, pricing of commodities,
organisation, etc., that is not related to some specific intervention
for pastoral development.
In some cases the problem has concerned organisation and
staff for specific pastoral development programmes. Staff have been
too few, poorly trained and inadequately motivated. Organisations
have been established with the wrong structure and work has been
poorly coordinated. Administrative procedures, e.g. for procurement
or financial control, have been inappropriate; and funds have either
not been provided at all or were provided too late to be of proper usej
However when all other causes have been taken into account
the fact remains that in many cases the failure to intensify pro
duction arose from attempting misconceived interventions and from a
failure to try interventions which would have succeeded. These errors
must have been due to ignorance. Ignorance can be put right by
general training and by research. It is to research and knowledge I
now turn.
Before looking at some specific issues there are some
general comments to be made. Three years ago I did a desk review of
about 30 livestock projects in dry tropical Africa which were financed
by one donor. Two main points of interest emerged. The first was the
low expenditure on research of any kind. In general agricultural
projects financed by this donor approximately 3 % of total project
expenditure was allocated to research. By contrast in livestock
projects in dry tropical Africa only 1.5 % was so allocated. The
second major point of interest was the indifference shown by project
planners to the technical base for their interventions. This was not
in general true of veterinary innovations. In the case of veterinary
interventions considerable attention was paid in the project appraisal
documents to their technical base, i.e. to the evidence that the
innovation proposed would have the expected results. But in the case
- 17 -
of other innovation* - especially range management innovations -
extremely little attention seems to have been paid to the technical
base (Sandford, 1981).
I now turn to some particular examples of interventions
which have taken place in pastoral development which are relevant to
our consideration of pastoral systems research (PSR) . I do not want
to preempt what my collegues are going to say about the nature of PSR
as a member of a general body of approaches more usually referred to
as farming systems research (FSR) . Let me however stress here four
aspects of PSR.
Firstly, PSR is concerned with a production system as a
whole. Although particular elements in the system may be the ones
on which attention is focused, nevertheless this is in the context of
a general understanding of the system as a whole and of the relation
of the elements selected to other elements. Secondly, in PSR consider-*
able attention is paid to understanding the whole system before
attention is focused on particular elements within it or before at
tempting to change these elements. Thirdly, PSR, in contrast to
approaches more dominated by technological or commodity interests, is
characterised by the attention paid to the pastoralists ' own points of
view. Fourthly, in PSR research is not only carried out on research
stations. A crucial element in PSR is the testing of proposed
innovations in pastoralists' own enterprises and under the conditions
which pastoralists usually face.
Three examples of relevance to a pastoral systems research approach
I now look at three examples of interventions where, I believe, a
PSR approach might have led to greater success or to less costly
failure.
In one country the pastoral areas were partitioned into
2
huge "divisions" - of the order of 6 000 km each - which were in turn
divided into "paddocks". The range management intervention proposed
was one in which a particular set of pastoralists would be associated
with a particular range division, water supplies would be developed
in each paddock, and the range management activities would then, with
the help of a committee of pastoralists, determine which paddocks
- 18 -
could be "open for use" at any one time and which "closed" in order to
implement a rest-and-rotation system. On the whole this programme
has not been successful; and one can identify at least three major
reasons for this.
Firstly, the programme did not adequately recognize the
enormous spatial variability of rainfall, and the absolute necessity
of allowing livestock to move from one division to another according
to recent rainfall rather than confining them to particular divisions
regardless of conditions. Secondly, the programme treated the live
stock as homogeneous and failed to recognize that rotation of
livestock between paddocks in a rather inflexible way prevented
herders from providing particular species or classes of animals the
particular types of vegetation, water and minerals which they needed
at particular times.
Both these problems arose from the failure of the authorities
to recognize the rationale behind the existing land use system. The
third main reason for the lack of success was that no research on
range management methods was carried out in that area prior to the
project's implementation. The explicit assumption was that range
management techniques for rest and rotation developed in the USA
would be an improvement on the existing system of land use and that
variations in these techniques could subsequently be made to suit
local conditions. However, no further local research was carried out
for at least the first 10 years of the project; and there is still no
evidence that the management techniques suggested - and which the
local pastoralists have ignored - would have raised land productivity.
My second example relates to the record of livestock re
search programmes in Africa in general rather than to a particular
project. In a number of cases trials have been carried out on
research stations on animal management practices, e.g. breeding
seasons, supplementary feeding, and as a result recommendations have
been made about practices which appear to be highly profitable.
Pastoralists and other livestock owners have not followed these
recommended practices because in their circumstances the practices
have been unfeasible or uneconomic.
- 19 -
In a number of countries governments have intervened in
marketing systems in order to improve their efficiency and provide
pastoralists with higher prices for their livestock. Common among
such interventions have been restricting permission to trade to those
traders able to fulfill certain license conditions, and insisting that
all trading takes place in formal markets, that sale be by auction and
that livestock to be sold first be weighed. Such marketing inter
ventions have usually been unsuccessful, firstly because no attempt
was made to assess objectively the efficiency of the pre-intervention
marketing situation - which was just assumed to be costly, inefficient
and subject to manipulation by traders for their own profit; secondly,
because often it was assumed that the function of livestock marketing
was to extract beef animals from rural areas for urban consumption
or export, and it was not realized that often the vast majority of
transactions were of breeding animals, or animals for draught
purposes - transactions for which sale by weight at auction is highly
unsuitable. Also the long time interval between market days which
may be suitable for the extraction of beef cattle from rural areas is
highly unsuitable for transactions in smallstock and in cattle for
other purposes. In other words the proponents of the interventions
failed to understand pastoralists' or other livestock owners' purposes
or needs in marketing stock and failed to have adequate regard for
their point of view.
Conclusions
PSR is relatively new in Africa, and I do not want to preempt the
discussions of the concluding sessions of this workshop by either
wholeheartedly supporting or disclaiming PSR. Let me end by
reemphasizing five points.
1. Although increases in livestock production in pastoral
areas of tropical Africa have occurred in the past these increases
came about by the extension of existing systems rather than through
their intensification.
2. Intensification of production has not yet been sussessful
in pastoral areas.
20 -
3. Part of the reason for this lack of success has been
ignorance about which interventions should be introduced.
4. Part of the reason for this ignorance has been the
generally inadequate volume of livestock related research.
5. However, part of the reason for the lack of success has
been the making of inappropriate interventions due to a failure to
carry out some of the stages included in PSR. In particular there
has been a failure to understand the present system - both in terms
of what people want to do (their ends) and why they adopt the means
they do. Also critical has been the failure to test innovations
developed elsewhere or those tested on research stations under the
specific conditions of the natural and social environment faced by
the pastoralists for which the particular innovation is planned.
References
De Montgolf ier-Kouevi, C. and Vlavonou, A. 1981. Trends and prospects
for livestock and crop production in tropical Africa.
ILCA Working Document No. 5, Addis Ababa.
Jahnke, H.E. 1982. Livestock production systems and livestock
development in tropical Africa. Kieler Wissenschaf tsverlag
Vauk, Kiel.
Jamal, V. 1983. Nomads and farmers: incomes and poverty in rural
Somailia. In Dharam Ghai and Samin Radwan (Eds) (1983)
Agrarian policies and rural poverty in Africa, ILO, Genena.
Jasirowski, H.A. 1983. Twenty years with no progress ? World Animal
Review 5: 1973.
Sandford, S. 1981. Review of World Rank livestock activities in dry
tropical Africa. Unpublished report. ILCA, Addis Ababa.
Wissocq, Y. 1978. Livestock development projects in Africa south of
the Sahara: a review of the period 19(1-1975. Unpublished
report. ILCA, Addis Ababa.
- 21
Expérience en matière de développement
Résumé
Le thème de ce document porte sur l'analyse de la recherche sur les
systèmes pastoraux dans le cadre du développement de l'élevage en
Afrique au cours du dernier demi-siècle.
Les zones arides sur lesquelles vivent 60% des ruminants de l'Afrique
tropicale couvrent 55% de la superficie du continent. La proportion
de la population totale de ruminants de l'Afrique élevés dans les
systèmes pastoraux se situe probablement entre 30 et 50%. Entre 1950
et 1970, la population humaine de l'Afrique a approximativement
augmenté au rythme de 2,5% par an mais la croissance de la production
carnée et laitière ne dépassait guère 2%, ce qui indique une baisse
de la production par habitant. De 1970 à 1975, le taux de croissance
de la population humaine a augmenté alors que celui de la production
annuelle diminuait, traduisant ainsi une chute encore plus rapide de la
production par habitant. En conséquence, on a observé une baisse des
exportations de bétail et de produits animaux parallèlement à une
augmentation des importations d'animaux et de produits de l'élevage.
Autrefois, le développement pastoral était synonyme d'un ensemble
hétérogène de programmes à caractère général. Après 1965, il y a eu
une accélération remarquable du rythme du développement, notamment avec
l'arrivée sur la scène de la Banque mondiale et de 1' Agency for
International Development des Etats-Unis (USAID), toutes deux grands
bailleurs de fonds dans le secteur du développement pastoral. Durant
la seconde moitié des années 70, la CEE/FED est entrée elle aussi
dans le club des grands bailleurs de fonds des projets de développement.
Mais dans l'ensemble, c'est à une extension plutôt qu'à une intensi
fication du système d'élevage qu'on a assisté. Celle-ci se
caractérisait par deux types d' améliorations : la lutte contre la
peste bovine qui permet d'accroître la population bovine et la mise
en valeur des ressources en eau qui permet d'augmenter la superficie
des terres exploitées. Malheureusement, dans la mise en oeuvre de
ces formes très adéquates d'amélioration pastorale, rien n'a été
envisagé en ce qui concerne la recherche sur les systèmes pastoraux.
- 22
L'intensification de la production animale en Afrique tropicale ne
s'est pas réalisée. Pour certains programmes de développement pastoral,
les raisons qui expliquent cet échec tiennent sûrement à la négligence
des responsables mais aussi aux contradictions et à l'incohérence des
politiques gouvernementales tout comme aux carences en matière
d'organisation et de personnel. Dans plusieurs cas, l'échec a été
causé par le choix d'interventions inopportunes à la place d'autres
interventions qui auraient pu être couronnées de succès.
Dans la recherche sur les systèmes pastoraux, l'accent est mis sur la
compréhension de l'ensemble du système avant celle des composantes
particulières du système ou avant toute tentative d'intervention sur
ces composantes. Contrairement aux approches qui privilégient la
technologie ou les produits, la recherche sur les systèmes pastoraux
se caractérise par l'importance accordée au point de vue de l'éleveur.
L'un des éléments les plus remarquables de la recherche sur les
systèmes pastoraux c'est qu'elle prévoit le test des innovations
envisagées au niveau de l'exploitation elle-même dans les conditions
dans lesquelles vivent habituellement les éleveurs.
Quoique par le passé on ait pu enregistrer des accroissements de la
production animale dans les zones pastorales de l'Afrique tropicale,
ceux-ci étaient plutôt le fruit de l'expansion des systèmes
existants que de leur intensification. L'intensification de la
production dans les zones pastorales s'est jusqu'ici soldée par un
échec. Cet échec s'explique dans une large mesure par l'incapacité
d'identifier les types d'interventions appropriées.
Cette lacune s'explique en partie par le volume généralement inadéquat
de la recherche sur l'élevage. Mais cet échec procède également
d'interventions inadéquates qui s'expliquent par le fait que
certaines étapes de la recherche sur les systèmes pastoraux ont été
négligées. En particulier, on ne s'est pas attaché à comprendre le
système actuel et à expliquer notamment les objectifs des populations
concernées et la raison pour laquelle elles adoptent les moyens
dont elles se dotent. Autre aspect important de cet échec: on n'a
pas essayé les innovations mises au point sous d'autres cieux ou
- 23 -
au niveau des stations de recherche dans les conditions spécifiques
de l'environnement naturel et social de l'éleveur auquel ces
innovations sont destinées.
- 24
Towards a framework for pastoral systems research
Cees de Haan
Deputy Director General, ILCA, Ethiopia
Introduction
The International Livestock Centre for Africa (ILCA) has adopted a
systems approach to research since its inception in 1975. This
strategy originated from ILCA's Foundation Report, which stated that
'technical answers are available to many of the specific problems
facing livestock development in Africa, but the major constraint lies
in introducing change into existing socio-economic systems, exacer
bated by inexperience in adapting technology to suit local conditions'
(Nestel et al,1973). The approach was reinforced by the growing
realisation as experienced in development projects that the results
of classical on-station research, whether in Africa or elsewhere, and
Western technology could not be transferred directly to African
traditional systems. A better understanding of those systems, and
adaptive research was necessary to identify relevant improvements.
When ILCA started its research no methodology for livestock
systems research (LSR) was readily available, and a considerable amount
of time and resources was spent on developing methods of studying and
understanding the complexity of livestock production systems, and on
defining ways in which available technology could best be tested
within a particular production system. With the emergence of a
farming systems research (FSR) methodology (Gilbert et al, 1980;
Byerlee et al,1980; Shaner et al, 1982) , ILCA has attempted to develop
a systems approach analogous to the framework used in FSR, although
the greater genetic variability of livestock, the lower degree of
control because of mobility, the daily rather than seasonal inputs
required for livestock, their multiple usages, and the greater influ
ence of individual management makes LSR more complex than cropping
systems research. Additionally in pastoral systems the greater
mobility of people and livestock, problems of land tenure, and the
- 25 -
limited scope for improvements, adds to the complexity in the
application of the approach.
This paper gives an overview of the ILCA research framework
and highlights some of the results obtained. It focuses on pastoral
systems of arid and semi-arid zones, but also includes the linkages
with cropping in agropastoral systems. It uses mainly examples from
ILCA's own field research in the Sahelian environment of Mali, and
the semi-arid areas of Kenya and Ethiopia but also incorporates
other situations in more humid areas or where ILCA was only indirectly
involved.
Objectives of livestock systems research
The objective of livestock systems research (as in FSR) is to assist
in the generation of relevant improvements for. the well-being of a
certain target population. These improvements can be technical and
socio-economic in nature. LSR is not responible for the extension
of a new technology, or for making complete baseline surveys or
management plans for a certain region. These distinctions should be
emphasised, as it immediately puts limits on the number of producer
units and the kind of parameters under study, the level of details
in observations made, the frequency of data collection, etc.
For example in a pastoral area, LSR would describe the
relationship between water resources and rangelands utilisation,
diagnose labour and sanitary constraints for the extraction and use
of the different types of water points, assess the effect of those
constraints on animal productivity and identify users' rights. It
would then use this information to design and test relevant technology,
which could overcome those constraints. It would not make a complete
inventory of all water points in the region, nor conduct a hy-
drological study of all underground water available. The latter
tasks are clearly the responsibility of a development agency, which
should use the technology developed and tested through LSR to design
and implement water improvement schemes for the region.
- 26 -
Stages in LSR
LSR as defined above consists of a sequence of stages, designed to
identify constraints in a particular system, and through experimen
tation or available knowledge identify possible solutions. These
solutions are then tested in real life situations, initially with
a high degree of scientific control, and as positive effects are
noted with increasing producer participation and decreasing scientific
involvement and control. Table 1 gives the different terms used in
each of the stages.
Table 1. The terms used in the stages of livestock systems research
Stage Object of observations
1. Descriptive
2. Diagnostic
3. Design
researcher managed/
executed
4. Testing
a. researcher managed/
farmer executed
b. farmer managed/farmer
executed
5. Extension
Natural, livestock and human
resources, production systems
Constraints, in order of priority
research requirements, chances
of success to overcome them
Possible solutions through on-
station experimentation or body
of knowledge
Producer management of improvements
Acceptability of improvements
Evaluation of technical and socio
economic impact of improvement
Source: Adapted from Shaner et al (1982) and Norman (1982).
Three points should be emphasised in this context :
1. The different stages can overlap, e.g. it is not necessary
to complete the whole diagnostic phase before any design or testing
activity can take place.
- 27 -
2. Although the classical approach is to start with the
descriptive phase, the research process can be initiated at any stage.
3. Continuous data analysis is essential for feed-back of
results and rapid progress from the descriptive/diagnostic phase to
the testing phase. Micro-computers can play a useful role in this
process. If no computing facilities are available, the data collec
tion system should be organised in such a way that calculators can
do the analysis.
Some considerations on sampling framework in LSR
With the increasing limitation on funds and the pressure from devel
opment agencies on research institutes to provide improvement the
emphasis is on using rapid and low-cost appraisal techniques. Sample
selection can be an important tool in enhancing the cost effectiveness
of the c.iagnostic and testing phase. ILCA's results indicate that
the following parameters are important in determining the management
strategy, and therefore the profile of possible improvements, and
should be considered in the sampling framework.
1. The livestock/man ratio ("rich vs poor"). It becomes
increasingly apparent that a different degree of livestock wealth
results in different management. For example, in the pastoral areas of
Niger, Wilson and Wagenaar (1983) found a marked decrease in offtake
with increasing herd size (see Table 2).
Table 2. Relation between herd size and offtake of WoDaaBe herds
in Northern Niger (1983)
Herd size Offtake* (%) per year
5-10 5.9
11-15 10.9
16-20 4.7
21-25 4.3
>25 315
* Sales, slaughter and gifts
Source : Wilson and Wagenaar (1982) .
- 28 -
In this example the category owning between 11 and 15 animals
also sold females which indicated that they were forced to sell. The
poorest category could not sell, and practised migratory labour.
Further examples of the importance of livestock/man ratio, and a
rapid method of assessing the relative wealth will be given by Bekure
and Grandin later in the workshop.
2. Men vs women. As shown by ILCA's subhumid programme in
northern Nigeria (ILCA, 1982) there is often a conflict of interest
in the male and female economic sphere particularly over milk offtake
and sales. Milk offtake normally benefits the women, whereas the
proceeds of livestock sales go directly to the men. Improvements
like supplementary feeding, which are financed by men, therefore
should not only result in increased milk offtake, but also in in
creased rates of calf survival, and at a later phase improvements
should benefit both groups in order to avoid future inequities. In
pastoral systems research, both groups should therefore be included
in the sampling framework.
3. 'On-road' vs 'off-road' producers. Especially in agro-
pastoral systems research activities tend to focus on the easily
accessible 'on-road producer. Access to a road, as shown in
Table 3 for the Ethiopian highlands, to a considerable extent affects
producers' strategy, and therefore should be one of the considerations
in selecting the sample.
- 29 -
Table 3. Average livestock holdings per household of on-road and
off-road producers* in the Ethiopian highlands
Livestock ' On-road' 'Off-road'*
Cows
Local breeds 2.1
Crossbreeds 0.4
Oxen 1.8
Sheep 4.9
Goats 0.2
Donkeys 1.0
1.6
1.6
8.5
0.3
1.2
* Minimum of 2.5 hours away from an all-weather road.
Source : Gryseels (per. coram .) .
Descriptive and diagnostic phase
The descriptive phase consists of a literature review and a general
reconnaissance of the resources and production systems of the target
region. It will generally provide an indication of the nature of
the constraints but not a quantitative assessment of their importance
or their priority ranking. The descriptive phase should be kept
short (one to six months) to move as soon as possible into the
diagnostic phase.
The different methods used by ILCA in the descriptive and
diagnostic phase for resource assessment, and the measurement of
vegetation, vegetation trends, animal productivity, importance of
diseases, nutrition, resource allocation within the producers' units,
and marketing will be presented in detail in this workshop, and are
therefore not reviewed in this paper. Some results which merit
special mention and which have assisted in developing ILCA's overall
programme are :
- 30 -
1. Herd and flock management practices by individual owners
seems to be one of the most important single factors affecting
productivity. For example, Wilson et al (1982) found in central
Mali the ratios of the production index (production kg per year per
kg metabolic weight of breeding female) given in Table 4.
Table 4. Ratios of production index of small ruminants in the
agro-pastoral system of central Mali
Source of variation Ratio Goats Sheep
Overall mean 1.47 2.31
Parity All parities/
first
1.40 1.23
Season of birth Best/worst 1.23 1.14
System Rice/millet 1.58 1.33
Flock (millet) Best /worst 2,.43
(rice) Best/worst 5 .58
Source : Wilson et al (1982).
Similar results were obtained in ILCA's Kenyan programme
both with cattle and small ruminants. This is all the more surprising
because grazing is communal to all and different owners have the same
access to this principal resource. Individual management skills
therefore seem to play a more important role than previously reported.
However, ILCA has not yet defined the components which determine
those management practices. In view of their overriding importance,
more attention is presently being given to quantify these factors.
Indeed if those parameters could be better quantified, and if through
extension the management of the below-average flocks and herds could
be raised to higher levels, considerable productivity gains could be
obtained.
- 31 -
2. Strong linkages already exist between livestock and crop
ping, and the enhancement of these linkages merits a high priority
in research, because they form a relatively easy entry point for
improvement .
On an Africa-wide scale Brumby (1983) observed a significant
correlation between the increase of total cereal production and
cattle numbers, e.g. that each extra number in the cattle population
was associated with an extra 0.25 ha of crop land and about 200 kg
of annual grain output. On a regional basis, the strong relationship
between livestock and cropping was shown by ILCA's aerial survey work
in several systems. For example, Table 5 demonstrates this inte
gration for central Nigeria.
Table 5. Relationship between livestock and cropping in central
Nigeria
% area
cultivated
Cattle density
wet season dry season
< 10
10-34
35-60
7.7 40.0
16.6 48.5
21.4 50.6
22.4 45.7> 60
Source: Miligan, 1980.
The beneficial nature of this relationship expresses itself
differently in the various situations. For example ILCA's systems
studies in central Mali demonstrated a very clear effect of livestock
ownership on millet yields, probably through the manure linkage
(see Table 6) .
- 32 -
Table 6. Relation between village field millet yields (kg/ha) and
livestock ownership in central Mali
Millet yield (kg/ha)
Year Households
cattle
; with Household without
cattle
1980
1981
679
637
*
440
Source : Fulton and Toulmin (1982).
439
A similar close relationship in the field of animal traction
was found by Gryseels (1983) in the Ethiopian highlands. His results
clearly show a marked increase in the crop area cultivated by each
family as well as a significant change in cropping patterns to
higher value cereals as oxen numbers increase (see Table 7). Similar
trends were observed by ILCA's central Mali studies (Fulton and
Toulmin, 1982).
Table 7. Relationship between area cultivated, cropping pattern and
oxen ownership in the Ethiopian highlands
No. of oxen Area cultivaited Area under Are.a under
per farmer (ha) cereals (%) pul;ses (%)
0 1.2 54 46
1 1.9 44 56
2 2.7 67 33
3 or more 3.6 92 8
Source : Gryseels (1983)-
- 33 -
These findings have directed to a considerable extent the
focus of ILCA's research thrust; for example, activities aimed at
strengthening the integration of crops and livestock (legume agronomy,
animal traction) are receiving approximately 150 % more funding in
1983 than in 1981.
Even more important than the specific results, although less
tangible, is the understanding ILCA staff have gained of systems and
systems research after five years of work.
The identification of improvements
Possible improvements are identified and designed by the systems team,
tentatively during the descriptive phase, but increasingly during the
diagnostic phase. It should be emphasized that the identification of
improvements and their priority ranking must be a team effort. Quan
titative models describing the input/output relationships at the
different systems levels are useful in assisting the team in the
decision-making process.
The plant/animal sub-component is adequately represented by
ILCA's herd productivity model. This model is being used to estimate
the effect of supplementation or management strategies, such as the
effect of early weaning, watering frequencies and distance to water
on livestock productivity, and can therefore be used to rank the
priorities for field research (Konandreas and Anderson, 1982;
Konandreas et al, 1983). A simple example of an input/output model
at the producers' unit level for the Ethiopian Borana system is
graphically represented in Figure 1.
- 34 -
Rain 600 mm Solar energy 750 kj cm2
80 ha wet season
50 ha dry season
LAND —
(100 ha)
T
WATER
400 1/day
200 ton dry matter1 (Exc. browse)
CATTLE SMALLSTOCK CAMELS
1 bull
9 cows
4-5 Q immature
4 Cf immature
6-10 sheep
7-9 goats
Purchase of grain
200-300 kg/yr
5-6 cows in milk
I
2000 1/yr
I
Family
consumption
1
Milk in dry season?
Sales
US$ 300
' Food requirement of cattle and smallstock about 45 tons DM/yr.
Figure 1: Input /output relationships of a Borana pastoral production
unit in southern Ethiopia
- 35 -
This particular example also illustrates the multi-
disciplinary approach to the identification of improvements. In the
Borana system the herd demography study revealed a significant differen
tial mortality between female and male calves (see Table 8) . This
is caused by a heavy reliance of the Borana on milk for subsistence,
which probably forces them to slaughter male calves to enable a
faster reproduction and to allow less suckling to increase the amount
of milk available for human consumption. However, this strategy
conflicts with government objectives aimed at increasing the meat
offtake from these rangelands. An obvious way therefore to reduce
this calf mortality, and to increase total offtake from this particular
system, is to increase either the total milk yield or the amount of
available grain to reduce dependence on milk. The joint ILCA/RDP
(Ethiopian Pangelands Development Project) team therefore will
shortly be testing the feasibility of millet cultivation around the
heavily manured permanent dry-season camps and has started work on
legume introduction for dry-season supplementation.
Table 8. Age I sex distribution (%) of cattle herds in southern
Ethiopia (dry and lactating herds combined)
Age/sex distribution (%)
Male
9.3
6.3
Source : Cossins (pers. coram.) .
The design phase
The assumption that the new technology for improving Africa's
traditional livestock production systems would be available, is
only partially true. Generally it can be said that in the field of
animal health and animal nutrition there exists a whole range of
improvements, which can immediately be tested under pastoral con-
Female
Calf 12.2
Immature 14.7
- 36 -
ditions, However, in the field of range management and in the
integration of forage production into cropping systems, the available
body of knowledge is still extremely scarce, and no clear-cut sol
utions to identified constrains are available. Production par
ameters have to be established, therefore, initially under controlled
conditions, e.g. in on-station research. It is therefore essential
that LSR has access to good research station facilities. One major
distinction between LSR and classical research station experiments
relates to the identification of the treatments. In LSR the
treatments are defined by the socio-economic and political framework
determined in the diagnostic phase. If for example the supply of
fertilizer or agricultural equipment is not expected in the near
future, such inputs would not be used in the experimental design.
The testing phase
Once the improvements have been formulated they have to be tested under
producers' conditions. The object of these tests is to obtain
scientifically valid conclusions regarding the effect of improvements
and to assess producers' acceptance. The first objective dis
tinguishes systems research from extension. In extension improvements
are normally introduced without control groups, and without built-in
scientific comparisons. In systems research the objective is not to
cover the whole target population, but to make a scientifically valid
comparison regarding the effect of certain treatments (improvements)
on a minimum required sample size.
This means that systems research should always have a with/
without comparison on a representative sample, simultaneously carried
out under similar environmental conditions. A before/after comparison
as carried out in many development projects will always have the
confounding effect of different environmental conditions.
For this comparison, matched pairs of animals with similar
characteristics within a herd or flock are preferred when inputs
are very easily administered. ILCA is following this experimental
procedure for example in the case of animal health inputs, and in
concentrate supplementation in northern Nigeria. If the improvements
-37
proposed require major changes in management strategies, or different
organisational structures, then comparisons have to be carried out
on a whole farm, or whole herd, basis. Examples of this procedure
are to be found in ILCA's programme testing the use of fodder banks
in northern Nigeria, and forage production and crossbreeding for
dairy production in the Ethiopian highlands.
The latter programme, although not a pastoral system,
illustrates the value of continued systems research. Although the
initial results (Gryseels and Anderson, 1983) showed a very positive
impact of the introduction of crossbreds and forage production (annual
milk yield increased sevenfold to 2300 kg, cash income fourfold to
US $ 800, food grain production by 40 %) , changes in the cropping
pattern clearly Remonstrated the need for more research on forage
production. It was shown that farmers: 1. replaced their crop
yielding the lowest gross margin (pulses) to grow forages with a
resulting negative long-term effect on soil fertility; 2. allocated
labour for soil preparation first to cereal crops, rather than the
forages; and 3. applied fertilizers only to cereal crops because of
recent increases in fertilizer prices. This in turn has caused ILCA
to allocate more resources to the search for leguminous pasture plants
requiring low labour and low fertilizer input. The first on-farm
trial with a perennial grass legume mixture will start in 1983.
Improved range management practices are particularly
difficult to test under pastoral conditions, and generally new
organisational forms are required a priori. Even so, a good
understanding of a traditional system frequently offers opportunities:
1. In the Malian delta, traditional forms of land tenure
offer the possibility of establishing pastoral units which would
enable the introduction of improved grazing schemes (ODEM/CIPEA, 1983
Wilson et al, 1983).
2. In Niger, building upon traditional water rights, a
promising grazing scheme developed on a government station will be
tested with a small number of experimental herders associations. Soon
to be established water points under the control of those herders
associations form the mechanism for testing this grazing scheme.
- 38 -
3. In the Kenyan group ranches, individual ownership of
certain pastures allows the introduction of legume fodder banks to
overcome the dry-seaon nutritional stress. ILCA plans to introduce
fodder banks in on-herd trials in 1983.
Extension phase
Once the improvement has been satisfactorily tested under producers'
conditions in farmer managed, farmer executed trials, it can pass to
the extension service for further implementation. This will generally
be carried out in the framework of a development project.
It is important to evaluate the improvement and, in this
phase especially, its socio-economic impact. The direct involvement
of a research institute in this evaluation is questionable. Ideally
the project or extension structure will include the capabilities for
the technical and socio-economic evaluation of the improvement. In
the absence of such a structure and the information still being
required, it should be national research institutes who take responsi
bility for this often politically sensitive task. The degree of
ILCA's direct involvement is an open question.
In any case, the evaluation procedures should be simple and
be focused on the specific improvement introduced. If a particular
improvement is to be successful, it should produce a considerable
gain in productivity and income, or added security. The data collec
tion system to study the impact could therefore be reduced in
sampling frequency and size. Qualitative information allowing the
identification of trends in this case might be given more emphasis.
ILCA's future role in LSR ,
Livestock systems research is location specific, focusing on a
particular target zone. It cannot be the role of an international
centre like ILCA to take direct responsibility for the whole African
production system. Rather, ILCA sees its role as 1. further developing
the systems methodology for each of the major ecological zones,
2. developing improvements for the constraints identified in those
zones, and 3. simultaneously assisting national research and develop-
- 39
ment agencies in establishing their own in-house capabilities. The
interest which has recently emerged is encouraging.
- In northern Nigeria, ILCA is actively cooperating with
scientists from the National Animal Production Research Institute
(NAPRI) in establishing an LSR group, which focuses on an agropastoral
system near Shika.
- In Mali, ILCA has recently signed a contract with the
Institute National de Recherche Zootechnique Forestier et Hydro-
biologique (INRZF) to assist in the establishment of a LSR group in
the institute to focus on smallholder mixed farmers in southern Mali.
- Negotiations are at an advanced stage with the Directorate
of Research and Special Services in Zimbabwe to assist them in the
establishment of an LSR capability, focusing initially on two mixed
farming systems, in high and medium potential areas respectively.
- The Government of the Cameroons has recently approached
ILCA to assist them in developing LSR capabilities in the Institute
de Recherche Zootechnique (IRZ) focusing initially on three different
production systems there.
Summary
The complexity of livestock production systems in sub-Saharan Africa,
and the multiple objectives of its producers requires that whole
systems rather than components be studied. Livestock systems
research, as practised by ILCA over the last year, represents an
approach to such studies. By applying this approach ILCA has gained
considerable knowledge of the constraints and opportunities in
pastoral and agropastoral systems.
The testing of improvements under producers' conditions is a
crucial and essential part of this approach. Livestock systems
present particular difficulties in implementing these tests. Some
experience has been gained but more methodology development is still
required for this phase.
ILCA sees its role as that of further exploring these issues,
assisting national agencies in developing systems research capa
bilities, and developing relevant improvements to the constraints
40 -
identified. With this combination of research thrusts ILCA believes
that a sustained gain in African livestock productivity and overall
food production can be achieved.
References
Brumby, P.J. 1983. Alternative strategies for dairy development.
Paper presented at the Seminar on Agricultural Research in
Kigali, Rwanda. ILCA, Addis Ababa.
Byerlee, D., Collinson, M.P., Perrin, P.K., Winkelman, D.L., Biggs,
S., Moscardi, E.R., Martinez, J.C., Harrington, L. and
Benjamin, A. 1980. Planning technologies appropriate to
farmers: Concepts and procedures. CIMMYT, El Batan.
Fulton, D. and Toulmin, C. 1982. Socio-economic study of an agro-
pastoral system in central Mali. Draft Report: ILCA, Addis
Ababa.
Gilbert, E.H., Norman, D.W. and Winch, F.E. 1980. Farming systems
research: A critical appraisal. M.S.U. Rural Development
Paper No. 6. East Lancing, Michigan.
Gryseels, G. 1983. Livestock in farming systems research for small
holder agriculture: Experience of ILCA's Highland Programme.
Paper presented at the Seminar on Agricultural Research in
Kigali, Rwanda. ILCA, Addis Ababa.
Gryseels, G. and Anderson, F.M. 1983. Research on farm and livestock
productivity in the central Ethiopian highlands: Initial
results. ILCA Research Report 4, Addis Ababa.
ILCA (International Livestock Centre for Africa), 1982. Annual Report
1982. Addis Ababa.
Konandreas, P. and Anderson, F.M. 1982. Cattle herds dynamics: An
integer and stochastic model for evaluating production
alternatives. ILCA Research Report 2, Addis Ababa.
Konandreas, P., Anderson, F.M., and Trail, J. CM. 1983. Economic
trade-offs between milk and meat production under various
supplementation levels in Botswana. ILCA Research Report 10,
In press.
- 41 -
Nestel, B., Pratt, D.J., Thome, M. and Tribe, D.E. 1973. Animal
production and research in tropical Africa. Report of the
Task Force commissioned by the African Livestock Sub
committee of the Consultative Group on International
Agricultural Research. Washington, D.C.
Norman, D. 1982. Institutionalizing the farming systems approach.
Research paper presented at the African Bureau of Agricul
tural and Rural Development Officers' Workshop at Ibadan,
Nigeria. USAID, Washington, D.C.
ODEM/CIPEA (Opération de développement de l'élevage dans la région de
Mopti/Centre Internationale pour l'Elevage en Afrique). 1983.
Recherche d'une solution aux problèmes de l'élevage dans le
delta intérieur du Niger au Mali. 5 vols., Addis Ababa.
Shaner, V.W., Philipp, P. F. and Schmehl, W.R. 1982. Farming systems
research and development. Guidelines for developing
countries. Westview Press, Colorado.
Wilson, R.T. (éd.). 1982. Livestock production in central Mali.
ILCA Bulletin 15, Addis Ababa.
Wilson, R.T. and Wagenaar, K.T. 1983. Enquête préliminaire sur la
démographie des troupeaux et sur la reproduction chez les
animaux domestiques dans la zone du projet gestion des
pâturages et élevage de la République du Niger. ILCA/Mali
Working Document AZ 80, Bamako.
Wilson, R.T., de Leeuw, P. and de Haan, C. (eds) . 1983. Systems
research in the arid zones of Mali: Initial results. ILCA
Research Report 5, Addis Ababa.
- 42 -
Vers un cadre de recherche sur les systèmes d'élevage
Résumé
Dès sa création en 1975, le CIPEA a adopté une approche par système
analogue au cadre utilisé dans la recherche sur les systèmes
d'exploitation agricole. Le présent document brosse un tableau
général du cadre de recherche du CIPEA, présente certains des
résultats qu'il a obtenus et met l'accent sur les systèmes pastoraux
des zones arides et semi-arides.
La recherche sur les systèmes pastoraux se propose de contribuer à
l'amélioration de la qualité de la vie d'une certaine population-cible.
Cette amélioration peut être technique et socio-économique. La
recherche sur les systèmes d'élevage est une série d'étapes conçues
pour identifier les contraintes d'un système particulier et les moyens
d'éliminer celles-ci par le biais de l'expérimentation ou par
l'utilisation des connaissances disponibles. Ces solutions sont par
la suite testées dans les conditions de l'exploitation, tout d'abord
sous le contrôle du chercheur, dont le rôle sera progressivement
transféré au producteur, à mesure que les résultats enregistrés
s'avéreront encourageants.
Avec l'accroissement des restrictions budgétaires et des pressions
qu'exercent les organismes de développement sur les instituts de
recherche en vue de la mise au point par ceux-ci d'innovations, il
faudrait privilégier l'utilisation de techniques rapides et peu
coûteuses d'évaluation. La sélection des échantillons peut constituer
un outil utile pour renforcer l'efficacité - coûts de la phase du
diagnostic et de l'expérimentation. Les résultats obtenus par le
CIPEA indiquent que les paramètres ci-dessous jouent un rôle
important dans la détermination de la stratégie de gestion et partant,
dans la forme des innovations éventuelles. Il faudrait donc en tenir
compte dans la base d'échantillonnage: rapport bétail/homme (riche/
pauvre); homme/femme; producteur situé à proximité de la route/
éloigné de la route.
- 43 -
Lors de la phase de la description,on procède à un examen de la
documentation et à une reconnaissance générale des ressources et des
systèmes de production de la zone-cible. Cette phase fournit générale
ment une idée de la nature des contraintes même si elle ne permet pas
de procéder à une évaluation quantitative de leur importance ou de leur
rang de priorité.
Cette phase donne lieu à l'identification et à la mise au point
d'améliorations potentielles par l'équipe étudiant le système. Ces
activités s'intensifient au cours de la phase de diagnostic.
Après la formulation des innovations, on doit procéder à des essais
dans les conditions de l'exploitation. L'objet de ces essais est de
confirmer scientifiquement les effets des améliorations et d'évaluer
le degré de succès des innovations chez les producteurs.
Une fois que les améliorations auront été testées dans les conditions
de l'exploitation au cours d'essais dirigés et exécutés par les
agriculteurs, on pourra passer à la phase de la vulgarisation pour une
plus large diffusion. De telles activités s'effectuent généralement
dans le cadre de projets de développement.
La recherche sur les systèmes pastoraux s'effectue en un lieu
spécifique, sur une zone-cible particulière.
- 44 -
Summary of Discussion Session 1 .
Chairman: Dr. L.J. Lambourne (ILCA)
Dr Choma referred to Dr de Haan's paper in which he mentioned possible
on-station work during the design phase. Where would ILCA do this
work in e.g. Kenya where it had no research station? Dr de Haan said
that an LSR team working in a particular area did not have to do the
on-station work itself but could involve other research groups in the
process. If the results of ILCA's diagnostic phase indicated that on-
station work was necessary, the co-operation of national authorities
would be actively sought. Dr Chema asked at what stage of LSR should
extension officers be involved; Dr de Haan said they should be involved
at all stages.
Dr Zulberti pointed out that in recent work on the main components of
project success the early participation of farmers in the descriptive
and diagnostic phases had been identified as one of the most important
ones. Also the role of extensionists was crucial especially when
testing was being done because of the multiplication effect of their
participation. Dr de Haan agreed. Dr Zulberti suggested that
participants used the term 'pastoral systems research' during the
workshop rather than 'farming' or 'livestock systems research'.
Dr de Haan explained that he had used the term 'livestock systems
research' because he wanted to highlight the specific difficulties
inherent in livestock as a part of an overall farming system. The
focus of the workshop was indeed on pastoral systems, although he had
given a broader scope in the opening paper.
In referring to Mr Sandford's paper, Dr Zulberti said that the
development of water systems was not necessarily a consequence of
machinery reducing the cost of earth movement, because the ILO had
data showing that the cost of earth movement was cheaper with hand
labour. Mr Sandford pointed out that in most pastoral areas it was
very difficult to hire people for manual labour and often pastoralists
had a low productivity in this regard.
- 45 -
Dr Abel questioned whether or not there had been a sincere development
effort in the arid and semi-arid areas - such areas were frequently
marginal ones politically, economically and ecologically with their
people poorly represented. Mr Sandford agreed that there were extreme
difficulties in channeling development efforts to pastoral areas - he
felt it was unrealistic to expect governments to commit large sums
over long periods to the development of marginal areas. Development
efforts needed to be self-supporting after as short an initial
investment period as possible.
Dr Sorbo said he felt that ILCA was confronted by two main problems
in the field of systems investigations: the problem of improving the
conceptual and data technical tools by which one approaches the
realities of production systems; and the problem of applying these
tools and techniques to finding solutions to decision-making in
practical tasks. He felt ILCA's activities needed to be focused on
both problems, not just the second. Dr de Haan agreed, but said that
under pastoral producers' conditions ILCA's main focus should be on
the testing of hypotheses e.g. making scientifically sound conclusions
regarding the impact of improvements, but not going beyond the sample
size and the inputs required to test the specific hypotheses. If
that focus were kept in mind, the distinction between research and
development would be clear.
Dr Chema asked Mr Sandford to clarify what he had said about the
control of rinderpest by quarantine. Mr Sandford said that quarantine
was very difficult to administer and was very costly in terms of both
government efforts and inconvenience to pastoralists. Tissue culture
vaccine was in every way a vastly superior method of control. He
said he had been making the point that quarantine had controlled
rinderpest first, but was not advocating a return to quarantine control.
Dr Rhissa said that he thought that the failure of projects to
intensify animal production was not primarily due to the negligence
of governments, the lack of adequate policies, bad co-ordination or
poor planning. But rather this failure was due to the approach used
in which the producer was not involved from the start of intensifica
tion programmes. Dr Rhissa felt that ILCA should intervene only after
- 46 -
the state had determined its precise needs; ILCA could then after
field studies apply a package of solutions which the state could
choose and adapt according to their own needs. Mr Sandford agreed
that development planners had tried to think and act too much on
behalf of producers instead of directly involving them in the planning
process. But he still thought that negligence, poor policies, and
poor planning had also been major problems. Mr Sandford thought that
there was a role for ILCA to intervene earlier than Dr Rhissa proposed.
ILCA ought to be able to make valuable suggestions which themselves
help populations and states to determine their precise needs.
Dr Abel said that the understanding between local people and foreigners
could be good and that in some countries the extension service was so
alienated from the local farmers that only an outsider had a fair
chance of getting useful information. Communication between farmers
and scientists was both useful and possible.
Dr von Kaufmann said that systems research was about change and was
iterative. One could not expect farmers to comment easily on some
thing new - their answers would be motivated by all sorts of reasons.
Iteration meant going forward and seeing the farmers' reaction, then
going back and adjusting the innovation, then putting it to the
farmers again. Evaluation was a continuous process.
Dr Diakite felt that Mr Sandford had not taken into account the problem
of financing - finances were in general inappropriate, inadequate
and conditional. Mr Sandford agreed that financing conditions were
often quite inappropriate and that one expected too much too soon.
But he doubted if the fact that one was dealing with 'traditional
populations' had much to do with it. The problem was that development
projects often tried to do foolish things rather than that traditional
populations hesitated to do wise ones.
Dr Diakite felt that Mr Sandford underestimated the importance of
international health agreements. Mr Sandford thought that they were
not necessarily well founded - indeed often they were manifestly not
in the interests of developing countries. He believed that disease
- 47 -
eradication or control policy should not be regarded as sacrosanct
but should be subject to review and justification from time to time.
Prof. Saka Nuru said that it was acknowledged that national institutes
had carried out research into various national livestock problems
over the years with considerable success. However, traditional
research had its limitations in that its direct application to the
field was not feasible. ILCA's role was different in that it not
only looked at the problem in its totality through its multi-discipli
nary approach, but also carried the applied research to the farmers'
level. However, ILCA had to continue to rely on the basic research
resources of national institutions and to relate its work to the needs
of the development agencies of each country. A closer link with
national bodies was therefore desirable.
Mr Sandford said that the way in which PSR had been described - and
so defined - may have given the impression that the only 'output' of
PSR was successfully tested improvements. There were other valuable
outputs - which were only spin-offs of PSR as defined by Dr de Haan,
but were the outputs of other ways of defining PSR. These outputs
were in terms of the better understanding by everyone concerned in
pastoral development. A piece of PSR which led to no successful
improvements could still be valuable in terms of increased understan
ding and consequently better oriented development efforts.
Dr Bekure commented that if one confined systems research to the
production of technology to increase productivity, one closed the door
to research on e.g. marketing which did not deal with the producer
directly but yet was important for the producers' incentives and for
an understanding of producers' perceptions. Dr von Kaufmann agreed
that testing technology was not the only subject for FSR. If the
diagnostic phase revealed extension as the bottleneck, then FSR would
focus on how many farmers should there be per extension officer, what
would be the logistic package and how could the package be best
presented to farmers.
- 48 -
Dr Grandin pointed out that ILCA was not involved in adaptive research/
extension for its own sake. ILCA's purpose was to understand the
process of doing PSR, to work out the necessary methods. This could
be done partially by other research activities which helped to
evaluate the role or process of PSR. Dr Lambourne said that ILCA had
a mandate which required it to "assist national efforts which seek to
bring about a change in the systems of production..." ILCA thus had
two roles: to look for ways of improving and changing existing
production systems, and to find those systems which were best left in
their present equilibrium because of ecological or social reasons.
Dr Habou suggested that most livestock projects were more interested
in the animal than in man - but in the long run man was the centre
of any production system. Mr Sandford commented that the disregard
in planning of the human element was much more obvious in the 1960s
than either before or after that period. It was perhaps too early
to judge whether the changes from the 1960s to the 1970s in this
respect had led to better projects.
Dr Ngutter suggested that ILCA should ask governments and research
institutions what role they would like ILCA to play in their own
countries rather than ILCA assuming they knew each country's problems
a priori. Dr de Haan said that the essence of systems research was
that the problems of a certain production system were not known a
priori.. ILCA would see its involvement along the following lines:
governments ask ILCA for a systems study in a certain target area
chosen by the government; ILCA then decides in discussion with the
government research and extension agencies the kind of co-operative
agreements required.
- 49 -
Résumé des débats de la première séance
Président: M. L.J. Lambourne (CIPEA)
Le Dr Chema a fait allusion au document présenté par M. de Haan
dans lequel celui-ci mentionnait des travaux éventuels au niveau de
la station au cours de la phase de conception. Où est-ce que le CIPEA
pourrait effectuer ces travaux, par exemple au Kenya où le Centre ne
dispose pas de station de recherche? M. de Haan a déclaré qu'une
équipe de RSP travaillant dans une zone donnée n'avait pas à effectuer
elle-même les travaux au niveau de la station et qu'elle pourrait au
contraire demander à d'autres groupes de chercheurs de participer aux
travaux. Si les résultats de la phase de diagnostic du CIPEA
indiquaient que les travaux au niveau de la station étaient nécessaires,
on chercherait activement la coopération des autorités nationales.
Le Dr Chema a demandé à quel niveau de la RSP devaient intervenir
les agents de vulgarisation. M. de Haan a déclaré qu'ils devaient
intervenir a toutes les étapes.
Le Dr Zulberti a souligné que dans des travaux récents sur les
éléments essentiels de la réussite des projets, la participation
précoce des exploitants dans les phases de description et de
diagnostic a été identifiée comme déterminante. Le rôle des
vulgarisateurs était également très crucial, notamment lorsque les
tests étaient effectués en raison de l'effet multiplicateur de leur
participation. M. de Haan a accepté ce point de vue. Le Dr Zulberti
a suggéré que les participants utilisent l'expression "recherche sur
les systèmes pastoraux" au cours du séminaire plutôt que celle de
"recherche sur les systèmes d'élevage" ou"d' exploitation agricole".
M. de Haan a expliqué qu'il avait utilisé 1' expression "recherche
sur les systèmes pastoraux'parce qu'il voulait mettre l'accent sur
les difficultés spécifiques inhérentes à l'élevage conçu comme un
élément du système global d'exploitation agricole. Il était évident
que ce séminaire avait pour principal objectif les systèmes
pastoraux bien que dans son exposé préliminaire, il ait parlé dans
une perspective beaucoup plus large.
- 50 -
Parian* du document présenté par M. Sandford, le Dr Zulberti a
déclaré que dans la mise en place de systèmes d'adduction d'eau, la
conséquence de l'utilisation de machines pour les travaux de
terrassement n'entraînait pas nécessairement la réduction des coûts,
car le BIT disposait de données montrant que le coût de ces aména
gements était moins élevé avec le travail manuel. M. Sandford a
souligné que dans la plupart des zones pastorales, il était très
difficile da recruter une main-d'oeuvre pour des travaux manuels et
que souvent la productivité des éleveurs dans ce domaine était faible.
Le Dr Abel a demandé si oui ou non il y avait eu un réel effort de
développement dans les zones arides et semi-arides, celles-ci étant
fréquemment des régions marginales aux plans politique, économique et
écologique, et leurs populations étant mal représentées. M. Sandford
a reconnu qu'il était très difficile de canaliser les efforts de
développement dans les zones pastorales. A son avis, il était peu
réaliste d'attendre des gouvernements qu'ils engagent des sommes
importantes sur de longues périodes pour le développement de zones
marginales. Les efforts de développement devraient être auto-entrete-
nus après une période d'investissement initiale aussi brève que
possible.
Le Professeur Sorbo a souligné qu'à son avis le CIPEA était confronté
à deux problèmes essentiels dans le domaine de la recherche sur les
systèmes. Le problème de l'amélioration des instruments conceptuels
et techniques de collecte de données qui permet d'aborder les
réalités des systèmes de production et le problème de l'application de
ces outils et techniques à la formulation de solutions en vue de la
prise de décisions dans des domaines pratiques. Il a estimé que les
activités du CIPEA devraient être axées sur la solution de ces deux
problèmes et non seulement sur celle du second. M. de Haan a exprimé
son accord mais a déclaré que dans les conditions de la production
pastorale, le CIPEA devrait se concentrer essentiellement sur le test
des hypothèses, par exemple travailler en vue de parvenir à des
conclusions scientifiquement valables en ce qui concerne l'impact
des améliorations,mais que l'envergure de tels tests ne
- 51 -
devrait pas dépasser la taille d'un échantillonnage et les facteurs
requis pour vérifier les hypothèses spécifiquement identifiées. Si
l'on garde cela présent à l'esprit, la distinction entre recherche et
développement devient claire.
Le Dr Chema a demandé à M. Sandford de clarifier sa pensée lorsqu'il
a parlé de la lutte contre la peste bovine par la quarantaine.
M. Sandford a répondu que la quarantaine était difficile à mettre en
pratique et qu'elle était coûteuse et pour le gouvernement et pour les
éleveurs. Le vaccin à base de culture tissulaire était dans tous les
cas une bien meilleure méthode de lutte. Il a déclaré qu'il avait
voulu souligner que la quarantaine avait été utilisée par le passé
pour lutter contre la peste bovine, mais qu'il ne tenait pas pour
autant à se faire l'avocat de la quarantaine.
Le Dr Rhissa a déclaré qu'à son avis, l'incapacité des projets
d'intensifier la production animale ne relevait pas essentiellement de
la négligence des gouvernements, de l'inefficacité des politiques
adoptées et de l'inadéquation de la coordination ou de la planifica
tion. C'était plutôt l'approche utilisée dans laquelle le producteur
n'était pas présent pendant la phase initiale de l'intensification des
programmes qu'il fallait incriminer. Le Dr Rhissa a estimé que le
CIPEA ne devait intervenir qu'après que l'Etat ait déterminé de
manière précise ses besoins; le CIPEA pourrait alors, après des
études de terrain, utiliser un ensemble de solutions que l'Etat
pourrait choisir et adapter en fonction de ses propres besoins.
M. Sandford a reconnu que les planificateurs du développement avaient
trop souvent essayé de penser et d'agir à la place du producteur
plutôt que de l'intégrer directement dans le processus de planifica
tion. Mais il persistait à croire que la négligence, de même que
des politiques et une planification inadéquates avaient également
constitué de sérieux problèmes. M. Sandford pensait que le CIPEA
pouvait intervenir plus tôt que ne l'avait suggéré le Dr Rhissa. Le
CIPEA devait être à même de formuler des suggestions valables qui
pourraient aider les populations et les Etats à déterminer leurs
besoins.
Le Dr Abel a déclaré que la compréhension pourrait régner entre les
autochtones et les étrangers et que dans certains pays, le service de
- 52 -
vulgarisation était tellement éloigné des préoccupations des agricul
teurs locaux que seul un étranger pouvait accéder à des informations
utiles. La communication entre les agriculteurs et les scientifiques
était à la fois utile et possible.
M. von Kauffmann a déclaré que la recherche sur les systèmes portait
sur les changements et qu'elle était itérative de nature. Il ne
fallait pas s'attendre à ce que les exploitants agricoles puissent
exprimer facilement leurs impressions sur quelque chose de nouveau.
Leurs réponses seraient motivées par toutes sortes de raisons.
L'itération signifiait l'observation des réactions de l'agriculteur
puis l'adaptation des innovations qui seraient de nouveau présentées
à l'agriculteur. L'évaluation était un processus continu^
Pour le Dr Diakité, M. Sandford n'avait pas pris en considération le
problème financier. Le financement était en général insuffisant et
conditionnel. M. Sandford a reconnu que les conditions de financement
étaient souvent réellement inadéquates et que par ailleurs, l'on avait
tendance à faire preuve d'un optimisme exagéré en ce qui concerne le
volume et les délais des financements. Mais il a déclaré qu'il ne
pensait pas que le fait qu'il s'agissait de "populations tradition
nelles" en soit la raison principale. Le problème était que le
projet de développement essayait très souvent de réaliser une
entreprise insensée et non que les populations traditionnelles
hésitaient à faire des choses censées.
Le Dr Diakité a déclaré qu'il estimait que M. Sandford sous-estimait
l'importance des accords sanitaires internationaux. M. Sandford
estimait que ceux-ci ne se justifiaient pas nécessairement et qu'en
fait, très souvent, ils n'allaient manifestement pas dans le sens des
intérêts des pays en développement. Il estimait que la politique
d' eradication des maladies ou de lutte contre celles-ci ne devrait
pas être considérée comme sacro-sainte mais qu'elle devrait de
temps à autre faire l'objet de révision et de justification.
Le Professeur Saka Nuru a déclaré qu'il était reconnu qu'au cours des
années qui venaient de s'écouler, les instituts nationaux avaient
- 53 -
effectué des travaux de recherche sur plusieurs problèmes nationaux
d'élevage avec beaucoup de succès. Toutefois, la recherche tradition
nelle avait ses limites en ce sens qu'une application directe sur le
terrain n'était pas faisable. Le rôle du CIPEA était différent dans
la mesure où, grâce à son approche multidisciplinaire, il étudiait les
problèmes dans leur globalité mais aussi parce qu'il menait à bien
des activités de recherche appliquée au niveau de l'exploitation
agricole. Toutefois, le CIPEA devait continuer à s'appuyer sur les
ressources qu'offrent les institutions nationales en matière de
recherche de base et à faire en sorte que ses travaux concordent avec
les besoins des organismes de développement de chaque pays. Il était
donc souhaitable que des liens plus étroits soient établis avec les
organismes nationaux.
M. Sandford a déclaré que la manière dont la RSP avait été décrite et
définie avait pu donner l'impression que la seule contribution de la
RSP consistait à mettre au point des améliorations testées avec
succès. Il y avait d'autres contributions valables qui n'étaient
que des effets secondaires de la RSP telle que définie par M. de Haan
mais qui découlaient d'autres manières de définir la RSP. Il s'agit
précisément de l'accroissement des connaissances de tous ceux qui
s'intéressent au développement du système pastoral. Des travaux de
RSP qui n'ont pas donné lieu à des innovations adoptées pourraient
néanmoins être valables, notamment en développant les connaissances
sur le sujet et partant, en permettant de mieux orienter les efforts
de développement.
M. Bekuré a souligné que si l'on confinait la recherche sur les
systèmes à la mise au point de techniques en vue d'accroître la
productivité, on fermerait la porte à la recherche sur la commerciali
sation qui ne s'appliquait pas directement au producteur mais était
toutefois importante en ce qui concerne la motivation des producteurs
et la compréhension de l'opinion des producteurs. M. von Kauffmann
a reconnu que le test de techniques n'était pas le seul objet de la
recherche sur les systèmes d'exploitation agricole. Si la phase de
diagnostic révélait que la vulgarisation constituait un goulet
d'étranglement, la RSA mettrait l'accent sur le nombre des exploitants
- 54 -
agricoles qui doivent être couverts par un agent de vulgarisation,
sur ce que devrait être la logistique à mettre en oeuvre et sur la
manière de présenter de la façon la plus avantageuse cette logistique
aux exploitants agricoles.
Mlle Grandin a souligné que le CIPEA ne participait pas à une
vulgarisation/recherche d'adaptation en soi. L'objectif du CIPEA
était de comprendre le processus de la RSP pour élaborer les méthodes
nécessaires. Cela pouvait s'effectuer partiellement par d'autres
activités de recherche qui ont contribué à évaluer le rôle ou le
processus de la RSP. M. Lambourne a déclaré que le mandat du CIPEA
consistait à "aider les pays intéressés dans leurs efforts en vue
d'introduire des changements dans les systèmes de production..." Le
CIPEA avait ainsi deux rôles: chercher les moyens d'améliorer et de
changer les systèmes de production actuels; trouver les systèmes qui
étaient les meilleurs dans leur état actuel d'équilibre, compte tenu
des conditions écologiques ou sociales.
Le Dr Habou a souligné que la plupart des projets d'élevage
s'intéressaient davantage a l'animal qu'à l'homme. Mais qu'à long
terme, c'est l'homme qui se trouve au centre de tout système de
production. M. Sandford a déclaré que la négligence en ce qui
concerne la planification de l'élément humain était beaucoup plus
évidente dans les années 60 qu'avant ou après cette période. Il
était peut être prématuré de dire si les changements intervenus entre
les années 60 et les années 70 à cet égard, on donné lieu à de
meilleurs projets.
Le Dr Ngutter a suggéré que le CIPEA devrait demander aux gouverne
ments et aux institutions de recherche, le rôle qu'ils voudraient
que le Centre joue dans leurs pays respectifs plutôt que de supposer
qu'il connaît les problèmes de chaque pays à priori. M. de Haan a
déclaré que l'essence de la recherche sur les systèmes était que les
problèmes d'un système donné de production n'étaient pas connus a
priori. Le CIPEA devrait s'engager sur la base des éléments
suivants: les gouvernements demandent au CIPEA d'effectuer une étude
de système dans une zone-cible donnée choisie par lui-même; le CIPEA
- 55 -
décide ensuite, à la suite de concertations avec les organismes de
recherche et de vulgarisation du gouvernement, du type d'accord de
coopération requis.
- 56 -
THE SURVEY AND DIAGNOSTIC PHASE OF PSR

Integration of remote sensing techniques
for resource evaluation
in pastoral systems research
1 . . 2
Peter N. de Leeuw and Kevin Milligan
Ecologist, Arid Zones (Eastern and Southern Africa) Programme, ILCA,
Kenya
2
Air Survey Coordinator, ILCA, Nigeria
Introduction
The leading assumption of this paper is that natural resource surveys
and inventories are an essential prerequisite for planned development
and should be integral components of the descriptive and diagnostic
phases of livestock systems research.
In the past, range resource assessment at ILCA was rarely
formally placed in this sequential framework, but awareness that such
assessment was needed has been increasingly felt. The trends in
ecological and vegetation research at ILCA is a convincing illus
tration. For instance, in Mali during the earlier years, emphasis
was placed on detailed ecological studies on the Niono ranch in order
to understand the vegetation dynamics of the upland Sahelian ecosystem.
This was combined with extensive vegetation mapping in the Niger Delta
and its surroundings according to the exacting procedures of the
Zurich-Montpellier school of phyto-sociology . Initially, there was
little concern about the application of this research to other
disciplines or to the production systems within which the research
was done. Such research was considered a laudable pursuit in its
own right. In recent years, inter-disciplinary research has become
prominent resulting in data flows from ecological research that were
more oriented to the needs of range and animal nutrition science,
without neglecting the pure research aspects of the work. This more
pragmatic and applied approach led to maps of the ILCA study area
that not only depicted vegetation types, but also contained complemen
tary information for carrying capacity estimations and assessment of
land use suitability for grazing, flooded rice and rainfed cropping.
- 57 -
This trend was not restricted to ILCA alone, but also
noticeable in the various national and international agencies engaged
in other surveys and inventories. As will be shown in the sections
on resource surveys and methodology, emphasis moved from heavy and
costly procedures to rapid, cost-effective and consumer-oriented data
generation. Although cost escalation reinforced this trend, the
knowledge that many of the earlier inventory reports remained shelf-
bound and never effectively influenced their designated audiences, was
more important.
This paper is designed to provide a short and selective
overview of resource surveys and their methodologies that seem most
appropriate to pastoral and agro-pastoral regions in Africa. Full
acknowledgement must be given to Dent and Young (1980) from whose
book several tables and figures are derived and on which substantial
parts of the text is based. The most recent applications of remote
sensing techniques are briefly summarised and put in the context of
rapid appraisal methods and early warning procedures within a livestock
systems research approach.
Types of surveys - a historical perspective
Soil survey
Soil survey involves the production of a pedological map, which shows
the distribution of soil units defined primarily according to their
morphology and their physical, chemical and biological characteristics.
These units are based on the description of soil profiles,
their comparison and subsequent grouping into a classification system
that is significant for the aims and purpose for which the survey
was intended. Stobbs (1970) classified soil surveys in three
classes from exploratory to detailed and indicated the appropriate
mapping scales (Table 1) . Classification of soil types varies
with survey objectives but usually follows either the FAO soil map
of the world (FAO-UNESCO, 1970-80) or the U.S. Soil Taxonomy (Soil
Survey Staff, 1975) and the selection of what system is used depends
often on the donor agency that has commissioned the survey.
- 58 -
General purpose soil surveys are most appropriate in sparsely
settled areas in developing African countries, where there was and
often still is a dearth of basic geographical information on resources-
Exploratory and reconnaissance surveys provide a first approximation
of such data. In densely settled areas small-scale surveys serve no
direct useful purpose and instead detailed surveys are needed (Table !)•
General purpose soil maps may serve as a basis for further stages of
interpretation, namely land evaluation (see below), which apart from
soil units includes other relevant physical, economic and sometimes
social factors.
During the 1960's a large number of soil surveys were
carried out in Africa. Although they were meant as a basis for a
land resource inventory, they have been criticised in that their only
purpose was to make a soil map, as a desirable end in itself. A
more tenable criticism is the cost and time involved in describing,
mapping and reporting. Dent and Young (1980) calculated that for
2
a survey at a scale of 250,000, 100 man-days /l ,000 km were needed or
2
about US$ 15/km at 1980 prices, of which the actual interpretation
of aerial photos (scale 1:40,000) would take 7 to 10 man-days. This
2
is somewhat lower than the 8 man-months/1 ,000 km needed for resource
mapping in the Sudan (Wilson, 1979). If these costs are applied to the
reconnaissance survey of the Amboseli-Kibwezi sheet in Kenya, this
2
project of 12,600 km would have cost USg 200,000.
Land systems survey
Although soil surveys have been used to produce derived information
on land resources, capability and suitability, more direct approaches
to natural resource inventory were sought, that would be more rapid
and less costly. In anglophone Africa the most common approach is
the use of the land system concept as the basic mapping unit for
subdividing and classifying land. A land system is defined as an
area with a recurring pattern of topography, soils and vegetation
and with a relatively uniform climate. This area is subdivided into
land facets which are the smallest areas that can be recognized and
delineated on aerial photographs and are usually linked by geomorpho-
- 59 -
Table 1. Types of soil survey
Type Map units Scale Purpose
Exploratory Associations of
phases of great
soil groups
Reconnaissance
Detailed low
intensity
Associations of
phases of soil
series or higher
categories (great
soil groups or
families)
Phases of as
sociations
1:1,000,000
and smaller
(schematic
maps)
.. To locate areas of
substantial soil
difference
(inventory)
2. Locate more de
tailed work
3. Test legend
1:62,500 to 1 . To survey areas
1:500,00 suited only to
extensive use
2. Pre-detailed
survey to locate
and define such
work
1:30,000 &
smaller
For forestry and
grazing development
areas
Source : Stobbs (1970)
logical process. A land system can be alternatively defined as an
area with a recurring pattern of genetically linked land facets.
As part of the integrated approach, derived information is
reported on vegetation, range and water resources, on actual and
potential land use, the latter being based on land suitability ratings
for each land system and its constituent facets.
Numerous integrated surveys have been carried out by the Land
Resources Division in Africa, in particular in Nigeria, Zambia and
Malawi (LRD, 1966-79). In the agropastoral zone in northeast Nigeria,
emphasis was put on assessing range resources, actual and potential
seasonal stocking rates and associated limitations to use such as
flooding, inaccessibility and tsetse challenge (de Leeuw et al , 1972;
60 -
de Leeuw, 1976) . From these inventories came recommendations on
seasonal de-stocking requirements if a more balanced utilisation
of pastoral land was desired (Tables 2 and 3) . Although these
recommendations were rarely actively implemented, they helped to
increase the awareness of the problems in ministries and planning
authorities.
In the subhumid zone in Nigeria land systems and soil
mapping was followed by a detailed land evaluation based on the
j
suitability for eight individual crops. Suitability was assessed
from climatic limitations, soil physical and chemical characteristics
and erosion hazards and this led to a map of crop options at a scale
of 1:250,000 (Hill et al, 1978).
The main advantages of the land systems approach is speed and
relative cheapness, the integration of different environmental factors
and a clear communication of results (although some reports run to
over 1,000 pages). Disadvantages are a high degree of generalisation,
variable and somewhat ill-defined mapping units and the static nature
of the information presented. Greater emphasis should be given to
the dynamic aspects of environment such as soil moisture regimes,
successive stages in the vegetation (cropping/fallows, effects of
fire on woody/herbaceous species balance) and the interactive effects
between range resources and livestock use.
Natural resource surveys and land suitability evaluation
As has already been shown above for the central Nigeria survey (Hill
et al, 1978), a basic natural resource survey can lead to the
production of maps and other data on which land evaluation is based.
The information needed for a range resource inventory concerns
amounts and seasonal distribution of dry matter production, digestible
protein, etc., but to define mapping units and to delineate their
boundaries calls for some combination of vegetation physiognomy and
species frequency according to established principles of vegetation
mapping.
- 61 -
Table 2. The estimated actual and potential wet-season cattle
population in northeast Nigeria (in thousands)
Zone Estimated Potential Difference
(%)
Tsetse-free 1,987 1,214 + 65
Tsetse 584 2,352 - 75
TOTALS 2,571 3,569
Source: de Leeuw (1976)
Table 3. Potential year-round population (1000 AUs)
Type of land Tsetse-free zone Tsetse zone
wet dry wet dry
season season season season
savanna 1,000 - 1,576 972
fallows 213 148 229 115
cereal residues - 363 - 180
alluvial/flooded land - 840 — 400
Sub-totals 1,213 1,351 1,805 1,667
Grand total 3 ,028
Source: de Leeuw (1976).
- 62 -
The fundamental descriptive features for livestock production
systems are the types and numbers of livestock, the livestock inten
sity and the degree of nomadism. The livestock intensity is the
number of livestock units per square kilometre of grazing land.
Where present, livestock systems may be classified as total nomadism,
semi-nomadism or transhumance (for definition, see Ruthenburg, 1980).
These descriptive headings can also be used for the livestock
component of mixed or predominantly arable farming. The activity
flow-chart of a land suitability evaluation is given in Figure 1.
Land Suitability Evaluation
Initial consultations
(a) objectives
(b) data and assumptions
(c) planning of the evaluation
Kinds of land use
Major kinds of
land use or land
utilisation types
I
T
A
\
. \
iteration/
/
/
Land use
requirements
and limitations
Comparison of land
use with land
I
Land suitability
classification
T
Presentation
of results
Resource surveys
Land mapping units
Land
characteristics
and qualities
Land
improvements
Source; Dent and Young (1980), p. 146
Fig 1 . Schematic* representation of activities in land evaluation.
- 63 -
The resource survey has the function of dividing the study
area into a number of relatively homogenous units and of providing
the information necessary to evaluate each of these units for the
kinds of land use under consideration. Thus, resource surveys should
not be treated as a separate activity, but as an integral part of the
evaluation. For example, in a semi-arid region that is dependent on
pastoralism, a survey effort would be directed towards range and
water resource surveys and the determination of livestock carrying
capacity. Seen in this light, the numerous surveys and mapping
carried out in francophone West Africa by IEMVT could be termed
pastoral resource surveys. Dent and Young (1980) enumerate the land
qualities related to livestock production as follows (Table 4) .
Table 4. Land qualities related to livestock production
Nutritive value of natural pastures
Nutritive value of improved pastures
Resistance of degradation of vegetation
Resistance of soil erosion under grazing conditions
Toxicity of grazing land
Availability of drinking water for livestock
Climatic hardships affecting livestock
Endemic pests and diseases
Access within the production unit
Size of potential management units
Location: - (a) existing access
(b) potential access
Source : Dent and Young (1980), pp. 166-168.
Remote sensing methodology
Remote sensing methods have been classified according to the altitude
at which imagery is taken, and include satellites, and photographs
and visual observations from low-flying aircraft. Satellite imagery
and conventional aerial photographs usually provide total cover,
whereas low-altitude aerial surveys are usually based on sampling
strips. The characteristics of these methods are summarised in
Tables 5 and 6.
- 64 -
Table5.Somcharacteristicsofthdifferentme hodsusnr m tesensing
Vehicle
Scaleofgeneralus
landrover,otpplicable
lighta r-non-photogrammetric
craftorf o
Usualcover
usuallyquadrat;
usuallystrips
total total
sampling sampling
250,000
,000 non-photogrammetric
1:1,000,000-1,N
1:20,000-1:100, 1:300-1:20,000 notapplicable
high-level
satellite aircraft low-level low-evel
aircraft aircraft
1 1
samplingusually
quadrats,fre
quentlylinked withlow-level photogrammetry
Sensors
PrimaryUse
various,b s cally
scanningradiometers
cameras cameras
humanobserversand
obliquesma l-format
cameras
humanobservers
usingara geof
resourcesurv y
equipment
mapping
mapping+overall
quantificationsouse
+useofpatt rns
quantificationsouse
+useofpatterns
photogrammetric
quantificationouse
+useofpatt rns
photogrammetric
'groundtruth'
Source:WatsonandTippet(1'1),Fig.,
Satellite imagery
The main source of imagery has come from the LANDSAT series, which
2 .
produces monochromic or colour-enhanced images of 34,000 km with a
resolution of 80 m (LANDSAT-3) at a frequency of every 18 days.
Thus, images at different seasons and successive years are readily
available and relatively cheap.
LANDSAT is increasingly used for rapid small-scale reconnais
sance surveys when mapping is envisaged at a scale from 1:1,000,000
to 250,000. Usually monochrome prints are used for landscape or
land systems mapping, while false colours are more appropriate for
vegetation and land use.
Interpretation of one LANDSAT image costs about USg 20/1,000
2
km (1980) and it is possible to map large areas in no more than a
few months. Mitchel and Howard (1978) reported that a land systems
2
map of Jordan (c. 100,000 km ) was mapped on a scale of 1:1 million
in 1.5 months of interpretation, one month of field work and two
weeks for report preparation and printing.
In addition to LANDSAT, satellites (or aircraft) have been
equipped with infrared detectors that record the far-infrared radi
ation of the earth's surface and can distinguish vegetation and soil
types by their thermal emission characteristics. Recently Tucker
(1983) used the thermal channel of the N0AA satellite to detect
reflectivity differences and found that variations in dry standing
biomass could be derived from remote-sensed temperature differences
(Table 6) .
Aerial photographs
Since the 1950s aerial photo interpretation has become an important
tool for soil and other surveys, in particular for reconnaissance
studies in Australia and developing countries. Black-and-white
photos are most commonly used but other types of film have been
tried for specific purposes : near infrared for tree differentiation
in forest evaluation and false-colour photography in vegetation and
land-use surveys and large-scale disease detection of crops and
forests. Colour photos have been tried for mapping desert landscapes
- 66 -
and urban fringe zones and also for sampling land use and
vegetation patterns from low-flying aircraft (Watson and Tippet 1981;
K. Milligan, personal communication).
Side-looking radar (SLAR)
Althought SLAR has very low resolutions (Table 6) , it is used for
mapping regions which are rarely cloud-free to permit other methods
of remote sensing. SLAR was the basis for a reconnaissance resource
inventory of the Amazon basin and for a country-wide land use and
vegetation map of Nigeria. The quality of this map is being tested
j
through comparison with low-altitude vegetation sampling as well as
with existing land-use maps based on aerial photo interpretation
(Hill et al, 1978; Milligan and de Leeuw, in press).
Spectral radiance
Linear or ratio combinations of two wavelength regions, red and
near infrared (see Table 6), measured at ground level (1-1.3 m) at
low altitude from aircraft (60-200 m) or from satellites have been
used to estimate green standing biomass (Tucker, 1980). For instance
in the Serengeti (McNaughton, 1979) and in the Amboseli ecosystem in
East Africa (Western and Grimsdell, 1979) satisfactory correspondence
was reported between clipped ground biomass and 'green machine'
measurements on the ground and from low-flying aircraft at least for
open grassland with a low woody cover. Estimating biomass in veg
etation types high in woody cover is less promising, although
Herlocker and Dolan (1980) in north-east Kenya found a good corre
lation between clipped green biomass of dwarf shrubs and hand-held
radiometer readings. Similarly, promising results were obtained
in the Sahelian zone in Senegal from radiometer readings of the
N0AA-7 satellite during the 1981 rainy season. From these readings
(calibrated with ground clippings) a map was made of cumulative
2biomass production for a 30,000 km region in classes of 200 kg/ha
in the range from 0-1,600 kg DM/ha (Tucker, 1983; Gaston and
Boerwinkel, 1982),
- 67 -
Table6.Somecharacteristicsofremo - en ingsyst ms
Resolution
veryhigh
high high high
(plusvisible,
0.4-0.-um)Wavelength 0.4-0.-um 0.--0.-um 0.3-0.-um 0.5-0.-um
Remotesensing
system
near-infrared false-colour
photography
normalair
photography photography true-colour photography
monochrome
1 i
Specialfacilities
Applicationsinsoil
orrelatedsu veys
thermalinfrared
side-looking
airborneradar
(SLAR)
--14um
low
0.--3.0cmverylow
chlorophyllreflects
strongly,water
absorbs
groundsurface
numerous
detectionofrainage,
impedancesurfacewater; differentiationoftree
species
urbanf ingesurveys,
display
emphasisesvariationsvegetationndl nd
invegetationusemapping,soilsurveys
inntertidalzone
recordsheatemittednone
bythegroundsurface
penetratescloud
possiblyreconnaissance
incloud-coveredareas
satelliteMSS
0.5-l.lumhighrelativetoun f rcoverageof
small-scaleimage,l rgeea
lowrelativeto
ground^^^_____________
coverageofve ylarge areas,initialoverview
inothersurveys
Source:DentandY ng(1'0),Table3.4,p.A-
Although Tucker (1980) concluded that 'for large area
surveys, the spectral method has been shown to work well and allows
for synoptic coverage of large areas from aircraft and/or satellite
platforms', several problems in the calibration techniques have to
be resolved (solar zenith angle, sunlight vis-a-vis overcast
conditions etc.) before a more general adoption of this methodology
can be recommended.
Low-altitude aerial survey
Aerial observations have become an integral part of reconnaissance
and inventory surveys. For a detailed discussion, reference
is made to Milligan and de Leeuw (1983) who have reviewed the
the aerial survey methodology within the framework of ILCA's
livestock system research programmes in Mali, Niger and Nigeria.
LIGHT AIRCRAFT
OBSERVATIONS1
r
impressions for
LANDSAT
Opt. Proc.
Imagery
interpretation
annotation for
identification
and delineation
SMALL SCALE AERIAL
PHOTOGRAPHY
J]
 
U-i
c
-a
Ground Sampling
Qual . Invent .
Estimation
Source : Thalen (1981), Fig. 4.
Figure 2. Possible combinations of complementary reconnaissance
rangeland survey techniques.
- 69 -
The integration of remote sensing techniques and ground truth
Introduction
In most resource surveys at the reconnaissance level, information is
gathered from space, air and ground and it is clear that information
quality improves from space to ground as do costs. Consequently,- a
balance needs to be struck between extensive and intensive methods
which is geared to the objectives of the survey undertaken.
Thalen (1981) reviewed the different method combinations for !
the evaluation and resource mapping of arid rangelands with the focus
on assessing what grows where and when and how good is it. He showed
that the most cost-effective approach for arid land vegetation surveys
was the use of low-altitude aerial surveys in combination with LANDSAT
imagery or detailed topographical maps. However, to produce a
2
rangeland map of the Kalahari desert in Botswana (240,000 km at a
j
scale of 1:500,000) LANDSAT imagery at different scales was used for j
orientation and initial delineation of rangeland types together with
complementary conventional aerial photography and aircraft observations
for more detailed type identification and description as well as for
stratification and subsequent ground truth sampling (Fig. 2).
A similar approach was used to describe, assess and map the
natural resources of Mali south of the 18° parallel. A broad land
systems map was produced from LANDSAT-2 false-colour composites
enlarged to a scale of 1:200,000. With the major land systems so
identified, representative sample transects were selected from false -
colour images and subsequently photographed with hand-held cameras
from light aircraft simultaneously at scales of 1:2,000 and 1:8,000.
The resulting slides were analysed for tree and grass cover, land
use and soil surface characteristics and correlated with false-colour
pixel distributions on the LANDSAT prints so as to relate the latter
to the real world. Transect photo data were then used for the
location of ground samples for data collection on soils, vegetation
and cropping. However, it is unfortunate that an appraisal of this
approach has to await the publication of the maps and reports.
70 -
An integrated appraisal approach
As shown by Milligan and de Leeuw (1983) the use of aerial surveys
within the different ILCA programmes has been less integrated than
was desirable. Data collection from aircraft was often done as an
independant exercise and data integration was carried out afterwards
by interested researchers (e.g. de Leeuw and Milligan, 1981).
Having gained experience of aerial surveys in the context
of livestock systems research (Milligan et al, 1982; Milligan,
1982), a better planned and more integrated approach seems worth
testing. Some work has already started in the Maasai system project
where aerial surveys are part of an inter-disciplinary study and direct
comparisons between a large volume of ground truth data and aerial
survey observations can be made (e.g. Peacock et al, 1982).
The proposed approach differs from earlier work in that it
is planned to execute a rapid appraisal of a target area (50,000 to
2
100,000 km in size) which should deliver a comprehensive description
of the resource base and the people that use it together with a
constraint diagnosis of the production system in the area.
As set out in Fig. 3, the appraisal is executed in time
sequence with the following steps :
1. From satellite imagery and other information, landscape
(landform or land system) units and vegetation zones are delineated
and areas of particular interest defined.
2. Systematic and/or stratified low-altitude aerial surveys
are carried out over the whole zone at the same or different
intensities according to strata. After an initial data analysis
and mapping of range, water livestock and human resource distributions,
the target area is subdivided in homogeneous 'land units', which
provide the base for further aerial survey and ground observations.
3. Range resources are further assessed by multi-scale low-
level aerial photography along transects and if possible, biomass
estimates are made from radiometer readings followed by ground
sampling of biomass for quality and quantity and species composition.
At the same time ground teams select representative households for
- 71 -
SATELLITE IMAGERY
T
ZONATION
Land forms
Vegetation
Land use
T
 
I
SAMPLE AREA SELECTION
\
SYSTEMATIC LOW-ALTITUDE FLIGHTS
±
VISUAL OBSERVATIONS
Range, Water, Animals, People
 
I
STRATIFICATION & SAMPLE SELECTION
I
TRANSECT FLIGHTS
Photography and
visual obser
vations
 GROUND SAMPLING
Biomass species composition
cropping system, herd pro
ductivity, labour and budgets
£
Wide-angle
farm areas
vegetation
water
I
Telephoto tree/
grass cover
species compo
sition cropping
systems
1
Radiometer
biomass
estimates
Fig 3. A rapid appraisal approach for pastoral systems research
- 72 -
rapid appraisal of human and herd demography, labour budgets and
income and expenditure. If large seasonal differences in resource
distribution are expected, a second aerial survey is planned to look
at all system attributes that are subject to large seasonal variations *
Early warning procedures
Forecasting of environmental conditions, in particular of range and
water resources relative to existing livestock and human requirements,
would be a prerequisite for the better long-term management of fragile
ecosystems that are subject to varying climatic events.
It is postulated that timely prediction is most urgent in
regions with a short single rainy season and with range resources
that largely consist of annual grasses. The need is much lower
in regions with bimodal rainfall patterns. While in the Sahel
two successive rainy seasons with poor rainfall will have a disastrous
impact on long-term livestock productivity due to its two-year
continuous time span, a similar event in East Africa (e.g. the
Maasailand) is of much shorter duration (10-12 months).
The same is true for recovery rates following droughts.
i
While it took nearly ten years for Sahelian cattle herds to reach
their pre-drought levels, livestock populations in Kajiado (south
east Kenya) recovered in half that time. More important than popu
lation recovery has been the long-term effect of the Sahelian drought
on livestock ownership patterns. Since the drought, ownership has
shifted markedly from herders to outsiders (traders and civil ser
vants) . This trend is a concomitant result of the unfavourable
price ratios between stock and grain during drought since many
producers not only lost stock through death but had to sell at low
prices to buy high-priced grain and became indebted to traders,
moneylenders and others. Such a 'drought syndrome' can be reduced
if pastofalists can be persuaded to react timely to the deteriorating
situation by early selling of stock and early purchase of grain.
Prediction in mono-modal rainfall regions can be based on
available resources at the end of the growing season. As disappear
ance rates of standing biomass are fairly well known^and if the size
- 73 -
of the livestock population dependent upon the resources is known
from aerial surveys, the balance between supply and requirements can
be calculated for the dry season. If the region covered is suf-
2
ficiently large (the NOAA satellite can cover up to 600,000 km in
one composite image) , cropping areas south of the pastoral zone can
be included from which potential crop yield estimations can be
derived.
Once the alert for pending drought is raised, adapted
management strategies can be put into operation (supplementary feeding,
mobilisation of additional water resources in under-utilised areas
etc.) combined with accelerating the flow of stock through marketing
channels and providing credit for early grain purchases.
The integrated approach could thus be extended to include
an 'early warning' component. To be effective it is essential that
biomass estimations through remote sensing are reliable for large
regions, and it thus seems worthwhile to accord a high priority to
developing this methodology.
References
Dent, D. and Young, A. 1980, Soil survey and land evaluation.
Allen and Unwin, London.
FA0-UNESC0, 1970-80. Soil map of the world. 1:5,000,000, vols.
1-10. UNESCO Paris.
Gaston, A. and Boerwinkel, E. 1982. Evaluation de la production
primaire herbacee sahelienne par mesures simultanees au
sol et par satellite, mise au point d'une technique
rapide et extensive. Coll. sur l'inventaire des ressources
et la coordination de la cartographie au Sahel. USAID,
Ibadan.
Herlocker, D.J. and Dolan, R.A. 1980. Comparison of different techniques
for the determination of large dwarf shrub biomas, IPAL
Techn. Rep. A-3, 30-40, Nairobi.
74 -
Hill, I.D. 1978. Land resources of central Nigeria, agricultural
possibilities. 1 vols. Land Resources Development Centre,
Surbiton.
Land Resources Development 1966-1980. Land resources studies (nos. 1-
31. Surbition.
de Leeuw, P.N., Lesslie, A. and Tuley, P. 1972. The land resources
of northeast Nigeria. Vol. 4. Present and potential land
use. Land Resources Division, Tolworth.
de Leeuw, P.N. 1976. Fodder resources and livestock development in
Northeast Nigeria. Savanna, 5; 61-74.
de Leeuw, P.N. and Milligan, K. 1981. Livestock resources. In:
Systems research in the arid zones of Mali. ILCA Systems
Study 5, pp. 36-54. ILCA, Addis Ababa.
McNaughton, S.J. 1979. Grassland-herbivore dynamics. In: Sinclair,
A.R.E. and Norton-Griffiths, M. (eds) . Serengeti, dynamics
of an ecosystem, pp. 46-81; Univ. of Chicago, Chicago press.
Milligan, K. , Keita, M. and de Leeuw, P.N. 1982. Recensement aerien
saisonnier du cheptel et types de paysage du delta central
du Niger au Mali. Report to ODEM/IBRD, Sevare.
Milligan, K. 1982. Aerial survey of human, livestock and environ
mental conditions in a central region of the pastoral
zone of Niger. Final Report. USAID, Niamey.
Milligan, K. and de Leeuw, P.N. 1983. Low altitude aerial surveys in
pastoral systems research. Proc. Workshop on pastoral
systems research in sub-Saharan Africa, March 21-25 1983,
Addis Ababa.
Mitchel, C.W. and Howard, J. A. 1978. Land systems classification:
A case hzstory, Jordan. AGTL Bull. 2/78, FAO Rome.
Peacock, C.P., de Leeuw, P.N. and King, J. 1982. Herd movement in
the Mbirikani area. Int. Rep., ILCA, Nairobi.
- 75 -
Ruthenburg, H. 1980. Farming systems in the tropics, 3rd edition
Oxford University Press.
Soil Survey Staff, 1975. Soil taxonomy. A basic system of soil
classification for making and interpreting soil surveys,
Agric. Handbook 436. Dept of Agric, Washington D.C.
Stobbs, A.R. 1970. Soil survey procedures for development purposes.
In: I.H. Cox (ed.). New possibilities and techniques for
land use and related surveys, 41-64. Geographical Publ.,
Berkhamsted.
Thalen, D.C. P. 1981. An approach to evaluating techniques for
surveying vegetation on arid rangelands. In: Low-level
aerial survey techniques, 85-98. ILCA Monograph 4,
Addis Ababa.
Tuckei , C.J. 1980. A critical review of remote sensing and other
methods for non-destructive estimation of standing crop
biomass. Grass and Forage Sc. 35: 177-182.
Tucker, C.J. 1983. Determining the spatial extent of deserts using
NOAA AVHRR digital data (In press).
Watson, R.M. and Tippet, C.I. 1981. Examples of low-level aerial
surveys conducted in Africa, 1969-1979. In: Low-level
aerial survey techniques, 35-58. ILCA Monograph 4, Addis
Ababa.
Western, D. and Grimsdell, J.J.R. 1979. Measuring the distribution
of animals in relation to the environment. Handbook
No. 2, African Leadership Foundation, Nairobi.
Wilson, R.T. 1979. Recent resource surveys for rural development in
southern Darfur, Sudan. Geogr, J, 145: 452-460.
- 76 -
Intégration des techniques de télédétection pour l'évaluation des
ressources dans la recherche sur les systèmes pastoraux
Résumé
Cet exposé s'articule autour de l'hypothèse selon laquelle les enquêtes
et les recensements sur les ressources naturelles constituent un
préalable fondamental au développement planifié et devraient constituer
une composante intégrale des phases de description et de dianostic de
la recherche sur les systèmes d'élevage. Le document est conçu pour
donner un aperçu bref et sélectif sur les enquêtes sur les ressources
et les méthodologies qui semblent les plus appropriées aux régions
pastorales et agro-pastorales de l'Afrique.
Les types d'enquêtes précédemment utilisés ont notamment porté sur la
pédologie, les systèmes fonciers, les ressources naturelles et les
caractères des sols. Une enquête pédologique implique l'établissement
d'une carte pédologique qui indique la distribution des unités
pédologiques définies en premier lieu en fonction de leur morphologie
et de leurs caractéristiques physiques, chimiques et biologiques. Dans
l'Afrique anglophone, l'approche la plus commune se fonde sur
l'utilisation du concept de l'unité territoriale considérée comme
l'unité cartographique de base pour la subdivision et la classification
des terres.
Les méthodes de télédétection ont été classées en fonction de
l'altitude à laquelle les photos sont prises; elles incluent l'utili
sation d'images transmises par des satellites, de photos et
d'observations visuelles à partir d'un appareil volant à basse
altitude. Les images recueillies par satellite et les photographies
aériennes de type classique fournissent en général une couverture
globale alors que les enquêtes à basse altitude sont généralement
basées sur des bandes d'échantillonnage. La source essentielle des
images recueillies par satellite provient des séries Landsat qui
2
produisent des images monochromatiques ou en couleur de 34 000 km avec
une résolution de 80 m, prises à des intervalles de 18 jours. Le
Landsat permet ainsi de disposer d' images peu coûteuses aux différentes
saisons et sur des années successives.
- 77 -
Depuis les années 50, la photo-interprétation aérienne est devenue un
outil important pour les enquêtes pédologiques et autres. Les photos
en noir et blanc sont les plus communément utilisées mais d'autres
types de pellicules ont été essayés à des fins particulières, y
compris l'utilisation de photographies de proche infrarouge et de
fausse couleur. Parmi les autres techniques utilisées, on note celles
du radar à vision latérale, de la radiance spectrale et des enquêtes
à basse altitude. Dans la plupart des enquêtes sur les ressources, la
qualité de l'information s'améliore et les coûts diminuent quand on
passe de l'espace au sol. Il convient d'établir un équilibre entre les
méthodes extensives et intensives en vue d'atteindre les objectifs de
l'enquête entreprise. Il est proposé une approche intégrée en matière
d'évaluation qui diffère de l'approche utilisée dans les travaux
antérieurs en ce sens qu'elle est conçue pour permettre l'évaluation
rapide d'une zone-cible qui doit donner lieu à une description
exhaustive des ressources de base et des populations qui les utilisent
ainsi qu'un diagnostic relatif aux contraintes du système de
production de la zone. L'évaluation est effectuée en séries
chronologiques avec les étapes suivantes:
1 . Délimitation des unités de paysage et des zones de végéta
tion et définition des zones d'intérêt particulier sur la base
d'images transmises par satellite et d'autres données;.
2. Enquêtes à basse altitude systématiques et/ou stratifiées
sur l'ensemble de la zone à la même intensité ou à des intensités
différentes selon les strates. Après une analyse préliminaire des
données, et le levé de la distribution des parcours, de l'eau, du
bétail et des ressources humaines, la zone-cible est subdivisée en
unités territoriales homogènes qui fournissent la base d'enquêtes
aériennes et d'observations au sol plus approfondies.
3. L'évaluation des ressources des parcours se poursuit par les
photographies à basse altitude à échelles multiples le long de
transects et, si possible, des estimations de la biomasse sont effec
tuées sur la base de mesures radiométriques suivies par le prélèvement
au sol d'un échantillonnage de la biomasse en vue d'en déterminer la
qualité, la quantité et la composition par espèce. En même temps, des
- 78
équipes au sol sélectionnent des ménages représentatifs pour une
évaluation rapide de la démographie humaine et animale, des budgets,
des revenus et des dépenses.
L'approche intégrée peut être élargie pour inclure une composante
"système de prévision avancée". Pour être efficace, il est essentiel
que les estimations de la biomasse par la télédétection soient fiables
pour de vastes régions et il semble par conséquent indiqué d'accorder
un rang élevé de priorité à la mise au point de cette méthodologie.
- 79 -

Low altitude aerial surveys
in pastoral systems research
1 2
Kevin Milligan and Peter de Leeuw
Air Survey Coordinator, ILCA, Nigeria
2Ecologist, Arid Zones (Eastern and Southern Africa) Programme, ILQA,
Kenya
Introduction
The recent development of the science of low-altitude aerial survey
has provided a rapid, reliable and cost effective method for assessing
the numbers and distribution of livestock and people over large tracts
of land.
In the past, descriptions of pastoral systems had to rely
upon indirect methods, such as vaccination counts or tax returns, for
obtaining information about livestock numbers. Such methods have
well-publicised errors and biases. However, it is rarely pointed out
that these indirect methods are usually based upon counts taken only
once a year and thus seasonal distributions are unknown. Also, because
the animals are generally congregated into camps for these counts,
their actual distribution in relation to resources during their daily
grazing orbit is not known.
The methods of low altitude aerial survey were initially
developed by ecologists in East Africa wishing to map the distribution
of wild animals in national parks. Specif ic strategies suitable for
assessing livestock and people are still being perfected and ILCA has
taken a leading role in exploring the possibilities. An area of
particular interest has been the inclusion of recordings of rangeland
resources. This information, especially on vegetation and water
availability, can be compared to the observed livestock distribution
and thus an understanding of resource utilisation can be included in
the descriptive phase of pastoral systems research. Such comparisons
often allow identification of specific constraints to livestock
distribution and thus a diagnosis of possible interventions to improve
the system.
- 81 -
This paper describes the basic methods now being used for
low altitude aerial survey and illustrates the cypes of results that
can be obtained and how these results are being presented. The
possibilities of linkage and integration between such aerial surveys
and other methods of remote sensing are covered in another paper at
this workshop (de Leeuw and Milligan, 1983) , and thus this paper
will discuss how low level aerial survey can be used to diagnose
constraints within pastoral systems and the relative cost effective
ness of such operations.
Methodology: survey procedures and data collection
Details of ILCA's methodology are given in previous papers (Grimsdell
et al, 1979; 1979a; Okali and Milligan, 1980) and reports (Milligan
et al, 1979; 1982; Milligan, 1980; 1981, 1983). What makes the
general method different from other remote sensing techniques is that
the aircraft is flown at a very low altitude, so that a team of
observers on board can make direct visual counts of animals, people
and range resources that pass by the aircraft.
Flight and sample procedures
ILCA usually carries out its aerial surveys using a systematic,
unstratified flight pattern (see Fig. 1), so that every part of the
study area is covered evenly and uniformly and results can immediately
be presented as distribution maps. It also provides a data base for
post-survey stratification.
Spacing between the systematic parallel flight lines is
usually decided by a balance between time and cost constraints and
the required levels of sampling and precision of results. Flight
lines based upon the 10 km UTM grid projection provide a useful
system for global reference. However, sometimes a flight pattern
that exactly corresponds to existing maps will be more useful when
ground truth work is carried out or planned in the study area.
Each flight line is divided into intervals down its length.
These intervals, together with the parallel flight lines, allow the
entire study area to be divided into a grid pattern of fixed dimen-
82 -
sions. All information collected can thus be related to the individ-
udal grids and the resulting maps show simple and clear distribution
patterns .
|
AIRCRAFT FLIES PARALLEL LIMES OVER STUDY AREA FIXED BAND SAMPLED ON EITHER SIDE OF AIRCRAFT
DISTANCE INTERVALS DOWN LINE ENABLE GRID CONSTRUCTION
Fig 1. Flight pattern, grid system and sampling.
The area actually sampled is restricted to a fixed band on
either side of the aircraft (Fig. 2) . This band is determined by
the projection, from the observer's eye to the ground, of two parallel
rods attached to each wing strut. The choice of a suitable flying
altitude is important. The higher the altitude, the greater the
sample area observed and thus the greater the expected precision of
results; however, animals and people become increasingly difficult to
see and count at higher altitudes. The selected altitude is usually
between 400 and 1 000 feet above ground level . Depending upon survey
design, overall samnle cover is usually between 5 to 20%.
- 83 -
 The strip sampled (w) oil either side of the aircraft
is determined by the projection, from the observers'
eye to the ground, of the two rods (a and b) fixed on
the wing struts. Strip width can be varied by
altering these rods and is directly proportional to
the aircraft height above the ground (h) . Typical
settings for cattle surveys are h = 1000 feet,
w = 500 metres.
Fig 2. Schematic representation of aircraft during observation flight.
Information collection and analysis
In the aircraft, there are two back seat observers who count and
photograph the numbers of animals and dwellings seen in the sample
band in each grid. Individual observer bias, which varies according
to certain factors such as the effects of vegetation structure and
time of day, requires careful calibration and correction. A front
seat observer, seated beside the pilot, is responsible for navigation
and recording ecological conditions. Depending upon survey require
ments, ecological zones and seasons, these would normally include:
- vegetation physiognomy;
- tree density per hectare, dead trees and indicator tree
species ;
- grass cover, height and greenness and extent of burning;
- water sources and extent;
- farm areas, extent and major canopy crops.
84 -
Ecological recordings are supplemented by hand-held side-
looking photography and occasional vertical photography.
The primary objective of an animal and human survey is
usually to answer the fundamental questions:
How many ? - this can be worked out from simple statistics
upon the recordings down each individual sample
flight line;
Where ? - this can be readily seen from the grid
distribution maps;
Why ? - relationships between animals and people and
their environment can be tested by a series of
multivariate analyses. Although simple corre
lations can be confounded by partial relation
ships, step-wise regressions or more preferably
analyses of variance or factor analyses can
display all the variables in terms of their
individual relationships to each other.
The opportunities for such analyses can be envisaged from
Fig. 3, which represents some data sets that could be collected from
a typical survey. Assuming that one variable such as cattle herds, is
of particular interest, this can be separated from the rest of the
data file. The general distribution of cattle herds can be seen by
examining the grid entries on this data element. The relations of
cattle herds to any of the data files can be examined statistically
and, in the case of fig. 3, cultivation, grass cover and distance to
water are all likely to influence cattle herd distribution. The final
analysis, which is particularly important to the identification of
actual constraints to cattle herd distribution, is to give a priority
rating to each of the data files in terms of their' sequental ability
to explain variation in the cattle herd distribution.
The total number of data files for analysis is not restrict
ed to information collected during flight, but can include other
information from existing maps. In this way, the importance of
.parameters such as rainfall, geology or administrative boundaries could
also be assessed.
- 85 -
CATTLE HERDS 
lolololo|o|l|2|3|4|4|3|2|l|o|o|o|
HABITAT TYPES
SlM
OlOiO 5 5 5 5 2 2 2 2 2 4 4 4 1 I ID
n n ^ s 5 i^^^iU * A 1 1 IB
0 0 c) 0 2 2 2 2 2 2 2 0 0 0 0 OlII
0 0 (>>2 2 2 2 2 2 2 0 0 0 0 0 ol 1 " 1 7 ?
L ; 1
^0 0 1 2 2 2 2 2 2 3 0 0 0 0 0 Tj |(2.l
1 \ 1
1 '
0 2 ! 2 2 2 2 2 2 3 0 0 0 0 321
0 0 () 0 0 2 2 2 0 3 0 0 0 3 3 Tj 6 ■■
0 0 <)T>1 0 0 2 0 0 3 0 0 3 3 3 jH
0 0 () 1 0 0 0 0 0 3 3 3 3 3 3 5
4 [j
1 0 3 3 3 3 3 4 4 3 3 5 5
E 3 2 I 8
2 -
1 0 3 3 3 4'• 433555
0 0 3 3 4 4 1 3 jSSSS - 3 ?
1 U
0 0 3 4 4 4 13 3 5 5 5 5
: s r
1 M
1 0 3 4 4 4 4 3 3 5 5 5 5 1 TH
1 ( ) 0 3 4 4 4 3 3 3 5 5 5 1 (M
1 SM
0 0 () 0 4 4 4 3 3 1 3| j 5 5 1 1 TJ
l(6_l0 0 (> 0 3 3 3 3 3 5 5 1 1 1 1 Tj HTMIlo 0 () 0 3 3 3 3 5 5 1 1 1 1 1 Tj
0 0 ) 3 3 3,5 511 1 1 1 1 1 1 l|
0 0 5 3 3 5 5 5 1 i i i i i i Tj
DATA FILE
DATA ELEMENT
BY CODING ALL ORIGINAL INFORMATION AT A COMMON FORMAT AND SCALE,
ANALYSIS (SUPERIMPOSING DATA, IDENTIFYING PROXIMITY OF RESOURCES AND
MATHEMATICAL WEIGHTING) ARE POSSIBLE WITH RESULTS IN A SPATIAL CONTEXT'
Fig 3. Typical data elements used in the computer data handling
technique.
Results: resources inventory and land stratification
To illustrate some of the results from low-altitude aerial survey and
the way these results are being analysed and presented, examples are
taken from ILCA's work in Nigeria, Mali, Niger and Ethiopia.
- 86 -
Animal and human populations: abundance and distribution
An initial analysis of data provides overall population totals for the
study area, including herd or camp sizes, and densities. Table 1
2
shows results from the entire 81,555 km USAID/NRL project area in
Niger, during two seasons.
Table 1. Seasonal livestock populations (+% SE) in the NRL project
area, Niger.
Heads Herds
Animal
type
Total
(Nos)
Stocking
(Ha/Hd)
Total
(Nos)
Mean
Size
Dry Wet Dry Wet Dry Wet Dry Wet
Cattle
Bororo
Azawak
Sheep/
goats
Camels
Donkeys
288,653(12) 376,533(16) 28 22
152,814(15) 246,900(17) 54 33
135,839(12) 129,632(17) 60 63
8,706(10) 10,043(12) 33 38
4,025(13) 5,444(16) 38 45
4,681(10) 4,599(10) 29 28
780,289(8) 1,147,914(11) 10 7 16,964(8) 16,513(10) 46 70
70,162(9) 155,708(32) 116 52 12,917(9) 7,829(13) 5 20
13,540(14) 23,375(12) 588 345 1,890(14) 4,086(9) 7 11
Such population and herd sized totals give information about
likely production levels and economy in the project area. A comparison
between the seasons suggests changing herding strategies. For example,
the population of the Bororo cattle was 62% greater during the wet
season than during the dry season, while the Azawak cattle breed
numbers remained stable between seasons; this suggests that the
Bororo cattle are more mobile and may be part of a larger-scale
transhumant pattern. Detailed distribution analysis detected: areas
of seasonal herd splitting; areas of total herd immigration; areas of
cattle influx linked to immigration of primarily sheep and goat owners;
areas of influx associated with livestock ownership by settled culti
vators; and a substantial wet-season immigration of camels.
Similar tables for the total pastoral population of the NRL
project area identified five distinct pastoral groups: three Tuareg;
87 -
one WoDaaBe Fulani; and one Arab. An immediately interesting result
was that, while the initial project preparation document for the area
estimated 50,000 people, the actual number of pastoralists alone was
about 175,000.
Although information about the total number of animals or
people in an area is valuable because it indicates the size of the
target group and thus the possible total regional benefits that might
result from successful intervention, information about distribution
;is often more important, in that it indicates the regional impact of
animals upon their resources, and guides follow-up research and i
development activities into areas of concentration. For example, an
'examination of the livestock and pastoralist distribution in this NRL
project area showed that the Bororo and Azawak cattle breeds had very
different areas of concentration and these corresponded to the
distribution of WoDaaBe Fulani and tented Tuareg respectively. These
differences in distribution immediately indicated that detailed in
formation about either breed or tribe would only be relevant to
certain parts of the zone, while the opportunities for research or
development contact with the two tribal peoples would be different.
The different transhumance patterns between the two groups suggest
different flexibilities in management strategies which are likely to
affect their long-term development paths and choice of interventions.
ILCA is at present using four different methods to represent
the results of distribution data (Fig. 4). As the original data is
collected on a grid basis, each grid may be shaded to represent the
density it supports (choropleta symbolism) . Alternatively, the
individual grid information can be represented as squares or circles
(proportional point symbolism), whose sizes correspond to the observed
densities. Such representation of actual recordings, on a grid by
grid basis, can often be more easily interpreted if variations between
adjacent grids are mapped as contours which can also be represented as
a three surface view.
- 88 -
choropleth symbolism proportional point symbolism
 
90
50
25
10 D
• ■ ■ • ■ • ■
■■■■■• •••••■■M» » ■ ■■■ a ■ ■■ ■
90
80 ■
60 "
50 .
♦0
30*20 8
2-dimensional surface 3-dimensional surface
 
 
Dry-season grass distribution in a central region
of the pastoral zone of Niger - May 1981
Figures shown are percentage grass cover
Fig 4. Methods of representing aerial survey distribution data.
- 89 -
A common feature of pastoral systems is the changes in
seasonal distribution patterns of livestock and people. Although it is
almost impossible, by aerial survey, to link these movements to indi
vidual specific herds (without radio collaring or herd marking) the
patterns of herd concentrations can be easily understood from repeated
seasonal surveys. If these distribution patterns are investigated on
the ground, by social interview, and movement histories are estab
lished, regional mobility strategies can be mapped and quantified.
As stated in the introduction, most of the ground census
data relates to one period of the year. For example, ground census
figures for the Jos Plateau of Nigeria, based upon vaccination and tax
returns, suggested a population of about 300,000 cattle, while aerial
surveys (Milligan, 1980) showed a dry-season population of 140,000 and
a wet-season population of 400,000. Often ground studies have provided
general indications of movements; however, the value of an actual
census can be illustrated from work by ILCA in the Nigerian subhumid
zone which was generally considered to be an area of dry-season
grazing, with animals returning north early in the wet season when
surface water becomes again available and the southern tsetse belts
expand. Table 2 shows that two of ILCA's four case-study areas had
high wet-season concentrations: an exactly converse pattern to that
expected.
Table 2. Cattle densities in four case-study areas in the Nigerian
subhumid zone.
Area Density Mean
(km2) (no. km2)
dry
size
wet dry wet
Kurmin Biri 2,500 17.3 4.2 51 54
Abet 2,475 37.4 22.7 48 61
Mariga 2,750 6.6 23.5 25 53
Laf ia 3,500 12.7 37.7 59 86
- 90 -
The Niger delta region in Mali is another area of marked
seasonal movements of livestock. This seasonally flooded region
within the Sahel supports, during the period of greatest animal
concentration, about 1,200,000 cattle and an additional 500,000 sheep
and goats. Maps of the seasonal distributions of cattle are illus
trated in Fig. 5.
October 1980
>■■■--
550
165
125
70
50
40
30
20
15
10
0
Cattle density per square kilometre
March 1981 June 1982
■ • ■
■ • • ■ ijml
■ ■ ■ i
a ■
a ■ ■ ■ ■
■ ■■■■• • ■
Fig 5. Seasonal cattle distribution in the Niger delta region of Mali.
91
These maps show, during October, high cattle densities in
the transition zone surrounding the delta and low populations where
water is lacking and also in the central basin which is inaccessible
due to flooding or because it is protected by river barriers. The
March pattern showed a concentration of cattle in the floodplain which
supported 72% of the total population. Two distinct sub-populations
were detected; one south of the River Niger in the Djenne region; the
other north, near Lake Debo. During June, cattle were leaving the
delta, and distinct movements westwards and eastwards from Lake Debo
were recorded.
Environmental conditions
All environmental variables .ecorded during a survey can individually
be illustrated as distribution maps based upon the aerial survey grid
pattern. Alternatively, average or total figures can be computed for
particular regions.
An example of these resource maps for the Mali delta region
shows that only about half of the survey areas can be termed 'upland
Sahel with little or no flooding1 (Fig. 6). Farming is of high density
in the Mema Jura to the east of Mopti and in the Sanari plains. The
rest of the area consists mostly of flooded plains and backswamps
(34%) with considerable land devoted to floating rice cultivation.
The highest levels of rice cultivation were recorded in the Macina
plains and those south of Mopti; these plains are broken by higher
level terraces, levees and point-bar systems, which are much less
inundated and show a mixture of rainfed and rice farming.
- 92 -
Level of flooding
• ■■■■■■ar
■ ■■■■■■■ ■!
 
•■■!
• ■ ■ ■ ■ ■
3s
■■•■
■■■■
■<
■■■■■
• ■ ■■
a. v h
• ■■ ■
100
5
i
S
S'
o
Rice cultivation Cereal cultivation
1 ■ ■
■ •
■ ■ ■ ■ ■ ■ •
■ ■
I
■ ■ ■ ■
m *
■■
» ■
m m ■ ■ ■ a
■
■ ■ urn, m ■ ■ ■
■ ■■■■■■ ■ ■a
■a■ ■ ■ ■ » ■■ ■ ■ ■ ■ ■ ■
■■■■■■■■ ■■■ ■■■■■■
• .1 ■ ■
■■■■■■■■■■ 100 ■ ■ ■ ■ a
■ ■
■
■■■ ■■■ ■■■■ ■ ■ ■ ■a ■ ■
■■
■■■ >
■■
■■■■■■■■a '•'a ■ ■■■
■ ■■ ■■■ ■■■ ■■ ■■■■
■■ ■■ ■■
■■■■■■■ ■■■
• ■■■»■
■■■ ■ ■ ■■■■■•
■ ■■■ ■ ■
■■■■ ■ MSB- ■
■■ ■ ■ ■ ■ ■ ■
■■■ ■
■ ■ ■
am ■ ■ ■■BMla
20.
,05-
■ ■ ■ ■ ■ ■
m ■ Baa*
)■■■■■■ ,■ m ■
!_ r- 0*
Figures shown are percentages
Fig 6. Levels of flooding, vice and ceveal cultivation in the Niger
delta region of Mali.
- 93 -
As for livestock, seasonal changes in environmental con
ditions can also be illustrated by a series of seasonal maps and
graphs. An example of a graphic application for the drier areas
surrounding the Mali delta is given in Fig. 7, for grass cover during
March and June. Such frequency distributions of grass cover levels
are often related to the levels of grazing: in general, curves with a
rightward skew indicate understocking, while a leftward skew indicates
overstocking, except where there are other causes for dry matter
disappearance .
50i
40
Co 30
o
o
20-
10-
o
i
March
June
 
"20 30 40
Grass cover (%)
50 60 70
Fig 7. Frequency distribution of seasonal grass cover levels in the
upland Sahel region bordering the Mali delta.
- 94 -
The actual environmental variables collected during a survey
will depend upon the objectives of the survey and the capabilities of
the survey team. Questions about specific environmental conditions
can often be answered; such as possible changes in degradation which
can be inferred from soil and erosion patterns, from grass cover, or
from the distribution of dead or indicator plant species. In Niger,
the distribution of a small shrub, Calotropis procera, was recorded as
well as the distribution of dead trees (Fig. 8); the former was in
response to USAID requests based upon observations that this species
had increased substantially in their area during the past few years
and the suggestion that it may indicate deteriorating conditions.
■ ■ ■ •
■ a • » ■ • ■
■ ■ ■ ■ *
• • • ■ ■ ■ ■•■■■■ ■■■ •
■•■■■■a•••
•a■ •■■■■■ •• • • • • ■
>■•■■•• ■■■■•■
■ ■ > • . ■ . . >,
■ • ■ ■ ■ ■ ■
t > • * ■ • • • . •
• ■ • • • ...
J
6
I
5
4
2
2
1
Dead trees Calotropis procera
Fig 8. Distribution of Calotropis prooera and dead trees in the NRL
project area in Niger.
Interrelations between animals and environment
The interrelations between animals or people and the various measured
environmental parameters can provide information that will directly
suggest appropriate management interventions into the pastoral system.
For example, if animals are usually concentrated into cultivated
- 95 -
areas, opportunities for better interactions with farmers may exist;
whereas, if cattle are usually away from cultivated areas and are
[concentrated in the natural rangelands and savannas, grazing reserves
jor improvement of range management practices may be feasible. Simi
larly, direct linkages to specific variables can indicate their
relative importance and thus whether they are likely to be a constraint
to the system. For example, if animals mostly graze near water sources
can be investigated.
While any set of two variables can be examined, for example
grass cover to distance from water, it is often more valuable to study
the system as a whole, considering the combined and cumulative effects
of many variables upon another independent variable. The objective
may be to identify, for example, which out of all the variables
recorded most closely corresponds to the distribution of cattle, on
ithe assumption that such a variable may thus be a limiting or con
trolling factor to cattle distribution.
This particular stage in the analysis is perhaps the most
valuable to pastoral systems research, as it moves away from the basic
descriptive phase of the system into the diagnostic phase, hypoth
esising specific constraints that can be validated and tested by
ground research teams. It is important to stress that results from
such analysis are necessarily "predictive" rather than "causal". For
example, results could indicate that animals are usually concentrated
near water, i.e. that a knowledge of water distribution allows one to
predict animal distribution. Water may, in fact, not be the cause of
the observed animal distribution, which could be due to factors not
even recorded during the survey, such as tribal or border conflicts,
or that a particularly desirable plant occurred near these water
sources. Clearly, a ground research team, equipped with the predic
tive hypotheses generated from an aerial survey, can quickly clarify
their validity.
In the USAID/NRL study area, the overall cattle population
distribution during May could be best predicted from a knowledge of
cultivated areas, grass cover and distance to various water sources.
During October, cultivation and grass cover continued to be most
important but predictions could also be made from the general veg-
- 96 -
etation structure and tree density maps. However, the two cattle
breeds were conspicuously different. Bororo distribution could nearly
always be predicted from grass cover, further refined by vegetation
and tree density distribution. Azawak were most concentrated in
cultivated areas and near water sources. Thus there was a clear
division between the two breeds in terms of their relationship to
natural vegetation versus man-induced conditions. Such observations
indicated that the basic options between rangeland development or
water development may have a different relative value for the two
breeds and the pastoral group associated with them.
In the southern Ethiopian rangelands, 19 variables, including
cultivation, grass cover, vegetation physiognomy, growing period
gradients and distance to three water source types, were included in
an analysis of livestock distribution. The distance to ponds was
nearly always the best predictor of livestock distribution. From such
a knowledge, the relationship between cattle and ponds was more closely
examined. Figs 9 and 10 show two ways that this information was
presented.
CKl
 
yells
Distance to water (km)
Fig 9. Cattle distribution in relation to water sources in the SORDU
project area of southern Ethiopia.
- 97 -
Cattle density
per square kilometre
205.
 
V■■ ■■
■ *■■ ■ /*f ■ ■■■■»
■■■ 1 ) "iP -■
Contours show distance (km) to ponds
25 kilometres
XfmV " " " " '■"i
*■•■■■■ ■ -■■■
 
y
Fig 10. Number of cattle and the distance to ponds in the SORDU
project area, southern Ethiopia.
Land stratification
Over large tracts of land, ecological conditions usually vary con
siderably and consequently so do opportunities for intervention. Thus
an effective diagnosis of the principal constraints within a pastoral
system will often require an initial stratification of land into major
vegetation or ecological units; followed by an assessment of probable
constraints within each strata; followed by an evaluation of whether
particular strata are likely to respond to the proposed interventions.
In Niger, geomorphology and landscape were used to divide
the NRL project area into six landtypes. Further subdivision into
- 98 -
management units depended on the supposition that, at the present
state of pastoral development, practical interventions require units
that are geographically distinct, with suitable operational bases.
The project area in Niger was finally divided into four major strata
with 10 units. Specific research and development possibilities could
be identified for each unit. For example, one of the units, the
Ighazer plains, was essentially a clay soil floodplain while another
was a sand dune grassland. While the development of surface water in
pools and ponds was likely to persist in the former, ground seepage
in the latter would necessitate construction of deep boreholes.
However, controlled animal stocking was likely to be of little value
in the floodplains where there was a low grass cover of scattered
perennial tufts, compared to the dunes which had a good cover of
annual grasses.
A similar approach to land stratification was adopted for
the Mali delta region, with breakdown into four strata subdivided
into 19 units, for each of which aspects of the animal and human
populations as well as the environmental conditions could be described
(Table 3).
Table 3. Cattle and selected environmental conditions in four land
management strata in the Mali delta region.
Land Area
(km2)
Cattl e population Environment
management Oct. March June Flood Rice Cereal
strata
Cooo) ('000) Cooo) (%) (%) (%)
Upland Sahel 13,197 226 63 116 0 0 19
Transition zone 5,811 237 91 102 10 7 14
Elevated plains 5,818 230 189 181 21 12 9
Inundated plains 11,117 114 871 410 72 24 3
- 99 -
Discussion
The general iterative sequence of pastoral systems research can be
seen to have four essential phases:
| > DIAGNOSTIC -i
DESCRIPTIVE TRIALS
L PROMOTE < 1
Low-altitude aerial survey is mainly valuable at the
descriptive and diagnostic phases. The descriptive phase provides an
inventory of the resources, together with an evaluation of their
structure and functioning within the production system. This descrip
tive phase needs to be open euded, as comprehensive as possible and,
most importantly, should be both rapid and cost-effective.
Thereafter, the diagnostic phase must first identify the
main constraints, devise ways of tackling them and then evaluate the
probability of the success of any proposed interventions. The remote
sensing nature of low level aerial surveys means that the constraints
to the system can only be proposed or hypothesised, and ground vali
dation is essential. Results of the work provide predictive state
ments, for example that animals tend to be found near wells, but the
actual reasons and mechanism for this may be more complicated: animals
may be near water for reasons other than water, while developing
alternative water sources may be either technically or socially
unfeasible.
Fig. 11 indicates how the essential flows from description
towards diagnosis can operate. The survey immediately answers the
resource questions: How many? Where? and When? while the
interrelationships provide the hypotheses: Why? These hypotheses
about possible causation and limiting factors to the system are posed
within the multi-variate animal, human, environment context.
- 100 -
z
o HOW MANY WHERE WHEN
E
5
u
ANIMALS ANIMALS ANIMALS
PEOPLE PEOPLE PEOPLE
LU
ENVIRONMENT ENVIRONMENT ENVIRONMENT
o
^V i
INTERRELATIONSHIPS
CONSTRAINTS
I
vt ENVIRONMENT ANIMAL PEOPLE
O GRASS COVER MOVEMENTS
HERO SIZES
STOCKING RATES
SPECIES RATIOS
ANIMAL x RANGE
MOVEMENTS
DISPERSION
MARKETS
ETHNIC RATIOS
PEOPLE x ANIMALS
z WATER
<
EROSION
5
FIRE
RANGE x CULT
Fig 11. Low altitude aerial survey.
Although low-altitude aerial surveys record a large number
of parameters relating to the animal and human populations as well as
their environment, numerous parameters cannot be recorded. For
example, while the animal population size and distribution are
recorded, age and sex structures are not. Similarly, baseline infor
mation about human demography, household and labour budgets and
decision making processes are not recorded, neither are environmental
factors, such as geology and rainfall, though these often exist in
published literature. While detailed range resource maps have been
produced from intensive photo-interpretation and ground truthing, the
general vegetation descriptions made during flight will often be
sufficient to indicate whether range resources are an important
constraint to animal distribution and thus whether further, more
detailed, research is in fact necessary.
- 101 -
A particularly useful advantage of low aerial survey is
its ability to guide aspects of the sampling design of ground inves
tigations (Okali and Mill igarij 1980) . If detailed animal production
studies require a stratified sample design based upon herd size, this
can be directed from the results of the aerial surveys. Subsequently,
aerial and ground data can be linked together and detailed results
from the ground can be extrapolated to the entire study area. Simi
larly, if household studies are to be based upon wealth strata and if
the wealth characterization can be linked to factors visible from the
air such as camp size, number of granneries, or cattle holdings, the
basic sample frame and its locations can be determined before the
ground work begins.
Outside its advantage as a multiple resource inventory
technique, low altitude aerial surveys are both rapid and cheap.
Although survey time will depend upon sample procedures and various
2
logistic considerations, an area of about 100,000 km could be covered
in three to four weeks and preliminary reports from the survey could
be made available to ground teams one month later. Costs are variable.
The actual flying operations themselves are usually substantially less
expensive than man-time costs, and thus analysis and reporting time
become important considerations. Normal costs, from planning to
2
reporting, are about USj5 1.00 per km , although costs would be higher
for particularly intensive investigations. This amount usually
represents as little as 1% of the total research and development
costs within a pastoral project.
References
De Leeuw, P. and Milligan, K. 1983. The integration of remote sensing
techniques for resource evaluation in pastoral systems
research. Invited paper, Workshop on Pastoral Systems
Research in Sub-Saharan Africa, ILCA, Addis Ababa.
Grimsdell, J.J.R., Bille, J.C. and Milligan, K. 1979. Alternative
methods for aerial livestock census. Invited paper,
International Workshop on Aerial Survey Methods, Nairobi.
- 102 -
Grimsdell, J.J.R., Sihm, P. and Milligan, K. 1979a. The role of
aerial surveys in livestock development projects. Invited
paper, International Workshop on Aerial Survey Methods,
Nairobi.
Milligan, K. 1980. Abundance and distribution of cattle on the Jos
Plateau, Nigeria. Report to the Federal Livestock
Department, Lagos.
I
Milligan, K. 1981. Aerial survey of human, livestock and environmental
conditions in a central region of the pastoral zone of
Niger. Report to USAID, Niamey.
Milligan, K. 1983. An aerial reconnaissance of livestock and human
populations in relation to land use and ecological con
ditions in the SORDU project area of southern Ethiopia.
Report to RDP, Addis Ababa.
i
Milligan, K., Bourne, D. and Chachu, R. 1979. Dry and wet season
patterns of cattle and land use in four regions of Nigerian
subhumid zone. Report to ILCA, Kaduna.
Milligan, K., Keita, M. and De Leeuw, P. 1982. Recensement aerien
saisonnier du cheptel et types de paysage du delta central
I du Niger au Mali. Report to ODEM/IBRD, Severe.
Okali, C. and Milligan, K. 1980. Ground/air linkage and the role of
socio-economic survey in interdisciplinary research for
intervention testing. Invited paper, Workshop on the Role
of Anthropologists and other Social Scientists in Inter
disciplinary Teams Developing Improved Food Production
Technology, IRRI/UNEP, Manilla.
- 103 -
Les enquêtes à basse altitude dans la recherche sur les systèmes
pastoraux
Résumé
La technique des enquêtes à basse altitude qui connaît depuis peu un
certain essor s'est avérée une méthode rapide, fiable et peu coûteuse
pour évaluer le nombre et la distribution des populations animales et
humaines sur de vastes superficies. Le document décrit les méthodes
de base utilisées actuellement pour effectuer les enquêtes à basse
altitude, y compris les procédures de vol et d'échantillonnage et la
collecte et l'analyse de l'information.
Des exemples tirés des travaux du CIPEA au Nigeria, au Mali, au Niger
et en Ethiopie sont utilisés pour illustrer certains des résultats qui
peuvent être obtenus avec les enquêtes à basse altitude et la manière
dont ceux-ci peuvent être analysés et présentés. Ces exemples mettent
l'accent sur l'importance et la distribution des populations animales
et humaines, sur les conditions mésologiques, sur les interactions
entre les animaux et l'environnement et enfin sur la stratification
territoriale.
Les enquêtes à basse altitude sont surtout importantes dans les phases
de la description et du diagnostic de la recherche sur les systèmes
pastoraux. Le recours que font les enquêtes à basse altitude à la
télédétection signifie que les contraintes aux systèmes identifiés par
le biais de telles enquêtes ne constituent que des hypothèses et que
la confirmation par l'observation au sol est indispensable.
L'un des avantages les plus importants des enquêtes à basse altitude
réside dans le fait qu'elles permettent d'orienter certains aspects du
plan d'échantillonnage des enquêtes au sol. Si les études détaillées
sur la production animale font appel à un plan d'échantillonnage
stratifié basé sur la taille des troupeaux, il est possible d'obtenir
un tel plan grâce aux résultats produits par les enquêtes aériennes.
Par la suite, il est possible d'établir la corrélation entre les
données issues des enquêtes aériennes et des observations au sol; ainsi
des informations détaillées recueillies au sol peuvent être extrapolées
- 104 -
pour l'ensemble de la zone d'étude. De même, si les études sur les
ménages doivent être basées sur la stratification de la richesse et si
la détermination de la richesse peut être liée à des facteurs tels que
la taille des campements, le nombre de greniers ou de
bovins visibles à partir d'un appareil, le cadre d'échantillonnage de
base et ses éléments peuvent être déterminés avant le début des travaux
au sol.
Les enquêtes à basse altitude représentent une technique multiple; en
outre, elles sont à la fois rapides et peu coûteuses. Une superficie
2
d'environ 100 000 km peut être couverte en trois à quatre semaines
et les rapports préliminaires relatifs à l'enquête pourraient être
mis à la disposition des équipes au sol un mois plus tard.
Normalement, les coûts (de la phase de la planification à celle de
2
l'élaboration du rapport) sont d'environ 1 dollar E.-U. par km ,
quoique pour les enquêtes particulièrement intensives, ce chiffre
puisse être dépassé. Les sommes consacrées aux enquêtes à basse
altitude représentent en général 1% seulement des coûts totaux de
recherche - développement des projets pastoraux.
- 105

Summary of Discussion Session 2.
Chairman: Dr Cees de Haan (ILCA)
Discussion led by Dr Noumou Diakite (Mali)
Dr Rhissa suggested that finance be sought for aerial surveys of all
the countries in the Sahel zone from international organisations such
as OAU, World Bank, UNDP. Dr Abel warned that in using remote
sensing there was one danger in particular that resulted from the
detachment of the observer from the object of the survey. Objects that
were visible on photographs and images may not be relevant to the
local users of the land. Land classification from remote sensing may,
because it ignored local use of particular resources, result in
development plans which adversely affected local land users. There
was also an unnecessary waste of existing information in the form of
local knowledge of the land and its use which had been tried and
tested over a long period. It was not, in Dr Abel's view, sufficient
to use the social scientists of an inter-disciplinary team to
counteract the inadequacies and dangers of remote sensing. He suggested
that if local land classification systems were described and used from
the outset, some of the dangers of inappropriate classification would
be removed. If different ethnic groups practised a variety of land
uses on the same land, each of their classification systems should be
incorporated into an integrated scheme, otherwise certain groups may
be placed at a disadvantage in the competition for land which commonly
accompanied planned changes in land use.
In commenting on the paper by Drs Milligan and de Leeuw, Prof. Saka
Nuru said that the remote sensing technique would be a useful
preliminary aid in the conceptual framework of LSR in terms of its
time-saving and cost benefit effect before more serious work was done
by scientists at a particular location. But he said that it was only
useful when used as a complement to ground surveys which were probably
more reliable in the identification of parameters of interest. There
were certain constraints to the technique, such as the difficulty in
identifying goats due to their small colour and size. Dr Milligan
pointed out that sheep and goats were not normally a problem because
of an aspect of their behaviour - they usually ran when an aircraft
- 107 -
approached and thus could readily be distinguished from cattle. It
was difficult to separate sheep from goats, and thus they were
normally classified as 'shoats'. Prof. Saka Nuru asked what was the
relative efficiency/sensitivity of the remote sensing technique vis
a-vis the ground survey. Dr Milligan replied that counting animals
from the ground was time-consuming and there were numerous practical
and theoretical problems. Dr Milligan agreed that ground truthing
was important, depending on the level of information that was
required.
Items estimated from the air could be checked and some biases
corrected from ground survey calibration. Also a knowledge of the
ground conditions helped onr to make correct observations from the
aircraft. Prof. Saka Nuru strongly supported the idea that ILCA
should be involved in a country-wide use of the remote sensing
technique in order to get a better picture of the distribution of
natural resources as an aid to future planning for livestock
development.
Dr Zulberti emphasised the need for a conceptual framework before any
information was collected to avoid gathering a large amount of data,
some of which might not then be used. Dr Chema also stressed the need
for a specific purpose to be clearly defined at the beginning of the
survey. Dr Milligan agreed, and stated that specific objectives were
indeed identified before each flight. The hypotheses and questions
that were stated at the start formed the basis for the kind of data
that was collected and the design of the sampling strategies.
Dr Hiernaux asked Dr de Leeuw whether or not he thought that the use
of satellite remote sensing methods and aerial surveys could result
in increased costs due to the many field observations that had to be
made to support such methods. Dr de Leeuw said that this highlighted
the need for researchers to be more clear about what data they really
needed from such surveys .
- 108
Résumé des débats de la deuxième aéance
Président: M. Cees de Haan (CIPEA)
Débats dirigés par le Dr Noumou Diakité (Mali)
Le Dr Rhissa a suggéré de rechercher le financement d'enquêtes
aériennes pour tous les pays de la zone sahélienne auprès
d'organisations internationales telles que l'OUA, la Banque mondiale
et le PNUD. Le Dr Abel a déclaré que l'utilisation de la télédétec
tion comportait un aléa particulier qui résultait de l'éloignement
de l'observateur de l'objet de l'enquête. Les objets qui étaient
visibles sur les photos et sur les images peuvent ne rien signifier
pour les utilisateurs locaux de la terre. La classification des
ressources territoriales par télédétection pourrait se traduire par
des plans de développement néfastes pour les utilisateurs locaux,
notamment parce qu'elle ne tient pas compte des utilisations locales
de ressources particulières. En outre, on ne tirait pas parti
d'informations disponibles sous forme de connaissance locale de la
terre et de ses utilisations et éprouvée par le temps. Aux yeux
du Dr Abel, il n'était pas suffisant d'utiliser les spécialistes
de sciences sociales d'une équipe interdisciplinaire pour pallier les
insuffisances et les dangers de la télédétection. Il a déclaré que
si les systèmes locaux de classification des terres étaient décrits
et utilisés dès le départ, certains des risques de classification
inadéquate pourraient être écartés. Si divers groupes ethniques ont
adopté diverses formes d'utilisation des terres sur la même superficie,
chacun de leurs systèmes de classification devrait être incorporé
dans un schéma intégré, autrement, certains groupes pourraient être
placés dans une situation désavantageuse dans la compétition pour les
terres qui accompagne en général les changements planifiés de l'uti
lisation des terres.
Dans son commentaire sur le document de MM. Milligan et de Leeuw,
le Prof. Saka Nuru a déclare que la technique de la télédétection
constituait un outil préliminaire dans la conception du cadre de la
recherche sur l'élevage en raison de l'économie de temps qu'elle
permet et de son effet coûts/bénéfices avant que des travaux plus
approfondis ne soient entrepris par des scientifiques sur un site
déterminé. Mais il a déclaré qu'elle n'était utile que lorsqu'elle
- 109 -
était utilisée comme complément d'enquêtes au sol qui étaient
probablement plus fiables dans l' identification des paramètres
importants. Cette technique comportait certaines contraintes telles
que la difficulté d'identifier les chèvres en raison de leur couleur
et de leur petite taille. M. Milligan a souligné que les moutons
et les chèvres ne constituaient pas normalement de problèmes à cause
d'un aspect de leur comportement: il s'enfuient généralement lorsqu'un
avion approche et peuvent ainsi être facilement distingués des bovins.
Il est difficile de distinguer les moutons des chèvres et ils ont
donc été normalement classés en "cavins". Le Prof. Saka Nuru a
demandé quelle était la fiabilité/efficacité relative de la technique
de télédétection par rapport aux enquêtes au sol. M. Milligan a
répondu que le dénombrement des animaux à partir du sol prenait
beaucoup de temps et qu'il posait de nombreux problèmes théoriques
et pratiques. M. Milligan a reconnu que la confirmation au sol était
importante, compte tenu du niveau d'informations requis. Les estima
tions faites à partir de l'appareil peuvent être vérifiées et
certaines distorsions corrigées par calibrage basé sur les enquêtes
au sol. La connaissance de la situation au sol a également permis
de faire des observations correctes à partir de l'appareil. Le
Prof. Saka Nuru s'est déclaré entièrement en faveur de l'idée selon
laquelle le CIPEA devrait participer à l'utilisation à l'échelle des
pays des techniques de télédétection pour avoir une image plus claire
de la distribution des ressources naturelles en vue d'une assistance
à la planification future du développement de l'élevage.
Le Dr Zulberti a mis l'accent sur la nécessité d'un cadre conceptuel
avant la collecte des données, pour éviter le rassemblement d'infor
mations abondantes dont certaines pourraient ne pas être utilisées.
Le Dr Chema a également souligné la nécessité de définir clairement
un objectif précis au commencement de l'enquête. M. Milligan a
accepté ce point de vue et a déclaré que des objectifs spécifiques
étaient en fait identifiés avant chaque vol. Les hypothèses et les
questions formulées au départ constituaient la base à partir de
laquelle les divers types de données étaient collectées et la concep
tion des stratégies d'échantillonnage mise au point.
- 110 -
M. Hiernaux a demandé à M. de Leeuw si oui ou non il estimait que
l'utilisation des méthodes de télédétection par satellite et par
enquête aérienne pourrait résulter en un accroissement des coûts en
raison de la multiplicité des observations qui devaient être faites
sur le terrain pour appuyer de telles méthodes. M. de Leeuw a
déclaré que cela soulignait la nécessité pour les chercheurs de
définir de manière très claire les types de données dont ils ont
besoin dans de telles enquêtes.
- 111 -

Les végétations et les resources fourragères
dans les systèmes pastoraux
Pierre Hiernaux
Ecologiste, zones arides (Afrique de l'Ouest), CTPEA, Mali
Le but des études menées sur la végétation et les ressources
fourragères dans les systèmes pastoraux est de détecter et d'analyser
les contraintes nutritionnelles pour l'élevage qui sont liées à une
insuffisance de ces ressources, et de proposer des solutions pour y
remédier.
Comme pour d'autres aspects du système, l'étude est
structurée en phases descriptives et de diagnostic, éventuellement
suivies de tests au niveau expérimental puis progressivement mis en
place dans les systèmes.
Une particularité des études sur la végétation tient à ce que
celle-ci est à la fois une des ressources alimentaires principales de
l'élevage qu'il faut quantifier et dont il faut analyser les processus de
production et de reproduction, mais c'est aussi un paramètre majeur de
l'environnement, et qui reflète, par sa structure et son fonctionnement,
les autres composantes de l'environnement. A travers sa structure
physique, floristique, ses variations saisonnières ou interannuelles, la
végétation est un indicateur synthétique des conditions de milieu qui
peut être utilisé pour identifier et caractériser l'environnement.
M. Bille traitant des méthodes utilisées pour l'étude de la
dynamique de la végétation, ma présentation se limite aux méthodes
d'inventaire et d'analyse des processus de production végétale en les
illustrant par des exemples extraits des travaux du CIPEA au Mali.
La phase descriptive
Buts: - Stratifier l'environnement dans lequel évoluent le ou les
systèmes pastoraux étudiés.
- 113 -
- Caractériser la production fourragère de chaque strate
et les variables écologiques qui la conditionnent.
Méthodes: Les relevés et les cartes phyto-écologiques
Les relevés
Un inventaire des ressources fourragères est réalisé sur le terrain
par relevés méthodologiques de la végétation et du milieu. Un relevé
consiste à noter systématiquement pour un site-échantillon les
valeurs prises par une série de paramètres de la structure de la
végétation et de l'environnement. Pour la végétation, ce sont des
paramètres de la structure des peuplements herbacés et ligneux
(recouvrement, densité, stratification...), la composition floristique,
la biomasse aérienne, etc. . Pour l'environnement ce sont les caracté
ristiques climatiques et édaphiques du site: nature de la roche-mère,
unité géo-morphologique, descripteurs du profil de sol... mais aussi
les indications sur l'occupation du sol -statut agricole et pastoral.
Pour faciliter les comparaisons et autoriser les manipulations
statistiques, la superficie des sites-échantillons est fixe et
l'emplacement est matériellement délimité sur le terrain. Dans les
2
etudes faites au Mali, la superficie des relevés est fixée à 100 m
(carré de 10 m de coté) pour le tapis herbacé et 2,56 ha (carré de
160 m de côté) pour le peuplement ligneux. Dans les deux cas, l'étude
des distributions aire-espèce a guidé le choix de la superficie qui
est proche de l'aire optimale phytosociologique .
La durée d'un relevé est comprise entre 2 et 6 heures pour un
relevé du peuplement ligneux et 1/2 à 1 heure pour un relevé phyto-
écologique (1 observateur).
L'échantillonnage des relevés
De l'emplacement et du nombre de stations écologiques choisies
dépendront la précision des résultats et leur représentativité. Un
plan d'échantillonnage fixe la liste dis situations à observer et le
nombre des répétitions à effectuer dans chacune d'elles. Le nombre des
situations est déduit d'une stratification hiérarchisée de l'espace
- 1 14 -
régional étudié en régions, secteurs, séries ou séquences et enfin en
stations écologiques. Il est égal au produit du nombre des stations
par celui des séries multiplié par le nombre de secteurs de chaque
région. Le nombre des répétitions est généralement compris entre 3
et 10. Le tableau 1 présente les paramétres de l'échantillonnage
réalisé au Mali. Au total, cinq cent relevés de reconnaissance (et
2
100 relevés ligneux) ont été effectués sur une superficie de 71 000 km ,
2
soit en moyenne 1 relevé pour 142 km (un carré de 12 km de côté) et
une moyenne de 3 répétitions par groupement végétal.
Tableau 1. Paramètres de Véchantillonnage stratifié des relevés
phyto-écologiques au Mali
Régions Glacis Delta Delta
Niveaux
de
perception
Ecologiques continental "mort" "vif"
Total
zone
d'étude
secteurs 4 (climat) 3 (climat) 9 (inon- 16
dation)
séries 8,5 (sol) 8 (sol) 6 (sol) 7,7
(moyenne /secteur)
stations l,78(artif.) 2,25(Artif .) 1 (Artif.) 1,7
(moyenne /série)
Nombre de strates • 60
Nombre de répétitions x 3
Nombre de relevés 180
Nombre de relevés
effectivement réalisés 167
Taux d'échantillonnage
superficie km /relevé 168
54 54 172
x 3 x 3
162 162 504
164 169 500
155 103 142
115 -
Le tableau d'échantillonnage fixe l'effectif des observations
et les localise grossièrement. Pour préciser leurs emplacements, les
documents cartographiques et les photographies aériennes sont très
utiles. Si l'échelle des photographies aériennes et leur actualité
sont appropriées, cette localisation peut être directement utilisable
sur le terrain au niveau duquel il ne restera plus qu'à vérifier la
correspondance avec les critères d'échantillonnage retenus et, pour
le détail de l'emplacement, à veiller à la représentativité de
l'élément choisi vis-à-vis de l'ensemble de la station écologique. La
photographie aérienne s'avère un instrument très précieux dans la
phase technique de l'échantillonnage, tant pour le contrôle de la
représentativité des sites que pour résoudre au mieux les problèmes
d'accès. La durée d'une campagne ile relevés telle que celle qui a
2
été faite au Mali est de 22 jours de terrain pour 10.000 km pour un
observateur (en moyenne 4 relevés phyto-écologiques/jour et 1 relevé
ligneux/jour) . A condition toutefois de choisir une saison propice,
surtout pour les herbacées (septembre à janvier au Sahel) .
L'analyse des données des relevés phyto-écologiques
Les données recueillies dans les relevés peuvent être présentées sous
la forme d'une matrice où les objets sont les relevés et les descrip
teurs à la fois les taxons (composition floristique qualitative ou
quantitative) et les états des variables écologiques observées ou
mesurées (Fig. 1). De cette matrice de base d'autres matrices sont
dérivées dont les objectifs sont les états des variables du milieu et
les descripteurs sont soit les espèces, soit les relevés groupés par
formation végétale. Les dimensions très élevées de ces matrices rendent
indispensable le recours à un ordinateur performant. Ce traitement
mécanographique des données réclame un codage des données et leur
transcription sur cartes perforées ou bandes magnétiques.
Après une série de tests pour contrôler l'échantillonnage
(mesure de l'interdépendance des variables sur la base des calculs de
probabilité, ou de l'information mutuelle sur les tables de contingence),
les groupements végétaux sont définis par une comparaison des compositions
floristiques des relevés. Ces groupements sont alors caractérisés par
les espèces et les états des variables écologiques auxquels ils sont
significativement lies.
- 116 -
MATRICE DE BASE
DES DONNEES PHYTO-ECOLOGIQUES
OBJETS:
R1
RELEVE \
R
%
D
E
S
C
R
I
P
T
E
0
R
S
El
11 * X1N
Especes
vegetales E
, X ....
- - - - E
P
pi •••• * XpN
Variables 1 Ml
• ylN
de
l'environnement L
1
\ yq r '-qN
\
x: presence-absence, abondance-dominance,
recouvrements, effectifs, biomasse....
de l'espece E dans le releve R.
y: etat de la variable L dans le releve R.
t>
Mesures de distance:
Groupements phyto-socio-
logiques
Mesures d' interdependan-
ce : structure du milieu.
Verification de l'echan-
tillonnage .
MATRICES DERIVEES Variable
Etat
K
L
Etat
1
Etat
W
ESPECE E U, . . .1 . . uK uNK U 1F. 1
RELEVE S R, . . . . . R • RNK1 K
11 frequence (ou participation) de l'espace E
dans les releves R, ou la variable L presente
l'etat K.
| PROFILS ECOLOGIQUES
ZZp> Frequences UK/RK
relatives:
Frequences UK.NR/
corrigees:
Probabilites : profil
ecologique
indice.
Fig. 1. Organigramme de ll'analyse des donnees phyto-ecologiques.
■. !,
- 117 -
Les calculs sont souvent complétés par une ordination en
espace réduit qui permet d'ordonner réciproquement entre eux groupements
végétaux, espèces végétales et variables écologiques. La projection
des groupements végétaux et états des variables de l'inondation sur
le plan des axes factoriels 1 et 2 dans une analyse factorielle des
correspondances faites sur les relevés phyto-écologiques du Delta est
un exemple des ordinations obtenues (Fig. 2).
Il faut compter un jour de travail pour coder et inscrire les
données de 10 relevés phyto-écologiques (1 jour pour 5 relevés ligneux),
2
soit environ 10 jours pour 10.000 km échantillonnés comme précédemment
pour un observateur. Il faut ajouter à cela la durée des calculs et
de l'interprétation qui n'est pas proportionnelle au nombre des relevés.
La carte
La caractérisation des groupements végétaux est complétée par une carte
dont le thème principal, synthétique corrélatif est celui des unités
phyto-écologiques.
La clef de cartographie est directement établie à partir des
résultats de l'analyse des données de relevés qui fournit en outre une
liste d'indicateurs. Pour lever la carte, on a recours à la télédé
tection par photo-interprétation des photographies aériennes (panchro
matiques, et infrarouge au 1/50 OOOème pour le Mali). Elle est d'autant
plus aisée que les photos ont servi à l'échantillonnage et aux
prospections de terrain. Dans l'approche empirique utilisée, la clef
de photo-interprétation est forgée progressivement au cours des
rapprochements répétés entre l'image et la vérité de terrain. La
télédétection peut procéder d'une démarche plus systématique; il
suffit d'adjoindre aux descripteurs de terrain ceux qui caractérisent
l'emplacement observé sur le document de télédétection: relief,
structure, texture et ton des images photographiques, signatures
spectrales sur les fiches de relevé. Des calculs identiques aux
précédents établissent la valeur indicatrice de ces paramètres ou de
leur combinaison, ce qui, dans le cadre de données géographiquement
cotées, permettrait une cartographie automatique. Il est cependant
douteux que les indicateurs multispectraux puissent répondre aux
caractërisations complexes et hétérogènes qui participent à la définition
des groupements végétaux.
- 118 -
 Opposition entre les inondations—
rapides et precoces x lentes et tardi^ es
«
€m iO
A
1
10
m
K)
<n ^J
~
" , ■j *" •»■ ■ ae r>
c
— o ?E
* • c £ ♦
1
"
5 •<
<£>
*J -1
■o
• to ■
• « "3
•
9
i ino" * ■
2
to
2 i
o
«M 2 Mu _ O
B o
Z
o A = o •W
u
-i
* ••-? '• c
£
** ""
e
io » IO c V
^
■D ■a
!h•o m <»— f\j o *■ <N» sm 1 <33w "2 £> o Q> -^
x?
— > & v-~J fe; .
rQ c\;
o a s- i^ a •^ 'a +s
0)» -
CN
•
00
•H
- 119 -
Une fois les clefs de cartographie et de photo-interprétation
établies, le lever d'une carte par les méthodes indiquées et à
2
l'échelle 1/50 OOOPnii- prend 60 jours/homme pour 10 000 km (25 jours
de préphoto-interprétation: 6 photos par jour, 10 jours de vérification
de terrain, 25 j . de corrections et de dessin).
La durée effective globale de la phase descriptive suivant
les méthodes décrites ci-dessus se chiffre à 100 jours/homme pour
2
10.000 km en comptant 10 j. forfaitaires pour l'établissement du
tableau d'échantillonnage (étude bibliographique, première observation
des documents photographiques) et 28 j. pour les calculs et l'interpré
tation des résultats.
La phase de diagnostic
But: - Détecter les contraintes à la nutrition animale liées à une
insuffisance quantitative ou qualitative des ressources
fourragères .
- Proposer des solutions techniques ou des voies de recherche
pour les résoudre.
Méthodes: Disponible fourrager, production, productivité et biomasse.
Les contraintes nutritionnelles dépendent des disponibilités fourragères
des principaux pâturages décrits en tenant compte de la fraction de
ces disponibilités effectivement accessible au bétail (fraction qui
dépend du parcours, de la nature du bétail et du mode de conduite de
la pâture, cf. G. Boude t 1975, P. Hiernaux 1982, H. Breman 1982). Les
disponibilités fourragères d'un parcours varient au cours du cycle
annuel, d'une année à une autre, et en fonction du mode de gestion. En
un lieu et instant donnés, elles résultent des phénomènes de
production, de prédation et de dégradation qui se sont produits jusque-
là. Aussi, les mesures de biomasse doivent avoir pour but l'analyse de
ces trois phénomènes sous pâture et sous mise en défens.
Les techniques de mesure de biomasse
Les techniques utilisables sont très nombreuses et sont l'objet d'une
documentation abondante. Nous nous limitons à indiquer quelques-unes
des méthodes pratiquées dans le cadre du projet du CIPEA au Mali.
- 120 -
■iomasse aérienne du tapis herbacé
Technique destructive - Quels que soient les organes et catégories
considérés (matière verte, matière morte, graminées, légumineuses,
taxon, strates physionomiques, stade phénologique. . .) la mesure est
faite par prélèvement pratiqué par coupe sur les placettes délimitées.
Le principal problème technique est celui de l'échantillonnage: il faut
fixer les formes, la taille, le nombre et la disposition de ces
placettes sur le terrain pour que les mesures soient représentatives
de la station.
Pour la forme de la placette, on sait que la forme circulaire
réduit au minimum les effets de bordure et que la forme rectangulaire
minimise la variance associée à la micro-hétérogénéité mais augmente les
effets de bordure. Finalement la forme carrée est retenue. Le nombre
et la surface élémentaire des placettes déterminent la précision de la
mesure mais aussi son coût. Si l'on tient compte du coût, la surface
2
optimale se situe suivant les cas entre 1 et 4 m . Le nombre de
placettes pour un seuil de précision requis est fonction de l'hété
rogénéité du tapis herbacé mais aussi du plan d'échantillonnage. Avec
une répartition aléatoire (ou systématique-aléatoire), une trentaine
de répétitions suffisant à atteindre une précision satisfaisante, la
précision sur la moyenne
s
x
P -
écart-type
moyenne
s.t
x n
n : nombre de répétitions
t : valeur du t de Student pour n-1
degrés de liberté au seuil de 0,05
se stabilise entre 5 et 25% selon l'hétérogénéité du parcours. La
disposition aléatoire ou systématique des placettes sur l'unité
étudiée n'est pas toujours aisée; elle est souvent remplacée par une
distribution subjective stratifiée moins fiable mais plus efficace. Il
faut compter une à deux heures par mesure de biomasse sur le terrain.
Technique non destructive - Les lectures des contacts sous
un point ou des intersections sous segment qui sont préconisées par
de nombreux auteurs sont des mesures indirectes: la biomasse est calculée
- 121 -
à partir de la densité de la végétation. Ces méthodes fournissent en
plus des informations très détaillées sur la structure du tapis végétal.
Mais elles exigent beaucoup de temps et seront réservées à des mesures
fines de la dynamique du tapis végétal (cf. présentation de J.C. Bille).
De la biomasse à la production et aux disponibilités fourragères
Dans une première approche, les compensations entre production et
dégradation sont négligées; on se contente de répéter les mesures de
biomasse et les analyses bromatologiques par intervalles au cours du
cycle annuel. Ces résultats permettent d'établir les premiers
diagnostics .
La courbe de la biomasse et de la teneur en protéine d'un
parcours sahélien à Schoenefeldia gracilis montre que même si le
parcours est protégé jusque là, le taux de protéine est inférieur à
6% une grande partie de l'année durant laquelle la qualité moyenne du
parcours est insuffisante (Fig. 3).
Dans une approche toujours globale, il est possible de comparer
diverses situations liées à l'utilisation du parcours, la différence
entre la parcelle pâturée et le témoin souvent appelée consommation
apparente traduisant globalement le bilan de l'ingestion de la
croissance modifiée et la dégradation sous pâture.
Cette approche peut être perfectionnée en faisant la part des
divers phénomènes qui concourent au bilan "biomasse", en différenciant
la production brute et la production nette dans un pâturage sahélien.
Les résultats obtenus au Sahel sont d'ailleurs variables, disent les
auteurs.
Sur un principe similaire, on mesure la production sous
pâture par prélèvements couplés à l'intérieur de cages fourragères
déplacées à intervalles rapprochés. Le tableau 2 donne un exemple où
sont comparées les productions sous pâture et sous trois régimes de
fauche d'un parcours à Echinochloa stoanina. La comparaison oriente
les recherches vers l'expérimentation de modes de gestion optimisant la
production par rotation des parcours.
- 122 -
Tableau 2. Production de repousses feuillees d'un parcours
Intervalle PRODUCTION DE REPOUSSES FEUILLEES (kg MS/ha)
entre les DU 28/12/80 AU 13/7/81
fauches
Prod.cumulee Prod, par fauche Prod. journal ier<
(semaines)
ou pature (p)
mama ma
1
2
4
4737 370 190 15
4807 311 400 26
4496 100 749 17
2346 2593 180 199
25,3 2
26,7 1.7
24,9 0,1
12 13,3
Biomasse
aerienne
(kg de MS/ha)
2000-
Debut des pluies
regulieres
pluies ^ I
precoces -
I I teneur en
I fin des pluies protéines
a )SdhoenefeIdia
1000
200-
 
Mois
Fig. 3. Evolution mensuelle de la biomasse et de la teneur en proteine
d'un pdturage herbeux a Schoenefeldia gracilis mis en defens
(ranch de Niono) .
- 123 -
Il est très difficile de fixer des normes de temps pour ce
type de recherches tant elles sont fonction de cas particuliers. Mais
d'une façon générale, elles portent sur un cycle annuel complet,
plusieurs cycles si les fluctuations interannuelles sont importantes.
A titre d'exemple - Pour le delta, les études menées sur 10 parcours,
3 années durant, occupaient une équipe de 5 personnes, 2 semaines pour
la mise en place puis une semaine par mois.
La phase d'expérimentation
But: Analyser l'impact d'une variable du système ou d'une innovation
technique sur le fonctionnement du système.
Mettre au point une technologie adaptée pour promouvoir un aména
gement préalablement testé localement.
Les expérimentations sectorielles
Elles sont menées sur quelques sites ou stations en conditions contrôlées
souvent assez artificielles par rapport aux réalités du système de
production.
Pour les pâturages, elles consistent principalement à analyser
l'impact de pratiques pastorales vraies ou simulées sur la production et
la dynamique d'un parcours. L'impact de la date d'entrée dans les
parcours du delta (simulée par la fauche) sur leur production et effet
de la précocité des incendies, illustrés par les essais menés sur un
pâturage à Andropogon gayanus (Fig. 4), sont un exemple d'expérimentation
sectorielle dont les résultats orientent les propositions d'aménagement
du système de gestion.
Les résultats obtenus sur les divers parcours amènent à
préconiser une première pâture ou un incendie les plus précoces
possibles après la maturité des herbacées.
Par ailleurs, l'impact des régimes de fauches répétées à
intervalles d'une, deux, quatre ou huit semaines sur la production de
repousses, figure dans le tableau 3 pour quatre pâturages du delta.
D'après les valeurs de production cumulées, il existe pour chaque
parcours un rythme de fauche optimal (entre 1 et 2 semaines pour Echinonhtoa,
2 et 4 semaines pour Andropogon . . . etc) .
- 124 -
Biomasse
(kg de MS /ha)
Legende :
Coupe precoce (CP) ,tardiv T / N. 1
.. i Brulage precoce (BP;/tardif (BT) / \ /
500-
/ ^^ / /\ / '
/ 1
CP / / \ / ■
/•• '. / ."
100- ■ / BT /
v7 ••""'7 */' *.,-
Date 17/11 17/12 17/1 16/2 18/3 17/4 17/5 17/6 17/7
Fig. 4. Effet de la date de fauche ou dHncendie sup la repousse d'un
parcours 5 Andropogon gayanus.
- 125 -
Tableau 3. Repousses feuillies de quatre graminees des plaines
d'inondation du Niger soumises 5. divers rythmes de fauehe.
Productions journali^res en "kg MS/ha.
Intervalle
entre deux
fauehe s
(semaines)
Repousses
Stagnina
80
m
Feuillees d ' Echinochloa
(dec. a juillet) kg MS/ha/j
81 82
mama
1
2
4
8
- - 25,3 2 28 8,9
31,8 1,6 26,7 1,7 37,2 4, 4
23,3 0,8 24,9 0,1 - -
26,6 0,3 - - - -
Intervalle
entre deux
fauches
(semaines)
1
2
4
8
Repousses Feuillees kg MS/ha/j
Andropogon Panicum Vetiveria
gayanus anabaptistum nigritiana
1980 1982 1982
m a m a m 0
- - 6,2 0,7
5,1 0,4 6,9 1,7 8,7 0,7
6,4 0,6 7,7 1,4 11,4 0,8
4,3 0,5 6,2 1 (15,1) (3)
- 126 -
Les expériences sectorielles peuvent aussi comporter des
essais de pâture dont les effets seront analysés à court et long termes.
C'est le cas des expérimentations menées sur le ranch de Niono pour
quantifier l'impact de la pâture saisonnière des parcours sahéliens.
Le test des innovations
Les innovations suggérées par les résultats de ces essais doivent être
testées sur quelques unités de production afin de vérifier quels en sont
les impacts sur le système et de mettre au point les technologies
appropriées.
Un test de la rotation des parcours de saison sèche est en
cours à Diafarabé. Les éleveurs de la Coopérative de Diafarabé ont bien
voulu se prêter à une expérimentation de rotation hebdomadaire sur trois
soles pour une partie de leurs troupeaux laitiers villageois. Le
protocole a été élaboré au cours d'une série de réunions où nous nous
sommes efforcés d'expliquer les buts techniques de l'essai aux éleveurs.
Les éleveurs assurent le gardiennage et la gestion de leurs troupeaux;
nous nous contentons de mesurer les effets sur la végétation et le poids
des animaux.
Nous n'avons pas vraiment franchi l'étape suivante dans le
projet Mali. Cependant un certain nombre de travaux réalisés dans le
cadre du contrat passé entre le CIPEA et l'Opération de développement
de l'élevage dans la région de Mopti s'intègrent à la phase de promotion
des innovations au niveau de l'ensemble d'un système de production. C'est
en particulier le cas des cartes des parcours, des capacités de charge
potentielle, et de la structure foncière des systèmes pastoraux du
delta intérieur du Niger et de ses marges sahéliennes ainsi que tout le
travail concernant la mise en place des unités pastorales pilotes,
autant de documents qui doivent servir à la réforme structurelle et
foncière indispensable à la promotion des innovations technologiques
proposées, en particulier toutes celles qui sont relatives à la gestion
des parcours.
- 127 -
Cet exemple illustre un aspect de la recherche sur les
■ystines pastoraux. De même que lors de l'échantillonnage, il a fallu
contidérer une hiérarchie des niveaux d'organisation de la végétation
et du milieu, de même les systèmes de production animale sont
structurés en niveaux d'organisation avec à la base les unités de
production, puis les communautés de base aggrégées en groupes plus ou
moins homogènes à leur tour intégrés dans des sociétés d'autant plus
complexes qu'elles sont vastes. Dans l'aménagement d'un système, il
faut veiller à ce que les proportions soient adaptées à chacun des
niveaux d'organisation et coordonnées entre elles. Même si l'unité de
production est le maillon fondamental du système et si la détermination
individuelle des éleveurs ou propriétaires d'animaux est indispensable
à tout aménagement, certaines décisions ne peuvent être prises qu'à un
niveau d'organisation supérieur. L'organisation de rotations pastorales
sur les terrains de parcours villageois nécessite une décision collective
(cas de la coopérative des éleveurs à Diafarabé) . La modification du
calendrier qui fixe l'entrée des animaux dans les parcours du Delta
doit être prise à un niveau encore supérieur puisqu'il concerne non
seulement les éleveurs de la plaine et du Sahel environnant mais aussi
les riziculteurs, les pêcheurs, et les services régionaux (vétérinaire,
santé, enseignement) .
Actuellement une décision est prise chaque année par une
commission paritaire organisée par le Gouvernorat de la Région. Les
documents cartographiques réalisés pour le contrat ODEM/CIPEA ne sont
certes pas très utiles à l'éleveur qui a une connaissance empirique
infiniment plus détaillée des parcours de son secteur, mais elles peuvent
jouer un role important dans le dialogue entre les éleveurs et les
cadres des organismes de développement pour le choix des aménagements.
- 128 -
Vegetation and feed resources in pastoral systems
English Summary:
The aim of studies on vegetation and feed resources in pastoral
systems is to detect and analyse the nutritional constraints to
livestock production and to put forward solutions to relieve them.
Vegetation is one of the main feed resources for animal production
which has to be quantified and its production and reproduction processes
need to be analysed. Vegetation is also a major environmental parameter
which reflects the other components of the environment.
The aim of the descriptive phase of studies on vegetation and feed
resources is to stratify the environment in which the pastoral system
or systems under study evolve. Such studies also aim to characterise
forage production within each strata and the ecological variables which
condition it. An inventory of forage resources is carried out in the
field using methodical surveys of the vegetation and the environment.
A survey consists of systematically noting for a sample site the values
recorded for a series of parameters covering the structure and the
vegetation of the environment. For vegetation these parameters cover
the structure of herbaceous and ligneous associations, the floristic
composition, the above-ground biomass. For the environment the
parameters are the climatic and edaphic characteristics of the site,
the nature of the bedrock, geomorphology, topography, description of
the soil profile and indications on land use, pastoral and agricultural
status etc.
The accuracy of the results depends on the position and number of
ecological stations selected. A sampling plan establishes the list of
situations to be observed and the number of replicates of each. The
number of situations is worked out from a graded stratification of the
area under study into regions, sectors, series or sequences and finally
into ecological stations. The sampling table establishes the number of
observations and situates them in rough terms. To situate them
accurately, maps and aerial photographs are very useful. Aerial
photography is a very useful tool in the technical sampling phase, both
for checking the representativeness of sites and for solving access
problems.
- 129 -
Data collected in surveys can be presented in the form of matrices
in which the objects are the surveys and the descriptors the taxons
and the states of the ecological variables observed or measured. The
large size of these matrices makes a computer essential. This kind of
data processing requires the coding of data and their processing onto
punched cards or magnetic tapes. The calculations are often completed
by a factorial analysis which allows the reciprocal processing of plant
groups, species and ecological variables.
The characterisation of plant groups is completed by a map whose main
theme is phyto-ecological units. The key to the map is directly
established from the results of the data analysis from the sample areas
which also provides a list of indicators. Remote sensing is used to
make the map; photo-interpretation of aerial photographs is easier if
the photographs have also been used for sampling and ground truthing.
The aim of the diagnostic phase is to identify the constraints to
animal nutrition linked with inadequate quantity or quality of feed
resources, and to propose technical solutions or lines of research.
The techniques which can be used for measuring biomass are numerous.
In destructive techniques measurements are made by cutting samples from
the small sample areas delineated in the field. The shape, size, number
and pattern of sample areas have to be predetermined in the field, so
that measurements will be representative of the ecological station.
One or two hours must be allowed for each measurement of biomass in
the field. In non-destructive techniques readings are taken from
contacts at a point or intersection at a segment. Biomass is calculated
using vegetation density as the basis. These methods also provide very
detailed information about the structure of the vegetation cover.
However, they are time-consuming and should be reserved for measurements
of the dynamics of the plant cover.
The aim of the experimentation phase is to analyse the impact of a
variable within a system or a technical innovation on the functioning
of that system, and to develop an adapted technology so as to promote *a
locally tested management scheme. Sectorial experments are carried out
on several sites or stations under controlled conditions which are
often rather artificial when compared to the realities of the
- 130 -
production system. For pasture they consist mainly of analysing the
impact of pastoral practices, real or simulated, on the production or
dynamics of a rangeland. Results obtained on different kinds of
rangeland lead to a recommendation to start grazing or to burn off as
early as possible after the grass has matured.
The innovations suggested by the results of these trials must be
tested on several production units in order to verify their impact on
the system and to develop an appropriate technology.
- 131 -

L'évolution du milieu
J.C. Bille
Ecologiste, programme des zones arides (Afrique de t'Est et Afrique
australe) , CIPEA, Ethiopie
Les causes de l'évolution
La description de l'évolution du milieu suppose une conception
suffisamment claire des lois biologiques qui s'appliquent aux systèmes
pâturés et des liaisons internes dans ces systèmes. On peut, en
première approximation, représenter un tel système par le diagramme
ci-après :
r - "1
Energie (
Atmosphère $ <
1
O o
J
8 *
?
ARBRES
eu HERBE Animal
X
VU
4-1
vu
Q.
a non
acces
sibles
c
o
0
Sol
i — .-l
 
Fig. 1: Représentation d'un écosystème pâture
- 133 -
On y remarque des compartiments pour la plupart reliés entre
eux, et ces liaisons symbolisent l'existence de flux susceptibles
d'aller d'un compartiment à un autre, généralement dans les deux sens.
Ce schéma a pour but d'attirer l'attention sur le problème le plus
habituellement rencontré lors des mesures de dynamique des pâturages:
on demande à un écologiste de décrire les modifications perçues dans
le sol et les plantes sans proposer aucune indication sur trois
éléments indissolublement liés aux précédents:
- les animaux, leur type, leur nombre, l'époque à laquelle
s'exerce leur actior , etc...
- la gestion du territoire, représentée sur le diagramme par
des points d'interrogation aussi multiples que les données
qui la composent;
- ce qui entre et ce qui sort du système, dans le meilleur
des cas exprimé sous forme d'unités monétaires.
Il est important de bien se rendre compte que la simple
notification des changements dans la végétation n 'apporte aucune
information si elle n'est pas accompagnée de mesures sur les divers
flux à l'intérieur du système et à travers le système. En outre,
l'intérêt réside habituellement dans les changements liés à 1' exploi
tation et non dans ceux qui sont induits par des causes naturelles
Cielles que le climat): c'est alors une méthode indirecte d'apprécia
tion de la gestion. Le premier point consiste donc à dissocier les
types d'évolution.
Types d'évolution
Il n'y a de changements climatiques tangibles et rapides que ceux qui
correspondent aux variations interannuelles des précipitations
habituelles aux zones tropicales. Cette variabilité est d'autant plus
élevée que le climat est plus aride, et se traduit presque immédiatement
sur la croissance végétale à des degrés divers. La sécheresse modérée
d'une seule année réduit la production primaire; une sécheresse intense
détruit une partie des plantes pérennes, ou certaines espèces seulement,
ou empêche les germinations. Il y a presque toujours un arrière-effet
climatique sur l'année suivante, soit par l'intermédiaire de la
- 134 -
production de diaspores, soit en raison des traumatismes subis par
les végétaux pérennes. Nous considérons cependant que ce type de
changement est à court terme, brutal, marqué et assez éphémère.
Si par contre on modifie à la fois la plupart des éléments
et des flux à l'intérieur du système, de nombreux phénomènes vont se
produire: on pourra avoir une réduction permanente de l'ensemble des
végétaux, ou simplement de certaines espèces, ou encore un transfert
de productivité des plantes herbacées vers les plantes ligneuses.
Tous les équilibres du système étant perturbés, il y aura aussi des
conséquences sur les sols (perte de fertilité, érosion, reprise de
pédogenèse) qui se feront sentir pendant 10 ans, 50 ans ou un siècle.
La cause la plus usuelle de ce second type de changements
est une sur-exploitation du système (c'est-à-dire que ce qui en sort
n'est pas compensé par ce qui y entre), en particulier par l'intermé
diaire des animaux domestiques. Ce sont des évolutions à long terme,
appelées à tort tendances évolutives car ces tendances ne sont
perçues que lorsque l'évolution a déjà eu lieu et qu'il est trop tard
pour intervenir.
Mesure des changements à long terme
Nous imaginerons à partir de maintenant que pour une raison quelconque,
on ne dispose pas d'une étude scientifique complète du système de
production à l'étude et qu'on ne veut pas réaliser cette étude, soit
qu'on ne dispose pas de l'expertise nécessaire, soit qu'on ait une
prévention innée contre ce type de travail. On souhaite cependant
mesurer ou évaluer ce qui se passe dans la végétation. Les possibilités
seront illustrées à partir d'exemples choisis au Kenya, où le CIPEA
s'est efforcé de réaliser un exercice de ce type.
Contraintes particulières à la mesure simple des changements dans la
végétation
Les paramètres les plus souvent utilisés pour la description des
formations végétales sont la diversité, la structure, la périodicité
des phytocénoses d'une part, et l'abondance, la fréquence, la constance,
la dominance, la fidélité, la vigueur et le couvert de chaque espèce
- 135 -
d'autre part. Certains de ces paramètres sont qualitatifs et mal
adaptés à la mesure de nuances subtiles; la mesure des autres demande
une somme de travail qu'on ne peut guère conseiller pour une inter
vention bon marché .
Ainsi, la détermination du couvert par espèce qu'on pourrait
aisément relier à la productivité est réalisable par représentation,
graphique de la végétation, par enumération des plantes associée à des
mesures sur leurs dimensions, par des méthodes de type P.C.Q. et par
des systèmes ponctuels à base de tiges fixées sur des cadres, sur des
roues et autres dispositifs. Considérant que la mesure d'un changement
se fait à partir d'une comparaison de mesures, et par suite suppose
des mesures répétitives, aucune des techniques citées n'est à la fois
assez rapide et assez précise pour répondre aux besoins.
Un autre type de contrainte est lié à la représentativité
des observations effectuées, et par suite à l'échantillonnage: les
procédés classiques de description de la végétation supposent toujours
qu'on travaille sur une unité homogène et non pas, comme c'est
nécessaire dans une étude de système de production, sur une unité de
gestion pastorale où l'hétérogénéité est une vertu capitale. Même si
l'on ne se sent pas concerné par les querelles académiques des
spécialistes, il faut admettre que la méthodologie appropriée n'est
pas décrite dans les manuels d'écologie qui d'ailleurs évitent pour
la plupart de traiter du sujet.
Dispositif de terrain utilisé au Kenya
Le dispositif qui a été mis au point au Kenya considère un territoire
relativement vaste et soumis à un type d'exploitation donné comme site
échantillonné. Les points-échantillons constituent un réseau permanent
et sont disposés de façon systématique (par exemple, un point tous les
2, 3 ou 5 km) le long d'un itinéraire fixe qui sera parcouru lors de
chaque évaluation.
L'un des points joue un rôle privilégié, en ce sens qu'il
est repéré au sol par un moyen quelconque (peinture, barres métalliques,
petites excavations, masses de métal enfouies,...) et qu'on y effectue
- 136 -
des mesures précises sur la végétation. Les autres points seront
décrits en fonction de ce lieu, et seulement sous forme d'estimations
comparatives. Les deux types de points de contrôle ont été appelés
respectivement Transect principal et Points secondaires, et l'ensemble
du dispositif est tel que présenté par le diagramme suivant:
Unité de production
TRANSECT PRINCIPAL :
Mesures sur
PS
 
PS
Nombre + couvert
des ligneux
PS
Couvert par espèce,
strate herbacée
sur lignes ou ban
Do ^n 5
50 P. S.
des permanentes.
POINT SECONDAIRE
+ si besoin, popula
Estimations
tion ligneuse.
Nombre d'arbres.
Dominantes (ar
bres, graminées,
non graminéens)
Couvert total.PS PS
Fig. 2: Plan de surveillance des terrains de parcours
- 137 -
Réalisation pratique des observations
En dépit des indications portées sur le diagramme, les mesures ne
présentent qu'une indication voisine du couvert, et ne prétendent pas
se substituer aux procédés plus scientifiques habituels. Elles
consistent en effet à poser un anneau de petite taille (de l'ordre de
2,5 cm) tous les 50 cm le long de deux lignes de 50 mètres et à noter
la présence de végétaux dans la surface, en restreignant le choix à
quatre possibilités: rien, un peu d'herbe avec une espèce, beaucoup
d'herbe avec 1 ou 2 espèces. Si le tapis est trop clairsemé, on se
contente de noter l'interception des plantes le long des mêmes axes.
Par ailleurs, deux bandes de 100 x 1 m incluant les mêmes
lignes servent à localiser les arbres qui s'y trouvent, en notant
leur position, leur diamètre et l'extension de leur couronne. Aux
points secondaires, on note si les arbres sont plus ou moins nombreux
qu'au point de mesure, si le couvert herbacé est plus dense ou plus
faible; on établit la liste des trois espèces qui paraissent dominantes
dans chaque strate, et ca note la pression de pâture.
Des tests sévères ont montré au Kenya que différents
observateurs donnent des informations comparables, que le contrôle du
réseau ne demande habituellement pas plus que le temps de parcourir la
distance, augmenté d'environ 4 heures, et que l'observateur le moins
spécialisé fournit des données acceptables. On peut remarquer que
ces résultats donnent des indications sur :
- la répartition des espèces végétales sur la surface, et les
changements dans cette répartition (espèces envahissantes,
ou qui disparaissent sous l'action du bétail ou pour une
autre raison, déboisement sélectif et autres modifications
qualitatives) ;
- le potentiel productif de la zone et sur
ses variations, en liaison avec l'utilisation des savanes,
et des renseignements immédiats sur la denudation ou
l'érosion.
- 138 -
Periodicite et resultats
Les resultats sont lies a la periodicité des mesures. Un controle
annuel ne revele le plus souvent que 1' influence du climat de l'annee
(connu par ailleurs, dut-on dans ce but inclure un pluviometre dans
le reseau, c'est-a-dire si aucune station méteorologique n'existe a
moins de 50 km). Un changement a long terme lie a l'exploitation
n'apparait generalement qu'apres 3 ou 5 ans. Les resultats qui suivent
donneront une idee du type d' informations recueillies au Kenya:
Tableau 1 . Exemple des resultats de strates herbacees obtenus au Kenya
Strate herbacee, transects
Espece
Digitaria milanjiana
Pennisetum stramineam
Themeda triandra
Sporobolus firribriatus
Cynodon plectostachyum
Microchloa indica
Chlovis spp.
Pennisetum mezianum
non gramineen
Total
litiere
% du couvert
Oct. 1977 Mars 1978 Mars 1980
3,9 +/-0,9 5,5 7
4,3 " 3,3 1,3 1,5
0,5 " 0,7 0,5 4,5
M " 0,5 2 0,5
0,5 " 0,7 2 1,7
2,1 " 0,8 0,5 -
- 4 2,7
0,7 " 0,8 0,8 T
5,5 " 0,6 3,7 0,3
20,2 22 19,4
9,1 15 13,7
Strate herbacee, points secondaires
Couvert relatif 1979 (%)
Espece
Digitaria milanjiana
Chloris spp.
Pennisetum spp.
Sporobolus spp.
Eragrostis spp.
Cynodon
Autres graminees
Transect Ranch
36,1 21,5
13,9 10,7
7,7 20
2,6 13,3
6,2 8
8,8 6,7
1,5 11,8
- 139 -
Mesure des changements à court terme
Sauf cas très exceptionnel, les changements dans le milieu au cours
d'une même année sont les variations dans la biomasse végétale. Il
est bien entendu qu'on ne peut en aucun cas assimiler la biomasse à
une époque quelconque avec la production primaire nette d'une année
ou d'une période donnée, même si on dispose d'un territoire en défens
et qu'on suppose résolus les problèmes de mesure des masses (énorme
variabilité dans l'espace, échantillonnage difficile, longueur et
coût des opérations de terrain) . Le graphique suivant illustre cette
affirmation:
Masses de matériel
végétal
Pluies
Biomas
protégees
3es,
\
s
/ ;
— - „•
(productivité primaire nette)
/ N.I'.P. cumulée
(productivité primaire cumulée)
/
l /
A
X
\
Biomasses, avec pâture
Mois
Fig. 3: Comparaison de la productivité de la biomasse observée sur
une période d'un an
- 140 -
Les trois courbes, obtenues au Kenya, diffèrent non seulement
par leur forme mais même par l'époque des maxima et minima s ils
existent. Le matériel végétal ingéré par le bétail ne peut pas non
plus être calculé à partir des courbes dont les variations incluent
les pertes par piétinement, l'action des décomposeurs et celle
d'éventuels autres consommateurs primaires. Les mesures de biomasses
végétales ne -peuvent dono en aucun cas être utilisées pour un autre
but que l'estimation de la quantité de matériel alimentaire offert
aux animaux.
Mesures sur le terrain
La technique des coupes réalisées sur de petits plots est bien connue,
mais est souvent pratiquée avec de nombreuses variantes:
- échantillons répartis au hasard à l'intérieur d'une
formation végétale déterminée: outre que le travail demande
l'existence d'une carte de végétation à grande échelle et
une certaine aptitude de l'opérateur à décider s'il se
trouve bien dans la formation voulue (des détails de
terrain peuvent ne pas avoir été cartographies) , le procédé
est très long et peu réaliste;
- échantillonnage volontairement biaisé: on ne garde que les
plots qui semblent représentatifs, mais il est alors interdit
de proposer un écart-type et un intervalle de confiance
pour le résultat;
- mélange de mesures vraies et d'estimations: on effectue les
estimations sur des plots répartis au hasard, et on vérifie
l'estimation par une mesure, par exemple une fois sur 10 de
façon à introduire un facteur correctif pour l'ensemble des
résultats. Le procédé réduit très sensiblement la quantité
de travail et a été utilisé au Kenya sur une surface où une
action mul tidisciplinaire intensive était en cours, justifiant
de tels efforts.
Le tableau reproduit ici donne une idée des résultats obtenus.
Pour toutes les valeurs, les mesures ont été arrêtées dès que l'inter
valle de confiance de la moyenne était inférieur à 20% de cette moyenne
qu'il faut donc lire: m +/-0,1 M.
- 141 -
Tableau 2. Exemple d'une biomasse herbacée mesurée pendant toute une
année pour quatre unités écologiques
Mois Biomasse herbacée
(kg MS/ha)
1 2 3 4
Oct. 80 355 180 660 4
Mars 81 275 80 495
Juin 81 560 440 1060
Oct. 81 465 360 910
Oct. 80 305 230 435 370
Mars 81 270 205 370 300
Juin 81 600 735 820 510
Oct. 81 455 370 430 310
Un autre type de mesures de terrain est la détermination
des taux de croissance de l'herbe, qui évite certains des reproches
adressés à l'utilisation des biomasses. Il faut disposer d'une
centaine de petites cages destinées à protéger les plots-échantillons
au cours de périodes assez courtes (de l'ordre de 2 semaines) pour que
la croissance végétale s'y effectue au même rythme qu'aux endroits
non protégés. On considère que la production à l'intérieur des cages
est assez proche de la productivité nette à l'extérieur pour lui être
assimilée.
Mesures indirectes
De très nombreuses méthodes existent, et on peut citer pour mémoire
les mesures par densitometrie sur photographies aériennes en noir ou
couleur, panchromatique ou infra-rouge; les procédés radiométriques
divers: on enregistre au sol, ou en vol à basse altitude, ou à partir
d'un satellite, la réponse du milieu dans certaines fréquences d'émission
micrométriques ; et plus récemment l'analyse des composantes de réflexion
du milieu qui demande l'usage d'un ordinateur et dont les deux tendances
sont actuellement:
- 142 -
- l'analyse par vecteurs construits dans un espace
multidimensionnel à partir des composantes de rayonnement;
- l'analyse multi-temporelle qui compare deux images
successives d'une même scène prises à des époques telles
que l'inclinaison du soleil soit différente.
Aucun procédé de mesure indirecte des masses végétales n'a
encore donné pleinement satisfaction, car le travail présuppose souvent
un terrain plat, une végétation homogène sur de vastes espaces (faible
définition des documents-satellites) et un accès aisé aux bandes
d'enregistrement des mesures, ce qui n'est jamais possible pour les
pays d'Afrique tropicale.
En outre, les mesures indirectes ne dispensent pas du
travail de terrain nécessaire pour calibrer les données de télédétection,
de sorte que les causes d'erreur s'ajoutent et que le résultat final
est peu fiable. Les observations aériennes à basse altitude qui
classent les couverts végétaux et la proportion de matériel vert en un
petit nombre de catégories, fournissent finalement des informations plus
précises et à moindre coût que les procédés les plus sophistiqués.
Utilisation des connaissances et choix du type d'investigation
Le texte qui précède montre en particulier que, si toutes les mesures
de changements sont des actions répétitives, les mesures de changements
à court terme sont infiniment plus coûteuses et plus contraignantes que
celles des évolutions à long terme. Si l'on compare la réalisation
d'une étude scientifique complète de base qui fournira dès le départ
tous les éléments de connaissances nécessaires pour juger simplement
des tendances du milieu (description des séries de végétation) , et la
seule mesure de la dynamique telle qu'elle vient d'être traitée, on
voit aussi que la différence réside surtout dans un étalement dans le
temps des efforts et des dépenses pour la seconde approche.
En fait, le choix d'une méthode dépend surtout des possibilités
d'investigation et de l'intensité des actions de développement prévues.
En cas d'intensification poussée de la production animale, les études
les plus complètes sont encore bon marché mais il serait inutile
- 143 -
d'effectuer un suivi des quantités d'aliments disponibles, par
exemple, si l'on n'a ensuite aucune possibilité d'intervention sur la
gestion du territoire et si l'on ne dispose pas sur le terrain a la
fois d'agents techniques compétents, d'une législation qui permette
d'agir et de l'infrastructure nécessaire à la mise en place d'une
intervention complexe.
En outre, la seule mesure des changements dans le milieu est
une activité sectorielle à la suite de laquelle on aura acquis très
peu d'informations sur les possibilités de cultures ou de reconversion
du système de production. Par contre, le suivi de la dynamique
végétale devrait accompagner toute intervention humaine sur les
systèmes de production animale, au même titre que le suivi de la
production secondaire ou des résultats socio-économiques. Il devrait
également être une composante automatique de toute expérimentation
pastorale, même lorsque le but de l'expérience est aussi spécifique
que l'amélioration génétique d'une Face animale ou l'accroissement de
la production laitière.
- 144 -
Vegetation trends
English Summary:
To describe vegetation trends presupposes an adequate and clear
knowledge of the biological laws applicable to grazed eco-systems and
the links within these systems. Merely noting down changes in the
vegetation provides no information at all if it is not accompanied by
measurements of the various flows within and through the system. The
starting point is to distinguish between the various types of
vegetation trend.
If the majority of components and flows within the system undergo
change, several trends may occur; there can be a permanent reduction
in the overall vegetation, or merely in some species, or a transfer
of productivity from grass towards ligneous species. Because the
overall balance of the system has been disturbed the soil will also
be affected; fertility is lost, erosion sets in, or soil development
begins again, for a period of 10, 50 or 100 years. The most usual
cause of this type of trend is over-utilisation of the system,
particularly by domestic animals. Such trends are long-term, wrongly
called evolutionary trends simply because they are noticed only after
the evolution has taken place, by which time it is too late to take
any action.
The classical procedures for describing vegetation assume that work is
being carried out on a homogeneous unit and not as is inevitable in the
case of production systems studies, on a pastoral management unit the
heterogeneity of which is its chief virtue. According to the approach
adopted in Kenya any relatively large area subject to a given utilization
is considered as the sampling site. The sampling points constitute a
permanent network and are arranged in a pattern along a fixed route to
be covered during each evaluation. The method uses main transect points
and secondary points.
Measurements with this technique offer only an approximation of the
vegetation cover. However the results give indications on the distri
bution of plant species over the area and changes in this distribution
and the productive potential of the zone and its variations, in
- 145 -
relation to utilisation of grazing land as well as immediate information
on degradation or erosion. Results obtained in Kenya are described.
Short term measurements of change are then described. Changes taking
place during the same year are variations in plant biomass. However
the measurement of plant biomass cannot be used for any purpose other
than for evaluating the quantity of feed on offer. The techniques of
hand sampling from small plots are well known. Variations include
sampling plots distributed at random within a specific plant formation,
deliberately biased sampling and a mixture of actual measurements and
estimates. Another type of field measurement is determination of the
grass growth rate.
Indirect measurements include densitometer measurements made from
black-and-white, panchromatic or infra-red aerial photographs by which
the response of vegetation in certain short-band wavelengths is
measured on the ground, by low level survey or by satellite. A new
method is that of computer analysis of the reflectivity of vegetation.
No indirect method of measurement has so far proved satisfactory.
Indirect measurements do not preclude the field work necessary for
calibrating remote sensing data, and as a result the sources of error
accumulate and the final result is unreliable. Low level aerial survey,
classifying vegetation and the proportion of green matter into a few
broad categories, provides more accurate and cheaper data than the more
sophisticated methods.
Measurements of change in the short term are far more costly and
cumbersome than those over the long term. The choice of method depends
primarily on what is feasible and the intensity of the development
effort planned. Measurement of change in the vegetation by itself is
a disciplinary activity. It provides very little information on the
possibilities of growing crops or on ways of transforming the production
system. On the other hand, the monitoring of vegetation dynamics should
accompany any human intervention in an animal production system, just as
the monitoring of secondary production or of the socio-economic results
should do. It should also be automatically a component of every pastoral
research project, even when the purpose of research is as specific as
the genetic improvement of an animal breed or an increase in milk
production.
- 146 -
Summary of Discussion Session 3.
Chairman: Dr Cees de Haan (ILCA)
Discussion led by Dr Ibrahim Gueye (SODESP, Senegal)
Dr Diakite pointed out that there was often a problem of transferring
research information to extension workers and herdsmen because of the
difficult language often used to describe research. Dr Nestel warned
that one shouldn't fault ecologists and forage scientists for
conducting elegant research in their own field of specialisation - that
was their role in a systems research team. One had to determine at
what stage one brought in social scientists to question the likely
benefit, if any, to be gained from pursuing a particular line of
research.
Dr Akilu stressed that rangeland management had social, socio
economic and political implications. Researchers and extension
workers couldn't solve these problems on their own. There was a need
to involve the politicians to decide on policies conforming to the
aspirations of herdsmen.
Dr Thomson asked if one always needed to do experiments on a research
station - could one not go directly to the producer and do on-farm
trials. Often there was a wealth of information on technologies
which aimed to solve some of the obstacles facing producers, which
were identified during the diagnostic phase. Did one need to repeat
this work? Dr de Haan said this depended on the situation - sometimes
one could leave out the experimental phase and go directly to the
producer, but sometimes one could not.
Dr Ngutter pointed out that work by vegetation/forage ecologists was
often not useful to the development planner unless such information
could help in determining the technical trade-offs possible between,
say, smallstock and large stock on a given range with a given botanic
composition, without affecting the botanic composition in an
undesirable way.
- 147 -
Dr Zulberti asked Dr Hiernaux if the bi-weekly or four-weekly grazing
systems mentioned in his paper could be applied or recommended in areas
where the annual variation in rain was 50% to 200% of the mean and
when pastoralists moved from area to area according to the availability
of pasture. Could not research on rotation be more oriented to these
conditions? Dr Hiernaux said that the proposals for rotational
grazing concerned the pastures of the entire Niger delta, the
productivity of which was high. The proposals were based on results
obtained on experimental sites. Trials were currently underway in the
field to test the relevance of these techniques to the existing
pastoral system.
Dr Barry pointed out that an understanding of the natural resources
entering into animal feed was important for maintaining the
ecological equilibrium, often highly fragile, in the Sahel countries.
Pastoral people and their herds were in a delicate situation due to
variability in rainfall, expansion of cash cropping and traditional
husbandry methods. Inventories of natural resources needed to be
applied to the real situation of pastoral people - thus one should
consider socio-economic factors. The problem was how to get these
people to adopt their knowledge of rangeland management in the light
of long periods of drought and considerable variation in rainfall.
- 148 -
Résumé des débats de la troisième seance
Président: Cees de Haan (CIPEA)
Débats dirigés par le Dr Ibrahim Guèye (SODESP, Sénégal)
Le Dr Diakité a souligné qu'il se posait souvent des problèmes de
transfert des données de recherche aux vulgarisateurs et aux éleveurs
à cause de la difficulté du langage souvent utilisé pour décrire la
recherche. M. Nestel a déclaré qu'il ne fallait pas blâmer les
écologistes et les agrostologues à cause des travaux complexes qu'ils
effectuaient dans leur domaine de spécialisation, ce qui correspondait
à leur rôle au sein d'une équipe de recherche sur les systèmes. Il
fallait déterminer le niveau auquel il fallait faire intervenir les
spécialistes des sciences sociales pour se pencher sur les avantages
éventuels, si tant est qu'il y en ait, à tirer de la mise en oeuvre
d'une voie de recherche particulière.
Le Dr Akilu a souligné que la gestion des parcours comportait des
effets socio-économiques et politiques. Les chercheurs et les
vulgarisateurs ne pouvaient pas tout seuls résoudre ces problèmes. Il
fallait l'intervention des politiciens pour prendre des décisions
conformes aux aspirations des éleveurs.
M. Thomson a demandé si l'on avait toujours besoin de mener les
expériences sur une station de recherche; ne pouvait-on pas aller
directement aux producteurs et effectuer des essais au niveau des
exploitations agricoles? Il y avait souvent des informations abondantes
sur les techniques qui visent à résoudre certains des problèmes
auxquels les producteurs ont à faire face. Celles-ci avaient été
rassemblées au cours de la phase de diagnostic. Etait-il nécessaire
de refaire ces travaux? M. de Haan a déclaré que cela dépendait de
la situation. Quelquefois, on pouvait sauter la phase expérimentale
et aller directement aux producteurs, mais d'autres fois, cela
n'était pas possible.
Le Dr Ngutter a souligné que les travaux entrepris par les écologistes
sur la végétation et sur les fourrages n'étaient très souvent pas
utiles pour le planificateur du développement à moins que de telles
- 149 -
informations puissent aider celui-ci à déterminer les avantages
techniques éventuels qu'il y a par exemple a élever du petit bétail
au lieu de grands animaux ou vice-versa sur un parcours déterminé, avec
une composition botanique donnée, sans affecter de manière négative la
composition botanique.
Le Dr Zulberti a demandé à M. Hiernaux si les systèmes de pâture
bi-hebdomadaire ou quadri-hebdomadaire mentionnés dans son étude
pouvaient être appliqués ou recommandés dans des zones où la variation
annuelle de la pluviométrie était de 50 a 200% de la moyenne et lorsque
les éleveurs se déplaçaient d'une zone â l'autre en fonction de la
disponibilité des pâturages. La recherche sur les pâturages tournants
ne pouvait-elle pas être mieux orientée vers l'étude de telles
situations? M. Hiernaux a déclaré que les propositions de pâturages
tournants concernaient les pâturages de l'ensemble du delta du Niger
dont la productivité était élevée. Ces propositions étaient basées
sur les résultats obtenus sur les sites expérimentaux. Des essais
étaient en cours sur le terrain pour tester l'applicabilité de ces
techniques au système pastoral existant.
Le Dr Barry a souligné que la connaissance des ressources naturelles
entrant dans l'affouragement de l'animal était importante pour le
maintien de l'équilibre écologique souvent très fragile dans les
pays du Sahel. Les peuples pastoraux et leurs troupeaux étaient dans
une situation précaire en raison de la variabilité de la pluviométrie,
de l'expansion des cultures commerciales et des méthodes d'élevage
traditionnelles. L'inventaire des ressources naturelles devait
s'appliquer à la situation réelle des populations pastorales; ainsi
les facteurs socio-économiques devaient être pris en considération.
Le problème était de savoir comment amener ces populations à repenser
leurs méthodes de gestion des parcours à la lumière des longues
périodes de sécheresse et des variations considérables de la pluvio
métrie .
- 150 -
Livestock productivity and management
1 2
R. Trevor Wilson and P. Semenye
Team Leader, Arid Zones (Vest Africa) Programme , ILCA Mali
2Animal Scientist, Arid Zones (Eastern and Southern Africa) Programme ,
ILCA, Kenya.
Introduction
Effective and realistic planning for total development in Third World
areas requires not only that baseline data be obtained relating to the
existing situation but also that future trends are predicted with a
reasonable degree of accuracy. While it is possible for economists
to derive output figures using a variety of methods for "with project"
situations it is usually the "without project" and the "base" situ
ation which are suspect, as often the base utilised is little better
than an educated guess.
Of all the elements utilised for the formulation of an
integrated development plan it has normally been the livestock
sub-sector which has been the weakest link. Perhaps there have been
good - even almost valid - reasons for this. The traditional reluc
tance of pastoralists and livestock owners in general to enumerate,
or to have enumerated, their animals and the mobile nature of the
assets in question have made accurate or even reasonably approximate
estimations of total numbers almost impossible to obtain. Indirect
methods of estimation such as those based on returns for tax, marketed
output, sales of hides and skins are also notoriously inefficient.
Aerial surveys have to some extent alleviated this problem but is
not only at the population or regional level that errors are apparent:
parameters at the level of the herd and of individual animals have
also been generally unknown. In the latter case, estimates of pro
duction - birth and death rates, growth, offtake - have been based on
factors derived from usually non-comparable situations in more
developed economies or data from experimental or livestock breeding
- 151 -
units where managment for "improvement" of local breeds (usually by
crossing with exotic, i.e. highly specialised and non-adapted stock)
takes place.
In recent years methods have been developed for the more
realistic estimation of livestock parameters in traditional systems.
These relate to estimates of total livestock in a given area as well
as to the determination of individual and herd production character
istics. By combining the two sets of data, precise figures for the
total livestock output of an area can be derived, with known
confidence limits. Once this point has been reached it is feasible
to manipulate the data for predictive purposes, taking into account
natural factors such as rainfall variability and effects of disease
and anthropogenic factors such as responses to market stimuli and
improved veterinary control.
Evolution of survey methodology
Low-level aerial surveys are particularly suitable for the enumeration
of total animal numbers related to both time and space. Sequential
surveys at fixed or varying intervals can be used to establish short-
term changes in livestock numbers, trends in total effective size of
the herd and seasonal or longer-term movement patterns.
The method is particularly suitable for the determination
of total animal numbers and, provided a series of flights over time
is flown, can also be used to establish the gross dynamics of an
animal population. It is not possible to obtain precise data on
recruitment (birth) and mortality rates from aerial surveys, nor is
it possible to establish population structure or sex ratios with
any accuracy - although in early surveys attempts at this were made
using size criteria based on photographic samples. Thus for estab
lishment of population dynamics and demography, as well as for
individual growth rates, production parameters for meat and milk
production and levels of offtake, some sort of ground survey is
necessary-
Until the early 1970s livestock production data for
traditional societies were almost invariably based on the work of
anthropologists and sociologists whose main interests were outside
- 152 -
the field of livestock production. Otherwise data have been compiled,
often by economists who themselves have usually had little experience
of livestock production, from questionnaire surveys. In general it
can be considered that both socio-anthropological and formal ques
tionnaire methods provide too little and too inaccurate data of the
kind required to established livestock population and production
parameters. There are exceptions, however, some of these types being
adequate for certain data: the good ones generally result from the
quality and knowledge of the observer rather than from the raw
information given. In addition (but perhaps to a decreasing extent
as social and educational horizons expand) formal questionnaires may
induce a response block in the more conservative pastoral societies.
There is no substitute for physical inspection - and physical contact
with - as large a number of animals as possible. Once the livestock
owner realizes that the researcher has a real interest in (and an
ability with !) the animals, then considerable information, as opposed
to the "data" of an elicited response caused by a questionnaire, is
usually forthcoming.
Thus, some years ago, attempts were made to establish a
"zootechnical" method of survey which would provide data of the kind
and of the accuracy required for detailed development planning. The
method, which is described fully in the next section, was primarily
designed for use with cattle, sheep and goats. It has also been used
in camel and donkey studies. Integrated development plans for several
semi-arid areas in northeast Africa have been formulated using live
stock data from these studies. These include plans for Darfur in
western Sudan in which livestock data were obtained from transhumant
Baqqara Arabs and indigenous sedentary cultivators; and for Tigre in
northern Ethiopia with data from short-cycle transhumant Afar in the
Rift Valley and sedentary Tigrean cultivators on the Ethiopian Plateau.
In West Africa a study of agropastoral systems forming part of the
ILCA semi-arid and arid programme used data obtained from transhumant
Fulani and sedentary Bambara cultivators in central Mali. Further
studies forming part of the ILCA subhumid and humid programme and of
their monitoring programme are under way in Nigeria and Kenya
respectively.
- 153 -
The ground survey
Three levels or depths of intensity of survey can be recognized:
1. pre-survey
2. initial survey
3. continued survey
The ultimate level of survey will depend on the kind of
information required and the use to which the data will be put. It
should be noted, however, that it is a prerequisite of the method that
the earlier stages must be carried out. It may be possible to tele
scope a continued survey either with one or with both earlier survey
stages but usually some initial work will need to be done.
Pre-survey
A knowledge of the natural environment, the social groups and the
general pattern of livestock distribution and ownership is fundamental
to any further survey. Initial stratification can be carried out
using LANDSAT imagery and (if an integrated aerial/ground survey is
envisaged) aerial reconnaissance. Some time, however, must be spent
on the ground in all parts of the survey area, preferably as part of
multi-disciplinary team. This should comprise at least the animal
scientist, a range ecologist and a social anthropologist or socio-
economist. Unless one of these is thoroughly familiar with the area
in question and/or the local language a guide/interpreter will be
required and in any case may be desirable or necessary for political
reasons.
During this phase it should be possible to select represen
tative areas and representative villages or other units for the initial
survey stage. An outline of the subsequent stages should also be
given to the probable participating units.
If an aerial census to determine total numbers is not to
form part of the subsequent stages of survey then available statisti
cal data should be collected at this point. In any case such data
may well be useful for comparative purposes and also may help to
establish household and individual ownership patterns. Market and
slaughterhouse figures also need to be collected.
- 154 -
At this point, however, we think it' would be useful to
illustrate some of the problems which are inherent in the published
data relating to these aspects. The FAO Production Yearbooks are the
most commonly quoted source for livestock numbers. These rely to a
considerable extent on national authorities for their data, although
they often make adjustments. They give little idea, unfortunately,
of the distribution of animals within countries and among people in
these countries. Some preliminary work may well be required to.
determine the importance of livestock in the area being studied - not
simply in terms of numbers but in the contribution of livestock to
the overall economy. ILCA is using a classification of "pastoral"
for people obtaining more than 50 % of their gross income from live
stock and "agropastoral" for those obtaining between 10 % and 50 %
from animals. Another problem is the continuing lack of importance
attached to small ruminants by national authorities and development
agencies. Yet in the seven countries which occupy the semi-arid belt
of northern tropical Africa from the Atlantic to the Red Sea, sheep
and goats together account for 15.5 % of the standing domestic
ruminant biomass. This is shown in Table 1 which also includes
figures for Kenya. In some statistics small ruminants are virtually
ignored altogether and we can quote here the Office du Niger in Mali
which, as Table 2 which is taken from the census records shows,
indicated only 12.0 % of families owning goats. The real situation
(Table 3) is very different and this table also shows another aspect,
not evident in Table 1 except for Niger and Kenya: in the semi-arid
zones, particularly in the agropastoral areas, goats are generally
of much more importance than sheep numerically and, on account of
their potential birth rate, are probably creating an ever widening
gap. Ownership patterns and flock sizes are one aspect that would
be useful for us to look into during the field studies.
155 -
Table1.Theimportanceofshe pandgoatsselectedAfricc untri s
ON
sheep&goats
per
inhabitant
Numberof
0.-3 6.41 2.13 2.10
1.N
2.13 1.65 0.-3
1.6-
No.ofsheep toonegoat
Ratio: 1.-5 1.56 1.04 0.40 1.00 1.31 1.35 0.-0 1.16
sheep&goats
Numbersof ('000head)
2-05 -200 114N -360 450-
N644
40N
-3-5 112-05
sheep&goats as%oftotal
Biomassof
12.2 2-.2
22.- 21.-
10.2
13.-
14.4
-.- 14.-
Senegal
Mauritania
Mali
Niger Tchad Sudan
Ethiopia
Kenya TOTAL
« P
W
£ 4i
•ri CO
Si
CO CO
tt•»i f *
+i Jl
to !h
•*i o■fa +i
5 CO
CO a,
rO £
a Cn•^ «
o•<^ •r>
S-, t-i
^ H
o «
Q i>»
+i
g
§2 s•*» w
'tl o
fc
O w
« rS1
t) -J^>
ti
s
^
•*s
•<J to
'Ti $^
i~J Q)
5 to
Ss
rV CO
n rQ
o 3
■p CO
CO
« w
» O
V v
M K
•
cu
r-l
X
CO
H
a o
•H CO <t CN H CI CM N CM CO CI CN CN CN CN CN
U 1 1 1 1 1 1 III 1 1 1 1 1
0} OJ i-li—1 i—t r—4 t—1 i—li—1 r-l r-l r-l 1—1 1—1 1—1 1—1 1—1 I—1
60 X>
CO c §
>> cfl 3
<u u a
c
o
Q 60
c
NO^ui-*mo>r<oco O i—i r-i r^ cti r~«
HO^oiOHNcn^mo O cn oo -* O a\
O incocivDNinifl^cnm vo ~* <f in r~ vo
s
•H CO O
U moommvoooom 00 o
0J OJ HvoCNinnvOCNtmoH r-l r-l St 00
60Xi I I 1 1 1 I I 1 1 1 III IIs e CNWlAinCNMinCNCNH O m m r-i r-i
10 rt 3 i—i
4-1 u c
cO
O
o
60 OOcr\cy,r-~omsi-r^Ln vo cn o oo oo cn
a
OOr^mmmoosrooo oo sf o cn r-i m
>
cooooor-~coi-^coincncn cn cn
o
.5 ■
2 5 O O O in m \o O60 -9
M q
m vo cn m co vo oo
1 1 1 1 1 1 1 1 1
r-l CO 00
1 1 1 1 1
a)
r-l m CO in CN CN CN CN in CN r-1 r-l
JS
tn
60
C
OOoosrr^mOOcnr-i O <r r-~ o cn cn
5 OOrHCNinr^oOcnm Oh H On«
0 n o\ co r~ oo n cn cn
c-1
•r4 M O O O oo o> m
OOOOmOcnOcnsj- <c si- m cn cn i—i
Nin^rocMinHrncnH r-l
60 -e
c a
I I I I I I I I I I 1 1 1 1 1 1
01 M c
i—Ir-li—1 CN i—li—li—1 CN CN i—1 CN CN tN r-l r-l r-l
1-1
4J
4-1
td
u
60 voOoo<fo>Ocsimr^r-( r^ CT> cm o -* t>i
i—lOmcnOmvocnoocn m cn m cn ^o cy.
i— or-^r^oor^ONincncn 00 vO 00 00 o> 00
s r-i
o
/*-\ /"\
co «*
T3 /-S r-l
r-i CT> /'V i—1
o /-v CN /—n v£> v-'
~-**, X s-\ en ^^ i—i cn ^s
S 01 0) O <*1 CN/-\^'S^NtyV
OJ 60 CO /-vr-lwcO^'O O CM CO
4J Ct) 3 CN -^ 3 St CO 60 v£> *~'
CO r-l O vO O cd 0) v—'03^— /—\ /—^ /"S /•^ /—\ /^N
p*. r-l J3 ^o^ede: 30 ct) m cn r^ cn m ct
CO •i-l 3 cri 3 -i-l 60 O J3 CUtj cn r^ cn m m cn
XI > H-l CN OOco-H33cflCUr-icO
4-» 60 g co c 60-h u a -i-t s
^-^ s^ \^ \^s \^ »»_•
3 o o
CO CD O tO -H CU -H X! -r-4 - 4Jcdcn VD CTv in i—1 r-l v£>
o r-l MsiWHWHwZZW <U 5S z ss S5 oa pq
z r-l O
^W' •H -i-l
S as
to
3
co
3
OJ
o
u
1)
60
3
13
OJ
a
•H
14-1
rJ
CW
B
aj
tn
>.
CO
X
3
co
CD
O
•H
CO
3
co
3
OJ
o
to aj
OJ >
•H •H
CJ 4-1
OJ tO
D, >-i
to u
to
r-l •r-l
r-l 3
nl •H
e
rJ Tl
o <
cw
B
TJ O
0) M
T) IH
3
r-l rd
CJ OJ
X 4-1
OJ to
>.
61) to
3 •O
•H 3
-a CO
r-l
o 4J
rC OJ
r-l
r-l r-l
•t-l ■r-l
z as
-1 CN
- 157 -
Table 3. Ownership patterns of sheep and goats in an agropastoral
area in central Mali
Irrigated rice
subsystem
Rainfed millet
subsystem
Goats Sheep Goats Sheep
Number of owners
Number owning sheep or goats
Number owning goats but no sheep
Number owning sheep but no goats
Mean flock size
. 2
Mean flock size
Range in flock size
26
27
15 16
16
12 7
1 0
9.0 6.4 38.2 7.1
9.3 11.5 38.2 12.6
0-23 0-64 2-91 0-58
Of all owners, i.e. irrespective of whether the holding of one
species is nil.
2
Of only those flocks in which animals are held, i.e. nil holding
excluded.
Initial survey
This stage is designed primarily to provide data on population struc
ture and individual animal growth rates. From these parameters it is
possible to postulate instantaneous individual and flock output
without any additional knowledge of birth rate, death rate or actual
levels of offtake. Indeed some of these parameters may be deduced
from the population structure at least as accurately as they can be
extrapolated from questionnaires.
Large numbers of each animal species are required to
establish realistic population means with a sufficient level of re
liability. In real terms this will probably involve not less than
1,000 head of each species (in Darfur data were collected in this
phase for 5,600 cattle, 1,900 goats and 1,200 sheep and in central
Mali almost 10,000 smallstock were involved) . Population classes
based on dentition - 4 pairs incisors worn, 4,3,2,1 pairs incisors
- 158 -
and milk teeth - while subject to some reservations, are likely to
provide data on demography just as valuable as can be obtained using
the more or less subjective ages in years given by most owners. For
the class of milk teeth some reasonably accurate age indications of
younger animals can be expected and can, with experience, be utilised
in conjunction with tooth growth and wear to establish sub-groups
within this class.
Fig. 1 shows the kind of population data which can be-
obtained using a questionnaire technique (this one was done in Tchad)
compared with that obtainable from a "handling and mouthing" exercise
(this was done in Mali) and relates to sheep. Fig. 2 shows data for
cattle at two extremes of the range of systems likely to be encoun
tered in the traditional sector, both these being obtained by the
latter method. We think it probable that for Mali we have sufficient
population data of this kind but it may be worthwhile doing some
demographic work in other countries. What might be useful in Mali
in the rice subsystem of the agropastoral sector would be to compare
the oercentage of work oxen in large herds with those in smaller ones.
Sample selection within the broader ecological/social strata
established during the pre-survey poses some problems. A frame
suitable for randomisation of the sample is unlikely to exist given
the differences which will probably be found in ownership patterns
and flock sizes. Some subjectivity might need to be accepted, after
the primary stratification governing locality and group size has been
adequately taken into consideration. A representative sample should
be the aim.
At this stage considerable additional information on the
animals may be obtained at little extra cost. In all cases this
might include details of external morphology: such physical data may
have little direct relevance to the primary objectives of the study
but nevertheless may well provide useful comparisons, particularly as
existing descriptions of African livestock, and especially of small-
stock, are generally inadequate and often of condiderable antiquity.
In some cases, and certainly where a continued survey is not envisaged,
it is worthwhile attempting to obtain individual reproductive his-
- 159 -
Transhumantgoats-Tchad
Transhumantsheep-Mali
I—r- N50
Nales
2-.3%
—T— 10
"E
chrAn
7*
6-7 5-6
5
A5 N4 N3
1-N 0-1■+-
0
Females
-1.-%
-T N0
105N5
Percentageofallanimals
 
Fig1.Populationdemogr phyofSaheliansmallstock
ONo
I I
1ede1taryassocia edwith irrigatedrice-Mali
Males
6N.8%
8GE
8HZ
Fe.ales
a.N%
20,,
Transhu.ant+se1dentary associatedwithrainf
.illet-1udan
aales
31.N%
AGE
8HZ
,,.8
8ercentageofall1i.als FigN.Sexanda ecompositionfthhr eprincipaldom stic
1I' ioe
Fe.ales
68.8%
,,
,
a
speciesresentedforslaughterNiono
tories of breeding females: if such data can be obtained then it is
possible to calculate such parameters as birth rate, frequencies of
multiple births and parturition histories.
Tables 4 and 5 illustrate the kind of information which can
be compiled with a lot of perseverance and not a little luck. Where a
continued survey is expected these latter parameters are probably of
too tenuous a nature to be worth the investment in time required at
this stage.
Table 4. Data on kidding for 42 flocks of sedentary goats in the
Niono area, (calculated from initial survey)
4 pairs 3 pairs 2 pairs 1 pair Milk All
Parameter incisors incisors incisors incisors teeth goats
Number ir sample 180 77 61 111 15 444
Type of birth
Triplet B 8
Twin •3 11 4 1 219
Single S7 139 81 77 5 759
Total births c68 150 85 78 5 986
Total kids born 887 161 89 79 5 1221
Average litter size 1.33 1.07 1.05 1.01 1.00 1.24
Parturitions
Mean 3.71 1.95 1.39 0.70 0.33 2.22
Mode 3 2 1 1 0 1
Range 0-10 0-5 0-3 0-1 0-1 0-10
- 162 -
in
QJ
H
'/>
M
a
0)
>>
14-1 CJ
c >
•r4
u 2 •U
CJ o a
& u 3
c X)
3 0
2
a.
OJ
CJ
>—
tO
'J
'—
o
1)
-O
S
3
2
OJ
CJ
>
ctJ
U
o
u
CO
QJ i->
60 C
< a
^ lT| r-» O CO co r^ m r^
CO o co in CO «* ~* <* a\
CM oj n CO —l cn vD
oo CO vO ^£> CO CO O CO
CN OJ U0 00 OJ 00 UO CN
1—1 r-^ ^-i CN ^
r^- CO -c CO r^
01 X CJ l-~ m
i—' 1—1 OJ r^ r-i
o
o
^j0
CM
oc
CO
oo
OJ
lA
Osl
CO
-* r-
CN .—I i—I O 00 CO
CO ,-H
cn oo o oo r-~
i—I CM
-* co r-i oo r>» cn
i—i <r m i—i
i— m nC co oo
tN OO m CN
cn
r^-ommoocnvocn
OO 00 O ^f CN r-l
i—I 1-1 CM r-t
r-i
I
CN
I i
-a- LA vO r>» 00 oo ^H
o
1 ! 1 I 1 1 1
I**
1 I +
^j CO UO CO OO o
co
vC
C
CN
3
H
C
H
- 163 -
In addition to normal help from the livestock owners a team
composed of the animal scientist and a recorder is the minimum re
quired. Providing there is a degree of confidence between these two,
that both are aware of the descriptive "shorthand" being used and both
understand the principal language of communication within the studied
community, good results can be expected. Up to 300 head of smallstock
or 80-100 cattle can be processed in a five-hour session.
A complete "initial survey" of this type might be expected
to take about six to eight man-months occupying a time span of four
to six months including full data analysis and report completion.
Continued survey
The continued survey is designed to provide full and reliable data
over the whole range of production parameters. These include birth,
death and offtake rate (the latter by components such as sales, gifts,
ritual and emergency slaughter) , individual weight gains by season
and year, meat and milk production and flock dynamics. Individual
identification of ani • _s is necessary. The most convenient method
from the point of view of the survey is by coloured and numbered
plastic ear tags. The application of these can be expected to create
some consternation and initial resistance in most societies, but this
has so far not proved too difficult to overcome. It does again,
however, condition sample selection and the aim should be a represen
tation of the generality of the areas. A minimum of ten units for
each species should be aimed at with not less than 300 animals of
each species being recorded initially. Very small units and
atypically large ones should be avoided. If the latter are selected
they may lead to inaccurate estimate of offtake, particularly if they
are operated on a commercial rather than a subsistence basis and if
the study is aimed at establishing production and identifying
contraints in the traditional sector.
At this first visit when each animal is identified, as much
information as possible on its physical characteristics and its
relationships to other animals in the unit should be obtained. All
this information should be transferred subsequently to the back of
an individual record card.
- 164
Recently we have analysed here in Addis Ababa, with the aid
of the ILCA computer, some three and a half years' data for Mali and
about two and half years' data for Kenya. Data from the highlands
programme and from the programme in Nigeria have also been, or are
in the process of being, analysed. So we now have considerable
expertise in field data collection and analysis of results including
programmes for analysis and modelling.
Physical and family relationship data enable positive
identification to be made should ear tags become accidentally lost;
family data are also useful in establishing the most prolific and
productive blood lines.
Subsequent visits need to be made at fairly short intervals
(two weeks or less) in the early stages of this phase. All events in
the flock in the intervening interval should be recorded and sub
sequently entered on the individual card. Newborn animals need to be
identified, cross-referenced to their dams and weighed: relevant
details should be entered on the dam's card and on a new card for
the young.
A canvas sling suspended from a spring balance of the dial
type is suitable for weighing smallstock but some kind of crush and
weighbridge will be needed for cattle. A crush is also required for
adult horses but a suitable sling and a 200 kg dial can be used for
donkeys and foals. Camel weighing poses problems and undoubtedly
requires some ingenuity (in the Darfur studies "live" weights were
established post-mortem in concurrent slaughterhouse studies) . We
were able to arrive at a formula for estimating the weight of camels
and there is another one available. These are given in Fig 3.
- 165 -
P = 53 TAH
where P = Weight in kilograms
T = Girth behind the breast pad
A ■ Abdominal girth over the hump in metres
H = Shoulder height
Y = 5.071 X - 457
where Y = Weight in kilograms
X = Girth in front of breast pad in centimetres
(best taken with camel in squatting position)
Fig 3. Estimation of camel weight from linear measurements .
The main problems in the early period of this stage are
associated with getting the livestock owner to appreciate the need
for full reports of all events. Outgoings in which the tag is
returned to the recorder are less of a problem than incomings. The
principal deficits in the latter respect are failure to report deaths
of newborn animals in which both birth and death have occurred
between visits; failure to report stillbirths or abortions; and
failure to report deaths of one or more siblings of multiple births.
It is not necessary to check every animal physically at each visit
but whole unit checks need to be carried out at regular intervals to
identify any discrepancies.
Once some confidence in the ability of both the researcher
and the owner to record and report all events has been achieved, the
frequency of visits can be reduced to something between four and six
weeks. Longer intervals may be acceptable where there is some
literacy in the group being studied and where a group member
undertakes to record data.
- 166 -
In order to achieve worthwhile and reliable results which
can be used for productive purposes or as aids to biotechnical
innovation this type of study should take a minimum of three years.
Such a period would minimize seasonal effects in the results other
than in exceptional cases. Normally, some useful indications of
production parameters would be forthcoming after the first 12 to
18 months.
A programme involving the three main species of domestic
livestock in 30 units totalling 1000 head in the region could be
expected to occupy a technician more cr less full time, with super
visory scientific and analytical work including report writing, in
the region of three man-months per year.
A comparison of some results from initial and continued surveys
In both Kenya and Mali in 1978 we carried out initial surveys of the
type decribed here with the intention in both countries of going on
to continued surveys. How much and what kind of information have we
obtained from the latter and how does it differ from that obtained
in tht former ?
For population structure the initial survey obviously is
the definitive one as we generally use much larger numbers of animals.
We have already had some examples of this in Figs 1 and 2 and we have
tried to indicate how population structure is affected by management
objectives.
We have also said that for establishing weight or growth
curves the initial survey can be satisfactory. This is true, however,
only for certain limited objectives, for example to find a general
ised growth curve and mature average weights. Fig. 4 shows direct
comparisons for Kali for initial and continued surveys for sheep and
goats from birth to 18 months. There is quite a lot of correspondence
but also some divergences. The main problems are at the older end
of this age range where, either the ages in the initial survey were
underestimated, or they were weighed at a time of the year which was
particularly favourable to them. What Fig. 4 cannot show, and what
can only be obtained from a continued survey, is a breakdown of
weights by type of birth and parity. With a continued survey we can
- 167 -
60
60
•H
>
 
— — Initial surveyCon nued survey
l r-
8 9
~i 1 1 1 1 1 1 1
10 II 12 13 14 15 16 17
Age (months)
Fig 4. A comparison of growth curves for the same population when
estimated from a single survey and calculated from long-term
data.
- 168 -
also have weight curves by season of birth among other things.
Initial surveys do not give seasonal weight variations either:
an example of these for cattle, using the same age classes by den
tition that we might use in an initial survey, is given in Fig 5.
This shows the magnitude of seasonal weight changes indicating the
main stress points, and where, for example, supplementary feeding
needs to be considered and to what classes of stock it should be
given. Continued surveys can also give a much better idea of the
variations in weight for age as shown in Fig 6.
Turning to reproductive traits, we saw in Table 4, taken
from an initial survey, the kind of data it is possible to extract.
Comparing tables 4 and 6 it can be seen that one reproductive trait
is fairly close for the goat in both surveys - the average litter
size. We also came very close, at 1.06, for sheep in the initial
survey although, as we have implied earlier, we think we were lucky.
What the initial survey will not tell us is the age at first partur
ition (although we can get some idea of this from dentition/partur
ition data) , the parturition interval or the distribution or
parturition intervals in relation to parity. The continued survey
will show us this, as can be seen from Figs 7 and 8; Fig. 7 also
shows the increase in litter size with parity, something which can
be deduced but with less accuracy from an initial survey as table 4
illustrates.
One of the principal drawbacks of an initial survey is the
lack of any firm data on deaths and offtake. Acisin it is possible to
compile some figures from population structure and individual breeding
female histories - total animals born, now in flock, died, sold etc.,
but these are subject to considerable errors of recall. A continued
survey enables us to establish precise figures for these parameters.
- 169 -
35Q-
340-
330-
320
310
300
290H
280
270
260
250
240
100
3 Female > pairs permanent ™.
IV Male » incissors
 
i i i 1 1 1 1 1 1 1 i 1
JFMAMJJASOND
Month
v//////mmm^m
cold hot late dry rains
dry
post cold
rains dry
Fig 5. Seasonal weight changes in cattle
- 170 -
uo-iivjndoddeei{svaojdcuznowcnoaB■pdq.vinoivomsmfrienivnp%aypwiJutDdB .dq.q.vosa q. moQ*98t£
00N1
0901
i
006
OSZi
0A
0St7i
ooc
I
051
0
'"-.o
6o|)8Sfr
AA
 
-01
A+
-0N
-0£Ofr
■09
<IDKTO$
Il
1-A
-)
Table 6. Some reproductive parameters for sedentary sheep and goats
in the Niono area
Parameter Sheep Goats
A. Age at first parturition
Age (days)
Standard deviation
Range
Number in sample
B. Interval between successive births
Interval (days)
Standard deviation
Ran^e
Number in sample
C. Multiple births: litter size
I. Number (%) of \irturitions
Single
Twin
Triplet
II. Number (%) of young
Single
Twin
Triplet
III. Average litter size (Il/I)
D. Number of young per annum (CII x 365)
B
470.5 484.4
90.5 105.5
341-673 275-739
52 54
253.9 271.3
69.1 79.5
150-491 156-638
225 280
557 (100.0)
528 ( 94.8)
28 ( 5.0)
1 ( 0.2)
587 (100.0)
528 ( 89.9)
56 ( 9.5)
3 ( 0.5)
1.054
1.52
668 (100.0)
521 ( 78.0)
144 ( 21.6)
3 ( 0.4)
818 (100.0)
512 ( 63.7)
288 ( 35.2)
9 ( 0.1)
1.225
1.62
Note: Calculated from data obtained over 3 years of ILCA continued
surveys .
- 172 -
a) Parturition interval
en
CO
T3
CO
c
0)
►J
320
300
280
260
240
220
200'
180- -1—
1-2 2-3
x x x Goats x = 261.14
O o O Sheep x = 240.92
 
3-4 4-5
Parity
—i 1—
5-6 9
b) Litter size
u
u
0)
D8
60
C
3
O
M-l
o
u
0)
1-35'
1-30
1-25
1-20-
115-
l-IO-
105-
100
x x x Goats x = 1.20
O O O Sheep x = 1.06
 
-1—
4
Parity
—r-
5 "9"
Fig 7. Components of small ruminant annual reproduction rate
- 173 -
to
a
o
• H
4->
•.-I
U
3
4J
>-l
CO
CX
<U
ed
a
a)
a
i-i
tD
3b-
34 X
32- X
30- X
38- X
26 X
24-
X
22- X
20- X
18- X
16- X
14 J X
12- X
10-
X
b-
X
6 X
4- (-
X
'£ Jx
x X
X
a) Sheep
s.d.
n
253.4
69.1
341
b) Goats
x
s.d,
n
265.1
79.5
331
XX
 
152 213 274 " 334 ' 395 " 456 ' 510
183 244 304 365 425 486 546
638
(0.49%)
395 456 510152 213 274 '334
183 244 304 365 425 486 546
Parturition interval in days
15-1
14-
13
121
II
10'
9-
8-
7
6-
5
4-
3
2-
PIx
X
s.d.
n
467.5
93.3
75
s.d.
n
466,
99
72
766
-0
x
m
X X
X XX
X
X
m XXX X ^^
""*" XXX xxx „_
287 C x x x xxx ^ 739
275 x x x x x x x x x
X X X X X x x x _ x X
x x x x x X X x x XX ^^
X X X X X x [xjx]X X X X X
365 425 486 546 607 669 730
335 395 456 516 576 638 699
304 365 425 486 546 607 669 703
335 395 456 516 576 638 699
Age at first parturition in days
Fig 8. Distribution of parturition intervals and ages at first
parturition for sheep and goats
- 174 -
The great advantage of the continued survey is its preciseness
and the opportunity it provides to calculate and utilise statistically
significant values. The combination of various parameters such as the
birth rate (a function of parturition interval and litter size) , death
rates, weights of females and young at weaning enable us to calculate
certain production indices. These are extremely useful for deter
mining the effect of management, biological and economic factors on
production. Table 7 shows the differences which can arise between,
for example, species and also between sexes, parities, seasons,
flocks and so on. These factors then indicate, by their magnitude,
their relevance and their importance in any proposals for improving
livestock production. Fig 9 illustrates a possible improvement
pathway for small ruminants. It has the initial advantage of costing
very little in financial terms and should provide real benefits in a
comparatively short space of time.
Table 7. Ratios of comparative advantages for sources of variation
in indices I, II and III in the agropastoral system
Source
Index I Index II Index III
Goats Sheep Goats Sheep Goats Sheep
Sex:
males to females 1.08 1.05 1.04 1.03 1.03 1.05
Parity:
all parities to first 1.54 1.29 1.42 1.25 1.40 1.23
Birth type:
twin to singles 1.37 1.23 1.48 1.45 1.55 1.43
Season:
best to worst 1.17 1.11 1.26 1.16 1.23 1.14
System:
rice to millet 1.82 1.45 1.53 1.31 1.58 1.33
Flocks:
millet best to worst 2.17 2.62 2.43
rice best to worst 9.34 5.30 5.58
- 175 -
f*Identify best flocks in both
sub-systems and determine
management factors governing
output
Preach and practise improved
management in all flocks
Isolate local strains of
epidemic diseases (e.g.
pasteurella) , prepare
autovaccines and conduct
mass campaigns
Study possibilities of trans
fer of rice sub-system bene
fits (water, crop residues,
and by-products) to millet
sub-system
Selection within flocks of
best individual males (growth
rate) and twins for breeding
<^>
t \
Institute seasonal breeding
if worthwhile advantage can
be demonstrated from this
practice at this stage
Further improvement in manage
ment (flock stratification) to
delay age at first parturition
and to prevent parturitions
occuring at intervals of less
than 240 days
-^
^
Veterinary treatment against >». T , . . , ,
- , , .? CVw. Individual veterinary treat-external (and external) ^O^^r
parasites B
ment
Manipulation of flock structures 01 r j- *
,.,,. - n , , n , , Supplementary feeding of out-(culling of old females and sale /v , J , b..
- , , , x of-season breeders and/orof surplus young females) to /-^ •>—1 - . , . ,
.. . . , r CT tT7 fattening for special purposesobtain maximum production of If
young (parturition internal +
litter size)
t. J
Fig 9. Interventions pathways for small ruminants in the agro-
pastoral system
- 176 -
Additional notes on weight and milk recordings
Livestock weighing
Weighing just after daylight on each weighing day is comparatively
easy on the animals and holds variation to a minimum. Weighing of
cattle later than the time they are usually released from their night
enclosure causes a lot of commotion, and should be avoided. Also to
be avoided is weighing when animals are coming back for the night
after grazing the whole day or soon after drinking water. However,
there are situations, as in Kenya, where calves are weighed in the
evenings in an effort to spread livestock activities throughout the
day. Calves in this case are weighed late in the evenings, because
they are left by their dams early in the morning after suckling and
they do not feed until the dams come back for the night. Whatever
time of weighing is picked, the key thing is consistency.
Frequency of weighing
Accurate weights should be obtained at all times. To ensure this the
scales being used should be calibrated frequently and test weights
should be used before any weighing and strategically thereafter.
For calves four weighings are necessary up to the age of two
years. The weights are at the ages of 1, 3, 7 and 18 months. "Birth
weight" should be taken within 24 hours of actual birth whenever
possible. On the other hand it weight at actual birth of the breed
in question is not known at all, it is then worthwhile to measure it.
We have weighed 53 calves in Kenya once every month up to
the age of four months. The correlation of 30 days with that of 60
days is 95 %, for 60 and 90 days is 95 % and for 90 and 120 days is
94 %. Because of this high correlation it is obvious that one can cut
down the number of weighings to two without losing accuracy.
For weighing calves up to the age of 120 days a weighbridge
is not necessary. Up to this age liveweights are easily taken by a
scale of 100 kg hanging from a tripod. Above this age weighbridge
is necessary. An ideal weighbridge is one with a yoke as it facili
tates tagging and mouthing and it restrains an animal for other
measurements. A weighing scale mounted on the back of a four-wheel
- 177 -
drive pick-up is better than one on a trailer. The former can be
moved faster and more easily from one site to another, covering
distances of over 100 km per day.
Two weighings are recommended within the first four months
as this is a critical period of the life of a calf. It is also the
period when calf growth is nearly entirely dependent on milk. It
therefore gives the best measure of milking ability. Weighing at the
age of seven months is taken as the weaning weight, for measuring the
mothering ability. It is not possible to weigh all calves at the
exact age of seven months and adjustment is necessary for calves
falling between six and eight months, to correct them to 210 days.
The final calf weight at 18 months after standardisation
for age differences is a better measure of genetic differences in
growth rate than earlier weights discussed above.
Monitoring of seasonal weight changes
If a measure of the productivity value of the range is desired as
dictated by seasons, mature steers or wethers are the choice. Steers
and wethers are preferred because changes in their liveweights are
not affected by physiological status of pregnancy or milk production.
In some situations one may require to study seasonal effects on
liveweight as affected by specific animal product, e.g. milk pro
duction, in which case the appropriate animal type is used.
Milk yield measurement
Zebu cows are known for refusing to let down their milk without the
stimulus of a calf sucking. So when computing lactation yield of
zebu cows milk taken by calves must be taken into consideration.
Several methods of measuring total milk yield have been
tried: weighing before and after suckling, oxytocin injection, bucket
feeding, partial suckling and liveweight growth rate milk equivalent.
Weighing of calves, lambs or kids before and after suckling
is a very sound method except that it requires a very sensitive scale
able to pick differences in the range of 50 g to 3.5 kg. Such scales
exist but are not suitable for weighing livestock which never remain
■v
- 178 -
still during weighing. The weight difference obtained in some cases
is negative due to the fact that before weighing after suckling the
young animal may have urinated or defecated. This method requires a
lot of labour, so that not many animals can be recorded and a very
small sample may to be taken, leading to unsatisfactory statistical
analyses .
Injection of oxytocin to stimulate milk let-down is possible
only on a research station. In some countries, e.g. Kenya, oxytocin
can be administered only by a qualified veterinarian. How many calves
and sampling times is it possible to study under free range con
ditions ? Bucket feeding and partial suckling methods are related.
In order to bucket-feed partial suckling is necessary first to
stimulate milk let-down and calves have to be taught how to drink
from a bucket. Complete milking-out under-estimates total yield
from milking plus suckling by as much as 18 % (Amble et at , 1965).
Calculation of the amount of milk equivalent to observed
growth rate is good only for the first three or four month of the
life of a calf, that is before it starts grazing. Necessary formulae
for converting liveweight and growth rate to milk equivalent are
available (Konandreas and Anderson, 1982; MAFF, 1975). The remainder
of the lactation yield is best estimated by computing the net energy
available for milk production, combined with milk offtake measurements.
With this method yield evaluation of many cows is possible and in
Kenya the method is used with over 600 cows. A cows are known to
attain their peak production within the first four months of lactation
this method is therefore adequate in estimating maximum milk yield
potential .
Number of milk measurements and yield rating
Milk offtake is normally recorded twice a day. Pastoralists normally
milk twice and, among the Maasai, approximately 80 % milk twice and
20 % once a day.
With the Kenyan pastoral tribe of Maasai a paired 't' test
was conducted between a.m. and p.m. yields on 2,939 milking days.
The difference between the two means (30 ml) was significant
(P 0.05) but not large enough to matter. It is therefore suggested
- 179 -
to do only one milk offtake measurement per day. This will cut down
the cost of evaluating zebu cows' potential for milk and beef pro
duction, leading to increased number of such evaluations for improve
ment purposes.
The minimum number of measurements is four per month,
preferably two in the morning and two in the evening. These should
also be made on days of average activities, rather than when cows
are on heat or following vaccinations or dipping.
Pastoral tribes rate their cows for their yield potential
by simple classifications into good, average and poor on milk offtake.
This classification is specially applicable to cows with a previous
lactation. Cows on their first lactation are rated provisionally on
their current yields. On analysis of 2,301 records, actual production
was found co conform with the yield rating of the owners; good cows
gave 1.2P, average 0.96 and poor 0.86 liters per day. The differences
were highly significant (P 0.005) . If breed improvement is envisaged
it would be easy to use pastoral peoples' rating to identify the good
producers, the best of which could be used as dams of young bulls for
progeny testing.
As an added variable to milk offtake it is recommended to
note the number of teats being milked on each measurement occasion,
and also the method of calf restraint.
References
Amble, V.N. et al. 1965. Estimation of the amount of milk sucked by
calves. Indian J. Vet. Sci. 35: 56-67.
Konandreas, P. A. and Anderson, F.M. 1982. Cattle herd dynamics: an
integer and stochastic model for evaluating production
alternatives. ILCA Research Report 2, Addis Ababa.
MAFF (Ministry of Agriculture, Fisheries and Food, United Kingdom)
1975. Energy allowances and feeding systems for ruminants.
Technical Bulletin No. 33, London.
- 180 - <y3
Productivité et gestion du bétail
Résumé
Cet exposé décrit l'évolution de la méthodologique d'enquête, y compris
celle utilisée pour les enquêtes à basse altitude, les enquêtes par
questionnaires et les méthodes "zootechniques" d'enquête. Il examine
ensuite les enquêtes au sol à trois niveaux: la pré-enquête,
l'enquête initiale et l'enquête continue.
La pré-enquête fournit des données sur l'environnement naturel, sur
les groupes sociaux et sur la structure générale de la distribution
et de la propriété du bétail. Au cours de cette phase, il devrait
être possible de sélectionner des zones, villages ou autres unités
territoriales représentatifs pour la phase "enquête initiale".
La phase enquête initiale est conçue tout d'abord pour fournir des
données sur la structure de la population animale et sur les taux de
croissance individuelle. A partir de ces paramètres, il est possible
de calculer la production instantanée des individus et du troupeau
même si l'on ne dispose pas de données supplémentaires sur les taux
des naissances et de mortalité ou sur les niveaux réels de
l'écoulement. Une enquête initiale complète peut prendre environ
six à huit mois/homme (et s'échelonner) sur une période allant de
quatre à six mois, y compris l'analyse complète des données et
l'élaboration du rapport.
L'enquête continue est conçue pour fournir des données complètes et
fiables sur l'ensemble des paramètres de production. Celles-ci
incluent les taux de naissance', de mortalité et d'écoulement
(ce dernier étant composé d'éléments tels que vente, don, abattage
rituel et in extvemis) , les gains pondéraux individuels par
saison et année, la production carnée et laitière et la dynamique
des troupeaux.
L'identification individuelle des animaux est nécessaire. Ce type
d'étude devrait durer un minimum de 3 ans pour produire des résultats
valables et fiables. Une telle période permettrait de réduire au
181 -
minimum les effets saisonniers sur les performances, sauf dans les
cas exceptionnels. Normalement, certaines données utiles sur les
paramètres de production apparaissent après les 12 à 18 premiers
mois .
Il est ensuite procédé à la comparaison des résultats obtenus dans le
cadre d'enquêtes initiales et continues au Kenya et au Mali; des
informations supplémentaires sur les poids et la production de lait
sont données. Il est recommandé d'effectuer le pesage juste après la
tombée du jour. Les animaux en souffrent relativement moins et cela
réduit au minimum les variations. On devrait pouvoir connaître les
poids précis à tout moment et les balances devraient être fréquemment
calibrées.
Les avantages relatifs de plusieurs méthodes de mesure de la produc
tion laitière sont examinés. Ceux-ci incluent le pesage avant et
après l'allaitement, les injections d'oxytocine, l'engraissement à
façon, l'allaitement partiel et l'équivalent en lait du taux de
croissance pondéral. Le prélèvement de lait est normalement
enregistré deux fois par jour.
- 182 -
Animal nutrition
12 3 4
L.J. Lambourne , M.S. Dicko , P. Semenye and M. Butterworth
2
Director of Research, ILCA, Ethiopia
2
Animal Nutritionist, Arid Zones (West Africa) Programme, ILCA,
Mali
3
Animal Scientist, Arid Zones (Eastern and Southern Africa)
Programme, ILCA, Kenya
Senior Animal Nutritionist, ILCA, Ethiopia
Introduction
Wilson and Semenye's paper outlined the procedures followed by ILCA's
field teams in surveying the productivity of livestock in arid and
semi-arid areas. They emphasised that quite large differences may
exist between herds and flocks within the same production system in
contrast to quite strong similarities in mean productivity among the
pastoial systems themselves.
Productivity is largely determined within a particular en
vironment by genotype, health and nutrition. For most of the
extensive pastoral areas we are concerned with there exist no poten
tially better breeds than the present well adapted zebu-based cattle
and the local strains of sheep and goats; but ILCA's systems studies
have shown, even in these adapted animals, the occurrence of serious
losses associated with disease. The nature of arid and semi-arid
grazing lands, with short and unreliable periods of rainfall,
necessarily means that animals and their keepers undergo long periods
of nutritional hardship.
Genotype
Current low reproductive rates and heavy calf mortality prevent any
deliberate selection for productivity in cattle apart from the
rigorous natural selection that this situation implies. In small-
- 183 -
stock the common pattern is to sell or slaughter for meat a large
proportion of the males at early ages - from six months to one year.
In the absence of any deliberate attempts to keep the fastest growing
males for future breeding, it is probable that they are the first to
be eaten, and that the smaller or slower growing are the more likely
to be left for breeding. There may thus be an unintentional, but
beneficial, selection for multiple births, of which the offspring are
always lighter at birth and slower growing. It is possible that
individual herders do exert selection for particular types of animals;
among the Maasai of Kenya for example it is basic policy to promote
both large and small framed cattle chosen from every recognizable
breeding "line" in the herd. This ensures a wide range of variation
in the herd and is held to maximize production in the wet season and
minimize losses in the dry season over the herd as a whole. In ILCA's
Kenya studies Sahiwal crosses are found to grow faster than local
Maasai zebu calves, and the pastoralists appear to be retaining
Sahiwal cross cattle for breeding rather than offering them for sale,
despite the higher prices that they fetch. Free-range grazing on
common land and use of common watering points mean opportunities for
animals to mate with males from another herd or flock and are
therefore inconsistent with a deliberate policy of genetic improvement.
Aprons are used by some pastoralists in range areas to prevent breeding
^ of sheep and goats until the best mating season, and castration of
?> inferior males is a recognized practice. The basic attitudes andmechanisms for genetic improvement are there to be used if clear-cut
methods for identifying the most productive animals for the environ
ment could be developed and ways of exploiting their merit could be
promoted.
Health
The interaction of nutrition and health in deciding overall pro
ductivity has been recognized in all ILCA programmes. Efforts have
been made from an early stage to diagnose the main diseases, study
their epidemiology and impact on production, and test available
control measures.
- li
In 1982 an agreement was reached with the research branch
of the Kenya Ministry of Livestock Development to conduct collabor
ative research on the incidence and impact of animal diseases on
Maasai herds and flocks. Through this arrangement the research branch
has made available one senior and one junior veterinarian. The former
is responsible for designing the research project (contents, sampling,
phasing, analysis and interpretation of data) in consultation with
ILCA scientists and for supervising the junior veterinarian, who is
responsible for carrying out the field investigation in collaboration
with the ILCA team. This component of the research was launched in
June 1982.
In Mali and northern Nigeria veterinarians have been
appointed to ILCA teams to carry out surveys of epidemiology of
common diseases in livestock being recorded in the main programme and
the causes of mortality in animals dying or slaughtered. The wide
spread practice of killing animals in extremis for meat, makes it
difficult to arrive at reliable statistics for the number of deaths
due to disease. In Mali it was found that most deaths occurred in the
overgrazed zone outside the live delta and major causes included tick-
borne diseases and pneumonia-like infectious diseases (Pasteurella,
Diaplococcus) . No evidence was found of protozoal parasites but
sarcoptic mange was common and many samples examined showed the
presence of the internal parasites Monieza, Eimeria, Paramphistomum,
Oestrus ovis, Echinocoecus, or Fasciola hepatica.
The Mali team has now set up a section of Animal Health
and Nutrition reflecting the close relationship of these two aspects
of animal productivity, and research in the present phase will give
more attention to disease and parasites as causes of the high
mortality in young animals, which averages 21% in calves up to 12
months, and 6.2% for the herd as a whole. Preweaning mortality was
as high as 35% in goats and 30% in sheep, varying between pastoral
systems (rice, millet) and with type of birth. Of the total deaths
recorded 16% were abortions, 22% were still births or occurred on the
first day, and a further 14 % in the first 7 days. Seventy percent
of all deaths occurred before 15 months.
- 185 -
Fewer figures are available from the newer Ethiopian
rangelands project concerning disease incidence, but similarly heavy
losses of young animals have been recorded and there can be no doubt
that many are disease related.
In all systems studied, internal and external parasites are
prevalent and tick-borne diseases are major problems. In some Fulani
areas ticks are regularly removed by hand, usually by children;
internal parasites are "treated" by use of traditional herbal prep
arations or by the annual "cure salee" or access to salt outcroppings,
which is believed to have curative properties. Most pastoralists are
aware of the symptoms of common diseases, indeed they also recognize
and classify diseases, by signs and by postmortem indications, into
groupings which can readily be translated into standard Western terms.
ILCA's systems studies have made it possible to assess major
diseases and to estimate their impact on productivity. Work now in
progress will make it possible to evaluate various approaches to
disease control, in economic and in practical terms, and their re
lationship to nutriti and management of the flock or herd. Wilson
(1982, and this wor'p >p) has outlined a programme for improvement of
smallstock productivity involving the upgrading of management, diag
nosis and control oi epidemic diseases, selection of males for growth
rate and twin-born females for retention as breeders, control of
breeding times, treatment, of parasites, careful culling of least
productive females and supplementary feeding for special purposes.
Nutrition
General
The most useful record for nutritional purposes is liveweight since
this is an integrated measure of the nutritional response of the
animal, and liveweight gains or losses give a sensitive indication of
nutritional adequacy or inadequacy. Weighing is possible, given
a weighing scale or tripod and clock face balance and necessary assist
ance, during both initial and continued surveys, as described by
Wilson and Semenye in this workshop, and yields very useful infor
mation. Where weighing facilities are not available, a very useful
- 186 -
approximation may be made by measuring heart (or chest) girth of
cattle and estimating the liveweight from these by means of one of the
published formulae. The relationship between heart girth and weight
depends on breed, and to some extent on condition and sex.
A more robust, easily applied but more subjective method
is to visually assess body condition. This is readily done on large
numbers of animals and, if clear criteria are set for distinguishing
grades of condition, it can be reproducible. Up to five grades are
easily distinguished; it is often claimed that nine or ten grades
can be used by experienced observers but reproducibility within, and
particularly between, observers may be rather poor. The fewer the
grades the more repeatable the grading, but the greater the differ
ences in body weight corresponding to each change of grade. In
frequent watering of cattle is likely to superimpose a two or three
lay cycle on liveweights, as also on visual estimates of condition.
Since individual cattle may take in up to 30 litres of water (30 kg
of liveweight) in five minutes one may question the interpretation
of measurements made either before or after watering. In these
conditions heart girth may be a more reliable measure.
More detailed information about nutrition can be obtained
from samples of blood, faeces, or body tissues. In ILCA studies in
northern Nigeria we have analysed samples of blood to correlate with
the mineral content of range grass and browse species, and samples of
liver to estimate tissue stores of copper and vitamin A in cases where
forage analysis has suggested these might be low. Blood plasma
inorganic phosphate analysis has often been used to assess the
adequacy of supplies of phosphorus, an element closely linked with
animal performance, but recent .work in the semi-arid tropics of
Australia has shown that more reliable information is obtained by
analysis of small samples of rib-bone obtained by simple surgery.
The skeleton acts as reserve of minerals, and bone is readily mobil
ised to maintain plasma concentrations when the P content of the diet
falls too low. We have analysed many samples of 'milk from the
Ethiopian pastoral programme, but milk composition seems to be deter
mined very much by genetic and physiological factors such as age and
stage of lactation. The "let-down" problem in zebu cows means that
- 187 -
it is difficult to obtain a truly representative sample of milk in
field studies. The wide variation we have found (2.5 to 70% fat)
may reflect this problem, but average figures of about 13.5% to
14.0% total solids, and fat percentage increasing from 4% to 6 or
7% during lactation, suggest that our samples are reasonably rep
resentative and consistent and that the Borana cow produces excellent
quality milk under pastoral conditions. Milk quality is of great
importance not only because of the need for good calf growth rate,
but also because pastoral families live on a predominantly milk diet
for much of the year. ILCA's nutritional studies include records of
weight and height for age of representative groups of pastoralists
and their families, together with observations of their food consump
tion and daily work routine. The very hard work of lifting water
from wells for stock comes during the late dry season, when milk pro
duction is at its lowest in a seasonally-calving herd, and when last
season's grain reserves are getting low. The supply of labour and
work capacity per person may limit the possibility of introducing any
management changes neer1 g extra work.
Because of -c self-evident importance of nutrition in all
the pastoral zones was decided early in ILCA's work to do more
intensive research into this component of the system.
Methods used
Amounts and composition of forage available
Liveweight or body condition measures the response of animals to their
nutritional environment integrated over weeks or months. For many
purposes it is necessary to know to which component of that environ
ment they are responding or how the factors of feed quantity and
quality, grazing, watering and walking are interacting.
- 188 -
-50%
faecal
production 3
(kg/d)
2-
digestibility
(%)
75%
<3f' \
.300
<x*
S+++++++ •+******
+ •
+ .'•f •
..•*•8' dry
++ matter
intake
(kg/DMI/d)
9 
liveweight
(kg)
200
-100
Fig 1, Relationship between faecal production (1) ,, digestibility (2) >
dry matter intake (3) and liveweight (4) of cattle.
Figure 1 shows the annual weight changes of cattle in the
agropastoral system of central Mali, and it is easy to distinguish
several main periods of the year :
- the wet season, when rapid gains are made on abundant
high-quality feed;
- the main dry season, when slow but steady weight losses
occur on an initially adequate but diminishing supply of poor quality
feed; and
- the late dry/very early wet season, when weight loss is
often more rapid. This last period may be so serious in its effects
that it is referred to as the 'crise de Juillet'. Several possible
causes for ill-thrift at this time have been suggested, including
leaching of the remaining forage, mycotoxins produced by fungal
- 189 -
growth following the first rains, excessive energy expenditure in
seeking the fresh grass or diarrhoea causing reduced digestibility
and apparent loss of weight.
Table 1 shows results of quadrat Sampling on pastures in the
Fulani agropastoral system of northern Nigeria's subhumid zone, in
which the main rainfall occurs between June and October. Because of
the good growth of tall grasses in this cattle area, the protein and
digestibility figures are generally low, particularly in the dry
period between October and May. Table 2 shows corresponding data
from a grass and browse system grazed by goats and sheep in northern
Kenya. The bimodal rainfall occurs in the period January to March
and in October, and a marked decline in availability and quality of
feed occurs during the intervening dry months of July, August and
September. Because of the high content of browse, the protein content
was notably higher throughout most of the year than shown in Table 1.
Table 1, Composition of forage available to traditionally managed
oattle in northern Nigeria.
Month : MJJASONDJFMA
available DM1 0.7 1.1 2.0 2.0 2.3 2.5 2.2 1.2 1.2 1.1 1.0 1.0
(t/ha)
CP2
(% of DM) 9.4 9.8 8.5 6.8 6.0 4.8 3.0 4.2 4.5 5.4 5.4 5.1
digestibility
(% DM) 61 60 54 50 48 46 42 52 47 53 53 53
grazing time
(h/d) 9.5 7.6 7.1 7.1 7.0 6.9 7.0 7.9 8.2 8.2 9.3 9.4
DM = dry matter
2
CP = crude protein
Source : Bello Sule, unpublished results, ILCA, Kaduna.
- 190
Table 2. Composition of forage and browse available to sheep and
goats in northern Kenya-
Month JFMAMJJASOND
available
DM1 1.5 1.20 1.50 1.20 0.90 0.80 0.70 0.55 0.70 1.20 1.25 1.50
(kg/head)
CP2(% of
DM) 17 21 18 14 22 22 17 14 14 21 26 21
digesti
bility
(% DM) 64 68 64 60 62 55 35 30 40 68 71 67
DM = dry matter
2CP = crude protein
Source : Smith et al (1982).
Preliminary analysis of the availability and nutritive value
of agro-industrial byproducts indicated that, though the oilseed cakes
and meals were excellent supplements for the low-protein pastures, the
quantities and reliability of supply were altogether inadequate for
any widespread use in pastoral areas (Dicko, 1980b) . Emphasis was
thus placed on better management of pastures and crop residues about
which not enough was known, and. a detailed study was therefore carried
out during 1979-80 of the grazing behaviour and feed intake of cattle
in the agropastoral system illustrated in Figure 1. Traditional
herders graze animals during the rainy season (July-October) on the
natural vegetation of the Sahel, largely composed of annuals like
Schoenfeldia graciliss Loudetia togoensis, and Zo'rnia glochidiata,
with Pterocarpus luaens as the principal browse species (Hiernaux,
1978). From November until December, they graze the standing stalks
and residues of the millet fields, combined with some grazing of the
191 -
regrowth of surrounding fields in fallow. In December in the particu
lar production system under study, they are moved to the residue of
the rice fields, with also a substantial regrowth of weeds. If the
rice straw and residues are completely graze,d the herd is moved about
April back to the millet fields, to await the rain in July. Obser
vations presented here were made over one complete year in a locally-
owned herd of 90 cattle. During a five day period each month the
grazing behaviour of four steers was recorded at 15-minute intervals
day and night while simultaneously the total faecal production of
other four steers was measured with a bagging technique specially
developed for free-ranging animals (Dicko, 1980a).
Composition and amount of feed eaten
Forage intake was estimated by the 'indirect' method, in which the
output of faeces was first measured (the indigestible part of the
feed eaten) by means of a harness and collection bag and this then
multiplied by a factor, ratio of total feed eaten to indigestible
part, or feed: faeces ratio. This is numerically equal to the ratio
100/100 % digestibility and the digestibility was estimated in several
ways.
1. By cutting or plucking by hand samples of forage eaten
by the animals. The digestibility of these samples was estimated by
chemical analysis. It is difficult to obtain a sample close to that
eaten by free-grazing animals. By close observation of the grazing
behaviour of each animal, samples could be taken in proportion to the
number of bites seen to be taken from each grass or shrub forage type
in the hope that the total sample thus obtained would be representa
tive of the day's diet,
2. By allowing the animal itself to select its diet, then
taking samples of the material grazed through a large cannula fitted
surgically into the rumen. This means physically emptying the
contents of the rumen by hand before the animal goes to graze, and
then taking samples from the freshly ingested material in the rumen
after 1-2 hours of grazing.
- 192 -
3. A technique used in later work in Kenya involves col
lecting samples of herbage grazed by means of a surgical fistula into
oesophagus, which allows feed being swallowed to pass out into a
suitable collection bag attached to the neck. This is less time-
consuming than the rumen method, and yields a sample which has not
been subjected to the fermenting effect of the rumen's active
bacterial population.
Two options exist for analysis - either in vitro digestion
with rumen liquor or purified enzymes, or else analysis for fibrous
fractions closely related to digestibility and the use of a predic
tive equation such as that of van Soest (1976) .
4. In other pastoral countries it has been observed that a
close relationship exists between forage digestibility and the com
position of the faeces derived from that forage. Data from digestion
trials have been analysed so as to predict forage digestibility from
either faecal fibre components (which are related negatively to
digestibility) or faecal nitrogen concentration (positively related
to digestibility). For che Mali study two equations were used, based
upon a wide range of native and introduced grasses and legumes grown
in northern N.S.W., Australia.
Results obtained
Grazing behaviour
Fig. 1 showed not only the weight changes but also the faecal output,
digestibility and estimated feed intake of the cattle studied. The
mean observed values are given in Table 3, while Fig. 2 gives
grazing times and other details as monthly means.
93 -
Table 3. Intake and performance measurements.
Month Mean Lwt DDMI
2
DCP1 MEI Distance! (km/d)
Lwt Gain kg/d kg/d MJ/d grazing walking
(kg) (kg/d)
A 237 0.75 3.91 .747 57.09 6.01 4.83
S 258 0.28 4.29 .631 60.37 7.84 5.09
0 266 -.14 2.62 .213 35.81 7.28 13.07
N 261 -.52 3.22 .240 42.98 9.44 5.10
D 248 0.13 3.83 .176 49.17 7.39 3.90
J 251 -.13 3.54 .299 43.39 7.77 2.32
F 248 0.08 3.28 .390 44.92 7.64 3.28
M 242 -.34 2.44 .176 29.95 6.93 3.94
A 236 -.32 2.87 .127 35.46 7.85 4.84
M 228 -.12 2.66 .253 34.04 8.03 3.38
J 224 -.15 3.74 .348 48.30 8.62 4.64
J 230 1.15 5.51 1.434 70.72 7.19 6.91
Digestible dry mat': -:_ intake.
2 . . .
Digestible crude protein intake.
3
Metabolisable energy intake.
Grazing time was lowest during and after the rainy season
(August to October) when pasture is relatively abundant and of good
quality. On millet residues and rice straw from November to December
grazing time increased steadily until April to June when biomass was
much reduced. Both in the rainy season and in the dry season signifi
cant negative correlations (r = -0.85 and ~0.92 respectively) were
found between standing biomass and grazing time. Covariance analysis
showed that grazing time was significantly lower for a given biomass
in the period July to October, when forage quality was high. This
is shown in Fig. 3.
- 194 -
rainy season
< >-*r
dry season
natural vegetation millet
P 1
J A
1979
*H H«Istubbl^
rice stubble millet
*K-
1
J F
1980
M
3 fields
M
20-
distance
walked
(km/d) 10-
grazmg
time
(h/d)
 
biomass
(kg DM/ha)
1
pooo- ppp,.
i
f ^^
0looo-
n
Fig 2. Relation between biomass availability and grazing behaviour.
- 195 -
Grazing time
(h/d)
A
10
8-
4-
2-
 
November
to
June
July
to
October
i i 1 1 1 1 1
400 800 1200 1600 2000 2400 2800
Forage biomass (kg DM/ha)
Fig 3. Relationship between forage biomass and grazing time.
Differences in grazing preference are shown in Table 4. The
low amount of time spent on millet residues, compared to that on rice
straw and fallow regrowth, points to the relative importance of these
latter agricultural residue resources particularly for cattle. Sheep
and goats spent far more time on fallow regrowth and on browse grazing
respectively. The distance walked was about 6 to 9 km during grazing
plus 3 to 12 km per day between camp, pasture and water.
- 196 -
Table 4. Estimation of annual grazing time of ruminants on different
types of forage.
Cattle Sheep Goats
Hours % Hours % Hours %
Total time 2 883 100 1 948 100 2 051 100
Time on browse plants 115 4 669 34 1 791 87
on pasture or fallows 1 519 53 1 142 59 215 11
on millet stems 179 6 135 7 45 2
on rice straw and regrowth 1 070 37 2 ■ - - -
Intake
The monthly mean dry matter (DM) intake figures for cattle are related
to the time spent grazing (Fig. 4) in the rainy season, intake was
approximately 1 kg D,M. per hour, in the early dry season approxi
mately 0.8 kg DM per hour, whilst in the late dry season this was
reduced to 0.5 - 0.6 kg DM per hour. The slightly higher values in
December and June appear to be associated with the grazing of rice
fallows. 
0.5 kg/h
grazing time
(h/d)
Fig 4. Relation between voluntary intake and grazing time in different
months.
- 197 -
The N content of rumen samples was consistently higher than
the plucked forage samples (Table 5) , perhaps indicating feed
selectivity even in the dry season. In the month of July, in the main
growth period, both were equally high. Both sets of values showed a
similar trend during the year but the difficulties of copying the
grazing selection of animals gives greater credibility of the analyses
of rumen samples, though these may have been influenced by saliva
contamination (Table 8) . Seasonal trends in protein content and
digestibility (Table 5) were similar to corresponding data from
Botswana (APRU, 1978) obtained by use of cattle with oesophageal
f istualae.
Table 5. Composition of forage eaten by grazing cattle; (a) Botswana
(April, 1978), (b) Mali.
Month FMAMJJASONDJ
(a) CP1
(% of DM) 11. 2 9,4 7.1 6.4 6.2 5.8 5.3 6.2 8.3 - 11.6 10.5
DM2 56 54 51 48 45 39 46 44 53 - 63 55
(b)CP %
(i)3 5.8 3.5 3.7 4.6 7.2 23.1 9.7 8.1 7.7 7.3 6.2 5.2
(ii) 10.6 7.8 6.3 8.6 8.3 23.6 16.6 14.6 8.8 8.7 7.6 7.6
DM % 51 53 53 48 53 62 66 62 48 52 55 55
CP = crude protein
2DM = dry matter percentage
3 (i)= samples plucked by hand
(ii)= samples obtained from rumen
Factors affecting intake
Intake of DM averaged 6.32 kg DM, or 2.6% of the mean cow liveweight
of 244 kg. Linear and multiple regression analyses showed that
- 198 -
forage protein (N x 6.25) content had much more influence on DM
intake (r = 0.68) than had forage digestibility alone (r = 0.48),
though the two were closely related.
Weight changes of cattle and their relation to intake
Regression analysis of liveweight changes showed a correlation of
0.68 with digestible dry matter intake (DDMI) , 0.89 with crude protein
intake (CPI) and 0.89 also as multiple correlation with DDMI and
CPI, indicating the major contribution of CP intake to the relation
ship. Metabolisable energy (ME) intakes were estimated from a
formula proposed for tropical grasses (INRA, 1978) . The linear
AW = 0.034 ME, - 0.450
A+P
R - 0.911
Liveweight
gain AW
(kg/d)
 
Fig 5.
0.034ME - 1 .513
- 0.879
Metabolisable
energy intake
(MJ/d)
Liveweight change as a function of total metabolisable energy
intake (MEj) and of metabolisable energy intake available for
activity and production (ME p) .
- 199 -
regression of liveweight change on ME intake (Fig. 5) gave:
Liveweight change (kg/d) = 0.034 MEI (MJ/d) - 1.513 (r = 0.88)
= 0.O34 (MEI - 44.5)
indicating a total maintenance energy expenditure of 44.5 MJ/d. If
the ME intakes are reduced by the estimated minimum maintenance
0.75
requirements of about 112 kcal/kg ' or about 31 MJ per day the
regression becomes:
Liveweight change (kg/d) = 0.034 (available ME - 13.2) (r = 0.91)
indicating that in this environment the additional energy expenditure
for grazing activities and walking amounted to 13.2 MJ/d, or an
additional 42% of the minimum maintenance requirement.
Evaluation of methods
Some discrepancies were noted among the various methods of estimating
digestibility, upon which this 'indirect' approach depends. Initial
comparison of the estimates suggested that the rumen samples gave
consistently lower values than the rest, and a mean value was used
which excluded the rumen samples, in calculating the values shown in
Table 3 earlier (Dicko et al, 1981). A later analysis showed that
liveweight changes were most closely correlated with intake estimated
via in vitro digestibility of rumen samples (r = 0.80), via in vitro
digestibility of hand-plucked samples (r = 0.78) and via the digesti
bilities calculated from faecal nitrogen content (r = 0.75 for
equation 1; r = 0.68 for equation 2). Estimates are shown in
Table 6. All estimates from the rumen samples were markedly lower
perhaps because of some initial digestion, and faecal nitrogen
regressions gave higher estimates than the rest.
- 200 -
Table 6. Estimates of digestibility by different methods.
Faecal nitrogen Van Soest equation In vitro digestion
Month Eq. 1 Eq. 2 hand rumen hand rumen
(a) (b) (c) (d) (e) (f)
A 63 61 61 46 66 64
S 66 64 - - 54 51
0 56 56 47 42 40 39
N 53 58 53 43 52 38
D 54 57 60 44 53 56
J 54 57 62 47 54 52
F 54 57 58 53 54 49
M 54 57 60 57 51 44
A 57 57 60 58 54 56
M 54 57 47 51 50 50
J 55 57 55 50 56 46
J 77
58.2
77
59.6
63
56.7
61
50.3
71
54.7
68
Mean 51.2
In later work in Ethiopia and Kenya the oesophageal fistula
method has been used. Although this method undoubtedly samples the
feed actually selected by the animal, not all food eaten is collected.
The proportion which comes out of the fistula depends somewhat on the
exact position and size of the opening and therefore differs between
cows. Table 7 shows fairly consistent and significant differences
among four animals given test feeds in known amounts, but no differ
ences between feeds of different physical types averaged over the
four animals. This indicates the need to 'calibrate' animals from
time to time, and to use a small group rather than rely on single
animals. The total weight of forage collected per hour of grazing
time cannot be simply multiplied by the number of hours of grazing
per day to calculate total daily intake, unless the percentage feed
recovery is known and is used to correct the weights collected.
- 201 -
Table 7. Percentage recovery of different feeds from oesophageal
fistulae in four cattle (no. observations in brackets) .
Feed given Percentage recove ry (%)
1 2 3 4 All
hay 59 39 17 53 41 (19)
grass 54 49 16 58 43 (20)
concentrates 30 31 29 84 41 (20)
mean 48 (18) 40 (11) 20 (18) 65 (12) 4. (59)
Table 8. Composition of material collected from oesophageal fistula
compared with composition of feed eaten.
Material DM
(%)
Nitrogen
(%)
Phosphorus
(%)
hay feed 50 0.87 0.21
OF2- total 55 1.00 0.52
- squeezed 51 0.97 0.34
grass feed 56 1.20 0.24
OF - total 58 1.32 0.43
- squeezed 52 1.28 0.33
concentrates feed 61 3.45 1.01
OF - total 68 2.92 1.04
- squeezed 62 2.97 1.01
DM = dry matter
2
OF = oesophageal fistula
Table 8 shows the effect of mastication and addition of
saliva on chemical composition of samples. More saliva is added to
hay than to grass or concentrate samples before swallowing, and the
- 202 -
effect of this on nitrogen and particularly on phosphorus content,
is quite clear. The phosphorus content of hay and grass was more
than doubled by the phosphorus contained in the saliva, which demon
strates the effectiveness of the recycling process by which ruminants
are able to thrive on poor quality feeds for many months.
In current work in Kenya Semenye is using Maasai cattle
with fistulae, and this is the method of choice where precise knowl
edge of botanical, and with less confidence chemical, composition is
needed, and where the large amount of work can be justified. For many
purposes it is now felt that close observation of grazing behaviour
may be sufficient, matched with analyses of forage samples carefully
plucked to mimic grazing selectivity. Additional information may be
obtained by analysis of faecal samples, which is closely related to
the characteristics of the feed eaten, at least in terms of nitrogen,
phosphorus and fibre contents.
The collection of faeces by harness and bag, as used here,
is much easier with animals which are accustomed to close herding and
regular handling than with free-ranging cattle. The standard chromium
oxide method would also be relatively easy with herded cattle, and
would make it possible to work with larger numbers and with young
growing cattle as well as with females. This is now being explored.
An interesting development of this work is the finding that
the average amount of forage taken in per 'bite' or per monthful, is
fairly constant. This has been calculated from careful recording of
the number of mouthfuls taken in a given time; che number of mouthfuls
per unit time or the intensity of grazing, is closely related to the
rate of walking while grazing. This is illustrated for several
conditions in Fig. 6 (Dicko, pers. coram.). At maximum grazing inten
sity an animal takes a number of monthful s per minute characteristic
of the pasture and when its rate of walking is slowest. As grazing
becomes less intense, the number of mouthfuls per minute declines
linearly with increase in walking speed. It is hoped that the
availability of fistulated cattle may make it possible to test this
idea under a wider range of conditions. Number of monthfuls is also
being recorded by a simple automatic device (Semenye, pers. coram.).
- 203 -
Conclusions
This component research study amply verified the nutritional limi
tation to cattle performance during the dry season, and showed that
this had several aspects. The main initial' constraint appeared to
be the low quality of the dry forage which caused the cattle to graze
selectively for long periods and expend extra energy in doing so.
Samples of forage obtained in various ways confirmed earlier findings
that the protein content of the forage was more closely related to
animal performance than was its digestibility. The close relationship
of forage N content to other quality attributes makes it difficult to
assert which one is of primary importance. In tropical pastures it is
well known that animal performance depends on maintaining forage
protein content above about 8% (Milford and Haydock, 1965). In the
present work protein content of forage was below 7.5% for most of
the period November to June during which liveweight losses occurred.
These findings emphasize the particular value of protein
or non-protein nitrogei supplements for grazing ruminants, and help
to explain the bette astained production of animals which browse on
high-protein legumii s plants during the dry season. If it is true
that protein content is the most valuable single attribute, and is
correlated with most other measures of quality, then improvement in
protein content should be sought even at the cost of some reduction
in gross DM production. Biomass protein content can be increased
even within the constraints of climate and soil fertility by manage
ment deliberately aimed at increasing the contribution of legumes,
herbs and browse plants rather than of lower-quality grasses, however
high their DM yield. Where pastoral systems rely on crop residues or
stubble grazing in the dry season, as is common, the inclusion of a
forage legume with the cereal crop combined with minimum fertilizer
input, would both provide supplementary protein to livestock and
enhance soil fertility so as to produce better crop yields and better
fallow grazing in later years.
Because of the great importance of nitrogen content and
digestibility in determining intake and performance of ruminants,
ILCA is collating these and other data on a wide range ol forage and
- 204 -
browse samples. The widespread practice in Africa of supplying salt
in various forms to livestock, suggests that soils and plants may be
deficient in particular mineral elements. ILCA has carried out a
wide sampling of forage plants on various soils and throughout the
year in northern Nigeria, and proposes to extend this survey into
the Malian and Niger Sahel and into the Ethiopian and Kenyan range-
lands. Analyses are, of course, most informative when they relate
to the species and the plant parts actually eaten by grazing animals;
this is why ILCA is interested in methods of sampling which mimic the
animals' selectivity or, as in O.F. methods use the animal itself to
obtain samples for analysis.
Knowledge of the nutritional features for which animals
select their forage intake has particular value for ILCA's simulation
modelling work. Mathematical descriptions and predictive models have
been developed for water use and plant growth (van Keulen et al, 1981)
and for animal performance (Konandreas and Anderson, 1982) but a weak
point in the sequence is the lack of knowledge about feed intake and
selectivity of African livestock under African pastoral conditions.
References
APRU (Animal production Research Unit) . 1978. Ann. Rept. APRU.
Botswana Min. Agric. Gaborone.
Dicko, M.S. 1980a. Measuring the secondary production of pasture:
An applied example in the study of an extensive production
system in Mali. In Browse in Africa: The current state
of knowledge. ILCA, Addis Ababa, p. 247.
Dicko, M.S. 1980b. Les sous-produits agro-industriels au Mali.
Paper presented at the AAASA/ILCA Workshop on the Utilisation
of Agricultural, Forestry and Fisheries Waste Products,
Douala, Nov. 1980.
Dicko, M.S., Lambourne, L.J., de Leeuw, P.N. and de Haan, C.
Voluntary intake and livestock productivity. Paper presented
at the 32nd Annual Meeting of the European Association for
Animal Production, Zagreb, Aug. - Sept. 1981.
I.N.R.A. 1978. In Alimentation des ruminants. France, Institut
National de Recherches Agronomiques . p. 579.
- 205 -
van Keulen,H., Seligman, N.G. and Benjamin, R.W. 1981. Agric. Symp.
6:159.
Konandreas, P. A. and Anderson, F.M. 1982. Cattle herd dynamics: An
integer and stochastic model for evaluating production
alternatives. ILCA Research Report 2, Addis Ababa.
Milford, R. and Haydock, K.P. 1965. Aust. J. Exp. Agric. Anim. Husb.
5 : 13.
Smith, G., Carles, A., Schwarz, J. t Blackburn, H.,Ruvuna, F. and
Cartwright, 1982. In Proc. Small Ruminant C.R.S.P. Workshop,
Nairobi.
van Soest, P.J. 1976. In Feed Energy Sources for livestock. Ed.
Swan and Lewis, Butterworth, London, p. 83.
Wilson, R.T. 1982. Livestock production in central Mali. ILCA
Bulletin 15, Addis Ababa.
- 206 -
Nutrition animale
Ré sumé
Les pâturages arides et semi-arides caractérisés par des périodes
pluvieuses courtes et irrégulières, assujettissent les animaux et leurs
conducteurs à de longues périodes de contraintes nutritionnelles. Les
travaux du CIPEA mettent en relief l'interaction de la nutrition et de
la santé dans la productivité globale. Des efforts ont été déployés
pour diagnostiquer les principales maladies, pour étudier leur
épidémiologie et leur impact sur la production et pour tester les
mesures de lutte disponibles. Les méthodes utilisées pour enregistrer
les données sur la nutrition sont examinées. Elles incluent des
méthodes relatives à l'enregistrement de la quantité et de la composition
du disponible fourrager et du fourrage consommé.
L'étude a confirmé les contraintes nutritionnelles qui font obstacle
aux performances des bovins au cours de la saison sèche et a montré que
celles-ci comportaient plusieurs aspects. La contrainte principale
semb1e être la faible qualité du fourrage sec qui oblige les bovins
à faire du pâturage sélectif pendant de longues périodes et à dépenser
beaucoup d'énergie. Des échantillons de fourrage prélevés de diverses
manières ont confirmé des découvertes antérieures selon lesquelles la
teneur en protéines du fourrage était plus étroitement liée à la
performance animale que ne l'était sa digestibilité.
Ces découvertes mettent en relief la valeur particulière des compléments
azotés protéiques ou non protéiques pour les ruminants et contribuent
à expliquer la meilleure performance des animaux qui consomment les
légumineuses à forte teneur en protéines au cours de la saison sèche.
S'il est vrai que la teneur en protéines constitue le facteur
déterminant et qu'elle intervient dans la plupart des mesures visant
à améliorer la qualité de l'affouragement, alors l'accroissement de la
teneur en protéines devrait être recherché, même au prix d'une certaine
baisse de la production brute de matière sèche. . La teneur en protéines
de la biomasse peut être augmentée,même malgré les contraintes
relatives au climat et à la fertilité des sols, par une gestion visant
délibéremment à accroître le rôle des légumineuses, des graminées et
des ligneux plutôt que celui des graminées de qualité inférieure, quelle
- 207 -
que soit l'importance de la production de matière sèche. Etant donné
le rôle que jouent la teneur en azote et la digestibilité dans la
détermination de l'ingestion et de la performance des ruminants, le
CIPEA a entrepris de recueillir des données sur celles-ci et sur une
vaste gamme d'échantillons de fourrage et de ligneux. La pratique
courante en Afrique de donner du sel sous diverses formes au bétail
suggère que les sols et les plantes peuvent être déficients en
certains éléments minéraux. Le CIPEA a procédé à l'échantillonnage
d'une vaste gamme de fourrages sur divers sols pendant toute l'année
dans le Nigéria du nord et propose d'étendre cette enquête aux pâturages
sahéliens du Mali et du Niger et aux terrains de parcours de l'Ethiopie
et du Kenya.
La connaissance des caractéristiques nutritionnelles qui déterminent
la sélection par les animaux du fourrage ingéré revêt une importance
capitale dans les travaux de modélisation du CIPEA. Des descriptions
mathématiques et des modèles de prévision ont été mis au point pour
l'utilisation de l'eau et la croissance des plantes ainsi que pour
la performance des animaux. Un point faible cependant dans cette
série: le manque de connaissances sur l'ingestion fourragère et la
sélectivité du bétail dans les conditions pastorales en Afrique.
- 208 -
Summary of Discussion Session 4.
Chairman: Dr Assefa Giorgis (Ethiopia)
Discussion led by Dr Samson Chema (Kenya)
Dr Chema asked what efforts ILCA was making to ensure a systems
research discipline by component researchers. He said there did not
seem to be a defined role for each component. Dr de Haan pointed out
that it was completely legitimate within the systems approach to focus
on one of the most critical components in order to understand the
basic processes. Animal nutrition was one such critical component.
Dr Ngutter commented that the neglect of small ruminants by the
authorities and development agencies stemmed from the condemnation by
colonial agricultural officers of the goat as the worst culprit in
ecological degradation. He asked what would be the cost of the
methodology proposed by Drs Wilson and Semenye in their paper, if
this was to be continued for at least three years. Dr Wilson said
that this would depend on the salary and benefit levels per senior
scientist, otherwise it could be calculated from the data given in
the paper plus two observers salaries plus vehicle transport at
5000 km per year.
Dr Diakite commented that many projects did not take sufficient
account of small ruminants, donkeys and camels. These animals should
be more closely considered in assessing pastoral management. Herdsmen
were often ahead of researchers in terms of understanding pastoral
systems - this too should be remembered.
Dr Zulberti referred to the information in Wilson and Semenye' s paper
comparing best vs worst for some of the variables. He thought that
these comparisons could be extended to larger numbers of animals.
Some other variables which could be included were- fertility, mortality,
disease control, multiple birth, seasonal breeding control, watering
interval and wealth.
- 209 -
Prof. Saka Nuru asked Dr Wilson if he had taken into account the
effect of nutrition type and management factors on dental eruption
and wear in aging the sheep and goats studied. Dr Wilson said that
the main variable was the sheep or goat - the management system had
a slight influence on the time of eruption but no other significant
effects. .
Dr Thomson asked if ILCA had now sufficient information from the
descriptive/diagnostic phase in Mali to enable it to concentrate on
the experimental/testing phase. Dr Wilson said that there was now
enough information from Niger and Kenya as well as Mali to move on
to the next phase. Dr Thompson asked Dr Lambourne if he was surprised
that protein had a greater effect on intake than digestibility, or
was he just confirming other work done in similar regions.
Dr Lambourne said that of course he was not surprised, but there was
a need to confirm that intake was affected by crude protein under
local conditions. This information would allow one to say which
forage species were valuable and which useless to the animal.
In referring to Dr Wilson's paper, Dr Grandin stressed that one had
to know a lot about an area, its important parameters and production
units in order to be able to choose a representative unit.
Dr Rhissa expressed his concern that the discussion on animal nutrition
had neglected the contribution of browse to the feed balance of
livestock in pastoral areas.
- 210 -
Résumé des débats de la quatrième seance
Président: M. Assefa Giorgis (Ethiopie)
Débats dirigés par le Dr Samson Chema (Kenya)
Le Dr Chema s'est informé des efforts déployés par le CIPEA pour
instituer une discipline de recherche sur les systèmes pour les
chercheurs sur les composantes. Il a déclaré qu'il ne semblait pas
y avoir un rôle défini pour chaque composante. M. de Haan a souligné
qu'il était tout à fait légitime, dans le cadre de l'approche par
système, de se concentrer sur l'une des composantes les plus
importantes en vue de comprendre les mécanismes de base. La
nutrition animale faisait partie de ces composantes importantes.
Le Dr Ngutter a souligné que la négligence des responsables et des
organismes de développement à l'égard des petits ruminants provenait
de l'accusation par les agents de l'agriculture coloniale de la chèvre
comme étant le principal responsable de la dégradation écologique. Il
a demandé de préciser le coût de la méthodologie proposée par
MM. Wilson et Semenye dans leur document, si celle-ci devait se
poursuivre pendant au moins 3 ans. M. Wilson a déclaré que cela
dépendrait des traitements et prestations dont bénéficierait chaque
scientifique de haut niveau et que sinon on pourrait procéder à un
calcul sur la base des données fournies dans le document, plus les
salaires des deux observateurs et un véhicule parcourant 5000 km par an.
Le Dr Diakité a souligné que plusieurs projets ne prenaient pas
suffisamment en considération les petits ruminants, les ânes et les
chameaux. Ces animaux devraient être étudiés de manière plus appropriée
dans l'évaluation de la gestion pastorale. Les pasteurs dépassaient
souvent les chercheurs en ce qui concerne la connaissance des systèmes
pastoraux. Il fallait également se rappeler cela.
Le Dr Zulberti a fait allusion aux données contehues dans le document
de Wilson et de Semenye, comparant pour certaines variables les
meilleures et les pires performances. Il a estimé que ces comparaisons
pourraient être étendues à des nombres plus importants d'animaux.
Certaines autres variables que l'on aurait pu inclure avaient pour noms,
- 211 -
la fertilité, la mortalité, la lutte contre les maladies, les
naissances multiples, le contrôle de la reproduction saisonnière, les
intervalles d'abreuvement et la richesse.
Le Prof. Saka Nuru a demandé à M. Wilson s'il avait pris en considéra
tion l'effet du type de la nutrition et des facteurs de la gestion sur
l'éruption et la chute des dents dans la détermination de l'âge des
moutons et des chèvres étudiés. M. Wilson a déclaré que la variable
essentielle était le mouton ou la chèvre. Le système de gestion n'avait
qu'une influence minime sur le moment de l'éruption et n'avait pas
d'autres effets significatifs.
Le Dr Thomson a demandé si le CIPEA disposait actuellement d'informa
tions suffisantes sur la phase de diagnostic/description au Mali pour
lui permettre de se concentrer sur la phase d'essai/expérimentation.
M. Wilson a déclaré qu'il y avait maintenant suffisamment d'informations
sur le Niger, le Kenya et le Mali pour que l'on puisse passer à la
phase suivante. M. Thompson a demandé à M. Lambourne s'il était
surpris que les protéines aient un effet plus grand sur la gestion que
la digestibilité ou bien était-il juste en train de confirmer d'autres
travaux effectués dans des régions similaires. M. Lambourne a déclaré
qu'il n'était certainement pas surpris mais qu'il était nécessaire de
confirmer que l'ingestion était affectée par les protéines brutes dans
les conditions locales. Cette information permettra de dire quelles
sont les espèces fourragères les plus valables et quelles sont celles
qui sont inutiles pour l'animal.
En se référant au document de M. Wilson, Mlle Grandin a déclaré qu'il
fallait avoir des connaissances exhaustives sur une zone donnée, sur ses
paramètres les plus importants et sur ses unités de production pour
être en mesure de choisir une unité représentative.
Le Dr Rhissa a souligné qu'il s'étonnait de constater que les débats
sur la nutrition animale avaient négligé la contribution des ligneux
à l'équilibre fourrager dans les zones pastorales.
- 212 -
Production strategies and pastoral man
Noel Cossins
Socio-economist and Ethiopia Team Leader, Arid Zones
(Eastern and Southern Africa) Programme, ILCA, Ethiopia
Pastoral Man in Africa
One of the main differences between range livestock systems in
countries such as America and Australia, and those in Africa, is
that in Australia people derive a living from the range while in
Africa people depend upon the range for life. In Australia if a
grazier loses all his cattle he may go bankrupt, but in Africa if
a pastoralist loses all his cattle he may also lose his life.
An African pastoralist may equally lose the capacity to
support the life of his family by losing only a portion of the pro
ductive capacity of his livestock. Remembering what is at stake to
the African pastoralist compared to the Australian grazier, is
essential in understanding a pastoralist's reaction to the introduction
of change and innovation.
2
In Africa about 13-16 million km , or nearly half of the
continent south of the Sahara, is desert or arid grassland and
savanna where cultivation is a high risk enterprise (Brown, 1971).
It is in these areas that almos.t all of the 20 million or so sub-
Saharan pastoralists live, subsisting wholly or almost wholly on the
products of their livestock (Helland, 1980).
Of all the secondary users of vegetation through animals,
the pastoralist is the only one who depends on milk and not meat.
All other secondary users including the advanced capitalist pastor
alist or large-scale rancher depend on meat. No Australian rancher
in his right mind in fact would attempt to produce milk commercially
in the type of semi-arid area in which nomadic pastoralists live.
- 213
And yet the African pastoralist has relied on milk and still
attempts to do so, except where that system is under pressure or
breaking down as is occurring with the Afar of the northeast
rangelands of Ethiopia. There is logic behind this reliance on milk
for it has created a system independent to an extent of farmer
neighbours with whom the pastoral system may have been hostile at
times, and which has the capacity to directly support, on a subsist
ence basis, far more people per unit area than any other arid area
production mode. Jahnke (1982) estimates for example that if arid
countries like Mauritania and Somalia organised their land use in the
form of modern ranching, their pastoral population would have to be
reduced by a factor of 50.
Measured in terms of survival, pastoral production systems
in Africa have been remarkably effective (Dyson-Hudson, 1982) . They
have in general achieved their major objective of providing a reliable
source of food for the population that actually operated the system,
a supply that has been able to be sustained at life support levels
throughout the dry stress periods in most years.
The number of people that any area is capable of supporting
can be termed the 'human support capacity', and it is this factor
that pastoralists seek to optimise. They do so through the production
and consumption of milk. The objective is the optimisation of human
numbers per unit area of arid land, while the strategy is milk pro
duction. This objective is the primary one for all pastoral systems,
with the second and equally important objective being to provide as
much security as possible for the dependent human population. This
is achieved by employing a number of specific strategies. These days
under the stimulus of development and change, there is also an
increasing interest in productivity and wealth in terms other than
livestock numbers.
General relationships between the natural productivity of
the land and the human support capacity have been established by
various workers for the agro-ecological zones of both East and West
Africa. For example, for East Africa they are as follows (Pratt and
Gwynne, 1977):
- 214 -
Agro-ecological zone Human support capacity
Annual (ha/person)
rainfall Type
(mm)
Very arid 189.0
200
300
400 Arid 48.0
500
600
Semi-arid 14.0
If the above criteria are applied to the semi-arid Borana
area of the Ethiopian southern rangelands a figure of 17.2 ha/person
results which indicates that the area has the capacity to support
21% n ore people than it presently does. The figure for the Afar
who live in an arid to very arid area in the northeast rangelands of
Ethiopia ranged from 11-50 ha/person which indicates that this pas
toral system is under considerable pressure and has more people than
it can reasonably support. Other data support these indications.
In all pastoral systems the consumption of milk or blood
seems to be steadily dropping, and there are few (if any) which rely
almost totally on milk or milk products. In some the reliance is
still fairly high. The Borana of the southern rangelands of Ethiopia
for example, with some seasonal variations, still consume up to 59%
of their diet as milk or milk products with the balance of the diet
being increasingly made up of grain. For the Afar, milk now probably
constitutes less than 20% of total energy requirements, and grain
again is increasingly the main food substitute. This increase of
grain and decrease of milk consumption is in fact more and more the
pattern in pastoral Africa. Nevertheless the African pastoralist is
- 215 -
still firmly oriented towards a milk production mode as far as
circumstances will allow and has not yet dramatically changed this
in favour of selling meat or growing crops.
In the 1960's, some ecologists (e.g. Brown, 1971) argued
that the dependence on milk by pastoralists in arid areas was ecologi
cally unsound, placing (as it does) the pastoralists in direct
competition with their calves, and in creating a herd population
structure where females make up 70% or more of the herd. In the
inevitable cycle of good years and drought years this high breeding
capacity led to rapid recovery after drought years and a very rapid
increase in numbers which spiralled upwards, while environmental
degredation increased with each drought period. In each drought, so
the argument went, animals did their damage before they died, and as
their numbers were great, so was the damage.
This is compelling theory in these times of environmental
consciousness but it does tend to simplify the rather complex set of
inter-relationships which exist between the pastoralist, his animals
and the resources available to these animals in terms of grass or
browse and water.
One of the functions of pastoral systems research is to
observe and quantify entire pastoral production systems. Given this
it is essential to consider pastoralists' objectives and the strat
egies they employ to attain these objectives. There are many miscon
ceptions about the reasons why pastoralists do the things they do.
Almost every study of pastoral systems over the last 30 years has
shown however that pastoralists follow observable behaviour patterns
which are rational with respect to their objectives, and that these
patterns can be described. It is sometimes easier to record the
effect (e.g. low weaning weight) first, and then determine the
behaviour (e.g. offtake of milk for humans as well as calves), but
the two, behaviour and effect, are linked and must be seen as a whole.
- 216 -
Pastoral man and production strategies
Almost everything a pastoralist does is the result of a deliberate
decision, from his rain-chasing nomadism to the maximisation of
females, which is an obvious strategy for a people one of whose
main objectives- is to maximise milk production. A similar case
exists for breeding practices where the wider the distribution of
the calf drop the more chance there may be of high mortality or poor
calf growth for those calves dropped in the dry season, and the
more likely there is to be a year-round supply of milk.
Nomadic pastoralism presupposes a high degree of organi
sational and spatial flexibility (Dahl and Hjort, 1979) . Households
constantly redistribute themselves over the terrain, and membership
of households changes as labour is allocated and reallocated between
different management units.
In any pastoral system a pastoralist has to construct
production strategies to cope with three series of constraints. These
are as follows :
1. Normal - the constraints placed upon the system by nor
mal, mostly seasonal, events. A pastoralist ' s year normally ranges
from times of plenty (the rains and post rains period) to times of
shortage (the dry seasons) . The year is punctuated by a series of
high and low periods where disease, parasite burdens, available
forage come and go as problems.
2. Disasters - at times some of the normal constraints may
assume disastrous proportions. The main disasters are epidemic
disease, range fires in the dry season, and drought.
3. Long-term changes - these are often irreversible and
consist of such events as the loss of dry-season grazing areas to
cultivation, the relative advantages which may accrue to richer pas-
toralists because of government policies, or loss of revenues from
caravans, raiding etc.
Not everything that a pastoralist does can be termed a
strategy, and some actions can better be described as tactical
decisions. In general, strategies are far less responsive to inter
- 217 -
ventions than are tactical decisions which can be changed and adjusted
fairly readily. In this paper I am using the term 'strategy' to
denote the conduct of a campaign or the response to a set of circum
stances in its long term and large scale aspects, while 'tactics' are
taken to mean the use of resources to the best advantage and with
respect to an immediate and short-term solution, generally at the
production unit level.
In the pastoral context we can define a strategy as the
consumption of milk, or the communal ownership of land, while a
tactical decision may concern the movement of animals to specific
areas at specific times, or a response to price differentials for
different stock types. For example, it generally pays a pastoralist
to retain a male animal in his herd until maturity. The reasons are
simply economic. The labour expended in herding male stock is small
compared Co that required for cows and calves, which has to be done
in any case, and the returns on keeping a male to maturity are
usually worthwhile (Dahl and Hjort, 1979).
In Borana, for example, the average price of an immature
male was Birr 198 compared to an average mature male price of Birr
313 (Negussie, 1983). The main labour bottleneck is at the wells in
the dry season but there, with a livestock to labour ratio of
50:1, 10% more or less males do not make a great deal of difference.
A change of relative immature /mature price ratios might bring about
a change of tactics however. Steers are also preferentially retained
to full maturity in systems which use blood as a milk substitute, as
in the case in the Borana 'dry herds', and a replacement food may
also have to be found before immature steers are readily sold.
Perhaps the most important strategy that pastoralists have
evolved is the response to the threat of drought. Drought or very
dry years are inescapable in most African pastoral systems. In the
southern rangelands of Ethiopia, for example, the main rains fail one
year in ten and the secondary rains one year in three in the Dolo
area. Similar figures can be produced for any rangeland or pastoral
system, and as there is no evidence to show that regular cycles of
drought occur (Bille, 1983), a pastoralist has to be ready at any
time for drought.
- 218 -
Knowing this, a pastoralist adopts several or all of a
number of strategies, or traditional forms of insurance against loss.
The most important of these (in East Africa) are as follows:
- maintain more than one species of livestock. Camels and
goats are for example more resistant to drought than cattle and sheep»
- divide livestock holdings into spatially separate units to
minimise the effects of localised drought. This requires a high
labour input or more than one household.
- establish and maintain social systems for resource sharing,
or for borrowing, lending and gifting.
- maintain large herds, or as large as possible, to maximise
the chances of having some left when the drought is over.
- during the drought or disaster minimise the reliant human
population by sending away all able-bodied people not required to work
the system. People have been most often sent to adjacent agricultural
areas (e.g. the Maasai to the Kikuyu areas in the smallpox/rinderpest
epidemics of the 1890' s), and contrary again to previous opinion,
contacts between farmers and pastoralists in East Africa seem to be
long established (Hjort, 1981).
Again these are reasonable strategies given the limitations
of the system and the fact that there has never been access to or any
reason to trust a formal banking system to build up reserves against
bad times, or an insurance policy which will sustain a pastoralist
through drought.
Drought effects do not always relate directly to rainfall
(Dahl, 1979). The same amount of rain may produce very different
subsistence conditions depending not only on the availability of dry
season pasture, but also on the number of people needed to make the
necessary movements, or raise water, and on the amount of milk, meat
or blood needed to feed them, or the amount of grain available for
purchase.
Drought effects are thus not solely an ecological phenomenon,
and socio-political aspects or market supplies are secondary but
important functions. The Afar, for example, are still responding to
- 219 -
the great Wollo drought of 1972-74 by switching from cattle to
selling smallstock. Our market figures show however that they bring
to market twice as many smallstock as the various markets can absorb,
and this inability to sell certainly affects the viability of the
system. The Afar response also reflects the fact that mortality rates
are markedly differentiated by age during a drought (Dahl, 1979).
Among the Afar Arapta clan during the beginning of the Wollo drought
of 1972-74 (Cossins, 1972) there were marked age group gaps in the
cattle herd particularly for older cows and heifers. When these gaps
were superimposed on the herd composition figures the graph shown in
Fig. 1 resulted.
Relative
Nos .
as %
^ milking age cow numbers.
\
 
Cumulative effect of
maturing calves born
after 1972
986
Fig. 1. Projection of future native milking age con numbers for the
Arapta clan.
- 220
Dahl (1979) found for the Isiolo Boran that there were
particular years after a drought that were bad from a reproductive
point of view. Fig. 1 shows that while there was likely to be a
marked recovery within five years, the worst was still to come for
the Arapta clan. Experience in the Afar system has vindicated this
prediction, and the Afar are still largely dependent on selling small-
stock for their subsistence. Provided another drought does not occur,
the Afar could be expected to move back into an increased reliance on
cattle from 1983 onwards.
There is a contradiction here, for smallstock are more ef
ficient at converting pasture into consumable meat and milk, and are
more easily marketed, so why should the Afar attempt to move back
into cattle? There are probably two main reasons. While goats in
particular are more resistant to drought and the recovery rate of
smallstock is rapid, smallstock are much more susceptible to disease
than cattle, and an epizootic can claim high death tolls.
Secondly, the present market system is not able to buy all
the smallstock that the Afar need or want to sell at reasonable prices,
nor dees it provide all the grain the Afar need to buy for subsistence.
Unless this is changed, it is a logical production strategy for the
Afar to move back, as far as is feasible into milk production from
cattle, even though this may not be an ecologically or economically
sensible strategy,
Probably the most emotive and contentious issue concerning
pastoralists is the numbers game. Range managers and ecologists
continually promote the need to destock African rangelands. The pas
toralists resist. Why they resist is the root cause of the problem
and one that tends to be overlooked. Brown (1971) was among the
first to identify the problem as being one of human numbers rather
than livestock numbers, and it is important to recognise this dis
tinction for it answers the question of what can be done about this
problem and why pastoralists will not destock. It has nothing to do
with the 'tragedy of the commons' argument (Harbin, 1968), and has
everything to do with common sense.
221 -
Let us look at the Borana for example. The Borana live in
a 600 mm rainfall rangelands areas, are still oriented to the pastoral
mode of production, and seem to be relatively well-off as a people, in
that hunger does not feature highly on their list of problems. We
know that the average family consists of about 3.5 adult equivalents
(it is actually slightly less than this from our household studies)
and we can estimate that their total energy requirement per day will
be about 33.7 KJ.
We know such a family owns or has access to about 18 head
of cattle which, using the herd structures derived from the well
studies, means about six lactating cows at any one time. We know also
that the average offtake per cow for human consumption over an average
ten-to eleven-in month lactation period is 312 litres or 2050 litres
per year. In terms of energy this is equivalent to about 7175 KJ.
Our market and household data show that the unit family
also sells two mature animals a year, and consumes slighty less than
one cull female or an animal about to die (fallen meat) . They also
keep six sheep and seven goats and the production from these is also
sold or consumed. The data also indicates that they consume about
150 kg of smallstock and fallen meat per year, and that part of
the income from all livestock sales is used to buy grain, coffee,
sugar etc., and that there is also a supplement of bush foods and
blood for the boys herding the dry herds. From these sources a total
of about 6 982 KJ of energy is potentially available 6075 KJ, if
no meat is consumed).
Putting the milk, meat, grain, coffee, sugar etc., together
gives a total available energy intake of about 14,157 KJ of which
milk constitutes about 51%- Based on the combination of adult
equivalents and estimated daily energy requirement, the average
Borana family requires 12,132 KJ per year. The difference between
these two figures suggests that we have either over-estimated the
contribution to the diet of other food, particularly meat and grain,
or the Borana eat well occasionally. Whatever the case, the above
exercise shows that the Borana generally have enough to eat, although
- 222 -
Rainfall
600 mm.
/
Solar Energy
2
750 kj cm
V
LAND 81 hectares wet season
48 hectares dry season
1
150 Ton DM.
J(excluding browse)
(actual requirements are about 100 T DM, so there is unused potential).
I
CATTLE (18)
9 Adult F-
4-5 Immat.
3 Immat. M
1 Bull
1 Castrate
WATER
1000 1 every
3 days
\
400 1/day
011a Plot Grain \
(Milk subs. &
increasing)
(could provide
upwards of 30%
requirements and
replace 2/3 cows)
S
SMALLST0CK
(13 - 18)
6-10 sheep
7-8 goats
(provide 20-30% food
energy requ. by sale
for grain or cons.)
A
N
OTHER
Horse
3 Camels
(camels provide?
of energy intake.
May be important in
dry season. May
substitute for cows/
small stock/ cash)
Zl
5 to 6 cows in milk
(plus 'fallen meat' or cull
cows and animals which die
also contribute 5% require
ments)
Offtake for family av.
2050 1/year (or 59% of
required energy intake)
Rest of milk into calves
Minimal sales
Sales offtake
2 males a year
2 matures or near
mats .
(minimum of 'l 6%
spent on basic ener
gy requirements)
/
Birr 626.00 Income
(equivalent to
cash requirements -
coffee, tea, sugar,
tobacco etc.)
Fig. 2. Family resource diagram for the Borana livestock system.
- 223
there may be seasonal shortages, with about 200-300 Birr per year
left over for clothes, tobacco, talla etc.; but the margins are not
great, and even one cow would be missed.
The above information is interesting in its own right, but
what we can also conclude from it is that there is no way that the
'average Borana family' can destock without substantially reducing
its standard of living, (e.g. no tobacco or talla or clothes etc.),
and that stock reductions of over 15% would put that family on the
subsistence borderline. Now the Borana are amongst the more fortunate
of East Africa's pastoralists, so that ic is quite clear that the
pastoralists ' resistance to destocking is based on a strategy of
survival, no matter how foolish it looks to the conservationist.
Jahnke (1982) describes the difference between the human support
capacity of African pastoral areas and the actual numbers as about
1:2, or 12 million versus an actual figure of some 20-29 million
people.
What is required to precede any destocking policy is not
an educational campaign which shows the pastoralist the damage his
animals may be doing to the range, but a change of pastoral mode.
If the 'average Borana family' as described in the above example were
to convert to a mainly grain/bean diet with some milk, and were to
maximise sales from all livestock, then that family's net income from
the same number of animals would at least double. Conversely, the
number of livestock could be reduced by half and the same standard
of living maintained.
This applies to most pastoral systems. Jahnke (1982)
suggests that the prevailing terms of trade for African pastoral
systems in general are 1,7 kg of grain for 1 kg of milk, and 4 kg
of grain for 1 kg of meat, so that a pastoralist significantly im
proves his subsistence basis by trading. Many examples of this can
be found in West Africa among the Fulani who trade milk particularly
for grain, while in East Africa the volume of this trade is rapidly
increasing.
Encampment food grain plots also provide an alternative. A
plot which yielded only 4 quintals per year would provide more than
- 224 -
30% of yearly food requirements for that same Borana family, would
allow at least a 15% reduction in the average herd size (from 18 to
15 cattle) , or would allow an increase in the offtake rate of at
least 6% (an increase to 18%) or add an additional 20,000 cattle to
the 40,000 presently sold out of the system.
All these figures are very tantalising to the planner and
developer, but there are substantial problems of pastoralists ' confi
dence and conservatism to overcome, just as there are for any farmer
anywhere when you want him to make a very considerable change in his
mode of production. There is also an equally large problem of pro
viding an assured and regular grain supply in a continent where grain
deficits prevail, and to provide an assured and regular market outlet
for livestock at acceptable prices.
My argument however, is not really about these aspects, but
merely uses the above to point out that overstocking is a problem of
human numbers, and that the pastoralist knows this and so resists any
pressures to destock unless alternative means of supporting his family
are available. And he has to have confidence in such alternatives.
The communal use of land is another pastoral strategy, and
it makes sense considering the variability and unreliability of
rainfall in African rangelands areas. In any one year rainfall may
vary within a pastoral system by as much as 200% from the mean, as
it may between years. As there also appears to be no cyclical effect
involved (Bille, 1983) so that a planned rotational system is not
feasible, it pays a pastoralist to have access to as large an area
as possible. Individual or group ownership of specific areas of land
precludes this strategy, and just so long as a pastoralist continues
to have his life at risk rather than his livelihood, and maintains
his bankable reserves as livestock, subdivision of land will be
counter to the ability of a pastoralist to survive.
Perhaps the last example to consider ig that of organisation
and cooperation. Pastoral systems generally require a high degree
of both if they are to function well. Where neither occurs, and small
units compete for resources instead of cooperating in their use, the
system begins to break down as is probably happening in the Afar system.
225 -
In the Borana system, organisation and cooperation is still
high and the system works well. Imagine for example, the degree of
organisation and cooperation that is required to work, say, the wells
at Bor Bor. There, in 1982 over three days, some 47,000 head of
cattle, 22,000 sheep and goats, and 2,000 other stock were watered in
groups of 50 to 100 at nearly 300 groups per day, in the same or
nearly the same sequence every third day, and some 780 people were
organised to work in one of 17 wells also every third day. This was
no mean feat under any circumstances. Thus a society which stresses
community over the individual, and which stresses common but organised
rights to resources, is more likely to succeed in a pastoral context
than one which stresses individual rights.
Production strategies and decisions affect almost every part
of any pastoral system. The decisions and strategy alternatives
facing a pastoralist are no less complex than those facing a western
rancher, and because he may have less control over his resources, and
the use of these resources involves far more human labour, the organ
isation and decisions required of a pastoralist may be even more
complex.
The idea of the pastoralist as a simple fellow pursuing a
simple mode of life is thus a false one. Outwardly, a milk-drinking,
cattle-owning pastoralist may appear so, but his system is highly
complex and the strategies he has devised to cope with it are equally
complex. Unless an intervention is aimed at the cause of the strategy
it may not succeed or be adopted, except where the intervention
affects a tactical decision rather than a strategy. It is important
to understand this when designing a research programme or deciding
on research priorities.
Determining management strategies
In the above I have discussed pastoral management strategies and
tactical decisions. In the past, pastoral management strategies were
relatively uniform with respect to a particular climatic zone, and
centred on a mobile human population, dispersed in small groups at
low overall densities, with an introverted milk-oriented subsistence.
- 226 -
This population experienced major fluctuations in numbers over time,
because of the unreliable distribution of rainfall over time and
space (Dyson-Hudson, 1982). Livestock numbers were also often limited
by the uneven distribution of water resources with respect to grass
and browse.
In general, multi-species herding was the norm as a response
to the probability of drought and to fully exploit the often mixed
grass land-shrubland-woodland of the environment (Dyson-Hudson, 1982) .
Modifications of the environment or essential resources included the
use of fire to control bush, stimulate the regrowth of grass, or to
control ticks, and the development of sometimes quite complex wells
and ponds. The major strategies however, concerned the dispersal of
small production groups over large commonly held areas, and a live
stock composite design which included a mix of species and a structure
where females predominated. This structure was remarkably uniform
for all species and even included horses in Borana.
Pastoral systems are under pressure. Their populations are
no longer allowed to adjust naturally to phenomena such as drought or
to expand in a traditional territorial sense through force. They are
increasingly being brought into the mainstream of development by
forces beyond their control, and into situations where traditional
pastoral strategies may become less efficient. These strategies then
become less useful to pursue, as Dyson-Hudson (1982) writes, and
pastoralism then becomes less likely to persist as a plausible pattern
of land use.
The strategies pursued by pastoralists are thus changing as
pastoral populations respond increasingly to development inputs,
political and administrative pressures, and to changing aspirations.
The main question to answer for any pastoral system and, by extension,
the importance of this question to any pastoral systems research pro
gramme, is what strategies are being pursued by the pastoralists under
study, and why, and how does one differentiate production strategies
from tactical decisions.
- 227 -
Seasonal movements reflect both strategy and tactics.
Access to as large a piece of communal land as possible in order to
exploit rainfall whenever and wherever it occurs is the strategy, but
the tactical decision is the determination of where and when to move
for a specific herd or herd group at a specific time. Given the
correct intervention, e.g. improved pastures in a specific area, it is
relatively easy to induce a change in the tactical decision. It is
not so easy to bring about a change in the strategy which is based on
the unreliability of rainfall in the arid areas, and the maximisation
of opportunity.
If a change from the traditional norm has occured, it will
tell us a great deal about the system and its possible future. The
Maasai for example are beginning to countenance individual ranch
holdings, which probably signals the beginning of the end for Maasai
as traditional pastoralists. The Borana on the other hand still
strictly observe the strategy of communal rights to all land in their
production sphere, and are still well within the traditional pastoral
orbit .
Mode of pro .action is also a strategy and can be determined
for example from herd structures. Borana herds consist of 74%
females which almost certainly means that the Borana follow a milk
production/consumption mode. However the herds also contain about
6% castrated male animals which indicates that animal sales are a
secondary strategy.
Pastoralists' drought response strategies can be determined
in a variety of ways. Species mix is one indication, and the extent
of this is also an indication of the frequency and severity of drought
in any area. In the southern rangelands, an area with a relatively
good and reliable rainfall (600 mm), the cattle/smallstock/camel
ratio is about 3.6/1.3/1, whereas in the much more arid northeast
rangelands area, the species ratio is 4.3/11.25/1. Both systems
have a species mix, but the Afar's reliance on smallstock indicates
a greater and more frequent drought risk.
- 228 -
In general terms the phenomena surrounding a tactical
decision respond readily to research inputs, whereas the response to
research relating to production strategies is another question. While
it is theoretically possible to change the whole production strategy
of pastoral reliance on milk for example, and so in the Borana case
convert a self-sufficient medium animal offtake (10-12%) system into
a self-sufficient higher offtake system (18-20%), this would require
considerable organisation and inputs on a supra-system scale, and
would also require a national net grain surplus.
Some general rules of thumb regarding pastoralists and the
strategies they have evolved to cope with their environment are as
follows. If the main national objective remains to use arid areas to
support the maximum number of people (maximum human support capacity) ,
then the subdivision of land into discreet units to be owned either by
individuals or groups will inevitably reduce the long-term human
support capacity of the area; social systems are best left to evolve
and look after themselves; removal of land at the fringes of pastoral
systems reduces the capacity of the system to survive, as this is
usually land whose value to the system far outweighs its physical
size (dry-season grazing or strategic area grazing retreats) ; and it
is worth rethinking the ideas of the past which have tended to empha
sise the superiority as a production mode of various modified forms of
ranching. Ranching cannot compete with pastoralism in terms of human
support capacity, and pastoralists can often produce meat nearly as
well as any ranch, and much cheaper in the context of the system they
already operate.
This paper has addressed strategies and tactical decisions
in terms of overall pastoral systems, but the individual producer unit
also operates a number of alternative strategies and is responsible
for almost all the tactical decisions that have to be made.
Dyson-Hudson (1982) has described East African pastoral
systems in terms of a two-tiered structure. The first tier is the
division of a region's natural resources into a small number of large
units and the second tier being a large number of small units which
control and manage the livestock resource in terms of defined herds
and flocks. The size of the first tier of units is posed as being
- 229 -
sufficient to accommodate most normal variations in seasonal rainfall,
and also to mitigate the effects of all but the most severe disasters.
The second tier of many thousand of small producer units is
essentially autonomous in terms of management decisions, and although
there is cooperation between these units, this does not, and in fact
cannot, extend to guarantees of survival. These small groups may be
linked, or controlled and governed to an extent through other bodies
which have been evolved as a mechanism to arrange security on a system
scale, solve disputes between units, distribute information, and
allocate resources. Such bodies may be based on kinship, locality,
age sets or other related aspects, but the basic structure remains
the two interdependent units which 'put humans into association with
livestock and with natural resources' (Dyson-Hudson, 1982).
Strategies and decisions are influenced by a variety of
factors at the producer unit level, and it is important to understand
what these factors are in terms of identifying constraints, and why
there are differences not only between systems and between seasons,
but also within a system between producers. For example we have found
in Mali marked differences in productivity between herds which can
only be attributed to management (Wilson, 1982), and in Kenya we have
found that richer Maasai pastoralists have different responses to
phenomena and management alternatives than do poor people (Grandin,
1982) . The richer pastoralists also have a greater ability to respond
tT opportunity, and in fact every parameter studied to date in the
Kenya programme has had a wealth rank strata.
The strategies and decisions which fall within the realm
of the individual producer deserve at least as much attention as those
of the pastoral system as a whole, and these are the subject of the
following papers.
References
Bille, J.C. 1983. The climatic risks to livestock production in the
Ethiopian rangelands. JEPSS Research Report No. 4. ILCA
Internal Document, Addis Ababa.
- 230 -
Brown, L.H. 1971. The biology of pastoral man as a factor in
conservation. Biological Conservation, 3(2)
Cossins, N.J. 1972. No way to live: A study of the Afar clans of
the northeast rangelands of Ethiopia. Report prepared for
the Ethiopian Government Meat and Livestock Board.
Dahl, G. and Hjort, A. 1979. Pastoral change and the role of drought.
SAREC Report No. 2.
Dyson-Hudson, N. 1982. Changing production strategies in East African
range livestock systems: An overview. Informal paper, ILCA,
Nairobi.
Grandin, B.E. 1982. The importance of wealth in pastoral production:
A rapid method for wealth ranking. Proceedings of the
workshop on Pastoral Systems Research. Sponsored by IDRC/
ILCA. Held at ILCA, Addis Ababa, 21-24 March, 1983. ILCA,
Addis Ababa.
Harbin, G. 1968. The tragedy of the commons. Science, 162: 1243-1248.
Helland, J. 1980. Pastoral map of tropical Africa. ILCA, Addis Ababa.
Hjort, A. 1981. Ethnic transformation, dependency and change. Journal
of Asian and African Studies 26 (1-2) .
Jahnke, H.E. 1982. Livestock production systems and livestock development
in tropical Africa. Kieler Wissenschaftsverlag Vauk, Kiel.
Negussie Tilahun 1983. Livestock marketing in northeast Ethiopia.
JEPSS Research Report. ILCA Internal Document, Addis Ababa.
Pratt, D. and Gwynne, M.D. 1977. Rangeland management and ecology
in East Africa. Hodder and Stoughton, London.
Wilson, R.T. (ed) . 1982. Livestock production in central Mali. ILCA
Bulletin 15. ILCA, Addis Ababa.
- 231 -

Les stratégies de production et l'éleveur
Résumé
Parmi tous les utilisateurs secondaires de la végétation, l'éleveur
est le seul dont la subsistance est assurée par le lait plutôt que
pai la viande. Le nombre de personnes qu'une zone peut accueillir
peut être appelé la "capacité de charge humaine" et c'est ce facteur
que les éleveurs cherchent à optimiser. Ils le font par le biais de
la production et de la consommation de lait. L'objectif est l'optimi
sation de la population humaine par unité de surface de zone aride et
la stratégie utilisée est la production laitière.
L'une des fonctions de la recherche sur les systèmes pastoraux consiste
à observer et à quantifier les systèmes de production pastorale. Il
devient donc vital d'examiner les objectifs des éleveurs et les
stratégies qu'ils emploient pour atteindre ces objectifs. Dans tout
système pastoral, l'éleveur doit mettre au point des stratégies de
production pour faire face à trois séries de contraintes:
- les contraintes normales. Il s'agit des contraintes qui
pèsent sur le système du fait d'événements normaux,
saisonniers la plupart du temps;
- les catastrophes. Quelquefois, certaines des contraintes
normales peuvent revêtir des proportions catastrophiques.
Les principales catastrophes sont les épidémies, les feux
de brousse en saison sèche et la sécheresse;
- les évolutions à long terme. Celles-ci sont souvent
irréversibles et se caractérisent par des événements tels
que perte de pâturage de saison sèche du fait de l'empié
tement de l'agriculture ou perte de revenus du fait de
pillage par les caravaniers, etc. .
Dans le contexte pastoral, nous pouvons définir la consommation de
lait ou la propriété collective de terres comme étant une stratégie
alors que les mouvements des animaux dans des zones spécifiques à des
moments spécifiques ou en réponse à des variations de prix pour
- 233 -
différents types de bétail peuvent constituer des décisions tactiques.
Signalons à cet égard qu'il est possible que la stratégie la plus
importante parmi toutes celles que les éleveurs ont mises ^u point
soit la réponse à la menace de sécheresse. La sécheresse est inévitable
dans la plupart des systèmes pastoraux africains. Fort de cette
connaissance, l'éleveur adopte tout ou partie d'un ensemble de
stratégies ou de formes traditionnelles de prévention contre les
pertes telles que :
- l'élevage de plusieurs espèces de bétail;
- la division du troupeau en unités spatialement distinctes
pour minimiser les effets des sécheresses localisées;
- la mise en place et le maintien de systèmes sociaux de
partage des ressources, d'emprunt, de prêts et de dons;
- l'élevage de grands troupeaux pour maximiser les possibi
lités de survie des individus après la sécheresse;
- la minimisation des risques pour la population humaine
au cours de la sécheresse ou pendant la catastrophe en
organisant l'exode de tous les bras valides dont la
présence dans le système n'est pas indispensable.
Les systèmes pastoraux sont soumis à de fortes pressions. Leurs
populations n'ont plus la possibilité de s'adapter à des phénomènes
naturels tels que la sécheresse ou d'assurer l'expansion de leur
territoire par la force. Elles participent de plus en plus au
processus général du développement du fait de l'action de forces qui
échappent à leur contrôle et sont plongées dans des situations où les
stratégies traditonnelles d'élevage peuvent devenir moins efficaces.
Ces stratégies deviennent alors moins utiles à poursuivre et la survie
du pastoralisme conçu en tant que mode praticable d'utilisation des
terres devient hypothétique. Les stratégies adoptées par les éleveurs
changent donc à mesure que s'accroît la sensibilité des populations
pastorales aux facteurs de développement, aux pressions politiques
et administratives et à de nouvelles aspirations.
- 234 -
Les stratégies et les décisions sont influencées par une variété de
facteurs au niveau de l'unité de production et il est important de
comprendre ces facteurs en vue de l'identification des contraintes
au système ainsi que les différences non seulement entre les
systèmes et entre les saisons mais également, au sein d'un système,
entre les producteurs. Les stratégies et les décisions adoptées
à l'échelon du producteur individuel méritent au moins autant d'intérêt
que celles de l'ensemble du système pastoral.
- 235 -

The importance of wealth effects
on pastoral production:
A rapid method for wealth ranking
Barbara E. Grandin
AntJiropologist, Arid Zones (Eastern and Southern Africa) Programme,
ILCA, Kenya
This paper discusses the importance of wealth differences between
producers in traditional pastoral production systems. It argues that
significant wealth differences exist, that these have a profound
effect on production strategies and that pastoral systems research
must pay attention to them at several stages from defining a target
population or recommendation domain to developing and testing inter
ventions. The paper then describes a rapid method for determining
the wealth rank of producers within a given community. Such a ranking
is an important tool to stratify a population of producers before
sampling to ensure the representativeness of the sample along this
important dimension. Alternatively, it may be used post facto for
assessing the representativeness of pastoralists already interviewed.
Wealth differences in pastoral production systems
For many years studies of traditional agricultural production systems
emphasised the essential homogeneity of producers. While it was
recognized that some differences existed, they were though, on the
whole, not to be significant, but purely a matter of scale.
In agricultural economics, for example, earlier research
results tended to be reported in terms of the average farmer, with his
average family, average cropping pattern and average yields. In
anthropology, to the extent that economic heterogeneity was researched,
emphasis was placed on "levelling mechanisms", which by definition,
functioned to counteract trends towards inequality. For example,
polygamy has been discussed as having a levelling function, as more
wives would mean more children and consequently the eventual fragmen
tation of the wealthy producers' assets.
- 237 -
In studies of peasant agriculture, there has been in
creasing recognition of the extent of wealth differences within
communities and their effects on production parameters. As Hill
(1972) noted in her pioneering work on inequality in a Hausa village
in Northern Nigeria, "it is not merely that a few farmers operate on
a much larger scale than others... but that there are many rich
farmers who have entirely different economic aims from many poorer
farmers." She observed that the tendency of scholars to ignore wealth
differences among African peasants had hampered our ability to under
stand their systems of production.
Today, researchers are paying far more attention to
intra-community differences between farmers. It is recognized that
wealthy farmers have differential access to land, labour, animal
inputs (traction and manure) and credit, to name a few important
production parameters. In addition with differential savings and
investment possibilities, wealthier farmers have different attitudes
to risk and innovation (Cancian, 1978) than their poorer counterparts
on the survival fringe.
For pastoral systems, recognition of the importance of
wealth inequality within communities has lagged behind. Much of the
household level research in pastoral production systems has been done
(and continues to be done) by anthropologists who have paid insuf
ficient attention to the issue of economic inequality in traditional
production systems. The ideology of equality which tends to predomi
nate in pastoral societies as well as the apparent similarites in
consumption levels of different households, (coupled with a theor
etical tradition which emphasised the homogeneity of traditional
communities) , have all contributed to this apparent bias of early
anthropologists in studying both peasant and pastoral societies.
Unfortunately, as Konezacki (1978) noted in his book
entitled The economicsof pastoralism: A case study of sub-Saharan
Africa, there are few data available on the distribution of livestock
ownership within pastoral communities; the little data which are
available however indicate that "the prevailing pattern of wealth,
and consequently income distribution among African societies dependent
on animal husbandry, is one of inequality."
- 238 -
Table 1 presents data from several African pastoral
societies on the distribution of livestock holdings between households.
These cases were chosen not because they represent extremes of
inequality, but rather because they are the few cases for which data
were readily available. In each instance there is a marked inequality
of livestock holdings, which can be taken as a close approximation of
wealth especially in a purely pastoral system where land is communally
held.
Table 1. Inequality in livestock distribution: a few African
examples.
Place Unit No. of
HH
Mean Median Range Date
1. Tuareg - A L.U 14 30 34e 0-58 1979
2. Tuareg - B Camels 31 21 17e 2-83 1982
3. Wodaabe Cattle 75 16 18e 5-45 1982
4. Somali L.U 36 132 70 4-660 1950
5. Sebei Cattle 42 21 12 0-100+ 1960
6. Maasai Cattle 41 109 58 4-499 1980
estimated
Sources : 1. Swift (1979)
2.3. Wilson and Wagenaar (1982)
4. Lewis (1961)
5. Goldschmidt (1976)
While there are some indications in the literature of
the extent of intra-community wealth differences in pastoral societies,
there are few, if any, systematic explorations of the effect of wealth
differences on production parameters. Intuitively, we would expect
differences in wealth to affect production in important ways in a
pastoral system. Wealth in the form of animal numbers both enables
(and necessitates) a large family which can be garnered through
polygamy, adoption, and the incorporation of poor relations as depen
dents. Greater wealth also means fewer cash flow problems, and
- 239 -
increased possibilities of access to purchased inputs, including drugs
and mineral supplements. Clearly a producer with 400 animals will
have different management strategies and possibilities than a producer
with only 4 animals. Wealthier pastoralists, particularly those with
sufficient labour within the household, have more management options
in terms of herd-splitting and lending of animals (both for risk
avoidance and to maintain useful social networks) . They would be
likely to have more control over grazing and watering than poorer
producers who are not able to herd independently. We would expect
breeding to be affected as poor producers might not have sufficient
access to a male for servicing whereas wealthy producers might have
difficulty with seasonal and other control of breeding in large herds
and flocks.
ILCA's research in several rangeland systems, particu
larly in Kenya, has focused on the effects of wealth on livestock
production. Table 2 presents preliminary findings from ILCA's on
going research in Kenya Maasailand. It shows that virtually every
important production parameter varies considerably with the wealth
rank of the producer. For ease of presentation, the data refer only
to the rich and poor (leaving out the middle stratum), and only to
one of the three group ranches under study.
Section A of Table 2 provides background information.
Of forty households on the ranch, 13 were classed as poor and 12 were
classed as rich. Of these, the ILCA sample included seven poor and
seven rich households. The poor households owned only 5% of the
livestock units on the ranch whereas the rich households owned 48%
of them.
- 240 -
Table 2. Production unit heterogeneity : example from a Maasai group
ranch.
A. Background
HH on ranch
HH in sample
% of ranches L.U. (total sample)
per household
B. Livestock holdings
1. Value of livestock holdings
2. No. of cattle
3. No. of smallstock
4. SS to cattle ratio
Poor Rich
13 12
7 7
4.9 % 48 %
0.7 6.5
Mean Mean
24,154 234,050
31 302
42 213
1.3 0.7
C. Livestock: offtake
1. Cattle: net offtake %
2. SS : net offtake %
3. Slaughter per capita, Kshs .
20
23
33
7
6
200
D. Livestock herding
1. cattle herd alone (%)
2. SS herd alone (%)
3. Cattle herding group size
0
29
160
57
57
372
E. Labour inputs
1. Total workers 6.3
2. No. of adult men 0.9
3. No. of adult women 1.7
4. Children in school: % 38
5. Ratio of cattle/worker 5
6. HR/day to LS per worker 3.9
7. HR/day per livestock unit per work 1.29
9.8
1.5
3.4
15
31
5.2
0.29
Income
1 . Income from LS/capita (Kshs)
2. Mean value of cattle sold
657/=
577/=
1420/=
971/=
241 -
Table 2 cont.
3. % value of commercial trans
actions in smallstock 18% 7%
G. Expenditure
1. Per cap. food/drink 195/= 465/=
2. Per cap. HH 165/= 238/=
Section B presents information on the livestock holdings of
the sample households. The figures are all means for the households
in that stratum. It shows that rich households have approximately
ten times the value of animals, largely accounted for by their having
ten times the number of cattle of poorer households. Poorer house
holds have proportionally more smallstock, with a smallstock to
cattle ratio of 1.3 as opposed to 0.7 in rich households.
The offtake lata presented in Section C continue to show
clear wealth differences. (de Haan has already mentioned the finding
of Wilson and Wagenaur in Niger that offtake rates are negatively
correlated with herd size). The same is true in Maasailand. The rich
stratum has net offtake rates of 7% for cattle and 6% for sheep;
for poorer households the figures are 20% for cattle and 25% for
sheep. Poor Maasai (as poor pastoralists in many areas of Africa)
(Wilson, 1980; Little, 1982) place far more emphasis on smallstock
than wealthier producers. They own proportionately more smallstock
and derive more of their income from smallstock (18% of livestock
sales rather than 7%). As we would expect, there is far more
slaughter per capita in rich households.
Section D shows that more rich households are able to herd
alone, thus having complete autonomy in decision-making with regard
to grazing. In addition, whether they herd alone or with others,
they benefit from an economy of scale by having very large herd sizes
for grazing, watering and dipping.
Section E presents a few data on labour. Again a clear
difference emerges; rich households have more members, but because
- 242 -
they have far more animals per worker (25 cattle per worker as opposed
to 5 in poor households), they send fewer children to school and work
longer hours on livestock management. Looked at from a different
perspective, however, we see that they benefit both from economies of
scale and also from their ability to marshall non-household labour.
Whereas each worker in poor households devotes over an hour a day
per livestock unit to livestock work, workers in rich households spend
only one quarter of an hour.
Section F shows that rich households have a per capita
income from animal sales which is more than twice that of poor house
holds. As rich households do not suffer the same cash flow problems
of poor households, they are able to keep steers to maturity, thus
fetching higher prices per animal. Again we see that poorer house
holds engage in more smallstock sales and purchases than rich
households .
Finally, Section G of Table 2 shows that higher income is
reflected in higher per capita expenditures on food, drink, and
general domestic goods.
Clearly, as we would expect, producers of different wealth
ranks differ considerably in a variety of ways; from their access to
the essential elements of production, to their herd/flock structures,
offtake rates, and management strategies .
Wealth ranking in pastoral systems research
There are several points in pastoral systems research at which it
would be important to take account of the wealth hetorogeneity in the
area under study. It would be useful in the early stages as part of
the identification of the recommendation domain. Interventions which
are useful to large herd owners might be unadoptable by poorer
producers; in fact, they could intensify and even solidify otherwise
transient economic inequalities by, for example, providing differen
tial access to credit. Secondly, during the verification survey, it
is important to be sure that the producers interviewed represent a
reasonable cross-section of the population for whom interventions are
to be designed.
243
As de Haan mentioned in his introduction, and as I have
tried to demonstrate, evidence suggests that the single most important
parameter for stratifying within a community is wealth rank. However,
among pastoralists wealth is a very difficult parameter on which to
obtain accurate data. For pure pastoralists, livestock holdings
represent a close approximation of wealth. While livestock holdings
can be used as a proxy for wealth in a pastoral production system,
animal censuses are often difficult, if not impossible, to carry out.
Producers are afraid to have their animals counted, due to fear of
taxation, other government interference, or solely on the basis of a
cultural taboo on the counting of animals. Even if the pastoralist
can be persuaded to have his animals counted, the logistic difficulties
of such an operation may be overwhelming. Animals may be scattered
into different management units, animals owned and animals kept are not
isomorphic categories, frequent movements and large distances add to
the costs of such censuses. Some of these logistic problems may be
overcome by counting animals at watering points, but young and ill
animals are likely to be missed. (For agro-pastoralists or pastor
alists, with significant non-pastoral occupational and investment
opportunities, the situation can be far more complex).
This paper contends that livestock censuses or other
objective measurements are not necessary to establish the wealth rank
of producers within a ..ommunity. The technique of having local
informants who rank members of their community according to wealth can
yield similar results at a fraction of the time and expense.
The technique of informant wealth ranking
Informant ranking was first used by Silverman in 1966 in a socio
logical study of prestige hierarchies. It has subsequently been used
by myself and several other anthropologists to elicit wealth ranks.
Informant ranking uses a card sorting technique in which the name
of each producer is written on a small card and several informants
are asked to place the cards in piles according to the wealth of each
producer. It has three basic requirements which will be discussed in
turn: 1. a list of producers to be ranked;
2. a few reliable informants;
- 244 -
3. The elicitation of the word or phrase in the local
language according to which the producers are to be ranked (in this
case wealth) .
List of producers
The technique is dependent upon ranking producers in relationship to
each other; obviously the ranking will reflect true wealth differences
in the population only to the extent that all producers are included.
This is not as overwhelming a task as it might first appear. Ways of
eliciting the list of producers within the community under study
include using censuses, tax roles, and/or voter registration, which
can be cross-checked with chiefs, elders or other community members.
If there is a known watering regime, names can be elicited at dry-
season watering points. With more sedentary populations, air photo
graphs can be used by "walking" informants through them. If, as in
Maasailand, producers have an area that they consider "home", whether
or not they are present, their names can be elicited on a neighborhood
by neighborhood basis. In one of the group ranches under study in
Kenya, for which there was no accurate list of resident households,
a neighborhood by neighborhood interview elicited the names of 206
household heads who normally use the ranch. This was done in an area
2
of 1350 km in less than a week.
The necessity of having as complete a list as possible and
having all the producers ranked, cannot be overstressed. One recent
study in Kenya (White and Meadows, 1981) proposed to examine wealth
differences in group and individual ranches in Kenya. The chairman
of each group ranch was asked to suggest 12 members for study: four
with few livestock, four with an average number of livestock and four
with more than average. The researchers were pleased to find signifi
cant differences between their wealth cohorts; they assumed that their
sampling technique had tapped wealth differences on the ranches.
Unfortunately that was not so. One of the ranches they studied is a
ranch on which ILCA has been working for two years. In checking the
sample households against the population (for which we had complete
livestock censuses), it was discovered that the three strata used by
White and Meadows represented the middle class, upper middle class and
very rich respectively. Poor households were entirely excluded.
- 245 -
This was not through the ignorance of the chairman as we
shall see later, but perhaps was due to his assuming that researchers
studying animal and especially cattle production would have no interest
in households with very few cattle. Table 3 shows the differences
between White and Meadows strata and those derived from a census of
all ranch households. Whereas their "poor" households owned a mean
of 51 cattle, the true population mean is 15. For the "average"
households, they show a mean of 218 cattle whereas the population mean
is 58. For the "rich" households the Meadows and White mean is 378
cattle while the true mean is 249. For the ranch, they calculate a
household mean of 215 cattle, whereas the true mean is only 109. Had
White and Meadows used the technique suggested here in which a complete
list of households were obtained, and ranked, a much more representa
tive sample and hence a better study would have followed.
Reliable informants
As the degree of inter-informant reliability on wealth ranking tends
to be quite high, only a few informants are required. For a small
community (i.e. up to 100 producers), where it is likely that everyone
kno,7.q everyone else fairly well, four or five informants (preferably
of different ages) should suffice. Larger communities may need to be
u voided into smaller sub-areas (defined by the local social structure).
Although I have not yet had the opportunity to do this myself, I see
no a priori reason why the results from different sub-areas should not
be merged, as long as there is sufficient overlap between the rankers.
For example if each of three groups of informants rank 250 producers
of whom each knows 100, the overlap should be sufficient to make a
composite rank.
Despite ideologies of equality, most pastoral producers
recognize that there are at least temporary differences in wealth
status of the members of their community. They also recognize that
people in dir "erent wealth strata are in qualitatively different
positions -;ir • regard to management possibilities. In the two com
munities wheri" I have done informant wealth ranking, I have had no
difficulty in finding a few informants willing to do the task, probably
because no precise information (e.g. how many cattle, how many wives)
is required.
- 246 -
Table 3. Distortion through poor sampling.
Mean household cattle ownershiLP in a Maasai group ranch
Poor Average Rich
Ranch
mean
Sample
Population
51
15
218 378
249
215
10958
Table 4. Scoring an informant's ranks.
Rank group No. of men Men plus
dummy case
Final score
1 (high) 13 13 6.5
2 15 16 21
3 9 9 33.5
4 (low) 13 13 44.5
Unknown 1
Local concept for wealth
In order to ensure the comparability of the data obtained through the
various informants, as well as to ensure that they are ranked
according to the criteria the researcher desires, it is important to
spend some time, preferably with one or two good interpreters,
eliciting the exact local cultural concept to be tapped during the
ranking.
Most cultures have a clear concept of wealth. Once the word
or phrase is elicited it is useful for the researcher to spend some
time determining the various elements that are considered in deter
mining the wealth rank of a producer. Having done this with
interpreters, it is also useful to elicit this information from
informants after they have completed the ranking. This ensures that
- 247 -
the informant has understood what was required; it also can provide
valuable insight into sources of wealth and differential access to
local resources.
In Maasailand, the concept used was "emali" a word meaning
"property" but most commonly used with specific reference to livestock
holdings. Both interpreters and informants felt that the primary
indication of a man's wealth was his livestock holding; subsidiary
points were smallstock to cattle ratio and family size. I was told
that the few Maasai who have been given individual title deeds to
land would have had that property included should they have been
included among the producers to be ranked.
Computation of the wealth rank
As noted, several informants are asked to rank the producers according
to wealth by putting the cards on which their names are written into
piles. The informant is allowed to choose the number of piles he
wants. Frequently in the course of sorting, the number of piles will
be increased as names appear of producers who don't quite fit into
existing piles. Informants need not be literate; the names can be
read to them after which they will just point to the pile where it
belongs. The cards of producers whom the informant is not able to
rauif are just placed to one side. When the ranking is finished, the
names in each pile are read to the informant. This allows for the
d tection of any errors; it also gives the informant a chance to
review the groups he has compiled and make any necessary changes.
On the basis of the sortings, a score is constructed for
each producer for each informant. Each producer is assigned a score
which is the equivalent of the midpoint value of the category in which
he fell. As the number of unknown producers varies between informants
a number of dummy cases equal to those of unknown producers are
inserted at equidistant intervals. Table 4 shows how the scores for
one informant were calculated on a Maasai population of 51 producers.
In order to arrive at a composite score for each producer, the scores
for that producer are simply averaged together.
- 248 -
Informant wealth ranking in Maasailand: A case example
In order to test the use of informant ranking in a pastoral population
(I had previously only used it with mixed farmers) , I had four inform
ants rank the 51 producers who live on a single group ranch in Kenya
Maasailand. The results of that exercise are reviewed here briefly
as a case study.
Table 5 shows the inter-informant rank correlations. The
ranking of each informant is compared to the rank of every other
informant. Informants 1 and 2 are young men of middle wealth. In
formant 3 is an older, poorer man who happens also to be the chairman
of the group ranch who had suggested the White and Meadows sample.
Informant 4 is middle aged and the richest man in the population.
The informants represented the four different neighborhoods on the
ranch. Despite the differences in the informants' backgrounds, the
inter-informant reliability is quite high (the Spearman's rho corre
lation coefficients range from 0.87 to 0.91). When the rank of each
informant is compared to the composite or average rank, the corre
lations are even higher 0.95 to 0.98). These are all Spearman's rho
and all are significant at P = 001). Clearly, despite age, wealth
and neighborhood differences between the informants they were in high
agreement about the relative wealth of the members of the ranch.
In the Maasai ranking, scores ranged from a high of 3.7 (the richest)
to a low of 49.5 (the poorest).
Table 5. Maasai wealth ranks.
Inter-informant rank correlation
12 3 4 Average
1 - 0.88
2 - -
3 - -
4 - -
N = 52
0.87 0.91 0.96
0.88 0.91 0.95
- 0.90 0.96
- - 0.98
- 249 -
Table 6. Informant-livestock unit correlation.
Livestock INFORMANT
unit 1 2 3 4 Average
A.
B.
0.92
0.95
0.95
0.94
0.96
0.97
0.96
0.97
0.97
0.99
N = 22
N = 20
All correlations are significant at P = 0.001.
Table 6 shows the results of a correlation between the
informant wealth rank and a ranking of producers based on an ILCA
census about six months previously. It refers to the subset of 22
producers with whom the census had been undertaken. The Spearman's
rho is 0.97 (Table 6A) . If the two cases for which it is known that
there were censusing problems (counting of more than one producer's
animals), the correlation rises to 0.99 (Table 6 B) (again both with
a significance level f P = 0.001). Clearly the informant results
gave an almost identical ranking of producers as the census. Fig. 1
plots the producers along two axes: livestock units censused and
informant wealth rank. It presents graphically what the correlation
coefficient has already indicated.
The high correlations are not an artifact of a small size.
Fig- 2 depicts the relationship between the wealth ranks (done in late
1982) with livestock units censused in mid-1980 for the 41 households
which were then resident on the ranch. Again the relationship is
quite clear, although there has not been time to correlate the coef
ficients. Had the Kenya team stratified it's sample by informant
wealth ranks, the results would have been identical to stratifying
on the basis of a complete animal census. In fact, in a certain
respect the stratification would have been better, as the initial
census contained a number of significant errors as one would expect
given pastoral resistance to censuses. Census errors are circled in
Fig 2. Some of these are minor (e.g. counting a poor relation's
animals with a rich man's herd) but others are significant.
- 250 -
J2
U
1
1-1
CO
o
u
o
o
10
Livestock units
i
99
ILCA census
514
Fig 1. Informant vs. census rank.
- 251 -
o
00
ON
60
G
c
u
en
3
to
c
<u
o
<
•H
®
®
X
X
X
X
X
-V
®
X
X
®
X
X
X
-V
X
®
®
s
X
X
X
X
Jr
X
X
X
|x
X
x O census error
LJ documented change
Poor Informant ranking (1982) Rich
Fig 2. Informant ranking (1382) vs. initial ILCA census (1980).
- 252 -
Having obtained an informant wealth ranking, the researcher
can use stratified sampling to enhance his precision. Informant
wealth ranking does not totally replace censusing. After stratifying
and sampling it would be critical to do an animal census to know the
holdings of the sample and generalize to the holdings of the community.
On the other hand, if data must be collected opportun
istically (i.e. with those producers who are amenable or accessible),
informant wealth ranking can still be used to gauge the representa
tiveness of the producers interviewed or censused.
Informant ranking techniques need not be limited to wealth.
Producers can be ranked on almost any variable including age, family
size, management ability etc. The technique, as I hope I have shown,
is rapid, simple yet accurate.
References
Cancian, F. 1978. The innovator's situation: Upper middle class
conservation in agricultural communities. Stanford Uni
versity Press, Stanford, California.
Goldschmidt, W. 1976. Culture and behaviour of the Sebei: A study in
continuity and adaptation. University of California Press,
Berkeley, California.
Hill, P. 1972. Rural Hausa: A village and setting. Cambridge
University Press, Cambridge. P.S.
Konczacki, Z.A. 1978. The economics of pastoralism: A case study of
sub-Saharan Africa. Frank Cass, U.K. p. 49-51.
Lewis, I.M. 1961. A pastoral democracy. Oxford University Press,
London.
Little, P.D. 1982. Risk aversion, economic diversification and goat
production: Some comments on the role of goats in African
pastoral production systems. In: Proceedings of the Third
International Conference on Goat Production and Disease,
Tucson, Arizona. University of Arizona, pp. 428-430.
253 -
Neadows, S.J. and White, J. M. 1981. Evaluation of group and individual
ranches in Kajiado district, Kenya, to economic development
and pastoral production strategies. Kenya, Ministry of
Livestock Development, Nairobi.
Swift, J. 1979. The economics of traditional nomadic pastoralism:
The Tuareg of the Adrar N Iforas (Mali) . University of
Sussex. (Unpublished, Ph. D dissertation).
Wilson, R.T. 1980. Indigenous goats and livestock development in
Africa. Manuscript.
Wilson, R.T. and Wagenaar, K. 1982. An introductory survey of
livestock population demography and reproductive performance
in the area of the Niger Range and Livestock Project. ILCA/
Mali, Arid and semi-arid zones programme, Bamako.
- 254 -
Importance de l'effet de la richesse sur la production pastorale: Une
méthode rapide de stratification de la richesse
Résumé
Cet exposé traite de l'importance des différences de la richesse
entre les producteurs dans les systèmes de production pastorale
traditionnelle. Il soutient qu'il existe des différences importantes
en ce qui concerne la richesse. Que celles-ci ont des effets importants
sur les stratégies de production et que la recherche sur les systèmes
pastoraux doit en tenir compte à divers niveaux, qu'il s'agisse de
la définition de la population-cible ou de la mise au point et de
1' expérimentation d' interventions.
Il décrit une méthode rapide de détermination des strates de la
richesse auxquelles appartiennent les producteurs dans une
collectivité donnée. Chez les pasteurs, la- richesse est un paramètre
sur lequel il est très difficile d'obtenir des données exactes. Pour
les éleveurs purs, la possession de bétail représente une quasi-
richesse. Mais les recensements des populations animales sont
souvent difficiles, voire impossibles à effectuer. L'étude soutient
que les recensements du bétail ou d'autres mesures objectives ne sont
pas nécessaires pour établir le rang (en termes de richesse) des
producteurs au sein d'une collectivité. L'utilisation d'informateurs
locaux pour classer les membres de leur collectivité en fonction de
leur richesse peut fournir des résultats similaires en moins de temps
et à un moindre coût.
L'informateur utilise une technique de tri de cartes dans laquelle le
nom de chaque producteur est écrit sur une petite carte et plusieurs
informateurs sont invités à ranger les cartes les unes au dessus des
autres en fonction de la richesse des producteurs. La technique
requiert l'obtention d'une liste complète des producteurs opérant au
sein de la collectivité.
Etant donné que le degré de fiabilité des informateurs pour le
classement en fonction de la richesse tend à être élevé, seul un
nombre limité de ceux-ci sera nécessaire. Pour une petite
collectivité où l'on suppose que tout le monde se connaît assez bien,
- 255 -
quatre ou cinq informateurs devraient suffire. Pour assurer la compa-
rabilité des données obtenues par le biais de divers informateurs et
pour s'assurer que le classement s'effectue selon les critères
souhaités par les chercheurs, il est important de consacrer quelque
temps, de préférence avec un ou deux bons interprètes, à l'étude de
concepts culturels locaux à exploiter au cours de la stratification.
Sur la base du tri des cartes, des points sont attribués à chaque
producteur pour chaque informateur. Chaque producteur reçoit des
points qui constituent l'équivalent de la valeur moyenne de la
catégorie à laquelle il appartient. Pour avoir la performance globale
de tels producteurs, on fait simplement la moyenne des points obtenus
par ces producteurs.
Après l'obtention de la stratification des richesses par les
informateurs, le chercheur peut utiliser des échantillons stratifiés
pour plus de précision. La stratification des richesses par les
informateurs ne peut Das remplacer complètement les recensements.
Après la stratification et l'échantillonnage, il convient de
procéder à un recensement des animaux pour déterminer la population
animale détenue dans l'échantillon et pour extrapoler à partir de
celle-ci la structure de la propriété au sein de la collectivité.
Les techniques de classification par informateur ne doivent pas
nécessairement se limiter à la richesse. Les producteurs peuvent être
classés sur la base de presque toutes les variables, y compris l'âge,
la taille de la famille, l'aptitude à gérer, etc. La technique est
rapide, simple et précise.
- 256 -
le regroupement des éleveurs ou des zones rurales selon divers
critères. Il a fait allusion au document de M. Cossins qui a mis
l'accent sur la tendance des Afars à réduire leurs troupeaux de petits
ruminants et a déclaré qu'en Afrique de l'Ouest c'était la tendance
contraire. La possession de petits ruminants permettait une plus
grande efficacité et une plus grande flexibilité que celle des bovins,
notamment parce que :
- la distribution entre les familles était moins concentrée;
- la perte en valeur unitaire était plus faible en cas de vente
forcée ou de perte;
- les petits ruminants étaient moins sensibles aux pertes
pondérales lorsque l'affouragement est insuffisant;
- la performance du petit bétail en matière de reproduction était
meilleure et permettait par conséquent de reconstituer le
cheptel beaucoup plus rapidement.
Le Dr Zulberti a mis en doute l'affirmation de M. Cossins selon
laquelle "l'éleveur est la seule personne dont l'existence se base
sur le lait et non pas sur la viande". M. Cossins a indiqué qu'il
faisait allusion aux zones arides et désertiques où (à l'exception
des zones irriguées) la laiterie était rare si tant est qu'elle y ait
jamais existé. Le Dr Zulberti a exprimé son accord sur l'affirmation
selon "laquelle l'éleveur cherche à optimiser le nombre de personnes
qu'une superficie donnée est capable d'accueillir" M. Cossins a
répondu que si les éleveurs n'avaient pas comme objectif d'optimiser
le nombre de personnes par unité de surface, ils n'avaient pas
besoin alors de faire du pastoralisme; le pastoralisme constituait
presque toujours un moyen d'accueillir plus de personnes par unité de
surface que ne le permettraient les autres modes de production dans
les zones arides. Dans ses réflexions sur le document de M. Cossins,
le Dr Diakité a souligné que des approches très différentes devaient
être adoptées pour étudier le pastoralisme en Australie et en
Afrique. M. Cossins a exprimé son accord et a déclaré qu'il espérait
aue dans son document, cette distinction avait été clairement établie.
- 261 -
Le Dr Timbo a indiqué que dans le contexte du Sahel, les stratégies
des éleveurs étaient régies dans une large mesure par la loi du plus
grand nombre, afin de combattre les problèmes d'ordre climatique ou
de santé animale. L'objectif primordial des éleveurs était la produc
tion des jeunes animaux/veaux et non la production de lait. Dans le
Sahel, la production de lait n'avait jamais constitué une alternative
de production au-delà de l'autosuffisance traditionnelle de la famille
pastorale.
Dans les observations qu'il a formulées sur le document de M. Cossins,
le Dr Rhissa a déclaré qu'en étudiant les éleveurs et leurs stratégies
de production, les conclusions devaient tenir compte des objectifs du
bénéficiaire de l'étude. L'objectif du gouvernement peut par exemple
être d'aider les éleveurs à mieux s'organiser, d'assurer la sécurité
alimentaire d'une zone et d'accroître la part des produits pastoraux
dans l'économie nationale. Les objectifs de l'éleveur étaient de
fournir du lait, d'assurer sa propre sécurité alimentaire et sociale.
Le Dr Rhissa a déclaré qu'à son avis, les objectifs de l'éleveur
étaient souvent négligés.
Le Dr Tilahun a fait allusion au document de Mlle Grandin et a indiqué
que dans l'étude sur l'inégalité, une distinction devrait être faite
entre le revenu du ménage et le revenu du ménage par tête. Mlle Grandin
a accepté cette observation et a déclaré que les études devraient se
pencher sur cette différence dans leurs analyses, notamment dans les
sociétés proches du niveau de la subsistance où une forte proportion des
dépenses était consacrée à l'alimentation qui augmentait proportionnel
lement avec le nombre de personnes dans la famille.
- 262 -
Household studies in pastoral systems research
1 2
Barbara E. Grandin and Solomon Bekure
1 2
Anthropologists Team Leader and Agricultural Economist, Arid Zones
(Eastern and Southern Africa) Programme, ILCA, Kenya
Introduction
This section of the workshop deals with household level data collec
tion in pastoral systems research (PSR) . It first briefly reviews
differences between crop production and pastoral production which
affect research possibilities. It then briefly reviews basic types
of methods that may be used to collect household level data. Finally
for three important areas of household level data collection in
pastoral production systems (i.e. livestock transactions, household
budgets, and labour use) it discusses specif ic problems as well as
special techniques which are available for data collection.
Farming vs. pastoral systems research
De Haan briefly mentioned some important differences between cropping
systems and pastoral systems of production. Important differences
also exist in research for these systems. It would be useful to
reiterate some of these, particularly as they affect the collection
of household level data.
Pastoral systems are marked by high mobility which causes
difficulties in sample selection and long-term survey work. Logistic
problems can largely determine possible sampling procedures both in
terms of the general location studied and specific households within
that location. Many areas are largely inaccessible to researchers on
the ground, or would be so at a prohibitive cost. Given the high
mobility, it is difficult to ensure that the same families can be
followed throughout a long-term study. It will be more cost effective,
ceteris paribus, to study more households in fewer communities than
a few households in a large number of communities.
263 -
Pastoral systems are marked by the communal sharing of major
resources such that the focus on the household as a unit of production
is often more problematic with pastoralists than with farmers. Gener
ally, in order to understand pastoral management practices and
strategies, far more attention must be paid to the relationships
between households in a system.
Pastoral systems, unlike cropping systems, are marked by
the need for continuous inputs as well as the production of at least
some continuous outputs. Farming systems research (FSR) studies tend
to concentrate on the agricultural growing season, which can be as
short as a few months. Thus in a fairly short period of time it is
possible to observe all the major inputs to and outputs from crop
production. This is not possible with pastoral production systems,
which have daily outputs and inputs, the nature of which can change
dramatically in the course of an annual cycle. It is not possible to
observe these different inputs and outputs in a short period of time.
(Parenthetically it might be noted that farming systems based on
certain tree crops would resemble pastoral systems in this respect) .
If the research suffers from time constraints, these will be far
more seriously felt in PSR. In the study of pastoral systems, at
least the major seasons and their transitions (Swift, 1982; Grandin,
1982) must be identified and considered. If they cannot be observed
nore reliance must be put on the ability of the producer to recall and
.. -port events.
Pastoral systems are marked by long cycles of drought and
post-drought recovery that are much less important in systems of
annual cropping. More attention needs to be paid to long-term cycles
in PSR, first in order to locate the population under study in a
cyclical time frame and subsequently to be able to evaluate the long-
term suitability and effects of proposed interventions.
Although it is a research difference rather than a production
system difference, it is important to note that testing raises more
serious problems in PSR than FSR. Some of these problems are due to
factors such as mobility which have already been mentioned. Other
relate to the virtual impossibility of simulating pastoral conditions
on research stations.
- 264 -
Summary of Discussion Session 5.
Chairman: Prof. Gunnar Sorbo (Norway)
Discussion led by Dr Sitta Barry (Upper Volta)
Dr Barry commented that CILSS (Comite permanent interetats de lutte
contre la secheresse dans le Sahel) was set up in 1973 by six
countries (Chad, Niger, Upper Volta, Mali, Senegal and Mauritania)
following the drought of 1968 to 1973. Subsequently Gambia and Cape
Verde had joined CILSS, whose mission was to
- increase the awareness of the international community to
drought and its effects;
- mobilise the necessary resources to combat drought and its
effects;
- co-ordinate activities to develop the Sahel and to combat
drought and desertification; and
- co-operate with other organisations having similar goals.
In the field of animal husbandry CILSS had established a strategy
consisting of six themes:
- increasing livestock feed resources;
- increasing livestock productivity using animal health control
methods;
* organising professional channels for livestock production;
- modernising and improving marketing channels;
- recruitment, training and extension work;
- research.
Dr Barry reviewed the papers by Drs Cossins and Grandin and commented
that demographic analysis should be carried out over homogeneous
environments which may mean the regrouping of pastpralists or rural
areas according to different criteria. He referred to Dr Cossins'
paper which emphasised the tendency of the Afar to destock small
ruminants, and said this was the opposite of the trend in West Africa.
The possession of small ruminants could be more flexible and effective
- 257 -
than that of cattle because:
- the distribution between families was less concentrated;
- the loss in value per unit was lower in the case of forced
sale or loss;
- smallstock were less sensitive to weight loss in case of
insufficient feed;
- the reproductive performance of smallstock was better,
therefore capital could be reconsolidated more rapidly.
Dr Zulberti questioned Dr Cossins' statement that 'the pastoralist is
the only one who depends on milk and not meat' . Dr Cossins pointed
out that he was referring to arid and desert areas where, with the
exception of irrigated areas, dairying was rare if existant.
Dr Zulberti disagreed with the statement that 'the pastoralist seeks
to optimise the number of people that any area is capable of supporting' .
Dr Cossins replied that if pastoralists did not have the aim of
optimising the number of people per unit area then they would not
need to pursue pastoral ism; pastoralism almost certainly evolved as
a way of supporting more people per unit area than any other production
mode in aride areas.
In commenting on Dr Cossins paper, Dr Diakite stressed that very
different approaches had to be adopted for Australian and African
pastoralists. Dr Cossins agreed and said that he hoped his paper
had made that very clear.
Dr Timbo suggested that in the Sahel context, the strategies of
pastoralists were governed to a large extent by the law of the
greatest number in order to combat climatic or animal health problems.
The primary aim of pastoralists was the production of young stock/
calves, not milk production. In the Sahel milk production had never
been a production alternative beyond the traditional self-sufficiency
of the pastoral family.
- 258
Dr Rhissa, in commenting on Dr Cossins paper, said that when studying
pastoralists and their production strategies the conclusions needed
to be qualified depending on whose behalf one had been carrying out
the study. A government's aims may be to help herdsmen organise
themselves better, to provide the food security of a zone and to
improve the contribution of pastoral products to the national economy.
A pastoralist' s aims were to provide milk, to attain his own food
security and his own social security. Dr Rhissa felt that the
pastoralists aims were often overlooked.
Dr Tilahun referred to Dr Grandin's paper and suggested that in the
study of inequality a distinction should be made between household
income and household per capita income. Dr Grandin agreed and said
that studies should investigate these differences in their analysis,
particularly in societies close to the subsistence level where a
high proportion of expenditure was on food which increased propor
tionally with the number of people in the family.
ill
- 259 -
Résumé des débats de la cinquième séance
Président: Prof. Gunnar Sorbo (Norvège)
Débats dirigés par le Dr Sitta Barry (Haute-Volta)
Le Dr Barry a déclaré que le CILSS (Comité permanent inter-Etats de
lutte contre la sécheresse dans le Sahel) avait été créé en 1976 par
six pays (le Tchad, le Niger, la Haute-Volta, le Mali, le Sénégal et
la Mauritanie) à la suite de la sécheresse 1968 - 1973. Par la suite,
la Gambie et le Cap-Vert avaient adhéré à l'organisation dont la mission
consistait à :
- amener la communauté internationale à prendre conscience de
la sécheresse et de ses effets;
- mobiliser les ressources nécessaires pour combattre la
sécheresse et ses effets;
- coordonner les activités pour développer le Sahel et pour
lutter contre la sécheresse et la désertification; et
- coopérer avec d'autres organisations ayant des objectifs
similaires.
Dans le domaine de l'élevage, le CILSS avait mis en place une stratégie
basée sur six thèmes:
- accroître les ressources fourragères;
- accroître la productivité animale en utilisant des méthodes de
contrôle de la santé animale;
- organiser des circuits professionnels pour la production
animale;
- moderniser et améliorer les circuits de commercialisation;
- recruter et former du personnel et effectuer des travaux de
vulgarisation;
- entreprendre des activités de recherche.
Le Dr Barry a fait réference aux documents de M. Cossins et de
Mlle Grandin et a déclaré que l'analyse démographique devrait être
effectuée sur des environnements homogènes, ce qui pourrait signifier
Lastly, PSR must differ from FSR because it suffers from
a dearth of reliable data on which to build. The bulk of research in
agricultural production in Africa has been on crop production; even
within farming systems little attention has been paid to the livestock
sub-sector. Gilbert et al (1980) noted with regard to the initial
stages of FSR, "The poorer the data and informational base is, the
more research at this stage becomes an art rather than a science and
the more on-farm studies are needed to describe and diagnose the
areas' characteristics and constraints". Yet as Eicher and Baker
(1982) noted, "In summary, research on the behaviour of livestock
herders in Africa is about at the same point where research was on
the economics of crop production some 20 years ago - many assertions
and a sparse supply of facts".
The need for household level data in PSR
PSR is aimed at increasing the productivity of pastoralists and
improving their welfare. As much resource management, inputs, and
outputs occur on the household level, data must be collected at this
level to understand these important production parameters. Addition
ally, in the final analysis interventions generated by PSR have to be
acceptable to and adopted by individual pastoral households. A good
understanding of the perceptions, goals, strategies, and decision
making of individual pastoral households is essential if researchers
are to effectively identify constraints and design and test inter
ventions which must be implemented at the household level. (This does
not negate the need to pay attention to extra-household relationships
as noted above) .
General methods available for PSR household level data collection
Although PSR will inevitably differ from FSR, there are general methods
used in FSR (and generally in survey research) which can be used in a
PSR framework. It is not in the scope of this paper to review these
methods in detail. A number of excellent texts are available for that
purpose (Collinson, 1972; Byerlee et al, 1980; Kearle, 1976; Bernstein,
1979). Nevertheless, it would be useful to list basic techniques for
gathering household level data and evaluate some of their advantages
- 265 -
and disadvantages particularly with respect to criteria used in FSR.
Within a pastoral equivalent of FSR, the first questions would be:
what exactly do we need to know and what is the most cost-effective
way of obtaining the necessary data? The answer to these questions
will depend on the nature and phase of the research project itself,
and will determine the specific data and methods required.
Types of data collection techniques which might be used in
a PSR context can be usefully scaled along two dimensions: formality
and frequency. They range from informal, open-ended interviews
normally done by the researcher himself, to highly formalised pre-coded
questionnaires normally administered by a trained enumerator. In
terms of frequency, techniques may require either single visits,
limited visits, or multiple visits which may be continuous throughout
the study period, or may use seasonal sub-samples. Multiple visit
surveys themselves may vary considerably in terms of the frequency
with which the data are collected from each producer.
Informal surveys
Informal surveys play a major role in the early stages of FSR. They
are conducted by the researchers themselves without the use of enumer
ators, questionnaires, or random samples, and hence are rapid and
qualitative. In the early stages of research, informal surveys
provide critical background information on the production system,
and form the basis for later, formal questionnaires. Even with the
wealth of data available on cropping systems, Byerlee et al (1980)
noted: "It is our experience that researchers rarely spend sufficient
time on this rewarding and essential task." In PSR, informal surveys
will need to play an even more important role.
Although informal techniques are often used primarily in
initial surveys, it is important to note that they can be and should
be more frequent. As Byerlee et al (1980) noted "(this) type of
informal researcher-farmer dialogue... should be a continuous process
through all phases of the programme". Informal surveys can elicit
types of information that are impossible to collect with formal
questionnaires. These include information on producers1 strategies,
decision-making, and social aspects of the production process. Sen
- 266 -
sitive information is also more likely to be provided in informal
contact with a researcher than with an enumerator. Generally,
informal surveys give the researcher the opportunity to interact
directly with the producer and explore issues as they arise, rather
than in a predetermined way.
Informal surveys by themselves have several disadvantages.
Firstly, as the producers are usually chosen opportunistically, the
representativeness of the sample may be a problem, especially in an
area where there is significant heterogeneity among producers. Sec
ondly, as informal surveys are qualitative, their validity is more
likely to be questionned by decision-makers and planners.
Case studies may be seen as a variant of the informal survey
in which a carefully selected sample of households are chosen
for intensive study, including frequent visits by the researchers
themselves. Case studies are particularly valuable in providing data
on the full range of production parameters for the sample households,
including data on goals and decision making, which are impossible to
glean from formal questionnaires. Through case studies, linkages
in the system are more likely to be understood.
Formal surveys
Formal surveys rely upon the administration of precisely designed
questionnaires by enumerators. Formal surveys have the advantage of
significantly expanding the number of households which can be included
in any study. They provide standardised and quantifiable data,
although there are inherent limitations in the type of data that can
be collected. While sampling errors are easily decreased through
formal surveys, measurement or observation biases inevitably increase.
Formal surveys also have the disadvantage of requiring significant
inputs for training and supervising enumerators, for coding, data
preparation and analysis.
The amount of measurement error in formal surveys depends
upon many factors, including the quality of the enumerators and the
questionnaires, the type of data to be collected, the frequency of
the visits, and the cooperation of the respondents.
- 267 -
Enumerators and questionnaires
The use of enumerators necessarily introduces biases in the data col
lected. While it is impossible to eliminate the 'enumerator effect1,
steps should be taken to minimize it through the proper selection,
training and supervision of enumerators as well as attention to
details of questionnaire design.
In addition to their suitability to conduct interviews,
the enumerators selected should, to the extent possible, have the
same ethnic background and if possible be known to the target
population. This facilitates their acceptance by the latter and
enhances their ability to explain the questions in terms and concepts
the respondents understand. The latter is very important for stan
dardisation, and will be improved upon even more if the questionnaire
format is translated into the local language.
Great attention should be paid to developing the question
naire, to be sure it taps the precise information required, to phrase
it so that it enhances the recall of the respondent, and to standard
ise it to lessen "enum:-ator effects". Evidence suggests that the
more precise the questions, the more likely the respondent is to
recall and relate the answer. On the whole this is more tedious for
both the enumerator and respondent, but if fatique can be avoided,
it is far superior in the accuracy of the data produced.
The accuracy of enumerators is enhanced by training them so
that their understanding of the terms used in the questionnaires and
the procedures for administering them is standardised. In training
enumerators, the ILCA Kenya team asked them as a group to translate
the proposed questionnaires into the Maa language. Considerable
debates followed which were instructive to both the enumerators and
the researchers. Once the questionnaires were standardised and
pretested, role playing was used for further training, with the
researchers trying to approximate all conceivable problems. Lastly
enumerators were observed while they administered the questionnaires
to non-sample households. If at all possible, one should start
training more than the required number of enumerators so that at the
end of training one will have a better selection on the basis of their
observed performance.
- 268 -
If a questionnaire is at all complicated, the enumerators
will still be learning in the initial phases of administration. Close
supervision will be required in checking the responses so that errors
(either in format design or administration) can be corrected before
they are made repeatedly. The first set of data collected by the
enumerators should be analysed as quickly as possible to assess its
quality.
One cannot over-emphasise the high degree of vigilance
required to ensure good quality data in survey work. (For a detailed
discussion of quality control and correction of enumerator bias using
analysis of variance, see Zarkovich, 1966).
Data type
Not all events are equally well remembered. In terms of probability
of good recall, events can be scaled along two dimensions: frequency
and regularity. Events which are regular are less likely to be
remembered individually, but producer estimates of their occurrence
are likely to be adequate. Irregular events which occur rarely (such
as livestock transactions) are likely to be remembered individually,
but frequently occurring irregular events will pose great difficulty
in recall and will necessitate high frequencies of questioning.
Labour inputs and expenditures fall in the latter category.
Although this section concentrates on recall, it must be
noted that information on certain types of events (e.g. treatment of
animals) might best be gathered by actual observation rather than
questioning. However, for most data, observation is far more time
consuming and consequently far more expensive than recall.
Frequency of visits and end effects
The frequency of visits needed depends to a large extent on the type
of event to be recalled (or observed) . The frequency will also depend
upon the depth of data desired on the event in question. For example,
if detailed questioning about decision-making is desired, greater
frequency would be helpful.
All surveys based on recall suffer from "end effect", i.e.
the difficulty of specifying the limits of the period for which
- 269 -
information is requested and the tendency of respondents to include
events from outside this reference period. In multiple-visit surveys,
the frequency of the visits will normally define the reference period,
which is normally the time lapsed since the last visit of the enumer
ator. End effect problems occur with the first interview as the end
point of the period of reference is often unclear. It is strongly
recommended that the data from the first interview be excluded from
analysis as it is invariably over-reported. Single-visit surveys
suffer particularly from end affect problems unless the period can
be clearly specified. Collinson (1972) advocates limited visits for
certain data collection and stresses the need to use time frames that
fit a local natural or cultural cycle.
Single-visit format surveys. Single-visit formal surveys form an
integral part of FSR methods as currently advocated by CIMMYT (Byerlee
et al, 1980). Their primary role is to quantify and verify the
hunches which have emerged from the initial informal survey. Single-
visit surveys are most appropriate for collecting information on
variables (such as herd structure) for which a single objective
measurement will suffice. They may also be used to collect general
data on a relatively small number of variables particularly when the
essential parameters of the production system are already well under
stood. For well remembered and registered events (e.g. livestock
transactions) single-visit surveys may be able to elicit time-depth
data on actual events. For non-registered events, single-visit
surveys rely more on the producer's ability to estimate, based on
general experience, rather than recall actual events; as such, they
produce averages rather than indicate variability, and are inadequate
when time series data are required.
Multiple-visit formal surveys. Multiple-visit formal surveys have
several advantages over single-visit surveys. They provide time
series data, are more likely to reflect actual events rather than
estimations, benefit from learning on the part of both the enumerator
and respondent. However, as the number of visits is increased, given
the same amount of resources, the number of households must be de-
270 -
creased. Again, careful attention must be paid to the nature of me
data required in order to decide between measurement and sampling
biases.
In literate societies, it is possible for respondents to
keep written records of the variables under study. This is almost
impossible in pastoral systems research. However, Swift (1981) be
lieves it may be possible to find, hire, and train local residents
to collect time series data, with a minimum input from senior re
searchers.
When time series data are desired, decisions must be made
as to whether the recording will be continuous or non-continuous.
For example, rather than recording continuous expenditures, it might
be possible with care to subsample within the important seasons.
Respondent cooperation and bias
Whatever the information collected, it is necessary to distinguish
between the respondent's ability and his willingness to respond. The
producer's ability to respond to the questionnaire is largely a
function of survey design: whether the questions are well phrased
and solicit information which it is possible for him to remember.
Multi-visit surveys appear to have a learning component, so that
some improvement in recall occurs, provided that it is not offset by
respondent fatigue.
A respondent's willingness to respond, however, is a func
tion of his general level of cooperation as well as the sensitivity of
the data solicited. FSR researchers frequently note value of early
interventions in increasing the general level of cooperation. Never
theless, there are still likely to be social/psychological factors
which encourage inaccurate, and especially selective, reporting. These
may be due to a desire for greater prestige, fear of shame, or a
cultural taboo on certain topics. Case studies and participant
observation are particularly useful to evaluate and adjust for such
reporting biases.
Having briefly reviewed types of data collection methods
for household level research, this session now turns to specific
271 -
topics and methods. We will discuss, in turn, livestock transactions,
household budget, and labour studies.
References
Bernstein, R. 1979. Design and management of agricultural research:
A guide for agricultural researchers. Draft Paper, IRRI,
Los Banos .
Byerlee, D. and Collison, M. 1980. Planning technologies appropriate
to farmers - Concepts and procedures. CIMMYT.
Collinson, M. 1972. Farm management in peasant agriculture. Praeger,
New York.
Eicher, C. and Baker, D.C. 1982. Research on agricultural development
in sub-Saharan Africa: A critical survey. MSU Inter
national Development Paper 1. Department of Agricultural
Economics, Michigan State University, East Lansing, p. 168.
Gilbert, E.H., Norman, D.W. and Winch, F.E. 1980. Farming systems
research: A critical appraisal. MSU Rural Development
Paper 6. Michigan State University, East Lansing, p. 146.
Grandin, B.E. 1982. Time allocation and labour inputs on a Maasai
group ranch: Preliminary findings from Olkarkar. Manuscript.
ILCA, Nairobi.
Kearle, B. (ed) 1976. Field data collection in the social sciences.
Agricultural Development Council.
Swift, J. 1981. Rapid appraisal and cost-effective participatory
research in dry pastoral areas of West Africa. Agric. Admin.
8(6): 485-492.
Swift, J. 1982. The start of the rains. Research Memo 2, Institute
of Development Studies, University of Sussex, Brighton.
Zarkovich, S.S. 1966. Quality of statistical data. FAO, Rome.
- 272 -
Les enquêtes sur les ménages dans la recherche sur les systèmes
pastoraux
Résumé
La recherche sur les systèmes pastoraux a pour objectif d'accroître la
productivité des éleveurs et d'améliorer la qualité de leur vie. Etant
donné que s'effectuent dans le ménage des échanges intersectoriels
et des opérations de gestion, il conviendrait de recueillir des données
à ce niveau pour comprendre ces importants paramètres de production.
Les interventions engendrées par la recherche sur les systèmes
pastoraux doivent être acceptables pour les éleveurs et adoptées par
les ménages pastoraux. Pour que les chercheurs puissent identifier de
manière effective les contraintes et mettent au point puis testent
les interventions qui doivent être introduites au niveau du ménage,
il faudrait qu'ils acquièrent une connaissance approfondie de
l'opinion, des objectifs et de la stratégie des ménages pastoraux
ainsi que du mécanisme de prise de décision.
Les types de techniques de collecte de données qui doivent être
utilisées dans le cadre d'une recherche sur les systèmes pastoraux
peuvent être déterminés selon deux critères: la forme et la fréquence.
Ils varient des interviews informelles à bâtons rompus effectuées
normalement par le chercheur lui-même, aux questionnaires précodés
de type très classique, remplis normalement par un enquêteur qualifié.
En ce qui concerne la fréquence, les techniques peuvent faire appel
à des visites uniques, à des visites limitées ou multiples qui
peuvent être continues pendant toute la période de l'étude ou a des
sous-échantillons saisonniers., Les enquêtes à visites multiples
peuvent elles-mêmes varier en ce qui concerne la fréquence de la
collecte des données auprès de chaque producteur.
Les enquêtes informelles jouent un rôle important dans la phase
initiale de la recherche sur les systèmes d'exploitation agricole.
Elles sont effectuées par les chercheurs eux-mêmes sans la participa
tion des enquêteurs et sont par conséquent rapides et de qualité
élevée. Lors de la phase initiale de la recherche, les enquêtes
informelles fournissent des informations importantes sur le système
- 273 -
de production et constituent la base des questionnaires de type
classique qui seront élaborés après. Les enquêtes informelles
comportent elles-mêmes de nombreux inconvénients. Etant donné que les
producteurs sont généralement choisis de manière opportuniste, la
représentativité de l'échantillon peut poser un problème, notamment
dans un domaine où il y a beaucoup d'hétérogénéité entre les
producteurs. Etant donné que les enquêtes informelles sont qualitatives,
leur validité est beaucoup plus sujette à caution pour les décideurs
et les planificateurs.
Les enquêtes de type classique se fondent sur la distribution de
questionnaires conçus de manière précise par les enquêteurs. Les
enquêtes de type classique ont l'avantage d'accroître de manière
significative le nombre des ménages qui peuvent être inclus dans
une étude. Elles fournissent des données harmonisées et quantifiable»,
quoiqu'il y ait des limites inhérentes aux types de données susceptibles
d'être recueillies. S'il est vrai que les erreurs d'échantillonnage
diminuent facilement danr les enquêtes de type classique, par contre,
les distorsions en ce qui concerne les mesures ou les observations
augmentent immanquable-ment. Les enquêtes de type classique ont
également l'inconvénient d'exiger des ressources importantes pour la
formation et la supervision des enquêteurs en vue du codage, de la
préparation et de l'analyse des données. La quantité des erreurs
de mesure dans les enquêtes de type classique est fonction de
plusieurs facteurs, y compris la qualité des enquêteurs et des
questionnaires, le type des données à rassembler, la fréquence des
visites et la coopération des enquêtes.
L'utilisation des enquêteurs introduit des distorsions au niveau des
données collectées. S'il est impossible d'éliminer l'effet de
l'enquêteur, des mesures doivent être prises quand même en vue de
les minimiser grâce à une sélection appropriée, à la formation et à
la supervision de l'enquêteur de même qu'à la vigilance dans la
conception des divers éléments du questionnaire.
La fréquence des visites nécessaires dépend du type de l'événement
dont il faut se rappeler. La fréquence dépend également de la
- 274 -
"profondeur" des données désirées sur les événements en question.
Le rôle primordial des enquêtes de type classique à visite unique
consiste à quantifier et à confirmer les impressions qui émergent
des enquêtes informelles initiales. Les enquêtes de type classique
à visites multiples ont plusieurs avantages sur les enquêtes à visite
unique. Elles fournissent des données sur les séries chronologiques
et permettent en général de recueillir des informations sur des
événements qui ont effectivement eu lieu plutôt que sur des estimations.
Ces données permettent d'apprendre beaucoup de choses sur l'enquêteur
aussi bien que sur l'enquêté.
- 275 -

Livestock transactions data collection
Barbara E. Grandin
Anthropologist, Arid Zones (Eastern and Southern Africa) Programme,
ILCAS Kenya
Introduction
This paper focuses on livestock transaction data. It discusses the
need for data on livestock transactions in pastoral production systems
and then considers several methods through which desired data can be
collected.
Livestock transactions are defined as all voluntary offtake
and acquisition of livestock, whether permanent or temporary. Thus
for acquisition they exclude birth, but include purchases and ac
quiring animals as gifts, on loan, as bride price or by any other
"socially mediated" means. For offtake they exclude death but include
such transactions as sales, gifts, exchanges of animals (for other
animals or any other product), slaughter, loans etc. Although births
and deaths are excluded by definition from livestock transactions,
methods used to collect data on transactions can frequently be ex
tended to include data on births and deaths.
Need for data on livestock transactions in PSR
Previous studies in pastoral systems have provided little quantitative
data on the range of livestock transactions. The anthropological
literature contains numerous descriptions of the ranges of livestock
transactions open to producers of different societies. On the whole,
however, ideal norms rather than actual behaviours are described. In
contrast, the economic literature focuses primarily on the "final
offtake rate". This is not consistently defined, but normally
includes only sales out of the area and in some cases slaughter for
home consumption (Eicher and Baker, 1982) .
This paper argues that in investigating a pastoral pro
duction system it is critical to study all transactions which affect
277 -
offtake and acquisition; whether or not they are commercial and
whether or not the animal leaves the area (or enters from outside) .
Firstly this data is necessary to understand how the producer manipu
lates his animals to fulfill both short and long-term needs and goals.
\s Gilbert, Norman and Winch (1980) noted with reference to FSR,
"The objectives of the farmer or farming families are directly incor
porated into the designing and testing of strategies. An attempt is
nade to understand the farmers' objective functions in the initial
descriptive or diagnostic stage."
Secondly, data on complete transactions, coupled with data
on births and deaths, provide a means of assessing livestock pro
ductivity which is complementary to the recording of direct production
parameters. If overall productivity is defined as the overall offtake
rate together with herd increase (or decrease), total offtake as well
as total acquisition must be assessed to properly interpret rates of
change both in the number of animals and the structure of the herd.
To underline the importance of studying total transactions,
Fig. 1 presents data obtained on one group ranch in Kenya. It shows
for poor and rich households the mean annual value of permanent
offtake. It also shows this value broken down by type of transaction:
sale, exchange, gift and slaughter for home consumption. While the
actual value of transactions varies greatly between the rich and poor
households, the percentage by transaction type does not. For both
groups, sales represented only slightly more than half of the trans
actions. Approximately 45% of the value of transactions was through
non-commercial channels. This emphasises the need to study the full
range of transactions. This information is critical not only for
understanding the producer's goals in herd accumulation, but also for
tapping his wider strategies. It is well documented that social
transactions of animals are important in African pastoral systems not
just as mechanisms of herd accumulation and disposition, but because
they serve diverse functions such as risk aversion and development
of social networks which may play a critical role in future access to
resources.
278 -
Ksh ('000)
3I-|
29-
27
25-
23-
21-
19
17-
15-
13-
II-
9
7-
5-
3-
I-
100%
55%
10%
21%
15%
Ksh ('000)
100%
56%
a total value
b sales
C exchange
d gift
e real slaughter
10%
18%
17%
Poor Rich
Fig. 1. Value of permanent offtake by transaction type: A Maasai
example.
Specific types of data required
As always in a PSR context, the first questions must be what exactly
do we need to know, and what is the most cost-effective way of
obtaining the necessary data. The answer to these questions will
depend on the nature and phase of the pastoral systems research proj-
- 279
ect itself. Thus it is not possible a priori to suggest specific
data required nor specific methods for every case. Despite this need
for goal-oriented specificity in research, it is nevertheless useful
to describe what information about livestock transactions is generally
needed especially during descriptive phases of research, as well as
possible methods of collecting it.
Important background information for understanding pastor-
alist's behaviour and strategies which can be obtained by informal
surveys include:
1. The general uses of livestock within the society, including both
economic and social aspects.
2. Within the economic sphere the general goals of the enterprise
(milk/meat; commercial/subsistence) .
3. The distribution of rights to animals and their products, which
are frequently not exclusive in pastoral systems.
4. Culturally available means of acquisition and offtake of animals.
For individual transactions, data should at least include
information on the transaction type, the species, age category and
sex of the animal. However, information collected on livestock
transaction is easily extended to provide data on a number of other
production/consumption parameters. For example, information on price
is obviously important for household budget studies. Information on
the person with whom the transaction was done (e.g. his residence,
relationship) provides important social network data, while infor
mation on what precipitated the transaction provides further insights
into production strategies.
Extensive survey methods in livestock transaction studies
For the more quantitative data on the totality of livestock trans
actions, the use of some recall techniques is essential unless one can
have trained recorders in every household. This paper will briefly
discuss multiple-visit techniques, then explore in more detail the
single-visit progeny history technique.
- 280 -
As voluntary offtake or acquisition is relatively in
frequent, important and discrete, its occurrence is usually well
remembered by pastoralists. To my knowledge there has been no study
comparing data obtained by enumeration at different intervals in
multiple-visit techniques for livestock transaction data. The Kenya
team has used twice weekly data collecting. The other Mali team,
as part of its continuous livestock productivity studies described
earlier in this workshop by Wilson, collects data monthly on a series
of tagged herds and flocks. The timing of the visits will depend on
resources available and on what other data are being collected. Thus,
for example, the high frequency in one Mali sample was chosen to
coincide with the collection of labour data rather than because it
was felt to be necessary in and of itself. The frequency will also
depend on the depth of data desired on the decision-making about
livestock transactions. While the more discrete details of trans
actions etc.) are likely to be remembered for a long time, details
on decision making (e.g. why that specific animal at that time) are
better elicited at more frequent intervals.
With single-visit techniques the reference period can be
quite problematic. As part of its initial survey in the group ranches,
the Kenyan team asked producers about their sales for the previous
year. The figures on the whole were several times those elicited
subsequently by more frequent visiting. It is clear that a major
problem was the reference period. As the out-off point "one year ago"
was not clearly distinguished in the pastoralists' mind; once they
started reporting sales they just kept on until fatigue or memory
problems intervened.
With single-visit techniques, it is recommended that the
reference be not a period of time. The progeny history technique,
which is discussed later, avoids the reference period problem by
using individual animals as the focus of questioning.
As livestock transaction events are relatively easy to
recall, perhaps the most important factors which encourage inaccurate
reporting are social/psychological ones. For example, in Maasailand
certain transactions are considered to be more prestigious than others.
Generosity is stressed in Maasai culture, whereas it is thought im-
- 281 -
proper to brag about the number of animals you have acquired. As a
result, producers tend to report giving more animal gifts than they
receive. For the same species/age/sex category, gifts given are also
appraised at a higher value than gifts received. In one study site,
this resulted in the value of animals reported received being only
43% of the value of animals reported given away.
Similarly, exchanges are probably selectively under-reported.
In Maasailand, exchanges of an adult castrate for a young female are
quite common. These exchanges are described as being initiated by
a person "begging" for the castrate because of some "problem" . Female
breeding stock are highly valued. It is more prestigious to be able
to increase one's breeding stock by acquiring a young female than to
have to sacrifice such an animal by exchanging it for a needed
castrate. In a parallel fashion, it is more prestigious to have a
surplus adult castrate to exchange out than to need such an animal
from outside the herd. Given this, it is not surprising that, whereas
in one year a random sample of producers reported exchanging in a
large number of immatur females (13 cattle, 25 smallstock), they
reported exchanging our very few (1 cow and 1 smallstock) . This
discrepancy is primarily due to selective reporting as partners in
exchanges are primarily other Maasai from the same or neighbouring
group ranches.
The progeny history technique
A useful method for collecting data on livestock transactions, es
pecially offtake, is the progeny history technique. It was origi
nally developed in Ethiopia by a team from the Ministry of Agriculture
and has been used extensively by several ILCA staff, including Cossins
in Ethiopia, myself in Nigeria (Grandin, 1981), and most recently by
Wilson and Wagenaar (1982) in Niger. It is also currently being used
by the ILCA Kenya team, although results from this exercise are not
yet available. This section describes the technique, then discusses
the extent to which it can be varied and intensified.
In the progeny history technique, with reference to each
adult female animal sampled, the producer is asked how it entered the
herd, then about the offspring to which it gave birth. Information
on the sex and disposition is solicited about each offspring in turn.
- 282 -
The progeny history method essentially depends on the ability of the
pastoralist to know, recall and be willing to tell you the life
history of his animals. I and other who have used the technique feel
that particularly for the major species of animals it is quite accu
rate.
For a very rapid appraisal, the choice of the animal can be
opportunistic. For a more in-depth study, a more systematic sample
of animals would be preferable. On the basis of this minimal infor
mation, it is possible to calculate roughly what percentage of the
animals of each sex leave the herd through death, sales, gifts, or any
other socially-mediated means of transacting animals.
A number of intensifications of the technique are possible.
These require more a priori knowledge of the system and more time to
elicit the information. One important way of intensifying the data
is to solicit information on the timing of events. Using local terms
for seasons it is frequently possible for the pastoralist to remember
at what season the particular animal was born, died or was otherwise
disposed of. This is particularly true for recent progeny. Further
more should a calendar of local events be available, in many cases it
is possible to determine the year in question. For cattle, a calendar
of 10 to 15 years would be sufficient, for smallstock a shorter one
would suffice, while for camels 15 years may be necessary.
Secondly, information about strategies and motivations can
be elicited. This may include social data on the party to a trans
action, reasons for disposition etc. In Maasailand, for example,
pastoralists will readily state what motivated a sale (specific need
for money, infertility of the animal etc.). Thirdly, for deaths, the
pastoralist' s diagnosis can be elicited.
The interview need not necessarily be done with the
producer or herd owner. Rather, it is preferable to interview the
person who is most likely to know the history of the animals. In
pastoral societies in which animals are allocated to women for milking,
or other reasons, it may be most useful to interview the women of the
household rather than the producer. For example in Maasailand ILCA
personnel have successfully interviewed older herd boys for progeny
histories. Particularly when herd boys are members of the family,
- 283 -
they tend to know a good deal about the history of the animals under
their care. This information is transmitted to them as part of their
pastoral training.
The progeny history technique has the distinct advantage of
providing time-depth information. Given the cycles of drought and
post-drought recovery which typify pastoral systems, this is an im
portant element of information-gathering. Although it is dependent
upon the recall ability of pastoralists, pastoral research specialists
have often noted that the history of animals and their progeny is
jwell remembered as it is considered critical information for manage
ment and selection purposes.
For example, older Maasai women have been able to recall
numerous details of the progeny of offspring allocated to them at
their weddings 20 to 30 years earlier. Certainly recall will be
better for animals still in the herd as well as for more discrete
events. Disaster often confuses memory of sequential events. Thus,
Maasai at least, have more difficulty in remembering specific events
which occurred during the last major drought seven years ago. They
will often say, "The dam had several offspring that died in the
drought but I don't remember the details."
To the best of my knowledge, however, this technique has
never been assessed for its reliability. As part of its ongoing work,
the ILCA team in Kenya is concurrently doing progeny history and
extensive data collection (by monthly interviews) on livestock
offake and acquisition. We will soon be in a position to compare
these two methods of collecting data on livestock transactions.
The progeny history technique is not quite as good for
collecting data on acquisition as it is for disposal. However, it
can be supplemented easily in order to get some time-depth. If the
dam was born into the herd, the origin of its mother is recorded.
Secondly, particularly if a complete herd enumeration has been made,
it is possible to find out the origin of all of the animals currently
in the herd. If there is no such enumeration available, it is pos
sible to ask the producer which of his animals were not born into the
herd and subsequently ask how and why they were acquired.
- 284 -
References
Eicher, C. and Baker, D.C. 1982. Research on agricultural development
in sub-Saharan Africa: A critical study- MSU International
Development Paper 1. Dept. of Agricultural Economics,
Michigan State University, East Lansing, p. 171.
Gilbert, E.H., Norman, D.W. and Winch, F.E. 1980. Farming systems
research: A critical appraisal. MSU Rural Development
Paper 6. Michigan State University, East Lansing, p. 6.
Grandin, B.E. 1981. Small cows, big money: Wealth and dwarf cattle
production in southwestern Nigeria. Ph. D. Dissertation.
Stanford University, California.
Wilson, R.T. and Wagenaar, K. 1982. An introductory survey of
livestock population demography and reproductive performance
in the area of the Niger Range and Livestock Project. ILCA/
Mali. Arid and semi-arid zones programme Bamako.
- 285 -
Collectes de données sur les transactions de bétail
Résumé
Cette étude examine la nécessité de données sur les transactions de
bétail dans les systèmes de production pastorale et envisage
plusieurs méthodes grâce auxquelles les données souhaitées peuvent
être recueillies.
Les transactions de bétail sont définies comme étant l'écoulement
et l'acquisition de bétail soit à titre temporaire, soit a titre
définitif. Les études précédentes sur les systèmes pastoraux ont
fourni très peu de données quantitatives sur les transactions de
bétail. Ce document soutient qu'en étudiant un système de production
pastoral, il est important d'étudier toutes les transactions relatives
à l'écoulement et à l'acquisition, que ce soit à titre commercial ou
non, qu'il s'agisse d'entrée d'animaux dans la zone ou de sortie
de celle-ci. Ces données sont nécessaires pour comprendre comment le
producteur manipule ses animaux pour satisfaire ses besoins et
réaliser ses objectifs à court et à long termes. Les données sur
les transactions conjuguées avec les données sur les naissances
et les décès complètent à merveille les relevés de paramètres
directs de production et constituent un moyen adéquat d'évaluer la
productivité de l'élevage.
Parmi les informations de base qui permettent de comprendre le
comportement et les stratégies des éleveurs figurent les
utilisations générales du bétail dans la société, les objectifs
généraux de l'exploitation, l'attribution de droits aux animaux
et à leur produits et les moyens culturels d'acquisition et
d'écoulement des animaux. Ces données peuvent être recueillies
grâce aux enquêtes informelles.
Pour les données plus quantitatives sur l'ensemble des transactions
de bétail, l'utilisation de certaines techniques de rappel est
essentielle à moins qu'on ne dispose d'observateurs qualifiés dans
chaque ménage. L'une des méthodes les plus utiles pour la collecte
des données sur les transactions de l'élevage, notamment en ce qui
- 286 -
concerne l'écoulement des animaux, est la technique de l'histoire de
la progéniture. Celle-ci est née en Ethiopie où elle a été mise au
point par une équipe du ministère de l'agriculture. Elle est
actuellement utilisée par l'équipe du CIPEA au Kenya. Avec la
technique de l'histoire de la progéniture, pour chaque femelle adulte
faisant partie de l'échantillonnage, on demande au producteur
comment elle est entrée dans le troupeau puis on lui pose des
questions sur sa progéniture. Des informations sur le sexe et sur le
caractère sont demandées ensuite sur chaque rejeton. La méthode de
l'histoire de la progéniture se fonde essentiellement sur la
capacité de l'éleveur de retracer l'histoire de ses animaux. La
technique a l'avantage distinct de fournir des informations qui
remontent à une période reculée. Etant donné les cycles de
sécheresse et de reconstitution qui caractérisent les systèmes
pastoraux, il s'agit là d'un élément important de la collecte des
données. La technique n'est pas aussi bonne pour la collecte des
données sur l'acquisition qu'elle l'est pour la collecte des
données sur l'écoulement. Toutefois, elle peut être facilement
complétée pour obtenir des données qui remontent à une période assez
reculée. L'équipe du CIPEA au Kenya effectue en même temps la
collecte extensive de données et utilise l'histoire de la progéniture
en ce qui concerne l'acquisition et l'écoulement de bétail. L'équipe
sera bientôt en mesure de comparer ces deux méthodes de collecte de
données sur les transactions de bétail.
- 287 -

Household income and expenditure studies
Solomon Bekure
Team Leader, Arid Zones (Eastern and Southern Africa) Programme,
ILCA, Kenya
Introduction
Very little is known about the economics of household level activities
in pastoral production systems (Eicher and Baker, 1982). Despite this,
interventions are frequently proposed which call for increased cash
expenditures. The implications of this for different groups of house
holds (poor, rich) needs to be assessed. An intervention may require
increased labour input. Will there be enough labour and if so will
sufficient food be available to sustain the energy requirements of
increased effort? An intervention may call for culling of non
productive and old animals from herds and flocks. What is the impli
cation of this for the security and viability of different groups of
pastoral households (rich, poor)? What needs of pastoral households
can be manipulated to provide incentives for culling? It is in such
a context that information on household income and expenditures is
required.
Testing and evaluating the impact of interventions as well
as assessing the welfare of pastoralists require bench-mark data on
income and expenditure patterns of different groups of pastoral
households.
Unlike agriculturalists, pastoral households depend more on
market transactions to satisfy their subsistence needs. Our own work
in Kenya shows that even during the wet season the Maasai obtain up
to 50% of their calorie intake from purchased food. This figure can
increase to 70% during the dry season. Household income and expendi
ture data are useful for determining the demand of pastoralists for
purchased goods and social services and for assessing the terms of
trade between pastoralists and the rest of the economy. A very good
example is given by Swift (1979) for Somali pastoralists. He analysed
- 289
the barter terms of trade between pastoral products sold by pastor-
alists and those they purchase by constructing a pastoral cost of liv
ing index. He concluded that by the early 1970S their terms of trade
had deteriorated and had "led the pastoral economy into a precarious
position".
Collection of households income and expenditure data
In addition to producing milk and meat for their own consumption,
pastoral households engage in a variety of transactions involving
livestock, livestock products, cash and other items (such as crops,
handicrafts etc ) , to fulfill different goals. Animals and their
products are sold to provide cash. They are given or lent to kin
and friends to strengthen social ties and ensure long-term security.
They may be similarly received. Animals may be exchanged for social
reasons or to increase the productive capacity of herds and flocks.
In order to determine the entire household budget, income
and expenditure studies should be designed in such a way as to include
not only cash income and expenditures but also these important trans
actions. Usually, the quantities of these transactions are known and
their values Can be determined by using prices the items would have
attracted had they been sold.
<
Sampling the target population
In any society the most important factor that influences patterns of
household income and expenditure is the wealth status of the household.
It is self-evident that the consumption of poor and rich households is
markedly different. It is therefore essential that study samples
adequately represent the gradient of wealth observed in the pastoral
group under study. This may not present a problem in situations where
a whole village or encampment or target population is studied as done
by the ILCA teams in Mali and Kaduna. Even then villages and camps
may be formed on the basis of social classes. Care should be taken
in selecting sample villages and camps so that results can be gener-
alisable to a known population type. In situations where the coverage
in area is more extensive one has to resort to sampling the population
as done by the ILCA teams in Kenya and Ethiopia. In that case a
- 290 -
stratification of the population into wealth categories is essential
(see Grandin (1983) for a detailed discussion of wealth effects and a
rapid method of wealth ranking) .
Given such a stratification of the target population, the
available resources for collecting data and the time-frame and nature
of the study, standard sampling procedures can be employed to determine
the size of the sample and choosing them (Cochran, 1963).
Types of data required
Inventory of resources
Once the sample households in the study have been determined, an
initial inventory of the human and livestock population needs to be
made. Here care should be taken so that animals owned by the house
hold but which are away from the main herd or flock at the time of
the inventory are included. Animals not owned by the household
but borrowed from others should be identified and recorded as such.
Similarly members of the household who are away at the time of the
census should be included and temporary visitors excluded. An
inventory of major household goods also gives a good indication of
investment and consumption patterns.
This information is vital for two reasons. First, it
quantifies the wealth status of the sample household. Second, it
provides the basic population data to perform per capita computations
without which meaningful comparative analysis cannot be made.
Household income and expenditure items
For designing the data collection formats background information is
required on the nature of items that form the income and consumption
baskets of the pastoral households to be studies as these vary from
culture to culture. A comprehensive list of these items should be
established from the researcher's personal knowledge or from informal
surveys involving a few pastcralists and shopkeepers in the area or
a combination of these. In addition to standardising the format for
enumeration, it is also a good device to facilitate recall by respon
dents .
- 291 -
The income items include :-
- livestock and livestock products (animals, milk, ghee,
hides and skins, manure etc.)
- agroforestry products (crops, wood, charcoal, honey etc.)
- cottage industry products (handicraft, beer, medicinal
herbs etc.)
- other forms of employment (trade)
- other cash inflows (remittance, borrowing)
Cash expenditure items can be grouped as:-
- food
- health and hygiene
- clothing
- transport
- livestock
- livestock inputs
- durable household goods
- others (cash outflows such as loans given) .
Frequency of data collection
Extracting information on household budgets, especially expenditures,
is extremely difficult because one has to rely on the memory of the
respondents to recall such data. Information on pastoral households'
income is by far easier to get because most of it is derived from the
sale of animals, which they remember very well. The fact that such
sales happen very infrequently facilitates recall. On the other hand
expenditures, especially on food items, occur so frequently in irregu
lar amounts that recall becomes difficult.
In collecting household income and expenditure data, the
shorter the time span the respondent is requested to recall the more
accurate is the information obtained. Researchers have used different
frequencies of collecting such data ranging from one-shot surveys
asking questions to estimate income and expenditure for a specified
period of time (e.g. per month or per week etc.), to continuous daily
recording for a long period of time (a year or more) . Within ILCA,
the Kenya team has used a frequency of once a month, the Kaduna and
Mali teams twice weekly. Of course the latter is more reliable but
- 292 -
is, however, very expensive. Unless it is combined with the conduct
of other research requiring daily observations, it may not be solely
justified on the basis of the precision gained. Even after using this
method in Mali, Swift (1983) writes:-
'It is likely that enumerators were only partially sucess-
ful in recording these details of household daily trans
actions. There was inevitably some resistance to such
detailed questioning and at times clearly false information
was given or important transactions were forgotten or
concealed. '
Our own experience in Kenya using a once-a-month recall
method proved that respondents could account only for about 70% of
their cash income. In conjunction with a nutrition study, which
required food intake data on a daily basis, household expenditures
were monitored for a month on a daily basis on selected households.
A comparison of the two methods showed that the monthly recall ac
counted for 73% of the total expenditures recorded on a daily basis.
Table 1. Comparison of monthly and daily recalls on four households
in Merueshi Group Ranch, Kenya (October, 1982).
ITEM
CASH EXPENDITURES FOR ONE MONTH ON
Food Livestock Others Total
Monthly
recall
Daily
recall
Monthly
recall
Daily
recall
Monthly
recall
Monthly Daily
recall
Mon.
rec .recall
KShs 1308 3047 1905 1980 1188 1015 4401 6042
Proportion of
total
0.30 0.50 0.43 0.33 0.27 0.17 1.0 1.0
Ratio of
monthly
recall to
daily recall
0 .43 0.96 1.17 0.73
Source: ILCA/Kenya (unpublished data)
- 293 -
However, this percentage varied significantly across dif
ferent expenditure categories as shown in Table 1. The monthly recall
method underestimates food expenditures substantially, proving that
frequent but irregular flows are hard to recall during a long period
of time. Researchers with wide experience in this field recommend
that a three-day span is the most that respondents can recall with a
high degree of accuracy (Nestel, pers. comm.)
The frequency of collecting household and income expendi
ture data should depend on the nature of the research. If a complete
understanding of the process is required the highly expensive more
frequent methods need to be used. If general orders of magnitude are
required the less frequent methods may be used. One may even sample
seasonally by concentrating interviews during specific months or weeks
in dry and wet seasons. Unfortunately, one may not encounter normal
wet and dry seasons, especially in East Africa. One must, therefore,
take into account the degree of representativeness of such samples in
the analysis of the data (see below).
Quality control in enumerator data collection
Invariably, enumerators are used to advantage to obtain income and
expenditure data covering a large number of households in one or more
geographic locations simultaneously. The use of enumerators, however,
introduces biases in the data collected as noted above. Careful
training and supervision is essential.
Analysis of data
Table 2 shows a summary of income and expenditure patterns of Tuareg
households in Mali. The dominant source of income is the sale of
animals, representing 42% of total income. Salaries earned are a
close second accounting for 35%. Income from cereals, other food and
consumption items and clothing is not from the sale of these items as
such, but represents the value of these items obtained in exchange.
The purchase of food items accounts for 49% of total expenditures,
the most prominent item being cereals (22%) . Expenditure on the
purchase of livestock and livestock inputs is also high (32%) followed
by expenditures on clothing (10%).
- 294 -
Table 2. Mean annual household income and expenditure of 16 Tuareg
households in Mali.
Income Expenditure
Amount (MF) % Amount (MF) %
Animals 92,350 42 66,194 32
Milk 1,069 - 6,600 3
Cereals 16,038 7 46,919 22
Other food and consumption 15,963 7 51,319 24
Clothing 1,813 1 21,506 10
Equipment and artisan work 15,150 7 9,556 5
Salarie etc. 78,250 35 9,425 4
TOTAL 220,656 100 209,519 100
Source: Swift (1982).
Tables 3 and 4 give a summary of per capita income and
expenditures of Maasai households in Mbirikani Group Ranch in Kenya.
In this case opportunities to earn income from employment outside the
pastoral sector are restricted. Livestock sales (mainly cattle)
represent a very high proportion (83%) of total cash income. The
only other significant source of cash income is trading in livestock,
which provides 11% of total cash income.
- 295 -
Table 3. Sources of oash income of Maasai pastoralists in Mbirikani
Group Ranch (July 1981 to June 1982)
cash income
Mean annual per capita
Source
Livestock sales
cattle
smallstock
milk sales
hides & skins
Trading
Remittance
Borrowing
OTHERS
KShs %
1523 83
1474 81
49 2
0 0
9 1
205 11
9 1
42 2
38 2
TOTAL 1821 100
Source: ILCA, Kenya 1982 (unpublished data).
Table 4 gives a more disaggregated picture of mean per
capita cash expenditure for food and non food items excluding pur
chases of livestock and livestock inputs. Cereals, sugar (consumed
in tea with milk) and fats/oils, which provide a substantial portion
of the calorie requirements of the Maasai, are the major items of
cash expenditure, followed by clothing.
From the above it is clear that the pastoral households
studied both in Mali and Kenya spend a very high proportion of their
income on food and livestock. The income elasticity of food expendi
ture is very low.
- 296
Table 4. Annual per capita expenditure on food and non-food items
by Maasai households in Mbirikani Group Ranch (July 1981 to
June 1982).
Food i I ems
Maize
Wheat
Sugar
Tea
Fat/oils
Potatoes
Vegetables
Meat
Other food
Expend Lture
KShs
% of total
food items
222* 49
20 4
55 14
30 7
29 6
7 2
2 0
6 2
6 2
Subtotal
Hotel food
Hotel drink
387
28
36
86
6
8
Subtotal food items 451 100
Non-food items
Tabacco
Kerosine
Soap/Omo
Transport
Medical
Clothing
Household items
Beads
% of total non-
food items
3 4
2 2
5 6
13 15
9 11
47 56
3 4
2 2
Subtotal non-food items 84 100
Source: ILCA, Kenya 1982 (unpublished data)
*74 kg per capita per annum.
- 297 -
Since the consumption of other items is almost negligible,
the implication is that any increased income will be largely spent on
the purchase of breeding stock unless attempts are made to increase
i«. "ioral i sts ' demands for other consumption and investment items.
A more sophisticated analysis of household income and
expenditure data can be performed using regression techniques. A
common representation of this is given as:
a=Po•M.^iV +&m*m
where E\.= Expenditure on the ith commodity or group of
commodities
^0= least square mean
p)jCj=',2 ,m) = coefficients of the independent variables
Xj(j-i;i >m) - independent variables specified as determinants
of expenditure e.g. x = household size
X = wealth or income group
x} = season of the year etc.
Usually, the log form of the above function is preferred
as the coefficients specified directly produce elasticities of expen
diture.
The main determinants of expenditures of pastoral households
are: (a) household size, (b) income or wealth status and (c) seasons
of the year.
Ceteris paribus, the larger the household size the higher
the expenditure, the higher the wealth status of the household the
bigger the expenditure, and the drier the season the larger the expen
diture, especially on food. Table 5 shows how the cash income and
expenditures on cereals of the Tuareg in Mali is affected by the
different seasons.
- 298 -
Table 5. Seasonal cash income and expenditures on cereals of 16
Tuareg households in Mali (1971-1981).
Season
Cash income Cash expenditure
Excluding salaries on cereals
MF % MF %
Hot (March-June) 39, 100 38 23 806 53
Rains (July-October) 44, 206 42 12 513 27
Cold (November-February)
TOTAL
20, 088 20 9 300 20
103, 394 100 45, 619 100
Source: Swift, 1982.
The use of expenditure baskets of pastoralists to calculate
terms of trade for pastoral products as calculated by Swift (1979) is
shown in Table 6 .
- 299
Table 6. Barter terms of trade for pastoral products in southern
Somalia
1847 1891 1951-53 1956-58 1971-72 1974-75
Barter value of 1 adult male camelCupels
rice (kg) 140 204 71 108 159 192
maize (kg) 978 380 472 165 421
sorghum (kg) 1110 978 490 460 284 1070
sugar (kg) 22 81 88 88 111 156
tea (kg) 25 22 27 49
dates (kg) 265 341
local cloth (yd) 83 25
American grey sheeting
(yd) 138
Indian printed cloth (yd) 576
cotton futa cLoth (yd) 39 43
Cattle Barter value of 1 adult ox or bull
rice (kg) 140 184 44 61 86 95
maize (kg) 880 235 267 164 418
sorghum (kg) 1110 880 305 260 154 648
sugar (kg) 22 73 55 50 60 156
tea (kg) 16 12 15 30
dates (kg) 265 306
local cloth (yd) 83 22
American grey :sheeting
(yd) 124
Indian printed cloth (yd) 518
cotton futa cl<3th (yd) 24 24
Sheep Barter value of 1 adult sheep
rice (kg) 21 10 12 16 41 25
maize (kg) 49 63 69 78 108
sorghum (kg) 167 49 81 68 73 167
sugar (kg) 3.2 4.1 15 13 29 24
tea (kg) 4.2 3.2 7 8
- 300 -
Table 6 cont.
dates (kg) 40 17
local cloth (yd) 13 1.2
American grey sheeting
(yd) 7
Indian printed cloth (yd) 29
cotton futa cloth (yd) 6.4 6.3
Goats Barter value of 1 adult goat
rice (kg) 14 5 9 12 32 15
maize (kg) 24 49 54 61 68
sorghum (kg) 111 24 59 53 57 105
sugar (kg) 2 2 11 10 22 15
tea (kg) 3 3 6 5
dates (kg) 27 8.4
local cloth (yd) 8 0.6
American grey sheeting
(yd) 3.4
Indian printed clc)th (yd) 14
cotton futa cloth (yd) 5 5
Ghee Barter value of 1 kg ghee
rice (kg) 4 5 2 4 7 5
maize (kg) 23 11 16 13 23
sorghum (kg) 29 23 14 15 12 36
sugar (kg) 0.6 2 3 3 5 5
tea (kg) - 0.7 0.7 1 2
dates (kg) 7 8
local cloth (yd) 2 0.6
American grey sheeting
(yd) 3
Indian printed cloth (yd) 13
cotton futa cloth (yd) 1 1
Source: Swift, 1979.
- 301 -
References
Eicher, C.K. and Baker, D.C. 1982. Research on agricultural develop
ment in sub-Saharan Africa: A critical survey. Michigan
State University, East Lansing.
Grandin, B.E. 1983. The importance of wealth effects on pastoral
production: A rapid method for wealth ranking. Paper
presented at the Workshop on Pastoral Systems Research in
sub-Saharan Africa, ILCA, Addis Ababa.
Cochran, W.G. 1963. Sampling techniques. John Wiley & Sons, Inc.
New York.
Swift, J. 1979. The development of livestock trading in a nomad
pastoral economy: The Somali case. In: Proceedings of
the international meeting on nomadic pastoralism, pastoral
production and society. Cambridge University Press.
Swift', J. 1982. In: Les systemes de production pastoraux au Mali.
First draft. ILCA, Addis Ababa.
- 302 -
Etudes sur les revenus et les dépenses monétaires des ménages
Résumé
On dispose de peu de renseignements sur l'économie des ménages dans
les systèmes de production pastorale. En dépit de cela, des inter
ventions qui impliquent l'accroissement des dépenses monétaires
sont fréquemment proposées. L'expérimentation et l'évaluation des
interventions et de leurs effets de même que l'évaluation de la
qualité de la vie des éleveurs nécessitent des données de base sur la
structure du revenu et des dépenses de différents groupes de ménages
pastoraux.
Outre la production de lait et de viande auto-consommée, les ménages
pastoraux participent à diverses transactions de bétail, de produits
animaux, de cultures commerciales, d'objets artisanaux etc., en vue
de réaliser leurs différents objectifs. Pour déterminer la totalité
du budget du ménage, les études sur le revenu et les dépenses doivent
être conçues de sorte à refléter non seulement le revenu et les
dépenses monétaires mais aussi d'autres transactions importantes.
Dans toute société, le facteur le plus important qui influence la
structure des revenus et des dépenses du ménage est la richesse du
ménage. Il est évident que la consommation est différente selon qu'il
s'agit d'un menage riche ou pauvre. Il est par conséquent essentiel
que les échantillons de l'étude représentent adéquatement le gradient
de richesse observé dans le groupe pastoral étudié. Une fois que les
ménages-échantillons de l'étude auront été déterminés, il faudra
entreprendre un inventaire initial de la population humaine et
animale. Un inventaire des biens essentiels du ménage donne également
une image fidèle de la structure des investissements et de la
consommation. Cette information est vitale pou. deux raisons: elle
quantifie la richesse du ménage - échantillon et fournit des
données de base sur la production pour effectuer, les calculs par tête
d'habitant sans lesquels aucune analyse comparative significative
ne peut être effectuée.
303
Pour la conception des formulaires de collecte de données, il convient
de rassembler des informations sur la nature des éléments qui forment
le revenu et la consommation des ménages pastoraux à étudier, étant
donné que ceux-ci varient d'un cadre culturel à l'autre.
Dans la collecte des données sur le revenu et les dépenses du ménage,
l'information recueillie est d'autant plus exacte que la période
de rappel est courte. La fréquence de la collecte des données sur
les dépenses et les revenus des ménages devrait être fonction de la
nature de la recherche entreprise. Si une compréhension approfondie
du processus est requise, ce sont les méthodes fréquentes, très
coûteuses qui doivent être utilisées. Par contre, si les ordres de
grandeur requis sont généraux, il convient d'utiliser les méthodes
les moins fréquentes.
Les enquêtes effectuées au Kenya et au Mali ont clairement montré
que les ménages pastoraux observés dépensaient une très forte
proportion de leur revenu pour acquérir des produits alimentaires et
du bétail. Puisque la consommation des autres produits était presque
insignifiante, on peut en déduire que tout accroissement du revenu
sera en grande partie consacrée à l'acquisition de reproductrices, à
moins que des efforts ne soient déployés pour accroître la demande
pour d'autres biens de consommation et d'investissement.
Une analyse plus sophistiquée des données sur les dépenses et les
revenus des ménages peut être effectuée en utilisant les techniques
de régression.
- 304 -
Labour data collection
Barbara E. Grandin
Anthropologist, Arid Zones (Eastern and Southern Africa)
Programme, ILCA, Kenya ,
Introduction
Labour plays a crucial role in any agricultural production system. In
a pastoral system where grazing and water are theoretically of free
access within a resource-sharing group, capital (in the form of ani
mals) and labour power are the essential means of production which
must be acquired and managed for a successful operation. This paper
addresses issues regarding the collection of data on labour inputs
for pastoral production. First it briefly considers the utility of
data on labour as part of pastoral systems research, and then dis
cusses research methods for data collection.
The need for labour data in pastoral systems research
We have already noted that unlike FSR, PSR on the whole suffers from a
paucity of baseline information and understanding of system dynamics.
This is perhaps more true of labour than of any other input to pro
duction. Anthropological studies, which comprise the bulk of the
research done on pastoral societies, have tended to focus on the
social organisation of labour while paying little attention to the
actual labour demands of the system. However, a proper understanding
of labour as a system component must pay attention to both its social
and technical aspects.
The energy demands of many tasks necessary for pastoral
production are much less than the energy demands for crop production.
A myth has arisen about the idle pastoralist, because inputs in energy
were confused with inputs in time. It has been assumed on the basis
of little evidence that labour rarely if ever constitutes a constraint
to pastoral production. Hence, traditionally, little attention has
been paid to the labour demands of pastoral production. Labour data
- 305 -
gathered by several ILCA teams, as well as other researchers, however,
are beginning to demonstrate that in fact there can be large demands
for labour time in pastoral production.
As labour is a critical input into pastoral production, it
is necessary to understand both the labour demands of the households
and the labour supply available. In the descriptive and diagnostic
phase of PSR informal surveys can be used to gather information on the
following aspects of labour:
1. The critical production tasks, their timing and
seasonality.
2. The culturally accepted age and sex division of labour.
3. The possible recruitment pathways for labour.
4. Labour demand/supply variations between households and
social mechanisms for adjusting these.
5. The constraints due to insufficient labour including its
effect on productivity.
Ideally there should be sufficient information to make some
estimation of the economic returns to labour.
Through the use of a labour profile which details the time
and amount of labour inputs into production activities by category of
worker and season, it may be possible to determine whether there are
labour constraints to production. It is also possible to determine
what the opportunity cost of proposed interventions might be in terms
of the shifting of time allocation from one activity to another.
These factors must be considered for designing interventions.
If proposed interventions require significant changes in labour inputs
or patterns, it is essential to know: Is labour available and/or can
it be obtained at the required time? What would be the opportunity
costs forgone with the desired changes of labour input? Will the
age/sex group normally responsible for a task or group of tasks be
willing or able to undertake a proposed intervention?
Occasionally, proposed interventions cross the accepted age/
sex division of labour. For example among Maasai , women have the
responsibility for fetching water: although they occasionally bring
- 306 -
water to the dwelling site for young or ill animals, their primary
responsibility is to fetch water for domestic use. It is the responsi
bility of the men to see to the watering of animals. Thus an inter
vention which suggests or recommends the carrying of water to the
dwelling site on a regular basis for young animals would most likely
meet with resistance on the part of women because they do not see it
as their responsibility to water animals except in extreme instances,
and on the part of men who do not see it as their responsibility to
bring water to the dwelling site.
In determining labour availability (particularly with a view
to introducing a new technology) it is important to collect data on
the household level. Householdsvary significantly in labour needs,
labour availability as well as in their labour need/supply ratio.
In one Maasai group ranch, poor and rich families have a
total work force of 6.3 and 9.8 people respectively. (Workers are
defined as all adults and children over the age of six years). How
ever, although richer households are much larger, their livestock
holdings are also much larger, and consequently the ratio of livestock
units per worker is much higher. For example in Olkarkar, whereas
rich households average 26 livestock units per worker, poor households
average only 6 livestock units per worker. These different ratios of
livestock units per worker affect the management strategies and time
allocation between rich and poor households, although they may be
offset by social and market mechanisms for acquiring labour.
With the ideas of building up a repertoire of methods for
gathering labour data, this section will briefly review some of the
more traditional methods available, and provide more in-depth infor
mation on a relatively new technique. The better-known methods
include: critical task analysis (which may include some direct
observation) and recall; the emerging technique is called the random
visiting or time allocation technique. The critical task technique
involves specifying the most important production tasks and trying
to quantify the time necessary for them. Recall techniques involve
interviewing various workers and asking them how much time they have
spent on a certain activity for a certain specified period. Recall
techniques range in intensity from seasonal to daily interviews. (In
- 307 -
the case of the seasonal interviews they are more like critical task
specification as it is impossible for the worker to remember actual
hours worked) . Lastly, the time allocation technique is based on
random visits at which time the current activities of all workers in
the household are recorded. These methods are not mutually exclusive;
as appropriate, they can enhance one another.
Critical task analysis
There is a particular dearth of labour input data for pastoral systems.
Swift (1979), Torry (1977) and Dahl (1979) have applied the critical
task technique to pastoral systems of the Tuareg of Mali and the Gabra
and the Borana of Kenya respectively. In critical task analysis,
through observations and interviews, the researcher develops a list
of the most important production tasks and the approximate amount of
time that it takes to complete these tasks. The quantification may
be done by direct observation (parallelling the work study approach
used in farm management studies). Dahl describes a fairly large number
of both the minimum livestock and domestic tasks necessary to sustain
a Borana household, but she does not attempt to quantify the time
involved, Torry, on the other hand, estimates only labour time re
quired for herding, milking, and watering the major species kept by
the Gabra, as well as labour needed to collect water for domestic
consumption. Using these species specific task estimates, he then
estimates the labour that would be required for an independent Gabra
household. He estimates that a total of 2,494 hours are required
each month for these critical tasks, and suggests that this labour
intensive system necessitates the pooling of the labour of several
households and thus joint herding. Swift also pays primary attention
to labour inputs to livestock management but he refines the technique
by estimating necessary inputs by season as well as by management unit
size.
These critical task approaches to livestock production
provide useful and interesting data in a cost-effective way. They
are similar in many respects to Collinson (1972) 's limited-visit
survey. As Collinson notes, "The same type of pattern is formed for
components of the labour profile by repetition from season to season...
- 308 -
limited visit techniques deliberately seek to exploit this pattern in
eliciting answers based on experience rather than on historical
recall of labour use for that particular season." They suffer, how
ever, from limitations inherent in the method. They deal only with
general descriptions of major tasks and provide no information on the
actual inputs of labour by various workers and variations in allo
cation of labour which might be related to season, skill or wealth.
Cossins (personal communication) uses a variation of the
critical task technique in his on-going study among the Borana of
Ethiopia. Preliminary survey results indicated that labour necessary
for watering animals in traditional Borana wells might be a key
constraint to productivity. All animals are watered from wells for
four to six weeks each year; thus the amount of water that can be
extracted from wells at this time is a limiting factor of production.
He is dealing with this particular labour input on several
levels. At the first level, for the whole production system, he
sought to determine in gross terms whether there is sufficient labour
available to work the wells, i.e. whether labour is a constraint to
water production in the dry season.
In the Borana study site, of the 30 or so major well groups,
24 were studied in detail and sufficient information was gathered on
the remaining wells to allow extrapolation. Data on total numbers
of workers required for each task (by age/sex category) were obtained.
This was compared with the size of the population for these age/sex
categories which had been obtained from a demographic survey. This
showed that in overall terms labour was not likely to be a limiting
factor. The second level, now underway, is to look at labour demand/
supply for each well group and then each well.
Extensive recall methods
Multi-visit recall techniques for collecting labour data are currently
being used by Swift in Mali and White in Niger. They use resident
enumerators to collect labour input data from all workers in sample
camps at twice-weekly intervals. As Swift (1982) notes, "The notion
of elapsed time created some difficulties." The use of watches was
- 309 -
not successful, so that the local concepts of time (broken into
seven categories) were used to estimate time spent on various activi
ties. Both Swift and White collect the labour data with a battery
of other data as part of a much broader study, so that the enumerator
costs were not high.
In designing a multi-visit recall study for labour data,
perhaps the two most important decisions are the frequency of the
visits, and whom to interview.
1. Frequency
Collison (1972) stated that "where work organisation is complex and
highly irregular even from day to day,... information for two or
three days previous may be too much to ask and liable to heavy
transfers." This is likely to be less of a problem in pastoral
societies where the bulk of the labour inputs are regular. However,
if detailed information is required, especially on tasks other than
herding and watering, twice-weekly recall is probably as necessary
with pastoralists as it is with farmers. This is the frequency being
used by Swift and 8Thite. As noted earlier, however, high frequencies
can lead to problems with respondent and enumerator fatigue.
2. Respondents
The decision on whom to interview depends both on the breadth of data
that is required and the cultural practices in the area under study.
The herd owner at least should be interviewed. He would probably be
able to say which people looked after each group of animals for the
previous few days, and in what major management tasks they engaged.
The researcher would have to be particularly wary of reporting biases,
e.g. if the herder really doesn't know, but won't admit the fact, or
if due to labour shortages, the household has non-conventional
patterns of labour input to which the herder would not want to admit.
For more precise information, each worker would need to be inter
viewed separately as Swift and White have done. This greatly
increases the costs of the survey (both in data collection and analy
sis), but would significantly increase accuracy.
- 310 -
Either limited-visit or extensive recall data may be able
to yield sufficient data for determining labour bottlenecks. However,
these methods have several drawbacks related to the fact that they tap
only major inputs into productive, and possibly domestic, activities.
First, as many smaller inputs are missed, it is not possible to cal
culate returns on labour in pastoralism (about which we know very
little as Eicher and Baker have noted). Secondly, the researcher is
not likely to know what other activities the workers are engaged in.
He will thus not be in a position to understand what activities would
have to be forgone should labour intensive innovations be suggested.
The time allocation technique
Because of these felt difficulties with critical task and recall
methods of labour data collection, in the Kenya Maasai study, the
ILCA human science team used a data collection technique called "time
allocation" recently developed by American anthropologists. The
method relies on randomly timed visits to a household during which
descriDtions of the activities of all its members are recorded. These
records produce a series of verbal snap-shots sufficient in number to
provide a thorough description of activities by such parameters as
age, sex and season. The time devoted to a particular task is ex
trapolated from the percentage of all activities devoted to that task
(see Grandin, 1982 for more details).
The requirements of the time allocation technique are the
following :
1. Sample households which are included in the study must
be censused so that the name, age and sex of all the workers in the
household can be determined.
2. Enumerators must stay near the homesteads so that all the
important hours of the day might be covered. Thus the technique is
difficult with a highly nomadic population in which encampmenLs or
households split seasonally.
3. A computer is needed for analysis as the number of snap
shots or records is quite large. Although the Kenya team has used
a main-frame computer, a micro computer would be sufficient.
- 311 -
The time allocation technique has one disadvantage in
addition to its basic requirements. It is not adequate for specifying
time spent on certain activities away from the home. For this
category of activity it is useful if it can be implemented by some
specific observations or at least interviews as is done in critical
task analysis. For example in the watering of livestock, if a pro
ducer leaves the household in the morning with animals to water, it
would be useful to know how much time he spent walking to the water,
how much time he spent waiting for the animals to be watered, how
much time the animals spent watering and how much time the man spent
socialising or gathering information. The time allocation technique
in and of itself cannot supply these data.
The advantages of the time allocation technique, however,
are numerous.
1. It easily covers all potential workers.
2. It is not plagued by the problems of recall since activi
ties are recorded on the spot.
3. There is little respondent fatigue; in fact almost any
person in the household can be interviewed. There is no requirement
for uniformity of respondents as long as one can be sure, within the
context of the specific society, that the person interviewed is likely
to know the whereabouts and activities of the various members of the
household. Recording of the information does not take very long.
Five to ten minutes is normally sufficient for a household of 10
people.
4. Apart from a household census, little pre-survey labour
data is required. It is not necessary to specify critical tasks be
forehand, and little needs to be known on the age and sex division of
labour.
5. The time allocation technique covers the complete range
of tasks whereas critical task techniques depend on the specification
of just the few most important tasks. Time allocation will record
minor but important tasks, the cumulative effect of which can be very
important for the production system. For example where critical task
techniques tend to concentrate on time to herding and watering, in our
- 312 -
Maasai study 70% a of the time devoted to livestock management (exclud
ing milking) was devoted to herding and watering. This critical task
specification would have missed 30% of livestock management tasks.
6. Although we have not done so in the Kenya study, it is
possible with time allocation to record multiple simultaneous activi
ties. Thus a woman who was sitting outside, watching a child while
making butter and keeping an eye on smallstock can be recorded as
doing all those tasks. Were she to be interviewed on what she had
done at that period of time, it is likely she would have said "making
butter" as that is the most discrete, least regular activity.
7. Lastly, and perhaps most important, the time allocation
technique provides critical information on production unit variation.
If the sample size is sufficient, households can be compared according
to their labour supply, wealth, neighbourhood or any other parameter
which is thought to affect labour input.
Although the time allocation technique used by the Kenya
team was continous for a period of just over a year, we see no
a priori reason why the technique could not be adopted to more limited
periods of time. With the help of the biometrician at ILCA's head
quarters, the team is doing retrospective analysis to determine the
minimum number of visits deemed to be necessary in a season in order
to provide a labour profile for that season. Preliminary results
suggest that little precision is gained by having more than 10 visits
per household per season. On the other hand, whereas a minimum of
10 people per person category desired for analysis would be required,
precision would be increased by increasing the number of people
studied. We suspect that frequent visits to a number of households
(for example two weeks of visits to each household four times a day)
should provide the necessary quantity of data to estimate production
inputs and other activities. Of course, in order to use seasonal
sub-sampling more background information on labour inputs and es
pecially constraints would be required to determine the optimal timing.
Sample results of the time allocation technique
This section presents a few preliminary results from the time allo
cation study by the ILCA team in Kenya. Visits were made to each
- 313 -
sample household twice a month at randomly chosen times between the
hours of 6 a.m. and 8 p.m. during which period the activities of every
member of the household were noted. Table 1 shows how the obser
vations were used to calculate the percentage of livestock management
tasks done by each age/sex category of worker. For example, there
were 1,121 observations of livestock management activities out of a
total of 3,530 observations. Thus 41% of all the observations were
of livestock management (representing approximately 4.3 hours per
worker per day). For watering, there were 123 observations, of which
15% were male children, 5% were female children, 75% were adult
males and 6% were adult females. While this table presents data on
the percentage of specific tasks done by various age/sex categories,
the time allocation data can also be used to calculate hours per day
spent on these tasks, either in total or by age, sex, season etc...
Table 1. Age/sex roles in livestock management at Olkarkar.
Pen: entage of tasks
each group
done by Livestock
management
Activity observation
Children1
M F
-,2 *3
Adult
M
s Total %
F %
watering 15 5 75 6 101 123 11
herding 47 45 3 5 100 641 57
dipping 5 5 68 21 99 19 2
boma livestock work 18 24 33 26 101 274 24
other livestock work 30 6 48 16 100 64 6
No. of observations 389 363 245 124 1121
% total observations
per age/sex 42 47 48 10 31
N.B. There were no differences in task percentages between wet and
dry months.
Children 6 years and above.
2Percentage errors due to rounding.
3 . . ...N = number of observations of livestock management activities.
- 314 -
Table 2 shows the time devoted to livestock management by
age/sex of worker, for all workers and finally by livestock unit. The
data demonstrate that whereas each worker in a rich household spends
more time on livestock management than a worker in a poorer household,
richer households devote far much less time per livestock unit held.
This is because rich households herd their animals in much larger
groups, thus benefitting from economies of scale; they also are able
to marshall labour from poorer households with whom they frequently
herd.
Table 2. Labour devoted to livestock management at Olkarkar.
Mean number o f hours of labour per day
1
renStratum Child Adu It:s All workers per
M F M F
livestock unit
5.02
A.621 6.3 0.8 1.29
5.72
2 7.1 6.0 1.5 0.60
6.42
3 7.6 7.9 1.5 0.25
Children 6 years and above.
2
Children spent a mean of approximately 2 hours a day at school.
These examples indicate just a few of the analyses possible
with the data gathered by the time allocation technique. Although the
time allocation technique is being increasingly used, to date there
are not, to my knowledge, any studies available comparing results
gained from this technique to those gained from recall or critical
task analysis. The Kenya team is in the process of designing a brief
study which will involve concurrent collection of recall and time
allocation data from a sample of households in the Maasailand. It is
hoped that the results produced will aid in assessing the validity and
reliability of the two methods under pastoral conditions.
- 315
References
Collinson, M.P. 1972. Farm management in peasant agriculture: A
handbook for rural development planning in Africa. Praeger,
London. P. 226, 229.
Dahl, G. 1979. Suffering grass: Subsistence and society of Waso
Borana. Dept. of Anthropology, Univ. of Stockholm,
Stockholm.
Grandin, B.E. 1982. The allocation and labour inputs on a Maasai
group ranch: Preliminary findings from Olkarkar. Manuscript
ILCA, Nairobi.
Swift, J. 1979. The economies of traditional nomadic pastoralism.
The Tuareg of the Adras N Ifras (Mali). Unpublished Ph. D.
dissetation, University of Sussex.
Swift, J. 1982. Les systemes de production pastoraux au Mali.
Internal document, ILCA, Addis Ababa, p. 308.
Torry, W. I. 1977. Labour requirements among the Gabra. Paper
Presented at the Conference on East African Pastoralism,
Nairobi, 1977.
- 316 -
Collecte de données sur la main-d'oeuvre
Résumé
La main-d' oeuvre .Joue un rôle très important dans tout système de
production agricole. Dans un système pastoral à caractère collecti
viste où les pâturages et l'eau sont théoriquement d'accès libre,
le capital (sous forme d'animaux) et la main-d'oeuvre constituent les
moyens essentiels de production à acquérir et à gérer pour une
exploitation productive.
Pour bien comprendre la main-d'oeuvre conçue comme une composante des
systèmes, il faudrait se pencher sur ses aspects sociaux et techniques.
Etant donné l'importance de la main-d'oeuvre dans la production
pastorale, il est nécessaire de bien connaître la demande et l'offre
de main-d'oeuvre des ménages. Dans la phase de la description et
du diagnostic de la recherche sur les systèmes pastoraux, des
enquêtes informelles peuvent être utilisées pour la collecte de
données sur les tâches de production les plus importantes, sur le
momevt de leur exécution et leur caractère saisonnier, sur la division
culturallement acceptée du travail par âge et par sexe, sur les
divers modes de recrutement des travailleurs, sur les variations
offre/demande de main-d'oeuvre entre les ménages et sur les
mécanismes sociaux d'ajustement de celles-ci et les contraintes dues
à l'insuffisance de la main-d'oeuvre, y compris ses effets sur la
productivité.
Il y a une indigence particulière de données sur la main-d'oeuvre des
systèmes pastoraux. Par le biais de l'analyse, d'observations et
d'interrogations, le chercheur dresse une liste des tâches de
production les plus importantes et du temps approximatif nécessaire
pour les mener à bien. La quantification peut s'effectuer par
l'observation directe* tout comme pour l'approche sur l'étude de la
main-d'oeuvre utilisée dans les études sur la gestion de l'exploitation
agricole.
L'analyse des tâches les plus importantes de la production pastorale
fournit des données utiles et intéressantes à un prix abordable.
- 317 -
Cette approche souffre cependant de limites inhérentes à la méthode.
Elle se contente dé descriptions générales des tâches les plus
importantes et ne fournit pas d'informations sur les travaux effectués
par divers travailleurs et sur les variations de la répartition du
travail qui peuvent être liées à la saison, à la compétence ou à la
richesse. Cossins utilise une variante de la technique de l'analyse
des tâches les plus importantes dans ses études sur les Borana
d'Ethiopie.
La technique de rappel fondée sur les visites multiples, telle que
celle utilisée par Swift au Mali et White au Niger se fonde sur
l'utilisation d'observateurs résidents pour collecter des données sur
le travail effectué par tous les travailleurs dans des campements-
échantillons deux fois par semaine. Dans la conception d'une étude
de rappel à visites multiples pour l'obtention de données sur la
main-d'oeuvre, les deux aspects les plus importants ont certainement
trait à la fréquence des visites et au choix des personnes à
interroger. Le rappel extensif ou les visites limitées pourraient
fournir suffisamment de données pour déterminer les goulets d'étran
glement en matière de main-d'oeuvre. Toutefois, ces méthodes
comportent plusieurs inconvénients étant donné qu'elles n'exploitent
que les facteurs les plus importants dans les activités de production
et éventuellement dans les activités domestiques.
La méthode de la répartition du temps se fonde sur des visites
programmées de manière aléatoire dans un ménage, au cours desquelles,
les activités de tous ses membres sont enregistrées. Ces enregistre
ments produisent une série de tableaux verbaux suffisants en nombre
pour fournir une description complète des activités par des paramètres
tels que l'âge, le sexe et la saison. Le temps consacré à une tâche
particulière est extrapolé sur la base du pourcentage de l'ensemble
des activités consacrées à cette tâche.
Les ménages-échantillons inclus dans l'étude doivent être recensés
de sorte que le nom, l'âge et le sexe de tous les travailleurs du
ménage puissent être déterminés. Les enquêteurs doivent rester à
proximité des ménages de sorte que les heures importantes du jour
puissent être couvertes. Une ordinateur est nécessaire pour
- 318 -
l'analyse, étant donné que le nombre d'entrées est assez important.
La technique de la répartition du temps comporte un inconvénient
en ce sens qu'elle ne permet pas de déterminer avec précision le temps
consacré à certaines activités effectuées loin da la maison. De telles
activités peuvent être couvertes par des observations ou par des
interrogations spécifiques comme dans l'analyse des tâches les plus
importantes. Les avantages de la technique de la répartition du
temps sont multiples. Elle couvre tous les travailleurs potentiels
et n'est pas Limitée par les problèmes que pose le rappel; les
enquêtés ne sont pas agacés et seul un nombre limité de données de
pré-enquête sur la main-d'oeuvre s'avère nécessaire; elle couvre
toute la gamme des tâches; elle permet d'enregistrer des activités
multiples et simultanées et fournit une information vitale sur les
variations au niveau des unités de production.
Certains résultats préliminaires de l'étude sur la répartition du
temps effectuée par le Groupe du CIPEA au Kenya ont été présentés.
Ils mettent l'accent sur quelques unes seulement des analyses
possibles avec les données collectées sur la base de cette technique.
L'équipe du Kenya est sur le point d'élaborer une courte étude qui
portera à la fois sur la collecte de données basées sur le rappel et
sur la répartition du temps dans un échantillonnage de ménages vivant
en pays Maasai. On espère que les résultats contribueront à évaluer
la validité et la fiabilité des deux méthodes dans les conditions
pastorales.
- 319 -

Summary of Discussion Session 6.
Chairman: Prof. Gunnar Sorbo (Norway)
Discussion led by Dr Befekadu Degefe (Ethiopia)
In referring to Dr Grandin's paper on Livestock Transactions,
Dr Wilson pointed out that off- take data are taken by animal scientists
as a matter of course. Indications of off-take and deaths can be got
from initial surveys, as was done in Niger.
Mr Sandford pointed out that labour intensity was different from
labour energy intensity, and both were different from the question of
whether labour was a constraint on pastoral production. Even if
people were totally idle for 11 \ months of the year, if labour was in
short supply for just \ month, e.g. for watering livestock from wells,
then labour shortage may be the critical constraint on the system.
Dr Grandin agreed and said that people might not work hard but in
fact were not idle as they may have to work long hours. If a labour
constraint was only for a short period of time, understanding the
dynamics of this would come from a critical task/season specification
and an analysis of the detailed labour input information.
Mr Sandford said that he suspected that the technique of the 'time
allocation study' might overestimate the labour requirements of a
system by failing to distinguish between those tasks which were truly
essential and those which were just undertaken to fill in time. In
the 'recall' technique, respondents were more likely to indicate
priorities by not mentioning the time spent on tasks which were
inessential.
Dr Zulberti warned that it was very easy for scientists to become data
addicts and to try to collect everything from everyone. Anyone could
look at his or her own work and recognise many occasions on which
data was collected and never used. Such data collection could
postpone real action in the field and could be an expensive endeavour
providing little or no effect on the well-being of the pastoralist.
He said that the final objective was not to know the pastoralist but
to know how one could be useful to him. Mr Sandford agreed that it
was important to be cost effective in research, but he pointed out
- 321 -
that there was a great danger in restricting one's collection of data
exclusively to those items which before the research started were
thought to be the most important. He said that sometimes breakthroughs
in knowledge occurred through the analysis of data which was collected
in a much more open-minded way. In project preparation, for example,
one may need to prioritise much more strictly. But in research one
needed to be much more open minded. This was the essence of the
systems approach - to find out what was important, not to pre
determine it.
Dr Zakary referred to Dr Solomon's observations that the expenditure
of pastoralists was mainly devoted to buying food and very little
was spent on systems inputs, and that perhaps parameters of
development should increase the demand so that pastoralists could
sell their cattle. Dr Zakary felt that although the latter might
improve the pastoral sector's contribution to the national economy,
at the same time pastoralists might then become slaves to demand -
this could be very dangerous during a drought. Dr Zakary felt that
the present system of using pastoral areas was good - developers
should above all allow pastoralists to organise themselves.
Dr Solomon agreed.
Dr Tilahun stated that the usual economic contention was that low
income households spent a large proportion of their income on food
and that this ratio declined as household income increased. But he
didn't think one could see such a relationship in pastoral societies.
Dr Solomon said that he had evidence from the Kenya study that the
wealth status of pastoral households greatly influenced the per caput
expenditure on food. Dr Solomon agreed with Dr Tilahun when he said
that information on decision making responsibilities of household
members on the use and disposal of livestock was important for
household income and expenditure studies. The analysis of such
information gave an insight into the extent to which management and
marketing practices varied due to household size and composition.
Dr Thompson said that the controversy regarding the need to understand
the complete system as opposed to part of the system in the
descriptive/diagnostic phase was resolved if one remembered that the
- 322 -
systems approach was iterative. This allowed one to return to the
initial phase to study either another well-defined aspect of the
system or to gather further information needed for designing and
testing solutions to a constraint previously identified. But the
time-span of an iteration must be short; thus a narrowly focused,
short duration diagnostic/descriptive survey was preferable to a
complete systems survey.
323 -
Résumé des débats de la sixième seance
Président: Prof. Gunnar Sorbo (Norvège)
Débats dirigé par le Dr Befekadu Degefe (Ethiopie)
Faisant allusion au document de Mlle Grandin sur les transactions de
bétail, M. Wilson a mis l'accent sur le fait que les relevés des données sur
l'écoulement sont effectuées par les zootechniciens de manière systé
matique. Des données relatives à l'écoulement et aux décès peuvent
être obtenues à partir d'enquêtes initiales comme ce fut le cas au
Niger.
M. Sandford a souligné que l'intensité de main-d'oeuvre était différente
de l'intensité d'énergie de la main-d'oeuvre et que ces deux éléments
n'intervenaient pas dans la question de savoir si oui ou non le travail
constituait une contrainte à la production pastorale. Même si les
travailleurs étaient totalement oisifs pendant II mois et demi et qu'au
cours de la dernière quinzaine de l'année il y ait carence de main-
d'oeuvre, par exemple, pour l'abreuvement du bétail au puits, la
carence en main-d'oeuvre constituerait la contrainte la plus grave
sur le système. Mlle Grandin a accepté cette idée et a déclaré que
la main-d'oeuvre ne travaillait peut-être pas dur mais en fait qu'elle
n'était pas oisive étant donné qu'elle pouvait avoir des journées de
travail très longues. Si une contrainte en matière de travail était
pour une courte durée seulement, on pourrait en saisir la dynamique
par une spécification saison/ tâche vitale et par l'analyse des données
détaillées du facteur travail. M. Sandford a déclaré qu'il avait
l'impression que la technique de la répartition du temps pourrait
surestimer les besoins en travail d'un système, notamment parce qu'elle
ne permet pas de distinguer les tâches réellement essentielles de
celles qui sont effectuées pour meubler le temps. Dans la technique
du rappel, il était probable que les enquêtes mettent l'accent sur
leurs priorités et qu'ils passent sous silence le temps consacré à
des tâches qui n'étaient pas essentielles.
Le Dr Zulberti a fait savoir qu'il était très facile pour qu'un chercheur
devienne un maniaque des données et qu'il essaie de collecter tout ce qui
est possible et imaginable. Nous pouvons tous trouver dans nos travaux
plusieurs exemples où nous avons collecté des données que nous n'avons
jamais utilisées. Une telle collecte de données pourrait retarder l'action
- 324 -
sur le terrain et pourrait constituer une entreprise coûteuse ayant
un effet limité ou nul sur la qualité de la vie de l'éleveur. Le
Dr Zulberti a déclaré que l'objectif final n'était pas de connaître
l'éleveur mais de savoir comment on pouvait lui être utile,
M. Sandford a reconnu que le rapport coût-efficacité était important en
matière de recherche mais il a souligné qu'il était très dangeureux de
limiter la collecte de données aux informations que l'on pensait être
les plus importantes avant le démarrage des activités de recherche.
Il a déclaré que quelquefois les progrès scientifiques intervenaient
à la suite de l'analyse de données qui avaient été collectées de
manière beaucoup plus générale. Dans l'élaboration des projets, il
est possible de fixer les priorités de manière beaucoup plus stricte.
Mais dans la recherche, il fallait être beaucoup plus ouvert. C'était
là l'essence de l'approche par système: trouver ce qui était important
et non pas le prédéterminer.
Le Dr Zakary a fait allusion aux observations de M. Solomon selon
lesquelles les dépenses des éleveuis étaient essentiellement
consacrées à l'achat de produits alimentaires, que très peu de
ressources étaient affectées à l'achat de facteurs de production et
que, peut-être, les paramètres de développement devraient accroître la
demande afin que les éleveurs puissent vendre leurs bovins. Le
Dr Zakary estimait que quoiqu'une telle démarche puisse augmenter la
part du secteur pastoral dans l'économie nationale, elle pouvait en
retour transformer les éleveurs en esclaves de la demande et cela
pouvait être très dangereux en période de sécheresse. Le Dr Zakary
a déclaré que le système actuel d'utilisation des zones pastorales
était bon, que les responsables du développement devraient par dessus
tout permettre aux éleveurs de s'organiser. M. Solomon a exprimé son
accord sur cette idée.
Le Dr Tilahun a déclaré que l'hypothèse économique habituelle était
que les ménages à faible revenu consacraient une proportion importante
de leurs revenus à l'achat de produits alimentaires et que ce rapport
diminuait avec l'accroissement du revenu du ménage. Mais il ne pensait
pas qu'on puisse observer une telle relation dans les sociétés pasto
rales. M. Solomon a déclaré qu'il avait des preuves émanant de l'étude,
- 325 -
selon lesquelles la richesse des ménages pastoraux influençait
considérablement les dépenses alimentaires par habitant. M. Solomon
a accepte le point de vue du Dr Tilahun lorsque celui-ci a déclaré
que les informations sur les membres du ménage investis du pouvoir
de décision sur l'utilisation et la disposition du bétail étaient
importantes pour les études sur les revenus et les dépenses des
ménages. L'analyse de telles informations donnait une idée de la
manière dont la gestion et les pratiques de commercialisation variaient
en fonction de la taille et de la composition du ménage.
Le Dr Thomson a déclaré que la controverse concernant la nécessité
de connaître l'ensemble et non pas une partie seulement du système
comme c'est le cas dans la phase de description/diagnostic était
dépassée dès lors que l'on se rappelait que l'approche par système
était itérative. Cela permettait de retourner à la phase initiale
soit pour étudier un autre aspect bien défini du système, soit pour
collecter d'autres informations nécessaires pour la conception et
l'essai de l'élimination d'une contrainte entièrement identifiée.
Mais la durée d'une itération doit être courte; ainsi, une enquête
de diagnostic/description bien délimitée et de courte durée était
préférable à une enquête complète sur les systèmes.
- 326 -
Livestock marketing studies
i z
Solomon Bekure and Negussie Tilahun
2Team Leader, Arid Zones (Eastern and Southern Africa) Programme, ILCA,
Kenya
2Economist, Arid Zones (Eastern and Southern Africa) Programme, ILCA ,
Ethiopia.
Introduction
Marketing is an important aspect of any livestock system. It provides
the mechanism whereby producers exchange their livestock and livestock
products for cash. The cash is used for acquiring goods and services
which they do not produce themselves, in order to satisfy a variety of
needs ranging from food items, clothing, medication and schooling to
the purchase of breeding stock and other production inputs and
supplies .
A major objective of pastoral systems research (PSR) is to
increase productivity and improve the standard of living of pastor-
alists. Interventions for increasing productivity generated by PSR
will have costs and returns associated with their adoption. It is
therefore essential that such interventions are not only technically
feasible and socially acceptable, but also economically feasible.
In other words, the incremental returns of the interventions must
out-weigh the additional costs incurred in adopting them. Both
input and output prices fluctuate over time. Researchers must,
therefore, establish the sensitivity of interventions by establishing
within what range of input and output prices they are stable. Time
series data on prices pastoralists are paid for their livestock and
livestock products as well as prices they pay for inputs are essential
for such analysis. Unfortunately, in most African countries while
time series data on input prices may be available, they are almost
non-existent for livestock prices.
The collection of time series data on product and input
prices, however, is not a function of PSR scientists. This should be
the responsibility of the ministries servicing the livestock and
agricultural sector. The purpose of this paper is to underscore the
- 327 -
necessity of studying livestock markets and routinely collecting time
series data on prices of livestock at local, regional and terminal
markets. Based upon the experience of ILCA economists and other
researchers in Africa it outlines a methodology, which can be easily
adopted by the relevant ministries of African countries in setting up
a systematic collection and analysis of livestock market information.
The second section provides the rationale for undertaking livestock
marketing studies. A simplified livestock marketing system model is
presented in the third section. A methodology for regularly collecting
time series data and for conducting in-depth studies on livestock
marketing is then given, followed by illustrated suggestions for the
analysis of data. A suggested national organizational framework for
collecting and analysing livestock market data is presented in the
last section.
Why study livestock markets?
As already stated, in most African countries there is a severe paucity
of time series data on livestock prices as well as on the performance
and efficiency of the livestock marketing system. Ironically, live
stock marketing happens to be a favourite sector, where African
governments choose to intervene in a variety of ways. These inter
ventions range from outright fixing of wholesale and retail meat (e.g.
Benin, Ethiopia, Togo) to monopolising the export market (e.g.
otswana, Kenya) . Yet in many instances policy decisions on livestock
marketing are taken in the absence of vital information on how they
affect livestock producers, traders, slaughter-houses, butchers and
consumers. Very often price fixing at unrealistic levels leads to
open black markets, where the real prices substantially differ from
those officially listed. In spite of this the official prices
constitute the price series data, which clearly distort any analysis
based on them.
Even when governments pursue price stabilisation policies
it is difficult, in the absence of livestock market data, to establish
to what degree their effects are transmitted to the level of producers.
The absence of data on the magnitude and seasonality of
supply as well as prices can frustrate the success of development
- 328 -
projects. The closure of the meat packing plant at Kotsi in Sudan
after a few months of operation is a case in point (Abbot, 1979).
It is often argued that the stratification of the beef
industry into areas of breeding, growing out and fattening should be
pursued; this is often included in livestock development projects
without much success. The comparative advantage of such a proposal
cannot be fully assessed without determining the long-term stable
price margins between the areas of stratification.
An important use of time series data is in assessing over
time the terms of trade between livestock producers and the rest of
the economy. This assessment can be made using weighted price indices
of a pastoralists' consumption basket and comparing it to the weighted
price indices of their sales basket (Swift, 1979). Such an analysis
should be made from time to time to gauge and temper the effect of
price policies on pastoralists, who depend on the market for subsist
ence much more than agricultural households.
Thus in any country, livestock marketing studies are
essential to provide vital information on the operations ana efficiency
of the livestock marketing system for effective research, planning
and policy formulation in the livestock sector.
A livestock marketing system model
A schematic representation of a livestock marketing system is shown in
Fig 1. The bottom part shows the flow of livestock from producers
to secondary (regional) and terminal (national) markets through one
or more primary collection markets. Livestock markets can easily be
differentiated by the type of sellers and buyers operating in the
market and the purpose for which livestock are purchased. Table 1
summarises these attributes for three types of livestock markets
(Ariza Nino et al, 1980).
- 329 -
Table 1. Characteristics of livestock markets.
Type of market Main Main Purpose of purchase
sellers buyers
1 . Primary Producers Other producers For stock replacement
collection or fattening
markets
Local butchers Slaughter
Traders Collection for resale
in larger regional
markets
2. Secondary Traders Local butchers Slaughter
distribution
markets Traders For resale in
terminal markets
3. Terminal Traders Local Slaughter
markets slaughter
houses
Traders Export
The top of Fig. 1 shows the external and internal factors
that influence the livestock marketing system. First on the supply
side the cash needs of producers, the strength of demand for their
livestock, and pastoralists ' expectation of the nature and length of
the dry and wet seasons influence the volume of the different species
of livestock on offer at any time. The higher the cash needs of the
pastoralists the greater the volume of livestock on offer. Their
response to market demand has been a subject of controversy in the
literature (Carlisle and Randag, 1970; Hill, 1970; Khalifa and
Simpson, 1972; Low et al, 1979; Jarvis, 1980). However, there is
growing evidence that pastoralists in fact dispose of their marketable
animals in a manner consistent with sound economic behaviour (Ariza
Nino et al,1980). In other words the stronger the effective market
demand as expressed by high prices, the greater the volume of
livestock supplied. Finally, pastoralists' perception of the climate
influences supply and hence the price of livestock. Anticipation and
- 330 -
occurrence of prolonged dry seasons induce more sales. The poor
condition of the animals plus the greater numbers supplied during such
times depress livestock prices. On the other hand the anticipation
and occurrence of good rains causes pastoralists to withhold animals
from the market so that they can put more weight and fetch better
prices later on.
Second, government policy through fiscal, regulatory and
development intervention affects the volume, flow and prices of
livestock in the marketing system. Favourable fiscal policies that
encourage livestock production and reduce costs to producers increase
the supply of livestock, e.g. subsidies, and price stabilisation
policies. On the other hand taxes and levies of all kinds tend to
restrict the volume supplied. The control of epidemic diseases, the
proper development of range areas and the development of trek routes
and livestock market facilities tend to increase the volume supplied
and reduce marketing costs. In general government monopolistic
tendencies and the fixing of artificially low prices stifle market
supply and demand.
Finally, market demand as expressed by the volume and prices
buyers are willing to pay for livestock influences the behaviour of
the markets at all links in the system. The efficiency of the market
as reflected by the marketing costs of the system and to what extent
price changes are transmitted through the marketing system strongly
influence the operation of the markets. The less efficient the
market the less responsive will supply be to changes in market demand.
Livestock market research methodology
Livestock marketing systems research involves a two-pronged approach.
The first involves the regular collection of time series data from
a network of selected livestock markets. The second deals with in-
depth studies of the performance efficiency of the livestock marketing
system at the various links of the chain as livestock move from
producers to the consumers.
- 331
EXTERNAL
AA
INTERNAL
Producerneeds
Narketdemand
Climate
NovernmentPolicy
-Fiscal,regulatory]
-Development
Narketdemand Narketsupply
Narketingcosts/
-Parastatal
Merchants
-Parastatal
TERMINAL (NATIONAL) MARKETS
Butchers
-Private -Private
REDISTRIBUTION
MARKETS Butchers
-Parastatal
Merchants
-Parastatal
SECOAARY (RENIONAL)
-Private -Private
•V w fc.)*■ w
s)
A
n
PRIMARYMARKETS
► Jw
V
-02
\) l \
9 *.
sJ
-0
I sJ
ra-N''
LIVESTOCK PRODUCERS
w
}
NARKET
INFLUENCES
NARKET
LEVEL
Figure1.Livestockmarke ingsystemmodel-
Collection of time series data
Identification and selection of livestock markets
The first consideration in setting up a livestock market data
collection network is the identification of livestock markets and the
selection of those markets to be included in the network.
Livestock markets in each administrative unit of a country,
classified in the manner described in previous section, can be put on
a map with arrows indicating the direction of flow of livestock by
species. Extension staff can be effectively used to provide infor
mation for such classification and estimating roughly the volume of
different species of livestock offered for sale on each market day.
Fig. 2 provides an example using cattle markets in Upper Volta.
The selection of which markets to include in the network,
like any sampling problem, is a function of the coverage desired and
the resources available for collecting the information. If there are
several primary collection markets feeding a secondary redistribution
market, it is not necessary to regularly collect data from all of them.
A few can be selected on the basis of location, distance from the
secondary market and the volume of supply. Seasonal observations
can then be used to correlate those left out from the network with the
secondary market in order to estimate their supply and prices.
Types of livestock market data to be collected
The volume of livestock on offer at the market is important as it
gives a picture of the supply as well as its influence on prices. If
possible the volume should be recorded by species, sex and age. Such
information is easier to obtain in fenced markets where there are
controlled gates. It may be difficult in open areas where livestock
are bunched together in mobs. In such cases number by species is
sufficient. The sex and age structure can be estimated by sampling
the mobs seasonally.
The number of sellers and buyers participating in the market should
be recorded to indicate the degree of concentration and hence
influence on the price of livestock and offtake.
- 333
—-i
MALI
J
 
aALI
r
I DEZ
•■v8
7
;
MARK1YE*,
G1R1M^:A.//NIGER
/*'Y3feT1i \
/^NZNAy
/
's f--_G°*™ __/gkSEYTENGA
™I°/'1)JIZO|\/-X—>toTera,Nia.ey
'IN7/
YZZ8\/
9™/
r<v./\
/
\
\
I/ \/^^\
K08D08G08*O^__:1Z772_.
toZouake
Abidjan
 
 
toNia.ey
\
DIA88G.
P
C
toCotonou
/
BENIN
—/ _1toLo.e/
to8aga,Z wku(T1G1
Ku.asi,a1dAccra""^i
-y>i.portandtransitrekoutes
_^.exporttrekroutes
.+do.estictrekroutes
«£pri.arycollection.arkets
□secondaryredistributio1.arke s
Qter.i1al.arkets
C1A1T
N.Majoroattlem rketsandflowsiUppVolta
1ource:8riza-Ninoetal( -10).
The price of livestock transacted should be recorded by breed, sex
and age either on a census or sampling basis depending on the volume
and the number of enumerators available. For large markets where
there are more than 200 animals sold per species a 25-30% sample is
desirable. In a market where one expects two breeds and their crosses
this yields a minimum of three to six observations per class of
animal by breed, sex and age.
A survey of prices producers receive should be made period
ically to establish their relationships with market prices.
The weight of traded animals in the sample if possible, should be
recorded. However, it may be difficult to weigh livestock in
unorganised markets, where there is no auctioning on a weight and
grade basis. In such cases visual assessment of body size (large,
medium, small) and body condition (good, fair, poor) coupled with
the age category of the animal can give a good indication. Weight
correlations with body size and body condition can be established
from the nearest abbatoirs.
The destination and purpose of traded animals should be recorded as
such information will give a good picture of the direction and
magnitude of flow and the purpose for which animals are purchased.
The mode of transport should be recorded as it will show the importance
of trekking, railing and trucking over time.
All of the above information should be recorded by field
officers or enumerators from the selected markets on the market days
predetermined by the market research officers. In addition to this
they should comment on conditions that influence the supply and demand
situation for that market day.
Frequency of data collection
Once the livestock markets from which data will be regularly collected
are determined, the next question is at what frequency should the
data be collected? This depends on the type of livestock market in
question and the frequency of market days. In primary livestock
- 335 -
markets livestock are usually traded once or twice a week. Market
days of secondary or regional livestock markets do occur more
frequently. The frequency of a terminal market-day may be as much
as six times a week.
It is desirable for time series analysis to have weekly
data but it is not necessary to collect data at each market day. It
can be collected on a sampling basis after establishing the represen
tativeness of the market days with respect to the week. Are all
market days held during the week similar or are some market days more
important than others in terms of volume offered and volume traded?
For instance if in a secondary livestock market both Wednesdays and
Fridays are market days and Fridays tend to be more important, one
can establish the factor (X) by which the volume supplied and the
volume traded is greater than on Wednesdays by conducting an initial
survey of the market as well as crosschecking the information by
interviewing buyers and sellers in the market. This factor can then
be used to estimate the market parameters that pertain to the
Wednesday markets from data collected on Friday markets.
In livestock markets where there is only one market day per
week a similar approach can be taken in sampling one week or two weeks
per month. However, one has to be cautious and take into account that
seasonal changes will be captured by such sampling.
In many African countries (e.g. Ethiopia and Nigeria) the
occurrence of major religious (Christian and Muslim) holidays has a
market effect on supply, demand and prices of livestock, especially
those of smallstock. Demand is high during these holidays and prices
can be 80% more than the annual average price (Okali and Obi, 1982) .
It is therefore essential to intensify livestock market data collec
tion during such holidays in order to accurately assess their impact
on the various market parameters.
- 336 -
Studies of market performance and efficiency
Studies of the performance and efficiency of the livestock marketing
system at the various links of the chain as livestock move from
producers to consumers (including the wholesale and retail trade of
meat) have to be conducted by a senior livestock market analyst who
is in charge of the entire livestock market data collection and
analysis with assistance in the field by his colleagues at head
quarters and the provincial livestock marketing officers (see Section
6) . The reason for this is two-fold. First the observations and
probings necessary to get the information require a high degree of
skill and experience. Secondly, the senior market analyst has to
acquire a first-hand insight into how the livestock market operates
in order to properly analyse and interpret the time series data being
collected by the field officers.
Studies of market performance and efficiency include two
major aspects of the livestock marketing system. The first is an
assessment of the degree of buyer concentration in the markets
selected for time series data collection and how livestock prices
are arrived at and purchasing is financed. Although the number of
buyers and how many animals they bought in the market can be recorded
by field enumerators, assessing the manner in which they operate,
whom they represent, in how many other livestock markets they trade,
and how they finance their purchase requires a considerable degree of
skill and market knowledge to elicit.
The second study involves establishing the cost of livestock
and meat marketing as animals change hands from the producer to the
primary markets, to the secondary markets and finally to the terminal
markets, where they are slaughtered for domestic consumption and/or
are exported live.
These marketing costs can be distinguished as costs of:-
(i) transporting (trekking, trucking and/or railing);
(ii) feeding (including grazing);
(iii) marketing levies and taxes imposed by local and national
authorities;
- 337 -
(iv) mortality or loss (some animals die during transit
because of diseases or other physical stress; some
might stray and not be recovered) ;
(v) slaughtering and processing costs;
(vi) capital as represented by the interest on the money
tied up by the livestock from the point of purchase to
the point of sale; and
(vii) the opportunity cost or salary of the operator (trader,
butcher etc.).
The above information can be established by interviewing
livestock traders and managers of slaughterhouses and spot checking
the information by actual observation on their operations. Livestock
marketing margins can be defined as the difference between the sales
price of the animal (meat) and the costs incurred by the seller
including the acquisition price of the animal (meat) . The less the
margins the more efficient the marketing system.
Analysis of livestock market data
The types of analysis of livestock market data are categorised under
four major headings :-
(i) supply of livestock;
(ii) destination of livestock;
(iii) price movements; and
(iv) market performance and efficiency
Supply of livestock
The sources of livestock supplying the particular market can be
analysed using frequencies of traded animals by area of origin by
breed, sex, age and total volume. This indicates what types of
animals are supplied by each area and the frequencies can be used
to estimate marketed offtake from the hinterland of the market.
A problem which frequently arises in such analysis is to
what detail origins of livestock need to be specified. One can
a priori section the hinterland by a functional criterion (geographic,
type of producer or source) and instruct enumerators to categorise
- 338 -
origins of livestock into the specified sections. Alternatively, one
can instruct them to record place names of the origins of livestock
and decide later how to categorise them. Table 2 shows an example of
such an analysis by type of producers.
Table 2. Source of cattle supply to the Emali market-
Source No. of cattle
Group ranches in Kajiado district 1143
Trading centres serving group ranches
in Kajiado district 292
Individual, ranches Kajiado district 23
Farms in Machakos district 9
77.9
19.9
1.6
0.6
TOTAL 1467 100
Source: Bekure et al (1982).
Destination of traded livestock
The frequencies of traded livestock by breed, sex, age and purpose of
purchase for major destinations reveals the relationships between
various livestock markets as well as the trade in livestock between
pastoralists, agro-pastoralists and agriculturalists. Tables 3 and 4
show such an analysis for selected destinations.
- 339 -
Table 3. Destination of cattle trade at Emali.
Purpose Destination No. % of
slaughter
% of
production
%
Total
Slaughter Ong'ata Rongai 510 32 n.a. 20
Dagoretti 214 13 n.a. 8
KMC - Athi River 242 15 n.a. 9
- Mombasa 9 1 n.a. 0
Mariakani 211 13 n.a. 8
Emali 25 2 n.a. 1
Others 380 24 n.a. 15
Subtotal slaughter 1 591 100 n.a. 61
Production Machakos district 612 n.a. 62 24
Group ranches Kaj iado 341 n.a. 34 13
Individual " 40 n.a. 4 2
Subtotal production 993 n.a. 100 39
TOTAL 2 584 100 100 100
Source: Bekure et al (1982)
The total volume and fluctuation of livestock supply are
of major interest as they show the degree of seasonality of supply.
This can be easily seen by plotting weekly or monthly supplies (e.g.
see Fig. 3). For long-term determination of seasonality several years'
time series data is essential. A standard statistical technique for
estalishing seasonality is the method of moving averages (Yamane , 1967;
Croxton et al, 1969) .
- 340 -
Table4.Char cteristicsofc ttboughtEmafordestinationM ch k sa dKaj adog up
individualranches'
Total
N-
166 '5 1 1 2 6 6
A3
female
-0 10 25
1 1 - - 26
Kajiadogroupnd
individualranches
castrate
24- N
N6
1 1 - -
N-
3N
male 61
A N - - - - N
female
2 35 ?7 - - - - N
castrate
Machakos
A A 13- - -
6 6 145 612
male 404 26 430
- - - -
430
Breed
Sahiwalcross
Maasaizebu Borancross NRAATOTAL
Immature
Adult
Subtotal Immature
Adult
Subtotal Immature
Adult
Subtotal
TOTAL
1800-
 
"~ — — Price of cattle
Head of cattle
60CH
0+
s
-I
0
1981
M A M
1982
Fig 3. Mean monthly supply and price per head at Emali market.
Source; Bekure et al (1982)
- 342 -
Livestock prices
The most important single parameter in collecting time series data on
livestock marketing is livestock price. For a given market and a
given period (week, month, year) mean livestock prices per head and
liveweight can be analysed by breed, sex and age. Table 5 gives an
example of such analysis.
The next analysis is that of determining whether there are
seasonalities in livestock prices. Seasonality of prices can be
determined by type of animals sold. The methodology is similar to
that of determining seasonality in supply of livestock. Fig. 3
illustrates an example. However, several years' data are required
to establish long-term seasonality.
Finally, a stepwise multiple regression analysis can be used
to fit a demand model for a given market relating price to breed, sex,
age, season, volume of supply and number of buyers in the market as
well as the interactions between these independent variables (Drapper
and Smith, 1966). The longer the time series data the better the
specification of the demand function. The value of the demand func
tion does not lie so much in its ability to predict future prices but
in its usefulness in quantifying the relationships between livestock
prices and breed, sex, age, season of the year, type of market and
other variables one may specify. Shapiro (1979) analysed price of
cattle in Upper Volta as a function of age, sex, season of the year,
region of the market, type of seller etc. The results of the
regression are summarised as follows:
- Prices for males increase at an increasing rate with age,
up to 5.7 years (average 5,000 CFA F per year overall);
they increase at a decreasing rate to age 11.4, where
they begin to fall.
- No strong age-price relation was found for females.
- A premium of 1,500 CFA F was paid for steers over bulls
at all ages.
- Only slight evidence of higher prices was found for sales
closer to major consumption and export centres.
- 343 -
- No difference was found between ethnic groups as sellers.
- Higher prices occur during the rainy season, lowest during
the dry season; males hold their prices better than
females during the dry season.
- The amount of seller market information had no significant
effect on prices.
- Higher prices were paid for males when they were sold to
butchers and traders; females brought higher prices when
sold to herders and farmers.
- The type of market had no significant effect on prices.
The foregoing analysis on time series data of livestock
markets can be done on district, province and national levels to give
information at various levels of aggregation.
Market performance and efficiency
Analysis of the studies of the operations of livestock markets yield
in the first instance qualitative and quantitative information on the
operations of the markets that is useful in analysing and interpreting
the time series data generated. There is no set way of analysing such
data. Market analysts have their own approach which is acquired
through experience. Two examples of calculations of marketing margins
are given in Tables 6 and 7. Studies of the well established tra
ditional marketing systems in West Africa, which also deal with
exporting live cattle from the northern pastoral areas to the coastal
zone, show that they perform efficiently (with gross margins of 15-
20%) despite their traditional base and complexity (Herman, 1979;
Staatz, 1979). Our own work (Bekure et al , 1982) and that of Matthes
(1979) in Kenya show that marketing margins are much higher (25-35%).
- 344 -
Table 5. Mean prices of cattle at Email by breed, sex and age -
September 1981 to September 1982.
Breed Sex
Immature Adult All ages
per per per
head
per per per
kghead kg kg(LWT) head
Small E.A. zebu male KSh 645 3.70 1 478 4.91 914 4.16
(N) (761) (15) (364) (9) (1125) (24)
castrate: KSh 866 4.76 1 603 5.70 1249 5.52
(N) (660) (26) (714) (112) (1374) (138)
female: KSh 886 4.83 998 4.87 984 4.87
(N) (62) (2) (446) (31) (508) (33)
All sexes: KSh 753 4.39 1 396 5.48 1 079 5.24
(N) (1483) (43) (1524) (152) (3007) (195)
% Grand total 48.5% 50% 98.5%
Sahiwal cross male: KSh 1284 - 2150 4.42 1284 4.42
(N) (5) - (2) (2) (7) (2)
castrate: KSh 1378 - 2159 6.08 1908 6.08
(N) (9) - (19) (9) (28) (9)
female: KSh - - 1683 5.0 1683 5.0
(N) - - (3) (1) (3) (1)
all sexes: KSh 1337 - 2099 5.71 1776 5.72
(N) (14) - (24) (12) (33) (12)
% Grand total - 1% 1.2%
Boran cross castrate: KSh 600 - - - 600 -
(N) (6) - - - (6) -
% Grand total - - - - - -
all breeds males : KSh 649 3.70 1482 4.82 916 4.18
(N) (766) (15) (366) (11) (1132) (26)
% Grand total 25,2 12% 37%
castrates : KSh 870 4.76 1617 5.73 1259 5.55
(N) (675) (26) (733) (121) (1408) (147)
% Grand total 22% 24% 46%
- 345 -
Table 5 cont .
females: KSh 886 4.83 1 002 4.87 908 4.87
(N) (62) (2) (449) (32) (511) (34)
% Grand total 2% 15% 17%
Grand total: KSh 758 4.39 1 407 5.50 1 087 5.27
(N) (1503) (43) (1548) (164) (3051) (207)
% Grand total 49% 51% 100%
Table 6. Wholesale butchers' margins in Ouagadougou (in CFA F per
head).
CFA F per head
Costs
Intermediary's commission 250
Holding fee 100
Slaughter tax
Condemnation loss
Apprentices' salaries
1,000
150
200
Purchase price 31,250
32,950
Revenues
Sale of meat 31,250
Sale of fifth quarter 5,250
36,500
Margin 3,550
Source : Ariza-Nino et al (1980)
One half percent of meat was condemned in Ouagadougou in 1976.
Average apprentice's salary 8,000 CFA F per month.
CCarcass weight 125 kg, 250 CFA per kg.
Meat sold at purchase price of animal.
eSales value of fith quarter is 42 CFA F per kg carcass weight.
- 346 -
Table 7. Cattle marketing costs at Email and, Ongx' ata Rongai-
KShs Kshs per
per head kg liveweight
Source: Bekure et al (1982)
Mean purchase price from producers 1 012 3.97
65 0.26
Market costs up to Email
- trekking KSh 20.00
- watering fee KSh 2.00
- food and lodging KSh 12.00
- transport KSh 4.00
- loss - trading KSh 10.00
- death (1/60) KSh 17.00
Mean sales price at Emali 1 396 5.48
Trader's mean gross margin at Emali 319 1.29
Mean purchase price at Emali 1 396 5.48
Markeing costs up to Ong'ata Rongai 119 0.47
- County Council fees KSh 7.00
- trekking fee KSh 20.00
- watering fee KSh 2.00
- food & lodging KSh 20.00
- transport KSh 12.00
- miscellaneous costs KSh 12.00
- loss - trading KSh 14.00
- death(l/60) KSh 32.00
Mean sales price at Ong'ata Rongai 1 919 7.6
Trader's mean gross margin at Ong'ata
Rongai 394 1.55
- 347 -
Organisation for collection and analysis of livestock market data
There cannot be a set way of organising a national livestock marketing
data collection and analysis. One can only make a general suggestion
which elaborates the main features. In the end the organisation to
be adopted in a given country will have to take into account the
prevailing conditions. What is suggested here is a flexible organ
isational framework which can be easily adopted with modification.
Figure 4 presents a schematic representation of a general
organisational framework for the collection and analysis of livestock
market data. It is based on the hierarchial structure most common
to ministries of livestock, or agriculture in many African countries.
Their field offices are usually hierarchially organized in conformity
with administrative units of their respective countries. Thus, the
ministries have field offices at provincial, district and occasionaly
at locational levels, where field officers with secondary school
education are posted.
It is suggested that the organisation for the collection and
supervision of livestock market data follows the same structure. At
the headquarters of the ministry a Livestock Marketing Analysis
Section should be responsible for organising the collection, super
vising and analysing all data on the livestock market system of the
country. This section should be located within the Livestock
Marketing Department so that it falls under the responsibility of a
department with a functional commitment to the task. In ministries
where there is no such department it may be placed under the Planning
Unit. The section should be headed by a competent senior livestock
market analyst with access to a statistician and data processing
facilities. How big and permanent the staffing of the section should
be is a function of the size and the importance of the livestock
sector in the country. It should therefore be tailored to the needs
of the country in question.
At the provincial level, it is suggested that a full-time
junior market analyst be made responsible for supervising and
partially analysing the data collected in the province. This will
enable prompt supervision of data collection as well as facilitate
- 348 -
c
c/>
c
s c/ls
Ph
rJ
Q
1
1
1
PlanningUnit
or
LivestockMarketing
Department
Livestockmarketing
Analysissection
1
m
c/>
S
►J
CO
S
rJ
PL, o
CN CN
S
3
Q
CN
1-1 i—i
a r-i
rJ 5
Pm o
CO
►j M
p oj> u
w u
►J CO
3
r-l cr
3 T3z CO
o 0)
M pa
H
< >%
tN u
H ■U
3 CD•r-l
o d
oi •H
o 2
co
0J
CJ
•H
CM
CJ
C
•H
>
O
rJ
Pw
 
CO
0)
CJ
•H
U-l
M-l
O
o
•r-l
r-l
■M
CO■r-l
Q
r*
o
o
■u
en
0)
>
co
l-i
o
co
•i-l
>
l-l
01 /->
P- Cfl
C/> 3
Q
co
01
CJ
0)
r*
M
• i-i CO
co o
cd
cj
o
►J
<n
4J
CU
u
a S
o hJ
O *J >-'
co
>
•r-l
rJ
to E
<3> +i
(35 co
c/> to
60
•r-l
.H
- 349 -
feed-back of market data to market participants at the local level.
For instance average weekly prices of important categories of live
stock in major district markets can be compiled at the provincial
office and promptly reported to the public by the mass media. This
in itself may increase the efficiency of the livestock marketing
system.
In addition to supervising and ensuring good quality data
collection at the district level, the provincial livestock market
analyst will be responsible for conducting in-depth livestock market
studies under the direction and supervision of the senior livestock
market analyst at the headquarters of the ministry.
At the district level, it is suggested that one of the
already existing personnel of the ministry be made responsible for
supervising data collection and passing the information to the provin
cial livestock market analyst. In many African countries district
level personnel are university graduates with sufficient background
to handle the task provided they are given adequate orientation and
on the job training. In cases where the district is located in a major
livestock trading region a full-time supervisor may be warranted.
At the market level it is our contention that most of the
time series data collection on livestock markets can be conducted and
supervised by existing personnel, who in many instances have offices
a few yards from these markets. In case livestock markets, where no
ministry officers are posted, are selected, other enumerators (e.g.
teachers, bussinessmen etc.) may be contracted on a part-time basis,
provided they are adequately supervised with unannounced spot checking.
We have used this approach at Ong'ata Rongai, Kenya with good success.
In conclusion, we would like to reiterate that the above
organisation is suggested only as a framework for consideration. In
small countries or in countries where the network of livestock markets
is small and caters for only one terminal market, as in Botswana, the
whole organisation may mean one senior and two junior analysts being
assisted by field enumerators.
- 350 -
References
Abott, J.C. 1979. Marketing issues in agricultural development
planning. In: R. Moyer and S. Hollander (eds) . Markets
and marketing in developing economics pp 87-116, Richard D.
Irwin, Publishers.
Ariza-Nino, E.J., Herman L., Makinen, M. and Steedman C. 1980.
Livestock and meat marketing in West Africa: Vol 1:
Synthesis Upper Volta. CRED, University of Michigan.
Bekure, S., Evangelou, P. and Chabari, F. 1982. Livestock marketing
in eastern Kajiado, Kenya. ILCA/Kenya Working Document No.
23, Nairobi.
Carlisle, K.R.M. and Randag, A.G. 19/0. Analysis of factors affect
ing meat packaging developments. In: A H Bunting (ed) .
Change in agriculture. Gerald Duckworth & Co Ltd., London.
Croxton, F.E., Cowden, D.J. and Klein S. 1967. Applied general
statistics. Prentice-Hall, Inc., Englewood Clifts, New
Jersey.
Herman, L. 1979. The livestock and meat marketing system in Upper
Volta: Summary of an evaluation of economic efficiency.
In: Kenneth H. Shapiro, (ed.). The summary report from
the livestock production and marketing in the Entente
States of West Africa Project, CRED, University of Michigan,
USA.
Hill, P. 1970. Studies in rural capitalism in West Africa. Cambridge
University Press.
Jarvis, L.S. 1980. Cattle as a store of wealth in Swaziland:
Comment. Amer. J. Agric. Econ. 62(3): 606-613.
Khalifa, A.H. and Simpson, M.C. 1972. Perverse supply in nomadic
societies. Oxford Agrar. Stud. 1: 47-56.
Low, A.R.C., Kemp, R.L. and Doran, M.H. 1980. Cattle as a store of
wealth: Reply. Amer. J. Agric. Econ. 62(3): 613-617.
Matthes, M.C. 1979. Livestock marketing in Kenya: Field draft
document. FAO/UNDP Marketing Development Project Ken
75-005, Nairobi.
- 351 -
Okali, C. and Obi, A. 1982. A preliminary report on the supply and
price of ruminants in selected markets, Oyo State, Nigeria.
(Unedited draft) .
Staatz, J. 1979. The economies of cattle and meat marketing in
Ivory Coast. Monograph II from the livestock production and
marketing in the Entente States of West Africa project.
CRED, University of Michigan.
Swift, J. 1979. The development of livestock trading in a nomad
pastoral economy: The Somali case. In: Proceedings of
international meeting on nomadic pastoralism. Pastoral
production and society. Cambridge University Press.
Yamane, T. 1967. Statistics: An introductory analysis. Harper &
Row, Publishers, New York.
- 352 -
Etudes sur la commercialisation du bétail
Résumé
L'un des principaux objectifs de la recherche sur les systèmes
pastoraux consiste à accroître la productivité et à améliorer le
niveau de vie des éleveurs. L'adoption d'interventions engendrées
par la recherche sur les systèmes pastoraux en vue de l'accroissement
de la productivité implique à la fois des coûts et des gains. Il est
par conséquent essentiel que ces interventions soient non seulement
techniquement réalisables et socialement acceptables, mais aussi
économiquement rentables. L'objectif du présent document est de
souligner la nécessité d'étudier les marchés de bétail et de procéder
à la collecte systématique de données chronologiques sur les prix
du bétail, sur les marchés locaux, régionaux et finals.
Dans la plupart des pays africains sévit une grave carence de
données chronologiques sur les prix du bétail de même que su: la
performance et sur l'efficacité du système de commercialisation du
bétail. Une représentation schématique d'un tel système qui montre
le flux de bétail du producteur au marché secondaire (régional) et
final (national) à travers un ou plusieurs marchés primaires est
donnée. Les marchés de bétail peuvent être facilement différenciés
par le type de vendeurs et d'acheteurs opérant sur le marché et par
l'objectif de l'achat du bétail.
La recherche sur les systèmes de commercialisation du bétail implique
une approche bi-directionnelle. La première fait appel à la collecte
régulière de données chronologiques sur un réseau de marchés de bétail
sélectionnés. La deuxième a trait à une étude approfondie de la
performance et de l'efficacité du système de commercialisation du
bétail aux diverses étapes de la chaîne qui conduit le bétail du
stade de la production à celui de la consommation.
Dans la collecte de données chronologiques, la première chose à faire
est de procéder à la mise en place d'un réseau de recueil de
l'information sur les marchés de l'élevage qui permette d'identifier
les marchés de bétail et de sélectionner les marchés à inclure dans
- 353 -
les réseaux. Les types de données des marchés de bétail à collecter
incluent le volume du bétail disponible, le nombre de vendeurs et
d'acheteurs, le poids des animaux échangés, la destination des
.r : ux échangés et le mode de transport utilisé. La fréquence de
la -jxiecte des données dépend du type de marché de bétail en
question et de la fréquence des jours de marché. Dans les marchés
de bétail primaires, le bétail est généralement échangé une ou deux
fois par semaine. Les jours de marché sont beaucoup plus fréquents
sur les marchés secondaires ou régionaux de bétail. Sur les
marchés finals, la fréquence des jours de marché peut atteindre six
par semaine .
Les études sur la performance et l'efficacité du système de
commercialisation du bétail aux diverses étapes de la chaine
production-consommation doivent être effectuées par un expert-analyste
des marchés de bétail chargé de la totalité des opérations de
collecte et d'analyse de données sur les marchés de bétail avec
l'assistance sur le terrain de ses collègues du siège et d'agents
provinciaux chargés de la commercialisation du bétail. Les études
sur la performance et l'efficacité des marchés portent sur deux
aspects essentiels du système de commercialisation du bétail. Le
premier a trait à une évaluation du degré de concentration des
acheteurs dans les marchés sélectionnés en vue de la collecte de
données chronologiques ainsi que sur la manière dont les prix du
bétail sont déterminés et sur le mode de financement des achats.
La deuxième étude fait appel à l'établissement des coûts de la
commercialisation du bétail et de la viande à mesure que les animaux
passent des mains du producteur au marché final, où ils sont abattus
pour la consommation locale et/ou exportés sur pied, après avoir
transité par les marchés primaires et secondaires.
Les types d'analyse des données sur les marchés de bétail sont
catégorisés comme suit: offre de bétail; destination du bétail;
fluctuation des prix; et performance et efficacité des marchés. Le
paramètre le plus important dans la collecte de données chronologiques
sur la commercialisation du bétail est le prix du bétail. Pour une
période et pour un marché donnés, le prix par tête et par poids vif
- 354 -
peut être analysé par race, sexe et âge.
Un cadre organisationnel souple pour la collecte et l'analyse des
données du marché est suggéré. Il se fonde sur la structure
hiérarchisée qui est si commune aux ministères de l'élevage ou de
l'agriculture dans plusieurs pays africains.
- 355 -

Summary of Discussion Session 7.
Chairman: Dr Jackson Kategile (IDRC, Kenya)
Discussion led by Mr Larry Ngutter (Kenya)
Dr Barry observed that marketing channels had developed since 1979 due
to Nigeria's decision to supplement its meat imports from outside
Africa by purchasing Sahel meat. However, to get to Nigeria,
animals had to go through Niger or Upper Volta. In 1975 a national
organisation was set up in Upper Volta - ONERA (Office national
d' exploitation des ressources animales) . ONERA' s headquarters in
Ouagadougou regularly gathered information on prices and numbers of
animals. This had resulted in a map of stock movements in 1980.
In reference to Dr Bekure's paper, Dr Chema pointed out that although
it was true that there was government control of meat prices in Kenya,
there was no government monopoly on the export of livestock, provided
that such exports were sanctioned by the government.
Prof. Saka Nuru suggested to Dr Bekure that the threat of epidemic
diseases (e.g. rinderpest) could be added to the list of factors
inducing the sale of livestock. Dr Bekure agreed, but said that
epidemics could not be considered a permanent feature of the system.
The model referred to was a simple one and was not expected to incor
porate everything.
Dr Tilahun referred to the ILCA/RDP livestock marketing study carried
out in Ethiopia in which the degree of responsiveness of buyers and
suppliers to price changes was measured. This could give an indication
of the market potential for additional livestock. The study revealed
that for cows the percentage change in quantity demand was 1.7 times
larger than the percentage change in price that brought it about.
The change in the price of female sheep and goats on the other hand
had little effect on the quantity demanded. For male goats and sheep
the quantity demanded was more responsive to change in price. For
small ruminants increases in price resulted in decreases in supply,
but for cows a 1% increase in price resulted in a 0.8% increase in
quantity supplied. The study also illustrated the effects of religious
- 357 -
festivals on livestock sales. Both Moslem and Christian festivals
influenced the sale of small ruminants in most of the studied markets
and Christian festivals influenced the sale of cattle.
- 358 -
Résumé des débats de la septième seance
Président: M. Jackson Kategile (CRDI, Kenya)
Débats dirigés par M. Larry Ngutter (Kenya)
Le Dr Barry a souligné que les circuits de commercialisation s'étaient
développés depuis 1979, notamment en raison de la décision du Nigéria
de compléter ses importations de viande en provenance de pays non
africains par l'achat de viande dans les pays sahéliens. Toutefois,
pour arriver au Nigéria, les animaux doivent traverser soit le Niger,
soit la Haute-Volta. En 1975, une organisation nationale, l'ONERA
(Office national d'exploitation des ressources nationales) avait été
mise en place en Haute-Volta. Le siège de l'ONERA à Ouagadougou
rassemble généralement les données sur les prix et sur les nombres
des animaux. Ces activités de collecte ont donné lieu à une carte
des mouvements du bétail en 1980.
Se référant au document de M. Bekuré, le Dr Chema a souligné que
certes le Gouvernement contrôlait le prix de la viande au Kenya, mais
il n'exerçait pas de monopole sur l'exportation du bétail qui était
toutefois soumise à son autorisation.
Le Prof. Saka Nuru a suggéré à M. Bekuré d'ajouter la menace de
maladies épidémiques (par exemple la peste bovine) à la liste des
facteurs à la base des ventes de bétail. M. Bekuré a accepté ce
point de vue mais a déclaré que les épidémies ne pouvaient pas être
considérées comme une caractéristique permanente du système. Le
modèle auquel il avait été fait référence était très simple et ne
pouvait donc pas tout inclure.
M. Tilahun a fait allusion à l'étude CIPEA/RDP sur la commercialisation
du bétail effectuée en Ethiopie, dans laquelle l'impact du
changement de prix sur les acheteurs et les producteurs avait été
évalué. Cela pouvait donner une idée de la capacifé du marché
d'accueillir de nouvelles têtes. L'étude a révélé que pour les
vaches, l'évolution du pourcentage de la demande quantitative était
de 1,7 fois plus importante que le changement de pourcentage du prix
qui en avait été à la base. Le changement du prix des ovins et
caprins femelles, d'autre part, avait eu très peu d'effet sur la demande
- 359 -
"V
quantitative. Pour les caprins et pour les ovins mâles, la quantité
demandée avait été beaucoup plus influencée par le changement du prix.
Pour les petits ruminants, l'accroissement du prix s'était traduit par
une baisse de la production mais pour les vaches, un accroissement du
prix de 1% a donné lieu à un accroissement de 0,3 de la quantité
offerte. L'étude illustre également les effets des fêtes religieuses
sur les ventes de bétail. Les fêtes musulmanes, tout comme les fêtes
chrétiennes, ont influencé la vente des petits ruminants sur la
plupart des marchés étudiés et les fêtes chrétiennes ont influencé
les ventes de bovins .
- 360 -
FEATURES AND CONSTRAINTS
IDENTIFIED IN PSR AT ILCA

Summary of Discussion Session 8.
Chairman and discussion leader: Dr Barry Nestel (U.K.)
Dr Rhissa said that in PSR he felt it was not only a question of
identifying constraints but it was also necessary to put forward a
package of solutions to governments. It was up to the governments
to analyse and adopt solutions relevant to their aspirations and
potential .
Dr Nestel asked Dr Cossins what were the time factors in a program
of work needed to reach the diagnostic phase. Dr Cossins said that
this would vary from system to system. In the case of the Ethiopian
rangelands study, an initial survey took three weeks in the field,
and an additional month and a half to prepare a research proposal.
Descriptive and diagnostic study programmes were then initiated and
only eighteen months later were the team at the point where they were
initiating some specific or component research programmes. Dr Wilson
said that his team produced systems studies in 3^ years in Mali and
then initiated component research on identified constraints. Some
component research on agronomy was however carried out from the
beginning. He felt it should be possible to identify some constraints
more quickly in zones of similar ecology and social organisation.
Prof. Saka Nuru thought that such a time-frame would very much depend
on the availability of the required information from the national
institutions and the government. With such information available,
the descriptive phase should not take more than a year to confirm or
amend what was known. Where such information was lacking then the
descriptive/diagnostic phase would take much longer. He felt that
the active co-operation of the national institutions, the government
and the pastoralists themselves was essential.
Dr Diakite, in referring to the situation in Mali, said that
researchers should not just leave the results of research to be
applied - one had also to ensure against poor use and interpretation
of data.
- 361 -
Dr Barry felt that the discussion was disturbing - surely the main
aim was to be useful to pastoral people? One constraint was that the
state did not help the pastoralist by instituting systems or
technologies which allowed the pastoralist to be exploited. National
priotities usually concentrated on grain production rather than
pastoral production.
- 362 -
Résumé des débats de la huitième séance
Président et directeur des débats: M. Barry Nestel (R.-U.)
Le Dr Rhissa a déclaré que dans la RSP, il pensait qu'il ne s'agissait
pas seulement de l'identification des contraintes et qu'il fallait
également proposer des solutions aux gouvernements. Il incombait aux
gouvernements d'analyser et d'adopter les solutions conformes à leurs
aspirations et adaptées à leurs capacités.
M. Nestel a demandé à M. Cossins quels étaient les facteurs temps
nécessaires dans un programme de travail pour atteindre la phase de
diagnostic. M. Cossins a répondu que cela variait d'un système à
l'autre. Dans le cas de l'étude sur les parcours éthiopiens, une
enquête initiale avait duré trois semaines sur le terrain et il avait
fallu 45 jours supplémentaires pour préparer une proposition de recher
che. Les programmes d'étude de diagnostic et de description avaient
ensuite été lancés et ce n'est que 18 mois plus tard que l'équipe a pu
démarrer certains programmes spécifiques de recherche sur les
composantes. M. Wilson a déclaré que son équipe avait produit des
études sur les systèmes en 3 ans et demi au Mali et qu'elle avait
ensuite commencé des recherches sur les composantes sur des
contraintes identifiées. Des activités de recherche sur les compo
santes dans le domaine de l'agronomie avaient toutefois été entreprises
dès le début. Il pensait qu'il devrait être possible d'identifier
certaines contraintes plus rapidement dans des zones où l'écologie et
l'organisation sociale étaient similaires.
Le Prof. Saka Nuru estimait qu'un tel cadre temporel devrait être
fortement tributaire de la disponibilité des données requises dans les
institutions nationales et gouvernementales. Avec de telles
informations, la base descriptive ne devrait pas prendre plus d'un an
pour confirmer ou infirmer ce qui était connu. Là où une telle
information faisait défaut, la phase description/diagnostic prendrait
beaucoup de temps. Il estimait que la coopération des instituts
nationaux, du Gouvernement et des éleveurs eux-mêmes revêtait une
importance capitale.
- 363
Se référant à la situation qui prévaut au Mali, le Dr Diakité a
déclaré que les chercheurs ne devraient pas simplement se contenter de
donner les résultats de leur recherche pour application. Il fallait
également veiller à se prémunir contre l'utilisation et l'interpréta
tion inadéquates des données.
Pour le Dr Barry, ce débat était génant. L'objectif essentiel était
certainement d'être utile aux populations pastorales. L'une des
contraintes était que l'Etat n'aidait pas l'éleveur en mettant en place
des systèmes ou des techniques qui favorisaient l'exploitation de celui-
ci. En général, la priorité nationale était la production de
céréales plutôt que la production pastorale.
- 364
IDENTIFYING THE SCOPE
FOR IMPROVEMENT IN PASTORAL PRODUCTION

The scope for improvement
12 1
Stephen Sandford , Frank Anderson and Addis Anteneh
1Economists, Livestock Policy Unit, ILCA, Ethiopia
Agricultural Economist and
Programme, ILCA, Ethiopia
2" Team Leader, Highlands
Introduction
This paper takes as its starting point that in the cycle of pastoral
systems research (PSR) the existing system has been adequately inves
tigated, described, and analysed and that the constraints on the
further development of the existing system have been identified.
The next stage in the cycle is to identify the scope for improvement.
Identifying the scope for improvement may imply different
things. To some people in some situations, there seems to be almost
no scope for improvement. Although the present system is one of low
productivity, it seems that nothing can be done to improve it except
at a cost which is far in excess of potential benefits. In such
circumstances identifying a possible improvement is a most challenging
task and enormous effort is spent in designing and redesigning re
search and development work in order to try to reduce costs and
increase benefits.
To other people in other situations there seems to be a
wide range of improvements, each of which offers promise. Identifying
the scope for improvement then appears to be more a case of selecting
from among the many alternatives those options which offer the
greatest promise. In such cases the main task is estimating the
likely consequences of alternatives.
Although this paper is relevant to both sets of circumstances,
it has a number of different alternatives primarily in mind. In both
cases decisions in principle are made on possibilities for improvement
and the ideas about the improvements are refined to the point where
they can be the subject of on-farm/range, on-research station,
365
•V
component research, or the subject of further study of some other
appropriate kind. This paper is concerned with the social and
economic aspects of the identification and assessment of improvements.
In particular situations potential improvements are likely
to induce some combination of changes in technology (technological
change) and in organisations, institutions, or in other economic,
administrative or social conditions (hereinafter collectively termed
"social change"). One extreme example, involving no change in
technology but substantial social change, would be the expropriation
of individually-owned and managed water points (on communally grazed
land) and the substitution of group-ownership and management by an
elected committee of pastoralists. Such a change could lead to
improvement both in overall productivity, through the establishment
of a mechanism to control the number of livestock permitted to graze
an area of land, and in the equity of distribution (between members
of a pastoral society) of access to water and so to grazing. At the
other extreme there may be a potential for substantial changes in
technology, e.g. by the introduction of an effective vaccination
against contagious bovine pleuropneumonia conferring immunity for
life, which will require no direct social change but which can be
applied through, for example, existing arrangements for anti-rinder
pest vaccinations.
Most potential improvements will require some combination of
substantial amounts of both new technology and social change. Both
the technological and the social changes required will usually be
of two sorts; firstly the direct intended change, which is a funda
mental part of the improvement, and secondly the indirect changes,
often unintended, unforseen and, sometimes, unwanted, which are
brought about by the direct changes. Direct technological changes
may lead to indirect social (as well as technological) changes,
and vice veicsa. The indirect changes may be as important, in both
their costs and benefits, as the direct.
- 366 -
Tasks in identifying the scope for improvement
This paper will discuss the tasks to be done principally from the
point of view of ILCA. However other organisations carrying out PSR
are faced with essentially the same obligations and constraints and
so, with minor amendments, what is written here about ILCA is appli
cable to other organisations.
In order to fulfill its mandate, ILCA needs not only to
identify potential improvements but also to make some assessment of
ILCA's mandate states that it should seek to increase livestock
production and to improve the quality of life in sub-Saharan tropical
Africa. In assessing potential improvements it needs to look at their
likely impact in terms of these two criteria. It also, in order to
prevent resources being wasted on research whose results are not
implemented, needs to assess the ohances of adoption of potential
improvements. Partly this is a matter of looking at the objectives
of the other parties involved in development, e.g. of the host
government or of the pastoralists of the area concerned, to see
whether what seem to be improvements in terms of ILCA's criteria will
also seem to these other parties to be improvements that they should
encourage rather than frustrate. Partly it is a question of assessing
whether, with the best will in the world, the parties and institutions
concerned are capable of introducing the improvements within a reason
able period of time. Finally ILCA needs to ensure that at least some
of the improvements that it develops will bear fruit rapidly. Host
governments, pastoralists and donors will all need early reassurance
of the capacity of PSR to yield useful results if their initial
enthusiasm and support is not to wane. Some improvements are not
only important but also, because of their very nature, have very long
pay-off periods. They need to be complemented by others which even
if not inherently so important can serve to sustain interest.
There are, then, four things to be done: to identify
potential improvements, to assess the probable impact of each, to
assess the probability that an improvement will be adopted and to
estimate how rapidly an improvement will bear fruit. It is convenient
for analytical purposes to classify them as separate activities
- 367 -
although in practice they will often be carried out more or less
simultaneously and by a single person, sometimes through a series
of iterations that modify initial ideas into something more appropri
ate and feasible.
In some cases the next step in the development of improvements
lies with ILCA (or other organisations carrying out PSR) alone, e.g.
where a new technology has to be devised or adapted for a specific
location and where ILCA can itself carry out the necessary technical
component research. In other cases some specific social change,
which can only be brought about by someone else, normally a part of
the host government, is required, either on its own or as a necessary
concomitant to an ILCA derived change in technology. Even where
someone else must take the next step ILCA needs to assess the likely
impact and chances of adoption of the improvement, and the probable
rapidity of its fruition so as to provide that someone with adequate
information on which to base their own decision. In some cases what
is required is research by someone else; in other cases further
research will not be useful and what is needed is the implementation
of development forthwith.
Identifying potential improvements
If, for example, an initial analysis of the system indicates that
marketing is a constraint then this signposts the need for an
improvement in marketing. "Identifying a potential improvement"
is the process of looking at the critical stage or steps in the
marketing process and tentatively selecting things which could be
done, in the way for example that Bekure, Evangelou and Chabari
(1982) have identified supply of credit, weighing and grading, and
sale by auction, as potential improvements in livestock marketing
in Kajiado, Kenya.
Assessing the impact of potential improvements
The assessment of impact of an improvement merits some further con
sideration. Essentially this is a predictive activity, or an attempt
to forecast something which may follow on from research; it is not
- 368 -
monitoring or evaluating something which is already taking place.
Assessment of impact has so far been described in terms of the likely
direct and derived technological and social changes which will be
associated with an improvement, but at this stage it is necessary to
go further and evaluate the changes according to ILCA's criteria, i.e.
in terms of their consequences on production and on the quality of
life. ILCA will not be interested, of course, in improvements which
simply maximize output regardless of cost. ILCA's mandate to increase
output implies the rider "at reasonable cost". On the other hand at
this stage in the PSR cycle precise prediction of costs of improve
ments in relation to benefits is impossible. The same is true of
other kinds of impacts. Since research has not yet taken place the
quantity of the potential benefits (yields) is not yet known and
the volume of costs (inputs) is equally obscure. Moreover, since
the improvement is still some way off in time from introduction to
pastoralists, the relative prices of outputs and inputs will probably
be subject to considerable changes, but ones which are largely unpre
dictable in direction or size. Analyses of various degrees of
sophistication and complexity can, and should, be done to explore
the combinations of yields, inputs, prices and other factors which
give rise to impacts which are on balance favourable or unfavourable.
These should permit an assessment of the probable long-term average
overall impact. Another element in the assessment of the increased
production is the extent to which increases in average (over different
sites or different periods) production is matched by increased risk
(variation) .
ILCA's mandate also requires it to improve the "quality of
life" - an expression which is in some ILCA documents rephrased as
"standard of living", although in common usage there are important
differences between the way the terms are used. The latter usually
implies much more emphasis on material welfare, particularly on the
consumption of goods and services, whereas the former embraces not
only material welfare (clean air, low infant mortality) but also
satisfaction of a less material kind, e.g. harmonious social re
lations.
- 369 -
In practical terms, ILCA can assess the probable impact of
potential improvements on the quality of life in a number of important
ways. The impact on the natural environment is one of these; a
second is to assess to what extent the potential improvements proposed
will benefit all sections of the community, i.e. all kinds of
households - "kinds" in terms, for example, of wealth and power,
occupation or ethnicity. Certain kinds of both technological and
social changes, even if their benefits are not intended to be restrict
ed to particular socio-economic classes, are in practice more likely
to benefit or be adopted by some classes and this may positively
injure others. ILCA, both out of a proper concern for social equity,
and from a need to avoid the resentment against itself which will arise
if it is thought to favour only certain groups, should aim to avoid
developing improvements which are likely to benefit only certain
ethnic groups in an ethnically mixed region, or the strong at the
expense of the weak. Thus, as far as is practicable, ILCA should
develop a package of improvements at least some of which should offer
benefit to every kind of household. An early assessment is required
of the likely impact on relative welfare of different groups, since
once research has been sucessfully carried out it will not be possible
to suppress its results even if their impact is inequitable. A third
important respect in which the impact of a potential improvement on
the quality of life can be assessed is in terms of the way it affects
the distribution of benefits and burdens within households - in
particular its effect on the nutrition of children and on the work
pattern of women. There are, of course, a number of other aspects
to the quality of life which may be relevant to particular potential
improvements, but those mentioned here deserve consideration in
respect of every improvement.
Assessing the chances of adoption of potential improvements
ILCA needs not only to assess the probable impact of improvements if
adopted, but also to assess the chances of desirable improvements
being adopted. It is not a sufficient condition for an improvement
to be adopted that it has desirable impacts in terms of output and
quality of life. There may be difficulties involved, for example
- 370 -
difficulties in the procurement and distribution of high technology
inputs, which are unlikely to be solved even if ILCA convinces
everyone concerned about the urgency of their solution. Social
constraints, in the form of traditional institutions and value
systems, have often been cited as causes of the failure to adopt new
technological improvements. The possibility of such social con
straints can not be denied, possibly in the form of fundamental
divergencies in values between, for example, ILCA and a progressive
government on the one hand and a traditionally-oriented pastoral
society on the other. But in many cases in the past where agricul
tural researchers and extension personnel have blamed such social
factors, the real problem has subsequently been found to be that the
researchers were advocating unprofitable technology or had failed to
grasp the full complexity, e.g. in respect of risk, of their clients'
decision-making process.
Assessing the rapidity of fruition
As already noted, ILCA needs to include in its package of improvements
some which will yield early evidence of the usefulness of PSR, even
if these are of only modest importance. Otherwise there is a danger
that essential support by other parties, i.e. host governments,
pastoralists and donors, will be withdrawn prematurely from more
important improvements which can be developed and yield results only
over a longer term. The early development and introduction of some
improvements, even if only modest ones, can help to sustain interest
and support while the more long-term, and possibly more important,
improvements are still being worked on.
Who should identify the scope for improvement?
The techniques for assessing the impact, probability of adoption, and
speed of fruition of potential improvements are overviewed in a
later section. This section examines who should do the identification
and assessment. A "natural" solution appears to be that as far as
improvements implying primarily technological change are concerned
the first task of identifying the potential improvements should be
by the natural scientist in that specialist field (e.g. soil science,
- 371 -
forage agronomy, genetics) most concerned, and that in the subsequent
task of predicting the indirect changes and the impact on production
and the quality of life, social scientists (including economists)
should also play a substantial part. The implication of this
approach is that it is primarily the staff and consultants of the
PSR organisation itself (e.g. ILCA) who will be involved.
However there are two shortcomings in this point of view.
Firstly the probability that a host government will take the necessary
steps that will allow a successfully tested improvement to be adopted
subsequently will be directly proportional to the degree of its own
involvement early in the process of decision making that led to the
development of the improvement. Such early involvement by the host
government will not only lead to its greater sense of commitment to
the introduction of the improvement and to an earlier awareness of the
institutional changes it may subsequently have to make, but also to
its providing ILCA earlier with information about the host govern
ment's own intentions that may affect the chances of adoption. That
is the positive aspect. The negative aspect is that unless the host
government's involvement in decision making on the improvements to
be designed and tested is to be merely token window-dressing, then
ILCA risks being prevented from developing improvements, which could
be highly beneficial, by host government officials who may not have
the training or time to be able to appreciate an improvement's real
potential. Host government involvement may be either a help or a
hindrance .
The other shortcoming is that an important argument raised
in favour of a systems research (FSR/PSR) approach is that it more
closely considers small farmers' /pastoralists ' point of views and
is more influenced by their values and opinions than are other
research approaches. Clearly this "advantage" is in danger of being
lost if, at a critical stage when decisions are being made about the
improvements to be tested or studied, farmers' /pastoralists ' opinions
are not directly canvassed but are only "represented" by what natural
or social scientists think farmers ought or are likely to think.
Consideration needs to be given as to how the opinion of pastoralists
might be incorporated into the selection of improvements for testing.
- 372 -
The "literature" on FSR is somewhat silent or opaque on this
subject. Most of the discussion in the literature about farmer's par
ticipation in FSR is about their involvement in identifying con
straints, in carrying out on-farm trials, and in subsequent adoption
of innovations. Literature dealing specifically with the issues of
the institutional devices required to incorporate farmers '/ pas to-
ralists ' opinions into the process of identifying improvements cannot
be identified.
Three points are worth making here. First the staff of many
government services for implementing development are unlikely to
represent farmers ' /pastoralists ' opinions any better than research
scientists. In many countries, indeed, especially where FSR/PSR
has been willingly adopted as an approach by the research services,
the researchers are more likely than are the agents of the often
'top-down' oriented extension services to be able to understand
farmers' points of view. Secondly, in seeking pastoralists' opinions
care has to be taken to obtain an adequate cross-section. It is not
enough to consult political leaders, or leaders of pastoralists'
associations, or "prominent" or "progressive" pastoralists. Such
people are likely to be drawn from among the better-off and are likely
neither to understand well the problems of the less well-off, nor,
if they understand them, to represent them if such views are contrary
to their own interests. Nor is it enough just to call a public
meeting to discuss selection of improvements. At such public meetings
the interests of the less well-off, even though they attend, are
unlikely to be strongly defended.
The third main point is that if pastoralists are to make the
contribution to identification of improvements that is needed of them,
then time and care has to be devoted to discussing with them what the
different improvements may involve. Of course, if asked "would you
like a new kind of livestock feed which will make your animals give
more milk? the response will be affirmative. That sort of consul
tation is mere window-dressing. what needs to be done is to thrash
out in some detail what the innovation may mean (including any
alternative options) in respect of, for example, cash cost, labour-use
profiles, risk, land tenure, loss of pastoralists' independence etc.
- 373
This is inevitably time-consuming and will not lead to published
articles in internationally referreed journals in the way that
conventional disciplinary research (even if subsequently ignored by
pastoralists) will yield. But it is more likely to lead to the
design and testing of innovations which will subsequently be adopted.
The point is that pastoralists cannot be expected to grasp immediately
all the implications for their pastoral system of a potential tech
nological change. Little in their own experience will have fitted
them to know what sort of questions one needs to ask about new tech
nology. But if scientists and pastoralists will sit down together
to thrash some of the issues out, from the union of their past,
separately inadequate, experience something useful may emerge. In
the case of the participation in decision making by both pastoralists
and officials of the host government's service for implementing devel
opment, care has to be taken to establish a proper institutional
framework for that participation and to ensure that participants are
adequately oriented and briefed so that their participation is both
genuine and has positive results.
Techniques for assessing potential improvements
This section of the paper deals briefly with the kind of techniques
available for assessing the relative merits of different potential
improvements once these have been identified.
Early on in the preparation of this paper the idea was
discussed that it ought to be possible to make generalisations, about
impact, about probability of adoption, about rapidity of fruition,
based on broad categorisations of, on the one hand, "types of impro
vement" and on the other "kinds of situations" in which the improve
ments are to be introduced: for example, "individual land tenure"
areas would be one kind of situation, "communal tenure" areas would
be another. Categorisation of improvements could be by a number of
different criteria. One of these would be in terms of western
concepts of disciplinary boundaries, and would lead to a categori
sation as follows.
- 374 -
1. Improvements to the primary productivity of the natural
vegetation leading to better animal nutrition.
2. The growing of introduced grass, browse and other forage.
3. The supply of supplementary feed (minerals, protein,
energy etc. ) from non-rangeland sources.
4. Improvements in animal health.
5. Genetic improvements (whether from imported genes or by
intra-area selection) .
6. Improvement in water supplies - leading to more frequent
watering, with higher quality water at less energy cost in watering.
7. Improvement in animal husbandry (e.g. breeding seasons,
weaning practices, housing etc.); i.e. improvements in the care of
individual livestock kept not subsumed under 1 to 5 above.
8. Improvements in marketing that lead to greater market
efficiency, convenience and equity.
Categorisation by disciplinary boundaries did not lead to
many useful generalisations. Possibly the only important one was
about rapidity of fruition. In many pastoral areas the full scope of
available animal health technology has not yet been exploited and
quite short periods of survey, followed by short trials on pastoral-
ists' herds, can lead to rapid results on a wide scale. Although
improvements on the health side may not substantially affect the
overall productivity of the pastoral system unless the nutrition
constraint is also overcome, they can provide early and dramatic
evidence of the efficacy of PSR as far as the productivity of indi
vidual animals is concerned.
A possible categorisation of improvements in terms of those
which are susceptible to testing by standard experimental techniques
and those which are not was also considered. The distinction is not
absolute but one of degree (more or less) . Most technological changes
are susceptible to standard experimental techniques when tested on
a research station, and some of these, for example, forage crop trials
on individually owned fields, weight gains of individual animals due
- 375 -
to supplementary feeding, can also be tested in this way "on-farm"
("on-range") . In other cases, however, because of difficulty of
measurement, e.g. in the case of milk yields of cows with calves at
foot, or because of difficulty of experimental control, e.g. with
different range management techniques on communally grazed land,
even technological improvements are barely testable in "on-farm"
experimentation. Most social changes are not susceptible to standard
experimental techniques of the kind applied to technological inno
vations because it is simply not possible to conduct a controlled
experiment with them. However, it should not be concluded that
experimentally testable improvements are legitimate and important
and that untestable ones are not. On the contrary, many experimen
tally testable improvements may offer only trivial advantages whereas
the major scope for improvement may be through untestable innovation.
Although it cannot be concluded that improvements which are
experimentally testable are, ipso facto, likely to be more or less
important than those which are not, they have some advantages when
it comes to reducing the risk of failure in large-scale implemen
tation. With improvements that are experimentally testable there
will be some points, short of full-scale implementation, at which
the impact of an improvement on increased production can be verified.
Of course, even in this case what will be measured, when doing
experiments, are changes in output under conditions of experimental
control or of supervision or guidance by scientists. These may
comprise rather different conditions than would be the case in the
event of large-scale adoption by pastoralists at a later time. On-
farm testing, if done for long enough and on a large enough scale,
can also in theory involve assessments of risk and the inter-class
and inter-household distribution of costs and benefits. In practice
it seldom can be done for long enough or on a large enough scale to
achieve this.
However the scope for success in the introduction of improve
ments will be determined more by the characteristics of the particular
situation (in time and space) on which PSR and development are
focused rather than by the category of improvement. Particular
situations differ from each other in so many different ways (environ-
- 376
mental, economic, social, political etc.) that useful categorisations
and generalisations are not possible here. Categorisation by one
criterion (e.g. rainfall) cuts across categorisations by another
(e.g. social structure) in a way that yields a myriad of sub-cat
egories. In other words, the assessment of potential improvements
cannot be done simply by reference to simple rules of thumb (gener
alisations applied to broad categories of improvements and situations)
but will require specific analysis in each instance.
Thus, in assessing potential improvements, broad generalisa
tions based on categories of improvements and of situations cannot
be relied upon to predict likely impact and probability of adoption.
Rather reliance will need to be wholly on case-specific predictive
models. In a few cases, if it is decided to select that improvement
for subsequent component research, some experimental evidence will
subsequently become available to shed light on the validity of the
original models.
The models can be of varying degrees of complexity depending
on the time and resources available to construct and test them and
on the input data available. At one extreme of simplicity, the
model may be no more than a "back-of-the-envelope" calculation
(simple "partial budgeting") to decide whether the average cash
cost of, say, a mineral block is likely to be exceeded by the average
value of extra liveweight gained as a result of using it.
A first step improvement on the "back-of-the-enveloppe
calculations model would be one which:
1. estimated probabilities of different values of net cash
returns, thereby taking at least partial account of the substantial
impact of variability in pastoral, systems on the pay-off from an
improvement.
2. included a cash flow exercise which estimated how the
period and financial deficit between the time at which cash costs
are incurred and cash returns are received might be bridged by
different classes of pastoralists.
- 377 -
3. showed the impact of a potential improvement on the
labour-use profile of different classes of pastoralists. In many
pastoral (and farm) systems, labour is as critical a limiting
resource as cash. It is, therefore, important to calculate whether
the extra labour demand generated by an improvement can be accommo
dated solely by sacrificing leisure (and if so whether the net
returns per extra man hour are comparable to those obtained by other
activities currently carried on) , or whether it will require the
displacement of some other productive activity, and if so with what
result. Such calculations are equivalent to the "gross margin" kind
of calculations carried out for conventional cropping enterprises
which show which activity is likely to yield the highest margin per
hectare and which activities are likely to be displaced by the more
profitable new one. Models that deal with labour-use are more complex
than "gross margin per hectare" analyses. For example in the case of
irrigated land whereas land can be allocated at most to three
different crops in succession each year, i.e. once every four months,
competition between activities for labour occurs on a much more
frequent basis.
The types of models discussed so far are extremely simple
and for the most part are economic models. If time and data allow far
more complex models can be used. In agropastoral enterprises, linear
programming maximizing models have already been used (e.g. by Eddy
(1979) and Delgado (1979)) but not yet for purely pastoral enterprises
except at an excessively aggregate scale, e.g. it has been done for
the Sahel as a whole (Picardi, 1974). Konandreas and Anderson (1982),
building on work by others, have devised a simulation model which can
be used to forecast some of the changes in livestock systems which
will spring from changes in technical parameters. Both these kinds
of models are essentially economic ones, are expressed in mathematical
terms, can be computerised, and deal with rather few relations and
variables at a time (the Konandreas and Anderson model contains some
25 key equations) . On the whole, the economic models developed so
far are best at predicting the impact of improvements on net output
(production less costs) and are not directly concerned with the
probability of adoption or the effect on the quality of life as spelt
- 378 -
out in this paper. Most anthropological models are verbal, not math
ematical, and are much more complex but correspondingly less precise.
They are much more concerned with questions of adoption and of inter
personal equity. A feature of all complex models is that they require
considerable time and effort not only to collect data but also to
manipulate the model and consider its results.
Concluding summary
There are four main tasks to be carried out when identifying the scope
for improvement: identification of potential improvements and then, in
respect of each of those identified, prediction of the likely impact
in terms of at least two criteria (increase in net production, change
in the quality of life), prediction of the probability of adoption and
of the rapidity with which the improvement will bear fruit if adopted.
These are not very radical suggestions but they are seldom practised.
Furthermore it has been suggested that when identifying scope for
improvement not only should the natural and social scientists of the
organisation practising PSR itself be involved, but also pastoralists
and officials of the implementing agencies of the host government
should participate as well. For such participation to be fruitful,
however, thought, trouble and time have to be devoted to ensure that
their participation is genuine and properly structured and informed.
In making predictions, generalisations based on categories of improve
ments and categories of situations in which the improvements are to be
introduced are of very limited use. Case-specific analysis will be
required which will use predictive models of varying complexity and
from different professional disciplines. Because the predictions are
being made in advance even of on-station research, they cannot be
precise but can only be estimates of probabilities concerning the
balance between benefits and costs or between positive and negative
factors .
- 379 -
References
Bekure, S., Evangelou, P., and Chabari, F. 1982. Livestock marketing
in Eastern Kajiado, Kenya. ILCA/Kenya Working Document 26,
Nairobi .
Delgado, C.L. 1979. The Southern Fulani system in Upper Volta.
A model for the integration of crop and livestock production
in the West African savannah. Michigan State University
African Rural Economy Paper No. 20, East Lansing.
Eddy, E.D. 1979. Labour and land use on mixed farms in the pastoral
zone of Niger. Centre for Research on Economic Development,
University of Michigan, Ann Arbor.
Konandreas, A. and Anderson, M. 1982. Cattle herd dynamics', an
integer and stochastic model for evaluating production
alternatives. ILCA Research Report 2, Addis Ababa.
Picardia, A.C. 1974. A systems analysis of nomadic herding in the
West African Sahel. Massachusetts Institute for Technology,
Center for Policy Alternatives.
- 380 -
Envergure des innovations
Résumé
Cette étude prend comme point de départ que le système existant à été
étudié, décrit et analysé de manière adéquate dans le cadre de la
recherche sur les systèmes pastoraux et que les contraintes relatives
au développement futur du système ont été identifiées. La phase
suivante dans le cycle consiste à identifier l'envergure des
innovations.
Il y a quatre tâches essentielles à effectuer dans l'identification
de l'envergure des innovations: l'identification des améliorations
potentielles et ensuite, par rapport à chacune des innovations
identifiées, la prévision des effets potentiels, compte tenu tout
au moins de deux critères (accroissement de la production nette,
changement de la qualité de la vie), la prévision de l'adaptabilité
de l'innovation et de la rapidité avec laquelle celle-ci portera
des fruits si elle est adoptée. Le CIPEA peut évaluer l'impact
probable des améliorations potentielles sur la qualité de la vie de
plusieurs manières. L'impact sur l'environnement fait partie de
celles-ci. Le CIPEA peut également évaluer la mesure dans laquelle les
améliorations potentielles profiteront à toutes les composantes de
la collectivité. Troisième aspect non moins important: la manière
dont une amélioration influe sur la distribution des avantages et des
inconvénients au sein des ménages.
En identifiant l'envergure de l'amélioration, il faudrait faire
appel non seulement aux chercheurs spécialisés en sciences sociales
et naturelles participant à la RSP mais également aux éleveurs et
aux responsables des organismes d'exécution du Gouvernement du pays
hôte. Pour qu'une telle contribution soit fructueuse cependant, il
faudra faire en sorte qu'elle soit bien structurée et que les
participants à cette recherche concertée disposent des informations
requises .
Dans les prévisions, les généralisations basées sur des catégories
d'innovations et des catégories de situations dans lesquelles les
- 381 -
améliorations doivent être introduites sont d'une portée limitée.
Des analyses de cas spécifiques, fondées sur des modèles de prévision
de complexité diverse , effectuées dans diverses disciplines seront
nécessaires .
Les modèles peuvent être de complexité diverse, compte tenu du temps
et des ressources disponibles pour les mettre au point et les tester
et compte tenu des données disponibles. Il peut être très simple mais
il peut également être assez complexe et estimer par exemple les
probabilités de différentes valeurs de revenus monétaires nets, inclure
des calculs sur la marge brute d'auto-financement déterminant la
manière dont le déficit financier qui intervient entre le moment
auquel les dépenses monétaires sont effectuées et celui où les revenus
monétaires sont perçus peut être comblé par les différentes classes
d'éleveurs et montrant l'impact d'une amélioration potentielle et le
profil de l'utilisation de la main-d'oeuvre de différentes classes
d'éleveurs. Si le temps et les données le permettent, des modèles
beaucoup plus complexes peuvent être utilisés.
Etant donné que les prévisions de tels modèles se font à l'avance
(même avant d'entreprendre la recherche au niveau de la station), elles
ne peuvent être précises. Toutefois, elles peuvent constituer des
estimations plausibles sur l'équilibre entre les coûts et les
bénéfices ou entre les facteurs positifs et négatifs.
- 382 -
Biotechnical options
1 2
L.J. Lambourne and M. Butterworth
ix Director of Research, ILCA, Ethiopia
2Senior Animal Nutritionist, ILCA, Ethiopia
Introduction
Let me begin by reminding you that ILCA is one of the International
Agricultural Research Centres with a mandate to carry out a programme
of research supported and extended by documentation and training "to
assist national efforts which aim to effect a change in production and
marketing systems in tropical Africa ..."
Let me quote also Baker (1977)'s selection from the ILCA
prospectus (1974). "The system approach is valuable in research
planning not so much in identifying problems, which are often self-
evident, as in selecting what are likely to be the most profitable
approaches to problem solving. This is especially relevant to
livestock production in Africa where improvement in yield or output
is as likely to come from social change or from range improvement as
it is from genetic change in the animals themselves."
The truth of this has been shown in the constraints to pro
duction and opportunities for improvement identified by the leaders of
our pastoral research teams in their papers to this workshop. One
might well argue that most, if not all, of the problems they have
specified could have been listed before ILCA's systems research pro
grammes began - indeed most of these problems were already identified
or identifiable from the large amount of research information then
available. What was lacking was knowledge of the best ways of solving
the evident problems, given that many optimistic attemps to introduce
new methods based on improved Western models had been dismal failures.
It was not clear whether the technology had been wrong, and just did
not work in the different physical environments of Africa, or whether
the social and economic conditions in which the technology had been
- 383 -
used had prevented it from being applied in such a way as to effect
the improvements of which it was potentially capable.
Sandford has outlined the considerations which seem to be
most important from the social and economic view points in identifying
the scope for improvement in pastoral production. We must also
consider the biological or technical possibilities which are available
for pastoral systems, and try to sort these into some ranking of
applicability considering not only the likelihood of technical success
but also their relevance and feasibility.
Technical features of pastoral systems
African pastoral systems are subsistence-oriented, and based on milk
production - milk, not only to rear the calves which will ensure long-
term continuity of the system, but also to form the mainstay of human
nutrition. These systems are not easily forced into the mould of
conventional Western beef production and many earlier attempts to
develop such enterprises have failed.
The zebu-type cow in these extensive grazing areas may
appear unproductive, indeed the milk yields recorded in ILCA pastoral
studies are quite modest. Consider a cow giving an offtake of 1 1/day;
in term:; of human nutrition she is producing as much as a castrate
gaining 1 kg/day*. Her biological effeciency is high judged by the
value of her output of high quality food in relation to the amount
and value of the forage she has eaten.
* 1 000 ml milk at 14% total solids = 140 g nutrients
1 000 Lwt gain = 600 g bodywt (30% skin, bone, inedible tissues)
= 600 x 0.70 edible tissues (70% water)
= 600 x 0.70 x 0.30 g edible DM
= 126 g nutrients
- 384
Table 1. Estimated yield (ml/day) of cows in ILCA pastoral programmes.
Milk offtake for
human use
(measured)
(ml /day)
Milk intake by calf
f stimated from
weight gains)
(ml /day)
Total
Country (ml/day)
Ethiopia 1140
720
730
2560
(2500)
3700
Kenya (3220)
Mali 1620 2350
Considering the harsh conditions under which they live
small ruminants too reach high levels of individual production. Even
though individual production may be high, cattle herd productivity
could be raised substantially if:
- fewer calves died, and more were reared to sell or to
milk;
- calves grew faster and matured earlier;
- cows calved first at an earlier age; and/or
- cows calved more frequently thereafter.
Table 2. Reproductive performance of sheep and goats in semi-arid
Africa .
Age at 1st
parturition(d)
Parturition
interval (d)
Litter
size
No. of
f email
young/
B/yr
Country
Goats Sheep Goats Sheep Goats Sheep
Mali 470
550
270 250
320 320
230 280
1.21 1.05
1.25 1.04
1.52 1.13
1.63
1.42
2.37
1.51
1.19
1.49
Kenya
Sudan
If more calves were reared better and either sold earlier or
calved earlier, lactating cows could form a higher proportion of the
herd, giving higher efficiency of production per unit of forage eaten,
or per unit area of grazing land used.
- 385 -
Milk yield is the key to most improvements since increases
in yield would make it possible to reduce calf losses, improve growth
rate of calves and thus to meet the implicit objectives of the
pastoralist while also making possible an increase in the turnoff
of surplus male cattle to meet national development needs.
The results of ILCA's systems analysis have shown the
importance also of smallstock - particularly their role as a rapidly
renewable liquid asset, easily converted into cash or exchanged to
meet short-term needs, and of a convenient size to provide meat for
domestic, religious or social celebrations. Their shorter gestation
and high fecundity provides valuable flexibility and their behaviour
makes it possible to exploit types of pasture or browse not eaten by
cattle. The almost universal use of young males for slaughter sale
or exchange has shown up clearly in ILCA surveys.
Table 3. Age and sex structure of flocks in Sahel transhumant systems.
 
Age Sheep (a) Sheep (b) Goats
(months)
Males Females Males Females Males Females
0-6 11 13 6 10
22 ) 23
7-14 6 12 9 19
15 - 21 4 11 3 8
6 ) 49
over 21 7 36 7 38
TOTAL 28 72 25 75 28 72
Just as the type of livestock kept and their role has strong
common features, so too the constraints identified can be grouped into
those which are common to most pastoral systems and some more specific
to a particular region or system. This distinction depends largely on
the overriding influence of climate, mainly rainfall and its seasonal
distribution; we shall examine how this dominates the husbandry of
cattle and small ruminants and contributes, directly and indirectly,
to many of our problems.
386
The effect of climate on nutrition
The ILCA team in Mali has recorded the amount and quality of forage
available, and the liveweight changes of cattle and small ruminants
in the systems relying on the Niger delta (Dicko et al, 1981) .
-X-
Liveweight (kg)
300
200- -
100-
wet season dry season
millet field of pre-irrigated
stubble rice stubble rice and millet
pastures
natural
>k >k ^ *£ =*
-J—I—i—i—.—i—«—.—>
» ND J FM AMJ
Metabolizable
energy
(MJ/d) 
20
n®
i_ld Lop _LT ja_M _i_' M ¥" J_
+ I.0-'
+0.5-
-0.5--
I.O-
Daily ga
(kg/d)
0
Fig 1. Seasonal trends in liveweight (1)
and metabolisable energy intake (2).
- 387 -
Fig. 1 shews the seasonal pattern of changes in liveweight
in cattle in Mali. Similar patterns have been found in many other
tropical climates, and it is evident that the annual weight increase
made during the short wet season when feed is abundant and of good
quality is partly lost over the following months of the dry season,
particularly in the few weeks immediately preceding the following
season's green growth. This means that yearly net gains are low,
maturity is delayed, and first calving is at four or five years of
age. Cows, which usually conceive towards the end of the wet season,
calve after a long period of under-nutrition during the following dry
season. It means also that cows which calve a little late and do not
conceive again before the dry season begins are likely to enter a
long period of lactational/nutritional anoestrus and will not conceive
until their calf has been weaned or until the following year's wet
season. Thus the herd will include too many cows which calve only
every second or even third year.
The classical solution to this predictable yearly problem of
feed shortage is to move, either between wet and dry-season grazing
areas where land is stiil plentiful, or to arable areas where crop
residues are available. Lack of water means that grazing areas
within reach of dry-season wells or ponds soon become overgrazed; the
high concentrations of livestock may help the spread of diseases, and
poor nutrition is made worse by the need to walk long distances each
day or two.
Options available for improvements
Problems arising from aridity
Not only mean annual rainfall and its variation, but also its distri
bution, greatly influences pasture growth and therefore secondary
productivity of livestock. Unreliable early rains can cause repeated
sequences of germination and seedling death; very heavy rains cause
massive runoff and fill surface ponds and streams, but poor infil
tration into the soil means that pasture growth is poorly sustained.
Rainfall distribution may be so unreliable that there is no possibility
of permanent use of land by livestock - Bille (1983) has suggested
- 388 -
that a monthly rainfall of less than 60 mm gives very little or no
grass growth (in the Ethiopian rangelands) and the probability that
rainfall will reach this minimum figure in any one month is the best
biological distinction between wet and dry months. Judged by this
criterion some areas of the northeastern Ethiopian rangelands, for
example, will have no effective primary production at least one year
out of ten except in natural hollows. Baker (1975) pointed out that
the Karamoja area of Uganda could expect one year in four to be so
dry as to seriously reduce crop and pasture growth and cause losses
of animals. This problem cannot be solved by the creation of more
watering points, and such areas can be exploited in the long term
only under a nomadic system permitting emergency access to areas which
have more reliable pasture growth. If this becomes impossible, it
will be necessary for governments to organise emergency drought
relief in the form of livestock or human food, or emergency purchase
of excess stock.
Areas with a well-defined but erratic bimodal rainfall are
betr.er suited to livestock grazing than to introduction of cropping.
Loss of browse through land clearing might even harm livestock
production more than it would benefit from the dry-season crop residue
grazing.
Rainfall cannot be controlled, but by systematic recording
over the pastoral area it is possible to arrive at valuable estimates
of reliability of monthly rainfall. This, in the long run, may make
it possible to plan more stable systems of utilisation and development.
This is clearly an option requiring governmental or regional projects'
initiative, and might well be linked to modern methods of satellite
remote sensing, to give advanced warning of failure of rainfall, and
to provide long-term and large-scale monitoring of rangeland con
ditions.
Problems of undernutrition
Inadequate nutrition in the dry season may occur as a long-term
problem because of progresive loss of dry-season grazing areas,
associated with increasing pressure of populations or with ethnic
or political rivalries. For such cases the options lie with govern
- 389 -
ments, through development planning or market and pricing structures
aimed at increasing the sale of surplus stock and thus reducing the
pressure on grazing resources.
Undernutrition occurs in normal years in most pastoral
areas, and many options are available to pastoralists for reducing or
avoiding its effects.
1. If it is not too serious, its effects can be minimized as
far as possible by sale of stock and by careful management, relying
upon the following wet season's gains to restore livestock to normal
weight and productivity. This depends on the reliability of the
compensatory growth phase and is the laissez-faire strategy adopted in
many areas in Africa and overseas. It fails if the rains fail, particu
larly in areas where lack of marketing facilities makes it difficult or
financially catastrophic to sell animals in a period of widespread
feed shortage.
2. Provide additional feed by improvement of grazing re
sources. Pasture improvement by fertilizing and oversowing legumes
is a well proven technique. Success depends on selection of suitable
seeds and provision, by pelleting or otherwise, of any necessary plant
nutrients and rhizobial inocula. Technically it is an attrative
option, but success depnds also on restricting first-season grazing
so that seedlings can establish, grow to maturity and set seed. This
v-annot be done on communally grazed land and this option, while techni
cally feasible, is therefore not applicable in most African pastoral
areas. It has been shown, however, that some systems recognize the
right of a person to reserve an area for his own use - as the
olepolole reserved for calves or milking cows in the, Kenyan Maasailand.
In northern Nigeria it has been found that pastoralist Fulani may
fence off an area of several hectares by agreement with landholding
sedentary farmers and that productive pastures of Stylosanthes species
can be grown on this to provide good quality forage for the dry season.
While this will not be possible for truly nomadic pastoralists it may
become possible for transhumant herders with a recognized main camp,
particularly under some form of herders' association or similar form
of regulation of grazing rights.
390 -
3. Supplementary feeding. The technology of using protein
concentrates (peanut or cotton-seed meal) or a cheaper non-protein
nitrogen source (urea) is well established in developed pastoral
countries. Compounded grain/molasses/urea supplements are easily made
by appropriately simple village or small-scale industry. These are
used in northern Nigeria, and the widespread practise of giving salt
to animals in the Sahel suggests that, after a brief research study to
define the nutrients most needed, herders might quickly adopt an
improved type of nutrient supplement particularly if its cost of
production or transport were subsidized by the government, as is the
case with fertilizers in many countries.
In Mali considerable use is made of cowpea and groundnut
haulms as supplementary feed, and ILCA has active programmes in both
Mali and northern Nigeria developing improved methods and testing new
varieties of legumes for use as an intercrop or in rotation with the
traditional sorghum or millet. Whatever supplement proves to be best,
economics require that it should be given to the animals most likely to
benefit and not wasted where no response is to be had. This means that
young females mating, calving or lactating for the first time should be
given preference, as also cows whose poor condition might prevent their
conceiving or rearing a vigorous calf. Newly weaned calves may well
repay extra feeding. This is an intervention available to the individ
ual herder when supplementary feed is available.
4. Modify grazing management. ILCA has shown that heavier
grazing in the early wet season will make more feed available in the
early dry season. This requires agreement or control over livestock
numbers and some right of exclusive use of the extra pasture thus
produced. The revival of earlier forms of social/territorial organ
isations, or the creation of new herders' associations may make this
option possible.
In many areas valuable browse species could be encouraged
by selective clearing of others. Browse could be used more efficiently
by careful selective cut t ing of branches to encourage regrowth, or by
judicious use of fruit in the dry season (Cisse, 1982). This, too,
might require some form of agreement among herders using a particular
area - there are precedents for such rational utilisation methods.
391 -
5. Modify livestock management. Where pasture is inadequate,
particularly if this is associated with bush encroachment it may be
better to reduce numbers of sheep or cattle and increase camels or
goats, which are better able to subsist on browse. The association
of goats with degenerated pasture land does not necessarily mean that
goats cause degeneration. Their selectivity in browsing enables them
to produce on land where sheep and cattle, mainly grazers, cannot
thrive. Their higher quality intake is obtained only at the cost of
lower quantity - Schwarz and Said (1981) found that nutrient intake
of goats in the dry season was only 1/4 to 1/9 their nutrient intake
in the wet season, but protein intake was 1/3 of the wet-season level.
Further research is needed to study grazing and browsing habits and
the effects of different methods of stocking on browse production and
regeneration.
More dry-season forage could be made available by the estab
lishment of more water points, ponds or wells to extend the pastoral
area available for grazing in the dry season. This requires government
or regional action, combined with careful assessment of the probable
effects on the vegetation of increased or more prolonged grazing
pressure. Where no great harm is to be expected and where some commu
nal control of livestock numbers can be assured this is perhaps one
of the most attractive options.
In agropastoral areas sedentary farmers have available new
short-cycle varieties of cereals which generally require reasonable
soil fertility. This can be maintained best by animal manuring, and
farmers are putting in extra wells in order to attract pastoralists
to bring their cattle to the farmers' fields in the dry season (Fulton
and Toulmin, 1982) . This interdependence between farmers and herders
will become more and more important in the future.
Within the power of individual pastoralists is the possi
bility of better matching feed requirements to feed supply by
controlling the times of mating and of weaning. To some degree
conception in cows is dependent on their response to improved
nutrition during the two to three months of the wet season and
calvings tend to cluster in the late dry season following. The need
of pastoralists for a year-round supply of milk probably accounts
- 392 -
for a more even spread of calvings seen for example in Wilson (1982) .
Whether it is possible to reduce neonatal and later deaths, and to
improve overall productivity by control of mating times or by sup
plementary feeding of bulls or of cows calving out-of-season needs
study in different systems.
Stock numbers could certainly be reduced by large-scale
sale of surplus animals in the late wet or early dry seasons. This
is a common feature of "stratified" or more highly developed pastoral
industries but depends on the existence of marketing, transport and
either abattoir or fattening facilities commensurate with the annual
surplus cattle turnoff. This needs attention by government or devel
opment project planners but in the long run is essential to a rational
exploitation of extensive pastoral regions.
The individual pastoralist could well be encouraged to cull
the least productive of his stock, rather than keep cows indefinitely
in the hope of getting a calf some day. This is logical if keeping
extra cattle is felt to be at no cost to the individual since the land
is communally owned but, in the longer term, pastoralists must be
persuaded that a chronically diseased cow or one of genetically poor
productivity is a liability and not an asset.
Separate management of different classes of cattle is already
widely practised and has some advantages in both nutritional and
health terms. It facilitates the gathering and holding for vaccination
or sampling of large numbers of stock, particularly if combined with
government establishment of dipping and similar facilities in the
vicinity of the wells near which animals congregate in the early dry
season. Dry stock could be grazed far from water in this season,
allowing pregnant and lactating cows and calves to be given closer
attention. Developments such as this call for some organisation of
pastoralists as well as government policy initiatives.
Specific problems of particular systems
Young stock mortality
Young stock mortality is a problem in Ethiopia and in the Sahel with a
two-year average of 28% deaths in calves. Research is needed to
393 -
establish to what extent this is due to undernutrition because of high
milk offtake for human consumption, particularly from cows rearing
male calves which are considered to be less valuable than female
calves. In small ruminants mortality was 32% before weaning. The
extent to which this was the result of disease or of other causes is
not known. This represents an important field of potential improvement
but requires further research. Losses in smallstock are similarly
high in Kenya, but losses of calves are much lower. In Ethiopia and
Mali the death rate in male calves is higher than in females,
substantially because of competition with humans for limited milk
supplies and preferential treatment of heifer calves. Improved
nutrition and milk production would lead to a substantial increase in
the number of surplus male calves reared for sale.
Bush encroachment on grazing lands
Research is needed into the long-term stability of grazing resources
in relation to grazing pressure, stocking rate and type of stock used.
Growth of productive browse shrubs and trees does not mean a reduction
in total primary production, and the spread of browse species that may
be used in the dry season should help to correct the lack of feed that
usually occurs then. Again, the fact that land is common-owned means
that no pastoralist is prepared to spend his time and effort to selec
tively clear undesirable species, or to safeguard young browse
seedlings during their vulnerable years by reduced grazing pressure.
Creation of pastoralists ' associations is a very necessary first step
in the better husbandry of grazing resources.
Epidemic diseases
Vaccines are available for some of the infectious or tick-borne
diseases that have in the past wiped out herds and flocks. Research
is continuing at ILRAD and there may one day be vaccines for East
Coast fever and trypanosomiasis. There is an urgent need for more
dipping or spraying and injection yards, better distributed over
major pastoral areas. In the absence of any regulation or control
of grazing, the development of rotational systems to reduce the number
of tick or helminth larvae is impossible. In other countries it has
been possible to introduce more resistant breeds of livestock, but
394 -
zebu-based cattle and local breeds of smallstock are probably the most
resistant livestock available. Apart from manual removal of ticks no
management or genetic means seem to be available, and pastoralists
must rely on veterinary and chemical methods of control, which in
turn depend on active, well motivated extension, advisory and
veterinary services in the zone affected.
Good versus poor herdsmen
Apart from identifiable single causes of production losses, evidence
from ILCA's studies in Mali and in Kenya shows that there are large
differences between herds and flocks, within the same general type of
husbandry.
Table 4. Production indices* for (a) Mbirikani group ranch, Kenya
and (b) two agropastoral systems in Mali (goats and sheep
combined) .
Production index
(a) (b)
Goats Sheep Millet Rice
Best flock
Worst flock
676
101
692
257
816
312
1 200
226
* Average litter size x survival to weaning x 150 day weight
Ewe weight postpartum
Research is needed to identify the features of individual
herd or flock management responsible for these large differences. In
Kenya Grandin has shown that productivity is associated with the
wealth of the "owner" or with herd or flock size, but it is claimed
that great skill is involved in grazing management in the early wet
season and that this has great influence on yearly productivity
(Swift, 1982). If these special skills can be described and under
stood, it might be possible to raise poorer managers to near the
- 395 -
level of the best, without introduction of any alien technology.
Wilson (1982) proposed an intervention pathway for small ruminants
combining many of these ideas. The series of improvements proposed
merits careful study, since it is based on many years' experience
in several countries.
r* T
Identify best flocks in both
subsystems and determine
management factors governing
output
t
*
Preach and practise improved
management in all flocks
Isolate local strains of
epidemic diseases (e.g.
pasteurella) , prepare
autovaccines and conduct
mass campaigns
Study possibilities of
transfer of rice
subsystem benefits
(water, crop residues,
and by-products) to
millet subsystem
Selection within flocks of
best individual males
(growth rate) and twins
for breeding ^
*
Further improvements in manage
ment (flock stratification) to
delay age at first parturition
and to prevent parturitions
occuring at intervals of less
than 240 dayc
-c^>
Institute seasonal
breeding if worthwile
advantage can be
demonstrated from this
practice at this stage
 
Veterinary treatment against
internal (and external)
parasites
<^
Individual veterinary
treatment
Manipulation of flock structures
(cullinq of old females and sale
of surplus young females) to
obtain maximum production of
young (parturition interval +
litter size)
<^7
i. «J
Supplementary feeding
of out-of-season
breeders and/or
fattening for special
purposes
Fig 2. Intervention pathways for small ruminants in the agropastoral
system.
- 396 -
Conclusions
The argument that a systems approach is valuable, not just in iden
tifying problems but in deciding the most effective means of problem-
solving, is amply borne out by ILCA's experience. The examples quoted
here all show the value of understanding the interrelationships among
the causes of lowered livestock productivity. This makes it possible
to direct attention to the root of the trouble, rather than to be
misled by superficial symptoms. Thus, calf deaths result from the
interplay of human needs and milk production, determined by level of
nutrition, which is itself the complex result of primary forage
production and its exploitation by individually managed herds and
flocks .
Increased numbers of water sources may solve an immediate
feed problem by making large areas available for dry-season grazing,
but this may be harmful in the long run to the stability of some
fragile plant associations.
Improvement of grazing resources by proven technology of
legume oversowing depends on the prior creation of some form of
associations of herders with an interest in regulation of livestock
numbers and grazing pressures on the area of land for which they are
responsible.
Because of the overriding influence of aridity, and the
communal use of land not many of the options of technology are
applicable in purely pastoral systems. Those that are relevant often
depend on prior government initiative or some form of social-territorial
organisation, but there are some ways in which pastoralists could help
themselves. More possibilities exist in agropastoral systems, and in
forms of transhumance which allow pastoralists to become responsible
for, and to benefit from, a defined area of land.
References
Baker, R. 1975. Development and the pastoral people of Karamoja, NE
Uganda. An example of the treatment of symptoms. In:
Pastoralism in tropical Africa. Ed. Monod, Oxford University
Press, Oxford, p. 187.
- 397 -
Baker, R. 1977. The administrative trap. In: East African
pastoralism. Proceedings of the conference, Nairobi,
August 1977. ILCA, Addis Ababa, p. 55.
Bille, J.C. 1982. Joint ILCA/RDP Ethiopian pastoral systems study.
Research Report 4. ILCA, Addis Ababa.
Cisse, M.I. 1982. Disponibilités en gousses d' Acacia tortilis pour la
nutrition des petits ruminants dans les ranches Maasai de
Mbirikani et Kimana. Programme Document no. A7 78, ILCA,
Addis Ababa.
Dicko, M.S., Lambourne, L.J. de Leeuw, P.N. and de Haan, C. Voluntary
intake and livestock productivity in the Sahel zone of Mali.
Proc. 32nd meeting EEAP August 1981, Zagreb.
Fulton, D. and Toulmin, C. 1982. Socio-economic study of an
agropastoral system in central Mali. Draft report, ILCA,
Addis Ababa.
Schwarz, H.J. and Said, A.N. 1981. Nutrition and systems of goat
feeding. ITOVIC - INRA International Symposium, Tours,
12-15 May 1981.
Swift, J. 1982. Internal Document, ILCA, Addis Ababa.
Wilson, R.T. ed. 1982. Livestock production in central Mali. ILCA
Bulletin 15, Addis Ababa.
- 398 -
Options bio-techniques
Résumé
Dans le document précédent, il a été procédé à l'examen des questions
les plus importantes aux plans social et économique en ce qui concerne
l'identification de l'envergure des améliorations de la production
pastorale. Il est également important de considérer les possibilités
biologiques ou techniques d'amélioration des systèmes pastoraux
et d'essayer de classer celles-ci par ordre d'applicabilité, en
tenant compte non seulement des possibilités de succès au plan
technique mais aussi de l'opportunité et de la factibilité des
techniques en question.
Le document passe en revue certains des aspects techniques des
systèmes pastoraux d'Afrique. Ces systèmes visent à assurer la
subsistance des éleveurs et se fondent sur la production de lait. La
productivité du troupeau bovin pourrait être augmentée de manière
substantielle si la mortalité chez les veaux régressait, si on élevait
plus de veaux et de génisses pour la vente que pour la production de
lait, si la croissance des veaux était plus rapide et s'ils atteignaient
la maturité plus tôt, si le premier vêlage des vaches intervenait à un
âge plus précoce et si les vêlages subséquents étaient plus fréquents.
L'équipe du CIPEA au Mali a enregistré la quantité et la qualité du
fourrage disponible et les changements de poids vif des bovins et
des petits ruminants du système du delta du Niger. Il est évident que
l'accroissement pondéral annuel enregistré au cours de la petite
saison des pluies lorsque le fourrage est abondant et de bonne
qualité se perd partiellement lors des mois de la saison sèche
suivante, particulièrement pendant les semaines qui précèdent
immédiatement la prochaine saison de croissance. La solution classique
de ce problème prévisible chaque année consiste à effectuer des
déplacements soit entre les zones de pâturage de saison sèche et de
saison des pluies, dans des zones où les ressources en pâturages
sont abondantes, soit dans des zones de culture où des sous-produits
agricoles sont disponibles.
- 399 -
Les variations et la distribution de la pluviosité moyenne annuelle
influencent considérablement la croissance des ressources pastorales
et partant, la production secondaire de bétail. Des pluies précoces
irréguliëres peuvent être à la base de séquences répétées de germi
nation de courte durée; les grosses averses peuvent entraîner un
ruissellement abondant et remplir les mares et les cours d'eau de
surface; mais une infiltration insuffisante dans le sol signifie une
croissance inadéquate des pâturages. Les zones à pluviomé
trie clairement définie mais irrégulière sont plus adaptées aux
pâturages qu'à l'introduction de cultures. La perte de ligneux par
le défrichage pourrait causer au bétail un préjudice que ne pourrait
compenser le pâturage des sous-produits agricoles en saison sècbe.
Une nutrition inadéquate pendant la saison sèche peut constituer un
problème à long terme à cause de la perte progressive de zones de
pâturage de saison sèche associée à l'accroissement de la pression de
la population ou aux rivalités ethniques et politiques. La malnutri
tion sévit dans les années normales dans la plupart des zones
pastorales et les éleveurs disposent de plusieurs solutions pour en
réduire ou éviter les effets. Ceux-ci peuvent être minimisés par la
vente de bétail et par une gestion appropriée, basée sur les gains de
la saison des pluies suivante pour restaurer la productivité du
bétail et améliorer ses performances pondérales. Du fourrage
supplémentaire peut être fourni par l'amélioration des ressources
pâturables - fertilisation et ensemencement de légumineuses par
exemple. La complémentation fourragère peut être pratiquée: la
technique qui consiste à utiliser des concentrés de protéines ou une
source non protéique d'azote moins coûteuse est bien établie dans les
pays à élevage développé. Le CIPEA a également démontré que
l'intensification du pâturage au début de la saison des pluies produit
plus de fourrage au début de la saison sèche. Lorsque les pâturages
sont inadéquats, en particulier lorsque cette insuffisance est liée
à l'empiétement des broussailles, il serait peut être plus approprié
de réduire le nombre des ovins et des bovins et d'augmenter ceux
des chameaux ou des chèvres qui supportent mieux un régime basé sur
la consommation de ligneux. On pourrait développer la quantité de
fourrage en mettant en place beaucoup plus de points d'eau, de mares
- 400 -
ou de puits en vue de l'extension de la zone pastorale disponible pour
les pâturages de saison sèche.
La mortalité des jeunes animaux constitue un problème en Ethiopie et
au Sahel ou,en deux ans, on a enregistré 28% des décès chez les veaux.
Il conviendrait d'entreprendre des recherches en vue de déterminer le
rôle joué par la malnutrition du fait des taux élevés de prélèvement
de lait pour la consommation humaine, en particulier pour ce qui est
des vaches ayant mis bas des veaux auxquels on attribue une valeur
inférieure à celle des génisses. Il est nécessaire d'entreprendre des
recherches sur la stabilité à long terme des ressources pâturables,
eu égard aux pressions sur les pâturages, aux taux de charge et aux
types de bétail utilisé?. La nécessité d'aménager des périmètres où
pourraient s'effectuer des bains détiqueurs, des pulvérisations et
des injections devient urgente, de même que celle d'une meilleure
distribution de ces périmètres dans les zones pastorales. Il est
nécessaire d'entreprendre des recherches pour identifier les
caractéristiques de la gestion des troupeaux de bovins ou de petits
ruminants qui expliquent les grandes différences entre les troupeaux
de bovins et les troupeaux de petits ruminants élevés dans la même
zone.
- 401 -

Summary of Discussion Session 9.
Chairman: Dr Samson Chema (Kenya)
Dr Rhissa felt that the discussion on the identification of possibili
ties for improving pastoral production had neglected the fundamental
point - the organisation of the pastoralists themselves. The herdsmen
were the principal agents of improvement and their contribution could
be better secured through training, information and meetings.
Dr Abel said that Mr Sandford was not encouraging about the
possibility of communicating effectively with pastoralists over the
introduction of new technology. He asked was there not scope for using
traditional pastoral institutions for this purpose. Mr Sandford said
that some traditional organisations were not well structured for*
carrying out this kind of dialogue - in most traditional organisations
it was an elite class who carried on the dialogue. They were probably
not representative of all groups e.g. the poor, women, subordinate
ethnic groups etc. In discussing new technological improvements they
would need to be adequately briefed on what the new technology
implied. They did not have adequate experience of introducing
technology to be able to spot the critical issues unless they were
adequately briefed. Dr Abel suggested that technology assessment
ought to include an appraisal of the ability of a new technology to
survive a drought, and the contribution it could make to a recovery
from drought. Mr Sandford agreed.
Dr von Kaufmann said that maximising milk production was not always
the pastoralist' s objective. If there were many people to feed with
few cattle then milk production was vital. But as herd size grew or
milk production increased then the percentage of milk off-take
decreased.
Prof. Saka Nuru said that the meeting had been informed that pastoralists
disposed of young male animals, particularly goats and sheep, in
favour of female ones. Yet the previous day the meeting had been told
that pastoralists sold male cattle at a later age to get higher
financial returns. There appeared to be a contradiction.
- 403 -
Dr Lambourne said that the decision by smallstock owners in the Sahel
to sell young males early was based on the fact that their growth
rate slowed down after 9 to 12 months. It was therefore good
business to sell them before this. On the other hand, the main cost
of rearing a small calf was the milk it had consumed. To keep a few
males an extra year cost little.
Prof. Saka Nuru pointed out that ILCA researchers in the field seemed
to neglect goat's milk - a vital livestock product. Dr Lambourne
said that goat's milk was used in the Sahel for family use by small-
stock herders, but this was not generally the case if cow's milk was
available. ILCA has considered the possibility of encouraging studies
of goat milk production, but lacked the staff to do it.
Dr Kalusa said that there seemed to be very little attention paid to
the role of donkeys in regions of e.g. Ethiopia. Yet the donkey
clearly affected the lives of the indigenous people. Dr Lambourne
replied that ILCA had a research programme on animal traction,
including cultivation and transport. Dr Wilson had worked on donkeys
in the past, and it was hoped that ILCA would do some more research
on this subject as resources permitted.
Dr Thomson illustrated the dangers of introducing supplementary
feeding policies into a livestock system with an example from Syria.
This was one of the reasons why she.p numbers in Syria were today
about double the number in the late 1950s, with consequent deleterious
effects on the natural grazing lands.
- 404 -
Résumé des débats de la neuvième séance
Président: Dr Samson Chema (Kenya)
Le Dr Rhissa a souligné que les débats sur l'identification des
possibilités d'amélioration de la production pastorale avaient négligé
un point fondamental: l'organisation des éleveurs eux-mêmes. Les
éleveurs étaient les principaux agents de l'amélioration du système
et leur contribution pouvait être plus positive par le biais de la
formation, de l'information et de réunions.
Le Dr Abel a déclaré que M. Sandford n'était pas optimiste en ce qui
concerne la possibilité de communiquer de manière efficace avec les
éleveurs sur l'introduction de nouvelles techniques. Il a demandé
s'il n'était pas possibile d'utiliser les institutions pastorales
traditionnelles à cet effet. M. Sandford a déclaré que certaines
organisations traditionnelles n'étaient pas assez bien structurées
pour servir de cadre à ce type de dialogue. Dans la plupart des
organisations traditionnelles, c'était à une élite que s'adressait ce
dialogue. Celle-ci n'était certainement pas représentative de tous les
groupes, par exemple, les pauvres, les femmes, les groupes ethniques
de caste inférieure, etc. Dans l'examen des innovations technologiques,
il faudrait que ces éleveurs soient correctement informés sur les
conséquences de l'introduction des nouvelles techniques. Ils ne
disposaient pas de l'expérience adéquate en matière d'introduction de
techniques nouvelles pour être à même de détecter, sans être correcte
ment informés, les questions cruciales. Le Dr Abel a suggéré que
l'évaluation technologique devrait inclure une évaluation de la
capacité d'une nouvelle technique de survivre à la sécheresse et de
sa contribution éventuelle à l'élimination des effets de la sécheresse.
M. Sandford a exprimé son accord.
M. von Kauffman a déclaré que la maximisation de la production laitière
n'était pas toujours l'objectif de l'éleveur. Lorsqu'il y avait
beaucoup de personnes à nourrir avec peu de bovins, la production
laitière était vitale. Mais a mesure que la taille du troupeau ou la
production laitière augmentait, le pourcentage de prélèvement de lait
diminuait.
- 405 -
Le Prof. Saka Nuru a déclaré que les participants à la réunion avaient
été informés que les éleveurs disposaient des jeunes animaux mâles,
en particulier des chèvres et des moutons pour garder les femelles.
Toutefois, le jour précédent, les mêmes participants s'étaient entendu
dire que les éleveurs vendaient les bovins mâles à un âge plus avancé
pour avoir des rentrées de fonds plus importantes. Il semble y avoir
ici une contradiction. M. Lambourne a déclaré que la décision des
propriétaires de petit bétail du Sahel de vendre tôt les jeunes mâles
était basée sur le fait que leur taux de croissance se ralentissait
après 9 à 12 mois. Il était par conséquent rentable de les vendre avant
cela. D'autre part, le coût essentiel de l'élevage d'un petit veau
était le lait qu'il consommait. Garder quelques mâles une année de
plus n'était pas coûteux du tout.
Le Prof. Saka Nuru a souligné que les chercheurs du CIPEA sur le terrain
semblaient négliger le lait de chèvre qui est portant un produit
animal vital. M. Lambourne a déclaré que le lait de chèvre était
consommé dans le Sahel par les familles des éleveurs de petit bétail mais
que ce n'était pas le cas en général lorsque le lait de vache était
disponible. Le CIPEA a envisagé la possibilité d'encourager des études
sur la production de lait de chèvre mais ne disposait pas du personnel
requis pour le faire.
La Dr Kalusa a déclaré qu'il semble y avoir très peu d'intérêt pour
le rôle joué par l'âne dans les provinces de l'Ethiopie par exemple.
Pourtant, il ne faisait aucun doute que l'âne contribuait à la vie
des populations locales.
M. Lambourne a répondu que le CIPEA avait un programme sur la traction
animale, y compris le labour et le transport. M. Wilson avait
travaillé sur les ânes auparavant et l'on espérait que le CIPEA
entreprendrait d'autres activités de recherche sur le thème lorsque
les moyens le permettraient. M. Thompson a souligné les dangers
inhérents à l'introduction de politiques de complémentation dans un
système d'élevage sur la base d'un exemple syrien. C'était là l'une
des raisons pour lesquelles le nombre des ovins était aujourd'hui
sur le point d'atteindre le double de son niveau vers la fin des
années 50 avec comme conséquence, la dégradation des pâturages naturels.
- 406 -
CASE STUDIES IN PSR

Design and testing procedures
in livestock systems research:
An agro-pastoral example
R. von Kaufmann
Agricultural Economist and Team Leader, Subhumid Zone Programme, ILCA,
Nigeria
Introduction
Farming systems research (FSR) seeks to employ the skills of
scientists most directly in the service of improving the welfare of
small farmers. Livestock systems research (LSR) is no different
except that livestock systems are two stage systems that necessarily
demand different approaches to some of the problems of research in
the field.
Rohrbach (1980) introduced a paper on FSR by stating that
'farming systems research is a philosophy and methodology of agricul
tural research for the development of improved technologies appro
priate to small farmer needs and circumstances'. He claimed that
there ore very few controversial issues in farming systems research
(due to a degree of agreement over the basic value and character of
this type of research. Those issues which remain most significant to
practitioners cannot he resolved in the process of debate per se.
These issues, which primarily relate to questions of methodology,
organisation and implementation, are in the process of being resolved
by experience.' In concurrence with that opinion, and in view of
the innumerable texts on farming systems research that already exist,
this paper will avoid repetition of argument and rather concentrate on
lessons drawn from the experience of the ILCA Subhumid Programme in
the practice of LSR. In order to keep the paper to a manageable
length it will concentrate on issues on which FSR and LSR differ.
Another author who is particularly appropriate to dis
cussions of ILCA's work is John Dillon because he reviewed systems
research for the Consultative Group on International Agricultural
Research (CGIAR) . He has stated 'that man, not cations or nodules or
rumen flora of crop varieties or livestock species or dollars,
- 407 -
consumates the system must be a basic text1 (Dillon, 1973) . He
stresses the need 'to take a teological view that effects may be
due to the purposes they serve and only a holistic approach, with
openness and teamness through interdisciplinary endeavour, can lead
to the capturing of adequate understanding of a system for purposes
of improving performance.' He points out 'the need for a structure
which will facilitate a synthesising, integrative, team-oriented
outlook rather than one that is analytical, compartmentalising and
disciplinary and that the agricultural system is a purposive one
involving physical, biological and social parts and that it operates
within an environment having significant purposive components.' He
also notes that 'adoption in the real world is a crucial factor and
hence implies consideration of communication and extension.'
Background to source of the case examples
ILCA subhumid programme
The subhumid programme is charged with a responsibility for re
searching ways and means of enhancing the welfare of sedentary
livestock producers through increased cattle and small ruminant
production.
The programme is based in Kaduna in northern Nigeria in an
ecological zone delineated by the 1 000 mm and 1 500 mm isohyets.
With its good rainfall and radiation (180 - 270 growing days) it is
an area of high potential production. However it is relatively under
utilised because of tsetse-borne sleeping sickness and trypanosomiasis,
though this situation is changing at a rapid pace. Farmers are moving
in, extending arable cultivation and at the same time reducing tsetse
habitats. Hard on their heels the formerly nomadic Fulani cattle-men
are settling and becoming mixed crop-livestock producers. Whilst
the settled agropastoral Fulani are the prime clients of the LSR
they are so closely interlinked with the arable farmers that both
communities have to be included in the research.
The 'pre-research model1 adapted by Kaufmann (ILCA, 1979)
from Johnson et al (1971), indicated that malnutrition is the single
most important constraint to range livestock production. It is also
the factor that is most sensitive to correction with available tech-
- 408 -
nology. Thus improving the nutritional status of the herds is the
paramount objective of the programme. Since there is an inadequate
supply of purchasable feedstuffs, the improved nutrition must come
from forage production.
Livestock systems research
The differences between LSR and FSR are brought about by practical
problems in conducting research rather than by differences in intent
and scope. The following schematic outline of an integrated research
programme drawn up by Harrington (1980) fits as well to LSR as it does
to FSR and is a suitable framework on which to hang the particular
aspects of LSR discussed in this paper.
ON-FARM RESEARCH
choice of
target
farmers and
research >>>>>>>»>»
priorities
• <<<<<<<<«<
POLICY
CONTEXT
national goals
input supply,
credit,
markets, etc.
1 . Plan
obtain a knowledge
and understanding of
farmer circumstances —F
and problems to plan
experiments .
2. Experiment
conduct experiments
in farmers' fields
to formulate improved—
technologies under
farmers conditions. D
3. Recommend
analyze experimental B
results in light of
farmers circumstances—
to formulate farmer
recommendations .
4. Assess
determine farmers'
experience with
technologies. K
new components
incorporated' into
on-farm research
EXPERIMENT
STATION
developing
and screening
new technological
components
eg. varieties
Fig.
Promote
Demonstrate improved
technologies to farmers.>>>>>
Overview of an integrated research programme.1
identify
problems
for station
research
- 409 -
Considerations in designing improved technology
Complementarity
LSR should freely use the research results and experiences of other
research and development organisations. LSR should also call on
these other institutions to carry out back-up work. The research
institutes can help overcome problems that require detailed on-station
experimentation. That does not, however, preclude LSR teams from
doing their own experiments which can not be done elsewhere for, say,
ecological or cultural reseaons.
The development agencies are essential to LSR as vehicles
for testing the proposed interventions and procedures in the real
world. There is no other way for LSR to test the adoptability of its
products .
Case example
ILCA's subhumid programme is closely associated with the National
Animal Production Research Institue (NAPRI) of Ahmadu Bello Univer
sity. NAPRI has been conducting research on animal production in
Nigeria for about 30 years and has a wealth of results, information
and scientific expertise on which ILCA has relied very heavily.
The programme is also linked to the Livestock Project Unit
(LPU) of the Federal Livestock Department. The LPU is responsible for
implementing a World Bank-assisted livestock development project. The
staff of the LPU provide an ever-present audience from the development
community which is necessary in the selection of priorities and for
the quick uptake of interventions. The association with LPU will
enable ILCA to assess the uptake, effectiveness and persistence of its
innovations in 'real life1.
An example of this three-way cooperation can be taken from
the nutrition trials conducted in the LSR programme.
On station. The rations were developed from data on feed requirements,
and forage and agro-industrial by-product nutritional values determined
by NAPRI scientists.
- 410 -
Researcher managed trials. Carefully controlled trials confirmed the
predictions and the Fulani appeared to accept the principle of feeding
certain animals certain amounts at certain times.
Farmer managed trials. The LPU then incorporated ILCA's findings into
a pilot smallholder dairy scheme that is now being actively promoted.
ILCA is continuing to obtain all the necessary records to assess the
uptake, success and persistence of the innovation through its close
association with LPU field staff.
Policy oriented
Being complementary also means that FSR must be policy oriented 'with
decisions relative to national research goals being fed downwards'
(Dillon, 1973). This is essential to the cohesiveness of the above
three steps.
Case example
The Director of the Federal Livestock Department, who is also on the
Board of ILCA, has appointed a technical advisory committee of dis
tinguished Nigerian agriculturalists to assist the team with advice
and guidance on both technical and policy matters. This committee
is chaired by the Director of NAPRI and has other members from federal,
state and University circles. Their input, both formal and informal,
is essential to keeping the team's work consistent with national
policy. If the team did not have this advice it not only could lose
vital support but it would also be much less likely to come up with
adoptable interventions.
Social responsibility
In most basic research the purposes and advantages to the end users
are not a major concern of the scientist. LSR, however, has a social
responsibility. For example technologies that may aid larger farmers
to the disadvantage of smaller farmers should be avoided lest they
defeat the prime objective of LSR, which is to conduct research for
development that does not exacerbate inequalities. That is not meant
to exclude interventions that may help both rich and poor alike or
even help rich without any effect on poorer farmers, since production
and paid employment are usually within government objectives.
- 411 -
Case example
Despite accusations of being patronising and attempting to keep
farmers backward ILCA has assiduously avoided the use of tractors
in the preparation of fodder banks because the extent of cultivation
is limited by the amount of available family labour, and tractors
could drastically alter the status quo to the disadvantage of the
poor.
Reviewing priority problems and opportunities
Once the objective of the LSR effort has been defined the LSR pro
grammes still need to be carefully focused or else the scientists will
tend to take on too much and, as a result, individuals may work hard
at appropriate but inadequately co-ordinated tasks. Each discipline
must be clear as to the contribution expected of it. If, for example,
livestock nutrition is the focus of the programme only those disci
plines necessary for resolving the nutritional constraints should be
employed. However, this will inevitably require a multi-disciplinary
team because the economic and social factors are likely to be as
problematic as the agronomic ones.
Case example
As indicated above the subhumid programme has focussed on the allevi
ation of malnutrition in domestic ruminants with forage agronomy at
the hub. The social scientists helped make it possible to grow
forages by determining how pastoralists could obtain the right to use
and fence fallow land belonging to arable farmers. There are a myriad
of other possible examples from all disciplines. For instance, of all
the possible diseases the veterinarian first concentrated on internal
parasites in young stock because it was reasoned that with seden-
tarisation and cattle continuously returning to the same spot the
worm burden in calves was likely to build up. This is likely to be
even more of a problem with the establishment of more or less permanent
fodder banks and, if true, will tend to negate the advantages of
better feeding.
Later it was argued that the establishment of fodder banks
might be more profitable if they were used by more productive animals
- 412 -
such as dairy crosses, but these animals are known to be more prone to
diseases. Thus the veterinarian is now studying the disease patterns
in a small number of experimental crossbred animals attached to Fulani
herds with fodder banks.
Appraising present techniques
This subject is possibly more difficult in livestock research because
there is typically even less contact between research stations and
pastoralists than with their farming counterparts. However, the LSR
scientists should still try and discover all the approaches that have
been attempted and how they fared. Then when considering alternative
ideas generated from within the team they should go back to the local
scientists and extension workers for advice. More often than not
there will be good reason why these ideas have not been tried success
fully before. Ultimately, however, the interventions will have to be
proven in the prevailing ecological and socio-economic environments.
Case example
The subhumid programme took all available advice before commencing
trials with undersown legumes but even then the results were somewhat
embarrassing; the seeds were washed out of the ridges, drowned in the
ridge bottoms, were weeded out in a 'surprise' third weeding and soon.
Ultimately it was found that the time of undersowing relative to the
planting of the main crop is very critical. If this is done too early
the crop is damaged if too late the yield of stylosanthes is also
supressed.
Similarly when crossbreds were first introduced they were
not tame enough for the traditional manual de-ticking and they con
tracted a range of diseases not prevalent in local Fulani cattle but
which are endemic to the area. Appropriate chemo-imunisation and
acracide spraying regimes had to be instituted.
Setting assumptions about near-term conditions
The team must take into account not only the present circumstances
but both recent past and immediate future trends. If it does not
it is likely that it will test interventions that will no longer
be relevant by the time they are proven.
- 413 -
Case example
The subhumid programme is aware of the trend towards increased areas
under cultivation and increased stock numbers. Both these factors
weaken the relative bargaining position of the pastoralists . There
will be more competition between them for less grazing. In these
circumstances it would be unwise to concentrate wholly on forage
production and the team is devoting considerable resources towards
crop-livestock interactions. This involves research into crop residue
production, storage and utilisation as well as growing food crops in
fodder banks by judicious rotation or by transplanting into rows hoe-
cut through the stylo.
Testing improved technology
The testing of improved technology is difficult to write up in the
form of an overview because so many of the details vary according to
the particular techniques employed by the different disciplines. For
instance the veterinarian will use quite different methods from the
agronomist. This section will, therefore, concentrate on discussing
problems of technology testing for improving livestock production.
The normal sequence used in explaining the phases of FSR work is from
researcher managed, researcher executed, through research managed,
farmer executed, to farmer managed, farmer executed trials. In effect
this means that the scientists first conduct their own experiments to
prove and explain scientifically how the technology actually works.
Then they run the experiment with the farmers' participation to see
if farmers have the technological resources to cope with the inno
vation and that it can work in their circumstances. Finally the
researcher takes a back seat and observes whether or not the farmers
actually adopt the innovation. If ultimately the farmers do not adopt
the innovation, or only with drastic alterations, the innovation must
be dropped or returned to the drawing board for further adaptation.
In pratice this is a somewhat simplified scheme of things
because there is constant feedback at all stages, whenever problems
arise the whole or part of the trial can be returned to an earlier
phase or occasionally leap-frogged forward. Generally speaking the
earlier the phase the greater the control and detail and the more
- 414 -
certain the science. The later the phase the greater the influence
of the 'real world' and the greater the assurance of relevance to the
system in question.
This tidy format is by no means automatic. For instance the
unwary scientist can very easily enter a farmer managed, researcher
executed phase when the farmers apply their considerable ethnoscience
in dealing with intruders with the objective of getting all they can
out of the researchers without any real commitment to the researchers'
objectives.
Researcher managed, researcher executed trials
When innovations are too uncertain and risky to try on farmers' fields
or they require examination under strictly controlled conditions they
should be tested in fields and herds wholly controlled by scientists.
These trials are usually carried out at national research institutes
or on sites controlled by the LSR teams in their case study areas.
The research at the national research centres will usually have been
done in the past and not specially for the LSR programme. Great care
must, therefore, be taken in extrapolating the results. The difference
between the circumstances of the original research and of LSR must be
identified and their effects determined. Occasionally they will have
to be repeated in the case study area.
Case example
NAPRI had considerable data on the productivity of Bunaji (White
Fulani) cattle. In view of the long generation inverval it was
extremely valuable to the team to find the data already existent.
However the data had been gathered from government livestock improve
ment and breeding centres and not from traditional pastoralists'
herds. It provided potential production parameters rather than base
line data.
NAPRI also had data on natural forage quality but it related to the
semi-arid border of the subhumid zone. The extrapolations have to be
validated with local data. NAPRI has data on legume cultivation and
Friesian-Bunaji crossbred productivity but both were supported by
mechanised farming which is not replicable in pastoral livestock
situations .
- 415 -
Researcher managed, farmer executed trials
Once the LSR team has designed an intervention and completed success
ful on-station trials it can set an hypothesis that it will be
beneficial for livestock, production in the selected district or
region. The next phase in the LSR cycle is to let the pastoralists
try out the intervention under the guidance of the LSR team: re
searcher managed, farmer executed.
Sample size
This is where LSR starts to diverge most sharply from normal research
station experimentation. It is the point where data collection
becomes extremely problematic because of the mobility of the stock,
their large unit sizes and the owners' emotional involvement with
their animals. Moreover, because of the long generation intervals
it is absolutely essential that records are kept of the same animals
for a number of years. The sample size has, therefore, to be large
enough to cope with losses and drop-outs over a long period.
Sample size is also determined by the need to have enough
farmers involved to eliminate effects of differences between individ
ual farmers. The effect of individual abilities, ambitions and
circumstances is likely to be stronger on livestock production than
on crop production because of the daily need for decisions over the
herd and the competition between herders for favourable grazing. At
the same time LSR can easily become too unwieldy and costly. For
instance a sample of only 20 herds may involve over 1 000 animals.
Sampling is also constrained by the need to work with pastoralists
who are willing to cooperate almost on a first come, first served
basis. There are too few herds within serviceable areas and too much
communication between pastoralists for the team to pick and choose
between them.
The subhumid programme was able to get over these conflict
ing demands in the early researcher managed phase by keeping control
animals in every herd. For instance only half the eligible animals
from any one herd would be included in any trial. The owners were
amenable to this because they preferred to have half the animals
better off rather than none at all. This technique could not,
- 416 -
however, be continued into later trials because once the farmers saw
the benefits, there was no way of preventing them employing the
innovation on the control animals at their own expense.
The question of cost effectiveness is dealt with below.
As for unwieldiness, it is true that all the catalysts for Murphy's
Law (remoteness, poor communications, almost illiterate enumerators
and huge amounts of data) are present in force. If it can go wrong
it surely will. This can, however, be considerably alleviated by
integrating the LSR programme so that, as far as possible, a single
pool of data is used. This optimises the use of available staff
since, for example, a calf is only weighed once, be it for the
breeder, feeder, vet, economist or whoever else needs calf weight
data. It also means that the various disciplines have to bargain
with each other and justify their use of resources in terms of the
team's overall objectives. The sample size is then determined by
the purposes of the survey.
Household economic studies are perhaps the most contentious.
Some economists argue for large random samples in order to achieve
statistical reliability but there appears to be a growing school of
thought, very evident in the criticism of earlier drafts of this
paper, that, in view of all the problem in supervising enumerators
and processing data in field programmes, small select samples may be
preferable. Innovations will only be adopted by herd owners if they
have marked effects on their welfare so it should be sufficient to
just know the major items of income, expenditure and time budgets
of a small representative selection of the various categories of
producers in the target population. Another advantage of the small
sample is that it can be done at the same time as the in-depth
study of household decision making processes that is an essential
part of most LSR studies.
Frequency of data collection
The frequency of data collection is perhaps most contentious in this
phase of the LSR cycle. In the previous researcher managed phase
the fact that the researchers are 'in charge' allows for smaller
sample size and greater reliability and fewer hidden factors. In the
- 417 -
next phase of farmer managed- farmer executed trials the technical
possibilities have already been determined and explained. The LSR
team is by then more concerned in determining to what degree the
farmers can cope with the innovations and what benefits they obtain
from following the recommendations.
In the researcher managed, farmer executed phase scientific
technical understanding and explanation is still required but the
researcher must allow for the farmer's independence. That means the
frequency of recording must take into account such factors as the
farmers natural reluctance to reveal sensitive information such as
sales prices. The farmers also have, deliberately and unavoidably,
very variable powers of memory recall. The more sensitive the topic
or the more aware the farmer is of the official, correct or expected
answer the less accurate his memory. To combat these factors it is
often better to observe the event rather than to ask about it. Some
degree of apparent over-kill in data collection may be necessary.
For instance since sales, purchases, deaths, births, slaughters and
losses can occur on any day it is probably as well to have an enumer
ator visit the herd daily. To avoid irritation at his presence and
because firm routines are essential to the management of staff who
are not expected to fully understand the reason or importance of
diligence, it may be as well for the enumerators to record milk
offtake daily. This gives them defendable purposes to be in the herd
whilst recording all the other activities.
Case example
The subhumid programme adopted all the above principles reasonably
effectively. Local school children were employed as enumerators to
record milk offtake, deaths, sales etc. on a daily basis. However,
difficulties did arise in data processing because the team had no
in-house computer and attempted to run all its data once a year on
the central computer at head office. This meant that the scientists
had the greatest difficulty in ensuring adequate supervision. Nor
did they have any way of doing preliminary or interim analysis to
test the reliability or suitability of the data. When finally
processing the data the remoteness of the computer meant that they
had no access to original records even for simple items like checking
- 418 -
eartag numbers. These problems led ultimately to an inefficient use
of computer and staff time. There is little doubt that systems
research teams ought to have their own micro-computer facilities from
the outset.
Duration of data collection
There can not be any hard fast rule about the duration of data collec
tion except that data collection should always be for a purpose and
once that purpose has been accomplished the data collection should stop .
Case example. The subhumid programme found it necessary to study the
grazing behaviour of pastoral herds as a means of determining the
type and quantity of feed available to the cattle throughout the year.
This was intended as a one year study but the extent of burning and
time spent on burn-regrowth had not been anticipated and was only
appreciated after regrowth had been a significant part of the diet
for a few weeks. The study had, therefore, to be continued for a
second year.
The basic herd productivity recording started at the outset
of the programme yet it is still continuing because after three years
there are too few data on calving intervals and age at first calving.
More time is, therefore, needed for the team to acquire data on such
basic parameters as age and weight at first calving in traditionally
managed herds, let alone what effect improved nutrition may have on
those parameters.
Farmer managed, farmer executed trials
Once farmers have successfully executed the procedures, or implemented
the interventions under LSR guidance, and the indications are that
they are happy with the design and wish to adopt it as part of their
normal husbandry practices, then the trials can move into the final
phase of the LSR cycle. The LSR team members must then stand back and
become passive observers so that they can test whether their brain
child can survive without them. The team needs to know if the inter
vention is adopted at all and at what rate (i.e. what is its
acceptability index) , how closely the farmers adhere to the original
- 419 -
design, what modifications they introduce, how successful the inter
vention is in terms of the goals it was designed to achieve, how
persistent it is and what side effects it has.
This phase of the cycle can only be accomplished in coop
eration with the extension and development agencies. It cannot be
done very convincingly by the team trying to simulate extension
officers. They can not be that uncommitted to their own concept.
It is essential, therefore, that the team develop a sufficiently
close relationship with an agency that will allow them access to the
necessary data both of extension inputs and from the participating
farmers.
The frequency of data collection can be very much less than
in earlier phases because it is no longer necessary to try and explain
what is happening at a micro-level. It is the macro-effect on farm
output and profitability that is important. The duration of the
experiment should be at least five to seven years in trials involving
large ruminants such as cattle but may be less for smaller species
with shorter generation intervals.
Case example
The cordial relationship ILCA has with the Federal Livestock Depart
ment and the Kaduna State Ministry of Animal and Forest Resources
provides ideal facilities for this phase of the LSR cycle. Indeed the
only problem is that some innovations are being adopted before ILCA
would normally be ready to move on from the researcher managed phase.
By working through the livestock service centres ILCA hopes
in 1984 to have, in effect, a network of testing sites spread
throughout the zone in Nigeria. These will provide ideal conditions
because they will be created under differing circumstances and well
away from the present sites in Kaduna State where ILCA's own influence
cannot be eliminated.
Applicability of LSR to national authorities
The advantage of the international teams, which must always include a
number of nationals and have close links with local institutions, is
that there can always be injections of fresh ideas and cross ferti-
- 420 -
lization from innumerable sources. There is no way, however, that the
tiny teams from international agricultural research centres (IARC's)
such as ILCA can do much more than scratch the surface of the problems
of livestock production, though they may help significantly in
developing new research methodologies, and in assisting the establish
ment of national LSR teams and possibly in setting goals and standards
for LSR.
National LSR teams should be established at all the major
research stations where they would help orientate their research
programmes towards existing field problems and speed the transference
of research results to the producers. The teams may be comprised of
scientists wholly engaged in LSR or it may form just part of research
scientists' duties. CIMMYT tried attaching one FSR economist per
research station to lead the FSR work and encourage the participation
of other scientists. It was hoped that this would improve the problem
solving relevance of the work of the stations. However, this was not
effective as it might have been because not all the resident scientists
had been persuaded of the advantages to them of FSR.
Case example
Having actively encouraged the ILCA subhumid programme for a number of
years NAPRI is now setting up its own LSR team that will be able to
capitalise on ILCA's experience and move into different ecological
zones. It will be able to focus more directly on Nigerian problems
and with access to the substantial research capability at NAPRI and
other departments of Ahmadu Bello University it is potentially a very
effective unit.
Cost effectiveness
LSR has the immediate appeal of not requiring the massive investments
in land, buildings, stock or equipment that is necessary for estab
lishing a research station. LSR is also adept at exploiting past on-
station research without further cost. LSR also has the advantage
that, with an appropriate inclusion of rapid 'down-stream' elements,
there is an almost immediate response in productivity. In other words
the benefit stream can be turned on certainly earlier than from on-
- 421 -
station research and even earlier than from most development projects
that tend to concentrate on infra-structural developments in the first
years.
Nevertheless, the outline of the problems connected with
livestock systems given above gives plenty of scope for very expensive
research. If each discipline expects to work with the same support it
would command in specialised research departments the costs of LSR
would be quite unreasonable. However, provided that those problems
that require very detailed research with, for example, expensive
laboratory equipment are referred to the appropriate institutions and
the LSR scientists are prepared to adhere to the goals of the LSR
programme and make the necessary compromises, LSR is not overly ex
pensive. In the ILCA subhumid programme it is estimated that it will
require only 120 herd-owners to adopt the proposed packages in order
to justify US $ 1 m of research. With a recommendation domain of
several hundred thousand livestock owners it should not be difficult
to defend the expenditure on LSR in social cost benefit terms provided
that the innovations really do catch on.
It is of course very rewarding when, as with the fodder banks,
producers in the vicinity of the LSR adopt the innovations for
themselves. It is also encouraging when pastoral organisations such
as the Myetti Allah Cattle Fulani Society follow the LSR and spread
the word. These actions will go a long way to justify the expenditure
on LSR but, especially for an IARC like ILCA, the prime clients for the
research results must be the extension and development agencies in the
host countries across the zone. It is their function to take the
innovations to the producers. If the LSR programmes can improve the
rate of return on the investments in development and extension schemes
there will be no question of the cost effectiveness of LSR.
Apart from the obvious need for thorough technical accounts
of the success and failures of interventions under test, the extension
workers require critical reviews of the circumstances and the back-up
support necessary to the success of the innovation. With this addition
the recommendations will be very much more useful than the jargon-
loaded reports in scientific journals or the limited instructions
presented in typical appraisal reports.
- 422 -
Case example
The LPU Smallholder Dairy Scheme grew out of the ILCA nutrition trials
where the Fulani had demonstrated that, with adequate guidance and
assurance of supplies, they would selectively ration their cows. By
doing so they improved the returns to the scarce resource of purchas
able feedstuff s. ILCA did not just hand over reports on the trials
but worked closely with the LPU staff in the design of a Livestock
Service Centre which could deliver the necessary advice, credit and
material provisions necessary to the success of the scheme. The pilot
Livestock Service Centre and its successors will serve as vehicles for
the dissemination of future proven innovations and research findings
as well. This will hopefully ensure that:
1. the producers get the proper advice;
2. the producers can obtain the necessary inputs as and when
they require them;
3. ILCA will have access to the records and the contact with
the Droducers that it needs to determine the uptake, adoptability and
persistence of its innovations in the final farmer managed, farmer
executed phase of the LSR cycle. This is particularly important in
range livestock work because, as noted above, there is tendency for
innovations to be picked up before they are proven. Thus it is essen
tial that a watchful eye be kept on them so that faults can be
detected quickly and timely corrective action taken.
Extension into development projects
As indicated earlier LSR can help bridge the gap between research
institutions and development projects firstly by establishing that
the innovations are acceptable to the producers and then by carefully
detailing when, where and how best to encourage the uptake of the
innovations. For instance, varieties, planting dates, seed and
fertiliser rates etc. are only part of what an extension officer needs
to know in order to encourage farmers to produce forages. He also
needs to know which category of farmer is most likely to respond,
which arguments are most effective in eliciting the response of the
farmer (i.e. those that are most closely allied to the needs and
- 423 -
interests of the farmer) and what back-up services the farmer will
require. The planners and decision-makers would also like to have
some idea about the likely rate of up-take, the optimum extension
officer to farmer ratios, the availability of inputs and the market
ability of the increased production. Obviously LSR can not provide
such a service to each and every project but a thorough analysis of
these factors in the process of conducting the trials will be of
immense benefit to those carrying out feasibility studies in other
areas. Instead of using blind hunches, project preparation teams
can look for similarities and variances and assess the consequences
of those factors that have been shown to be important to the success
of the innovations.
Naturally the closer the link between the LSR teams and
development project staff the smoother will be the transference of
innovations from research to development. This will apply to all
development projects, not just the ones physically associated with
the research, because of the establishment of an empathy with the
problems of development. As indicated above this can best be promoted
by cooperation at the farmer managed, farmer executed phase of LSR.
References
Dillon, J.L. 1973. The economics of systems research. Agric.
Systems Research Conference, Massey Univ., Massey, New
Zealand.
Harrington, L.H. 1980. Initiating applied farming systems research
in developing countries. AID-USDA Symposium on Farming
Systems Research, Washington, D.C.
ILCA, 1979. Livestock production in the subhumid zone of West Africa.
ILCA Systems Study 2, Addis Ababa.
Johnson, G.L.,Dean, R., Halter, A.N., Marvin, L.H.,Kellog, B.,Manesch,
T.J. and Kioumars, P. 1971. A generalised simulation
approach to agriculture sector analysis, with special refer
ence to Nigeria. East Lansing, Michigan State University.
- 424 -
Rohrbach D.D. 1980. A discussion of issues relevant to the develop
ment and implementation of a farming systems research
program. U.S. Dept. of Agric, Office of International
Coop, and Dev., Washington, D.C.
- 425 -
Conception et essais en matière de recherche sur les systèmes
d'élevage: un exemple agro-pastoral
Résumé
La recherche sur les systèmes pastoraux devrait s'inspirer largement
des résultats et des expériences des autres organisations de recher
che et de développement. La recherche sur les systèmes pastoraux
devrait également faire appel à de telles institutions pour effectuer
des travaux d'appui. Les institutions de recherche peuvent contribuer
à résoudre des problèmes qui requièrent une expérimentation poussée
au niveau de la station. Les organismes de développement sont
essentiels à la recherche sur les systèmes pastoraux en ce sens qu'ils
représentent des moyens de tester dans les conditions de l'exploitation
les interventions et les méthodes proposées.
La recherche sur les systèmes pastoraux doit s'inspirer de politiques
clairement définies qui tiennnent compte des implications sociales
de ses résultats. Par exemple, des techniques susceptibles d'aider
les grands exploitants agricoles au détriment des petits exploitants
agricoles devraient être écartées pour ne pas contrecarrer l'objectif
primordial de la recherche sur les systèmes pastoraux qui est de mener
à bien des travaux de recherche en vue d'un développement qui
n'entraîne pas l'aggravation des inégalités.
Dès que l'objectif des activités de recherche sur les systèmes
pastoraux d'élevage aura été défini, des programmes de recherche sur
les systèmes d'élevage devront alors être clairement définis pour
éviter de disperser les efforts des chercheurs par manque de
coordinatioTi . En examinant les idées nouvelles proposées par un ou
plusieurs membres de l'équipe, les scientifiques participant à la
recherche sur les systèmes pastoraux devraient retourner auprès des
chercheurs locaux et des vulgarisateurs en vue de recueillir leur
opinion. L'équipe doit non seulement prendre en considération la
situation actuelle mais également l'évolution récente et les tendances
futures immédiates, faute de quoi il est vraisemblable qu'elle sera
amenée à tester des innovations qui ne seront plus appropriées au
moment où leur validité sera confirmée.
- 426 -
La chronologie normale utilisée pour décrire les diverses phases de
la recherche sur les systèmes se présente comme suit : a) essai
dirigé et exécuté par les chercheurs; b) essai dirigé par le chercheur
mais exécuté par l'exploitant; c) essai dirigé et exécuté par
l'exploitant. Cela signifie que le scientifique effectue d'abord
ses propres expériences pour prouver et expliquer scientifiquement
le mode de fonctionnement de la technique. Puis il effectue
l'expérience avec la participation de l'exploitant pour voir si celui-
ci dispose des ressources techniques nécessaires pour assimiler
l'innovation et si celle-ci peut s'adapter à la situation. Pour
terminer, le chercheur se retire et observe l'exploitant pour voir si
celui-ci adopte oui ou non l'innovation.
Lorsque les innovations ne sont pas du tout sûres et qu'il s avère
risqué de les essayer sur les champs des exploitants agricoles ou
qu'elles requièrent un examen dans des conditions soumises à un
contrôle strict, elles doivent être testées dans des champs et dans
des troupeaux entièrement contrôlés par les chercheurs. Ces essais
sont en général effectués par les équipes de la RSP dans la zone
d'étude.
Une fois que l'équipe de la RSP a conçu une intervention et effectué
des essais fructueux au niveau de la station, elle peut poser comme
hypothèse qu'elle sera profitable à la production animale dans le
district ou la région sélectionnés. La phase suivante de la RSP
consiste à laisser les éleveurs tester l'intervention sous la
direction de l'équipe de la RSP dans le cadre d'essais dirigés par
les chercheurs et exécutés par l'exploitant agricole.
La fréquence de la collecte de données est peut-être très aléatoire
lors de cette phase du cycle de la RSP. Dans la phase précédente
gérée par le chercheur, le fait que les chercheurs soient responsables
permet d'avoir un échantillonnage plus petit, une fiabilité accrue et
un nombre plus limité de facteurs cachés. Dans la phase suivante
gérée et exécutée par l'exploitant, les possibilités techniques ont
déjà été déterminées et expliquées. L'équipe de la recherche sur les
systèmes pastoraux se préoccupe alors de déterminer le degré auquel
- 427 -
les éleveurs peuvent assimiler les innovations et les avantages qu'ils
peuvent tirer de la mise en oeuvre des recommandations. Dans la
phase gérée par les chercheurs et exécutée par l'exploitant, la
compréhension et l'explication techniques sont encore requises. Mais
le chercheur doit donner à l'exploitant une certaine indépendance.
La fréquence de l'enregistrement des données doit donc tenir compte
de facteurs tels que la réticence naturelle de l'exploitant lorsqu'il
s'agit de révéler des informations aussi névralgiques que le prix de
vente.
Une fois que lès exploitants ont suivi les recommandations comme il
se doit ou mis en oeuvre les interventions sous la direction
d'éléments de la RSP, qu'ils sont satisfaits des innovations et qu'ils
souhaitent Ses adopter comme éléments de leur pratique quotidienne en
matière d'élevage, on peut passer à la phase finale du cycle de la
RSP. Le cycle gestion et exécution des essais par l'exploitant ne
peut être accompli sans la coopération des organismes de vulgarisation
et de développement. Les activités de collecte de données peuvent
être nettement moins fréquentes que lors des phases précédentes
parce qu'il n'est plus nécessaire d'expliquer les paramètres de
manière intensive.
La RSP est intéressante parce qu'elle ne fait pas appel à des
investissements importants pour l'acquisition des terres, des
bâtiments, du bétail ou du matériel nécessaires pour la mise en place
d'une station de recherche. La RSP fait également appel à
l'exploitation de la recherche passée au niveau de la station, ce
qui n'implique pas des coûts supplémentaires. La RSP a également
l'avantage de contribuer à un accroissement immédiat de la producti
vité en cas d'inclusion appropriée d'éléments rapides en aval.
La RSP peut contribuer à combler le fossé entre les institutions de
recherche et les projets de développement, tout d'abord en établissant
que les innovations sont acceptables pour les producteurs puis en
déterminant clairement la manière, le lieu et le moment les plus
appropriés pour encourager l'adoption des innovations.
- 428 -
Des exemples qui s'inspirent des travaux du programme du CIPEA/zone
sub-humide basé à Kaduna dans le nord du Nigeria ont été donnés.
- 429 -

Summary of Discussion Session 10.
Chairman: Dr Samson Chema (Kenya)
Discussion led by Prof. Saka Nuru (Nigeria)
Prof. Saka Nuru questioned the use of the term 'agro-pastoral' by
Dr von Kaufmann as applied to people deriving most of their revenue
from livestock. Dr von Kaufmann said that the term suited his
purposes in the paper because it indicated the crop-livestock
interaction that existed. Prof. Saka Nuru asked how could the ILCA
team continue to record performance (five to seven years had been
suggested), and did this include researcher managed, farmer executed
trials? Dr von Kaufmann said that the various research activities ran
concurrently - there was no need to wait for the study of basic
production parameters to finish before testing interventions.
Dr von Kaufmann agreed with Prof. Saka Nuru when he said that LSR
was not to be regarded as a 'water-tight' concept that had to be
executed as it had been thought of initially. It was a dynamic
process and one corrected one's mistakes as one went along. Prof. Saka
Nuru then suggested that the terms 'researcher postulated/proposed'
and 'farmer initiated' could be used to replace 'researcher managed'
and 'farmer managed'. Dr Kaufmann said that the term 'farmer initiated'
appealed to him, but explained that ILCA was relatively new to the
field of systems research, and had suffered in the past from trying
to develop its own terminology. Only by understanding terminology
use could one influence the development of definitions. Prof. Saka
Nuru said that it was difficult to compare the cost-effectiveness of
LSR and traditional research findings. Dr von Kaufmann agreed and
said that he had only wished to point out that LSR could be very
cost-effective - in fact each form of research could help to increase
the cost-effectiveness of the other.
Dr Bekure stressed that case studies could not take the place of
extensive surveys. Without extensive surveys, case studies remained
unique without wide applicability to the population. Dr von Kaufmann
agreed in principle with this and explained that he had only been
commenting on the situation in his own area of work, where he had
- 431 -
found oral data to be very unreliable - he therefore preferred to
record actual observations.
Dr Okali said that the categorisation of producers in the areas the
meeting was concerned with was not important. Regardless of the
terms used there was a continuum along which producers lay, and it
was not essential to have strictly defined categories.
Dr Diakite, commenting on Dr von Kaufmann's paper, said that the
exchange of information between ILCA's team in Nigeria and ODEM was
worthwhile and could be strengthened.
- 432 -
Résumé des débats de la dixième séance
Président: M. Samson Chema (Kenya)
Débats dirigés par le Prof. Saka Nuru (Nigeria)
Le Prof. Saka Nuru s'est interrogé sur le bien-fondé de l'utilisation
de l'expression "agro-pastoral" par M. von Kauffman en référence à
des populations dont la majeure partie du revenu provient de l'élevage.
M. von Kauffmann a déclaré que le terme était utilisé de manière
appropriée dans le document parce qu'il indiquait l'interaction élevage-
agriculture qui existait. Le Prof. Saka Nuru a demandé comment
l'équipe du CIPEA entendait continuer à enregistrer les performances
(5 à 7 ans ont été suggérés) et si cela incluait les essais dirigés
par les chercheurs et exécutés par l'exploitant. M. von Kauffmann
a déclaré que les diverses activités de recherche s'effectuaient
simultanément et qu'il n'était pas nécessaire d'attendre la fin des
études sur les paramètres de production de base avant de tester les
interventions. M. von Kauffmann a reconnu avec le Prof. Saka Nuru
que la RSP ne devait pas être considérée comme un concept rigide.
Elle ne devait pas nécessairement être exécutée comme elle avait
été initialement conçue. Il s'agissait d'un processus dynamique et
l'on corrigeait ses erreurs au fur et à mesure qu'on les commettait.
Le Prof. Saka Nuru a alors suggéré que l'expression "proposée/avancée
par le chercheur et démarrée par l'exploitant" soit utilisée pour
remplacer les expressions "dirigée par les chercheurs" et "dirigée
par l'exploitant". M. von Kauffmann a déclaré que l'expression
"démarrée par l'exploitant" lui plaisait mais que le CIPEA était,
toutes proportions gardées, un nouveau venu dans le domaine dp la
recherche sur les systèmes et qu'en outre.il avait souffert par le
passé de la tentative de mise au point de sa propre terminologie.
Ce n'est qu'en comprenant l'utilisation de la terminologie qu'on
pourra influer sur la conception de définitions .Le Prof. Saka Nuru
a déclaré qu'il était difficile de comparer 1' efficacité-coût de la
RSP et les découvertes de la recherche traditionnelle. M. von Kauffmann
a exprimé son accord et a déclaré qu'il souhaitait simplement indiquer
que la RSP pouvait avoir un caractère efficacité-coût et qu'en fait,
toute forme de recherche pouvait contribuer à accroître l'efficacité
des autres formes de recherche.
- 433 -
M. Bekuré a souligné que les études de cas ne pouvaient pas remplacer
les enquêtes à grande échelle. Sans les enquêtes à grande échelle, les
études de cas n'étaient pas tellement applicables à la population.
M. von Kauffmann a exprimé son accord de principe avec cette idée et a
fait savoir qu'il n'avait fait que formuler des observations sur la
situation dans sa propre zone d'étude, où il s'est rendu compte que les
données orales n'étaient pas fiables. Il a par conséquent préféré
enregistrer des observations ou bien des faits réellement observés.
Le Dr Okali a déclaré que la catégorisation des producteurs dans les
zones qui préoccupait tant les participants à la réunion n'était pas
importante. Quels que soient les termes utilisés, il y avait une
continuité qui s'imposait aux producteurs et il n'était pas essentiel
d'avoir des catégories strictement définies.
Dans ses réflexions sur le document de M. von Kauffmann, le
Dr Diakité a déclaré que l'échange d'informations entre l'équipe du
CIPEA au Nigéria et l'ODEM revêtait une grande importance et qu'il
pourrait être renforcé.
- 434 -
A proposal for pastoral development
in the Republic of Niger
1 2
P.N. de Leeuw and C. de Haan
Ecologist, Arid Zones (Eastern and Southern Africa) Programme, ILCA,
Kenya
2Deputy Director General, ILCA, Ethiopia
Introduction
For several years there has been a growing tendency to consider a
pastoral production system as one of the least rewarding targets for
development. Although many pastoral development projects have been
launched, they appear to have failed to come to grips with the intri
cate problems posed by this mode of production: the end results of
such efforts have been frustrating and the final returns of input have
been usually low or negative. Several analyses of the reasons behind
the high failure rate of pastoral development projects have been
published (Ferguson, 1976; Horowitch, 1979; Goldschmidt, 1980) and
their conclusions need not be repeated here. However, some positive
thinking to offset the negative balance sheet of ex post pastoral
project appraisal may be worthwhile. Thus, it may be useful to focus
on a defined area, e.g. the pastoral zone of the Niger Republic, for
which a well thought-out development programme has been formulated and
proposed for funding.
USAID-supported research and preliminary development work in
the pastoral zone started in 1979. The first phase of the Niger Range
and Livestock (NRL) Project will be completed in June 1983 and be
followed by a development project termed the Niger Integrated Live
stock Production (NILP) Project. The contract for the design and
implementation of NILP was awarded to Tufts University as the prime
contractor, with ILCA, North Carolina A & T University, New Mexico
State University as sub-contractors.
The design team (also called the joint enterprise group)
assembled in Niamey in early January 1983 and submitted their first
- 435 -
draft in early February. Several design sub-teams were created to
write the various components for the project paper. C. de Haan (ILCA)
and A. E. Sollod (Tufts School of Veterinary Medicine) were respon
sible for the animal production component, including animal nutrition
and health, while P. N. de Leeuw (ILCA) and G. Greenwood were charged
with ecology, range and water resources and their management. G.
Greenwood was a consultant representing New Mexico State University.
This paper presents a summary of the design team's findings
and proposals in the fields of range management and animal production.
However, it should be realized that the development of these technical
components are part of an integrated developmental strategy aimed at
increased food production, income and security for Niger herders. Thus
the project will promote the participation of pastoralists in the
development process by interweaving the activities of herders,
herders' associations and the services provided by the Government.
The primary objective of creating local "institutions" is that devel
opment should serve human needs and must therefore be defined in terms
of human objectives. The goals in the technical sector are inter
mediary objectives, by which the primary objectives can be attained.
At the time of writing, the authors had not yet received the
final project papers on the other components of the design (on
"institution building" and associated socio-economic research, human
health and education and training) . However, for an overview of the
integrated approach that underlies the proposed development strategy,
reference can be made to a draft discussion paper on strategy by
Swift (1982), who is the coordinator of the socio-economic unit of the
NRL phase, and to a summary of this paper (de Leeuw and Swift, 1983).
The authors wish to acknowledge the contributions of their
colleagues A.E. Sollod and G. Greenwood without which this paper could
not have been written and wish to thank J. Swift for providing the
necessary background on the NRL project.
Description of the production systems
2The target area is 80,000 km in size and lies within the pastoral
zone of the Niger Republic. It has a population of some 170,000
pastoralists, the majority of which are Tuareg, the minority tribes
- 436 -
being WoDaaBe and Arabs. The area supports about 330,000 cattle,
930,000 sheep and goats and 100,000 camels (Table 4) . Mean annual
rainfall varies from 200 mm in the north to 400 mm in the south, most
of which falls in the two- t» three-month long rains between July and
September. Inter-annual variations in total amount, distribution and
length of the growing season are great (Table 1) .
Table 1. Annual rainfall (mm) distribution over 60 years (1920 - 1981)
for Agadez (16°40,NJ 8'OO'E) and Tahoua (14°40'N, 5°30'E).
Annual rainfall distribution
(as % of 60 years)
Region 100 mm 100-200 mm 200-300 mm 300-400 mm 400-500 mm 500 mm
Agadez
Tahoua
22 57 22
17 37 34 12
Source : Milligan (1982a) .
Range resources vary with annual rainfall and its distri
bution. End-of-season standing biomass ranges from 200 to 500 kg DM/ha
in the north to 500 - 1 500 kg DM/ha in the south and consists
mainly of shortlived annual grasses and herbs (Table 2) . Woody cover
is variable but on average is low (Milligan, 1982a) . Herbage crude
content rises to 10 - 15% in August, dropping to low values of 3 to
5% towards the end of the dry season. The greatest variation in
herbage quality occurs during the transition period (from late May
to early August, its duration varying between years), when there is
a mixture of scattered new growth and old standing forage and litter.
The same seasonal trends occur for herbage digestibility (Fig. 1).
- 437 -
14
12-
10-
8
—r—
s
T-
0
—I—
N
-i—
D
-1-
M
-T-
MJ ASO DJFMA
Figure la. Range of crude protein content (%, DM) -in annual grass
herbage.
80 -i
70
60-
50
40
t ~T 1 I 1-
A S 0 N
-r-
F M
-r-
M
Figure lb. Range of digestibility (1 DM) in annual grass herbage.
- 438 -
Table 2. Total annual rainfall and maximum standing biomass on four
sites in the pastoral zone of the Niger Republic
Total annual Maximum Ratio
Site Year rainfall
(mm)
standing
(kg/ha)
kg/ha/mm
Aderbissanat / 1980 108 660 6.1
(15o40'N,7050'E)
J 1981 153 500 3.3
( 1982 151 730 4.9
Gadabeji / 1980 - - -
(15°N, 17°10'E) <1981 159 190 1.2
U982 150 1130 7.5
Ibeceten / 1980 192 970 5.1
(15°20'N, 6°88'E) J 1981 210 640
3.0
' 1982 160 590 3.7
Dakoro / 1980 238 1 500 6.3
(14°30IN,6°40IE) 5 1981 236 750 3.2
V 1982 240 1 200 5.0
Source: Wylie et al (1982).
The question arises whether range resources within the area
are adequate to sustain the 500,000 TLU (1 TLU = 250 kg) in particular
during periods of drought. To answer this question, theoretical
carrying capacity calculations have been made based on the assumed
relationship that 1 mm rainfall produces 2.5 kg of dry matter. From
Table 2, it is clear that this relationship is poor and actually
ranges from 1.2 to 7.5 kg DM/mm rain, because of differences in
rainfall distribution and intensity, and soil characteristics.
Depending on rainfall, the estimated carrying capacity ranges from
400,000 to 700,000 TLU (Table 3), which demonstrates that in average
years the existing population can be sustained, but a serious fodder
shortage may develop when in the north the rains are below 100 mm and
in the south below 300 mm, events that occur during one year out of
five (Table 1) .
- 439 -
Table 3. Carrying capacity (CC) estimates of the NBL zone in relation
to rainfall and biomass production.
Northern part (40,000 km ) Southern part (40,000 km )
Rainfall Biomass CC Total Rainfall Biomass CC Total
(mm) (kg/ha) (ha/TLU) (TLU) (mm) (kg/ha) (ha/TLU) (TLU)
100 250 36 111 250 625 15 267
150 375 24 167 350 875 10 400
250 625 15 267 400 1 000 9 444
Source: Penning, Vries and Djiteye (1982).
Livestock density increases from north to south in line with
2 2
range productivity, from about 4 TLU/km to 7.5 TLU/km (Table 5).
During the rainy season there is an influx of cattle and camels into
the northern part of the area, which return to the south from late
September onwards (Table 4). Although the Tuareg have a much more
diverse species mix (Wilson and Wagenaar, 1982), they appear less
mobile than the cattle-owning WoDaaBe. The greater mobility of the
latter is reflected in the much greater seasonal fluctuation in
WoDaabe herds (Table 3). Within this general context of movement,
differing goals and constraints lead to different production strat
egies which involve different breeds and species of animals maintained
in herds, reliance on different types of dry-season water sources,
attachments to particular dry-season geographic zones, and the use of
different salt cure zones.
- 440 -
Table 4. Livestock populations in the NRL zone.
Livestock population (x 1000)
May 1981 Sept 1982 Oct. 1982 Mean TLU (%)
Cattle 289 331 376 332 249 (50)
Bororo (153) (192) (247) (198) (148) (30)
Azawak (136) (139) (130) (133) (101) (20)
Small stock 780 831 1 148 930 140 (28)
Came 1 s 70 156 89 98 99 (20)
Donkeys 13 23 20 18 9 (2)
TOTAL TLU 410 540 554 497 497 (100. 0^
Weighted annual mean.
2
Cattle = 0.75, smallstock 0.15, camel 1.0, donkeys = 0.50, TLU
(250 kg LW) .
Compiled from Milligan 1982a, 1982b.
Table 5. Seasonal livestock densities (no. /km ) in the NRL zone
(1981-82).
Livestock density (no. /km )
cattle smallstock zamels donkeys Total TLU ha/TLU
Northern part
Wet season 1.5 6.5 2.5 0.4 4.8 20
Dry season 1.1 6.7 1.2 0.2 3.2 31
Southern part
Wet season 6.4 10.2 0.6 0.3 7.1 14
Dry season 5.5 13.6 1.3 0.4 7.7 13
Source : Milligan (1982a)
- 441 -
Constraint diagnosis
Range resources
The single major constraint to which all producers within the zone are
subject is that of a unimodal range resource base. This implies that
the productivity of the system hinges on a two- to four-month rainy
season during which nutrient supply is sufficient for maintenance and
production in terms of incremental herd growth (liveweight gain plus
reproduction) . Once the growing season is over, the range resource
may remain adequate in quantity but it deteriorates gradually in
quality over the nine-month long dry season. This situation is in
contrast to many other West African production systems that are capable
of breaking the deteriorating fodder situation through short or long
distance movement to perennial floodplain grasslands (e.g. the inner
delta of the Niger River, the shores of Lake Tchad, the floodplains of
Benin or Nigeria), to crop residue resources or to burned savanna
woodlands. None of these options are available to the WoDaaBe and
Tuareg pastoralists in the ILP zone, at least not for those with
cattle herds, while for small flocks and camels, the options are some
what greater through the exploitation of browse.
Animal nutrition
Protein
It appears that protein deficiency is the most limiting dry-reason
factor in grazing animals (cattle and sheep) . Early in the rainy
season the protein level (nitrogen) of grasses is at its highest (8 -
12% digestible protein) . As the rainy season ends and the dry season
progresses, this level drops steadily until the end of the dry season
when it averages 1% (Louis, 1982). Selective grazing may result in
an actual intake slightly higher than this, but the protein intake
would still be well below maintenance requirements for cattle and
sheep. In cases of severe protein deficiency subcutaneous edema
(bottle jaw) develops due to hypo-proteinemia; this has been observed
in sheep in the project zone (Glazier, pers. comm.).
- 442 -
Microminerals, vitamin A and phosphorus
Research completed late in the NRL project has demonstrated the soil
to be severely deficient in cobalt (Co), copper (Cu) , and selenium
(Se) . Critical points of soil mineral concentrations where deficiency
starts in pastured ruminants are four times that found in the NRL soil
for Co, 22 times that for Cu, and ten times the critical soil concen
tration for Se. Local salts from Bilma and Tigguidan Tessoum do not
contain adequate concentrations of these elements to be useful as
supplemental sources. Striking clinical features of these deficiencies
have been observed in animals in the zone. The deficiencies appear to
be significant constraints to animal production and health, but the
production responses to supplementation must still be tested since the
zone's slow-growing animals may respond only minimally; however, as
general nutrition is improved through other project interventions, the
animals' requirements for microminerals will increase.
Empirical observations of clinical signs and responses to
treatment, both in the project zone and elsewhere in the Sahel, have
demonstrated severe vitamin A deficiencies at the end of the dry
season; therefore, dry season vitamin A supplementation will be one
of the first large-scale interventions in the ILP project.
Forage analysis for phosphorus is presently unreported, but
it is anticipated that some deficiencies do exist.
Water points
Uneven water point distribution may result in underutilisation of some
areas while there is overgrazing in other areas which have an abun
dance of water points. Dry-season water points with very high output,
such as bore holes, may attract more herds than the localised biomass
might be expected to support; however, the degree to which these
negative effects may occur in the project zone has not been adequately
quantified. In an intensive study on a single WoDaaBe herd, Glazier
(1983) found evidence that the time required to draw water from a well
detracted from the potential foraging time for the herd, and it may
have adversely affected the animals' feed intake.
- 443 -
Water quality varies considerably throughout the zone.
Herders appreciate the value of clean, fresh water, and they will go
to great lengths to obtain it. They are quite aware of the human and
animal health problems that are associated with some forms of water
poiuts, such as ponds and some types of cement troughs (Sollod, 1981;
Knight, 1981; Loutan, 1982). Herders are also quite concerned with
the mineral quality of the available water (Maliki, 1982). They
commonly seek water sources which are believed to have a positive
effect on human and animal health, and they avoid other water points
which are known to have undesirable qualities.
Infectious diseases
In the past, the single most important infectious disease hindering
increases in animal production was rinderpest. The JP-15 rinderpest
campaign conducted in the 1960s failed to eradicate this disease but
greatly reduced its incidence, and today it continues to be controlled
by widespread vaccination. However, knowledge of other important
animal diseases prevalent in the region is limited and, except for
contagious bovine pleuropneumonia, no efforts have been undertaken to
systematically control them.
The NRL project conducted two forms of animal health re
search. A veterinary consultant made several trips into the project
zone and through extensive interviews, herd examinations, and
necropsies was able to assess the f imal health status of the herds
and to formulate a tentative list of disease constraints based on
their relative economic importance (Sollod, 1981).
Field research on animal health was also done by some
members of the socio-economic team. Their work emphasised herder
perceptions of diseases, specific causes, approaches to treatment
and control, and their relative economic importance. It was found
that the herders had a fairly sophisticated understanding of diseases
and epidemiological relationships (Maliki, 1982).
Production indices
Sahelian pastoral systems generally utilise their range resources as
efficiently as other production systems in similar arid environments;
however, because of the constraints described above, production is
- 444 -
very slow as reflected in low growth, maturation and reproductive
rates. There are also high young stock mortality rates and low
offtake rates which reflect both the selective use of milk for human
consumption and uncontrolled infectious diseases. Table 6 gives the
main production parameters of the existing systems.
Table 6. Estimates of important production parameters in the zone.
Net animal Mortality
Age at
first
parturition
(years)
reproduction birth to
weaning
weaning to
first par
turition
(%)
adults
(%)
Species rate
(%) (%)
Cattle 50
100
110
25
30
35
20
4
8
7
7
10
5
8
7
4
4
Sheep 1.25
Goats 1.25
Camels 40
40
5
Donkeys 20 4 3
Source- Wilson and Wagenaar (1982).
Project output
Testing and extension of range management strategies
The programme involves three phases:
- the technical and economic evaluation of proposed strat
egies under controlled conditions;
- on-herd trials of strategies shown to be effective under
controlled conditions (contingent upon the removal of watering
constraints and the formation of experimental herding units);
- the extension of proven strategies through the herders'
association (HA) network through trained HA range management auxili
aries chosen on the basis of their traditional range skills.
- 445 -
Deferred dry-season grazing
Deferred dry-season use trials will be conducted at two ranches
during the 1982-83 and 1983-84 dry seasons. Every day and every other
day watering regimes will be investigated, in part to simulate local
conditions. The animal performance will be followed after the end of
the experiments until the end of the following rainy season to investi
gate compensatory gain.
Centripetal grazing
Experimental work on the Ekrafane ranch has shown that centripetal
herd management can reduce weight losses in immature steers (Klein,
1981) . Centripetal grazing is in fact a more elaborate system of
deferred dry-season use, coupled to a variable watering regime. The
advantages of the outside-in system are thought to be 1. maintenance
of higher quality feed through deferred use, 2. reduction of animal
energy expenditures in walking at the moment of greatest stress, and
3. higher intake and digestibility due to more frequent late dry-
season watering. As with deferred dry-season use, this technique
could result in better animal condition and lower calf mortality in
the late dry season.
Under the project it is proposed that :
1. A modelling exercise be performed in order to estimate
the effect of centripetal grazing on cattle.
2. One on-ranch trial of the centripetal system be estab
lished during the dry season 1983-84. The trial will include two
stocking rates plus a control herd with a total of 60 cattle. Daily
and alternate day watering during the hot season will be superimposed
on the stocking rates.
The question of watering frequency and the allocation of
labour to animal herding versus watering is considerably more complex.
According to the only animal behaviour survey conducted by NRL
(Glazier, pers. comm.), such a system may be precluded by a lack of
labour to simultaneously water the animals and herd them away from
forage that is to be conserved for later use. Furthermore, given the
present watering technology, the imposition of daily watering would so
446 -
reduce the time available to the animals for grazing that herd
productivity would fall rapidly.
Thus, the feasibility of applying centripetal grazing in the
project depends on:
- a reasonable degree of control of a water point;
- the allocation of a certain amount of additional labour to
herding the animals to the desired ranges;
- a reduction in the time spent watering the animals;
- the effects of increased water frequency on animal
performance, on the rate of forage disappearance, and on the stocking
and movement strategies of the herder.
Watering technology studies
That the provision of water during the dry season is an important
component in the production system has been recognized early during
the NRL phase. Several studies have been conducted but have not been,
or have been only partially, published. Data from the waterpoint
survey are currently being analysed, while labour inputs on a house
hold basis are incorporated in socio-economic studies, the report on
which is in preparation.
For the quantification of the effects of current watering
technology on herd and labour productivity, the sample used is the
same as for the animal production studies (see below) and is strat
ified to distinguish between Tuareg and WoDaaBe ard between households
with low and high ratios of animals to people.
Particular attention will be paid to the relationship
between the amount of time spent watering, the quantity of water
consumed per head, the time and resources available for grazing, and
herd productivity. With this information, a better prediction can
be made of the effects of changing the watering regimes in the project
zone.
This project consists of a follow-up to the NRL water point
survey (Knight, 1981) that aimed to carry out the following within
selected HA areas:
- 447 -
- the preparation of accurate water point maps;
- the delineation of water deficit and surplus zones through
measuring water point outputs and requirements by season;- and
- labour inputs and efficiency for water extraction.
During the survey, the constraints and limitations of
existing technology will become evident, while more efficient inno
vations and techniques will be identified. Potential innovations
include:
- well mouth reinforcement (already tested by NRL) ;
- better harnesses for draft animals;
- installation of concrete screens in the aquifer to prevent
slumping of aquifer material, to increase the volume of water present
at any moment and to facilitate recharge;
- development of well covers to permit temporary closure of
wells;
- methods of directing water into traditional watering basins
without detaching the tire d'eau; and
- development of water-trapping structures in sandy areas.
Animal production and health
Introduction
Objectives. The objectives of the animal production and health
component of the ILP project will be:
1. to identify economically feasible and socially acceptable
production increases in the cattle, sheep, camel and goat production
systems of the pastoral zone through:
- improvements in per animal productivity by increasing
milk production and by reducing mortality with strategic feeding
practices;
- improvements in production efficiency with herd health
programmes .
- 448 -
2. to develop, with the assistance of the government of
Niger, animal production and health input delivery systems to the
herders' associations and Veterinary centres through:
- the training and continuing education of one veterinary
auxiliary for each herders' association;
- support to the annual livestock service vaccination
campaigns;
- the construction of veterinary posts at pastoral centres
and the provision of animal restraint facilities in the zone; and
- introduction of a process approach to field research in
the livestock service.
It is expected that the ILP project will have its greatest
impact on animal production by introducing strategic inputs of protein
and animal health commodities.
Institutional framework
Animal production and health activities will be undertaken within the
institutional framework of the HA and the government livestock
service with the support of private markets and the national veteri
nary pharmacy. In an effort to extend animal health care to a grass
roots level, training of veterinary auxiliaries was initiated by the
livestock service and the NRL project in 1981.
The animal health study of NRL (Sollod, 1981) pointed out
that herders were often not using the optimal combination of commodi
ties in animal care. Under ILP it is anticipated that a more cost-
effective population approach to animal health may be carried out by
utilising a better selection of animal care commodities and by working
through veterinary auxiliaries who are members of the HA. Animal
production inputs, mainly protein supplements and nutritional
counselling will also be introduced through the system of auxiliaries.
The ILP project will begin to develop a process approach, or
an "evolutionary operational programme", to field research whereby the
results of research are rapidly tested through field trials and
socio-economic evaluation, and they are then extended and monitored
under the supervision of the researchers themselves. It will be a
- 449 -
major objective of the ILP project to introduce and begin to
institutionalise the evolutionary operational programme in the
livestock service.
Output
The project will produce three major outputs, all of which will focus
on improving per animal production in the production systems of the
zone. Emphasis will be placed on solving the most severe, and the
most easily addressed of the constraints described in the previous
section, namely, seasonal protein deficiency, vitamin/mineral de
ficiencies and infectious diseases. Improvements will also be made
in environmental hygiene (such as water point decontamination) and
animal reproduction (through early pregnancy diagnosis and the
identification and treatment of causes of infertility in cattle and
goats .
The first output, the evolutionary operational programme, is
a management system in which new technology is gradually introduced
through a logical sequence of research and implementation activities.
New inputs are carefully evaluated for their technical, economic and
social impact before being extended on a broader scale.
The evolutionary operational programme has five stages which
will carry the project activities through a logical sequence. These
stages include field research, field trials, socio-economic evaluation
(under partial implementation), full implementation and monitoring. .
Field research. Interdisciplinary research will be carried out to
provide base-line data on herd productivity, animal nutrition, habitat
utilisation, epidemiology and reproduction. The frequency of data
collection, which can be accomplished by single visits to each herd,
will be limited to four times per year in order to reduce costs and
to minimize disruption to herders and their livestock. Visits will
coincide with the key seasons which modulate pastoral production,
that is early wet, full wet, early dry and late dry.
Herd productivity. Except for a retrospective survey (Wilson and
Wagenaar, 1982) no animal production data are available from the
traditional systems of the project zone. The project will therefore
- 450 -
implement a data collection system for the two production systems:
WoDaaBe and Tuareg. The following parameters will be collected:
1. herd/flock composition, covering the main categories by
species, age, and sex. Through interviews, information will be col
lected on animal births, deaths, sales, gifts and transfers;
2. milk offtake per individual animal;
3. calf growth (long bone growth and body weight) up to
approximately 50 kg liveweight and small ruminants up to approxi
mately one year of age;
4. condition of all cattle in the herd, through the
establishment of a simple system of condition scoring.
Animal nutrition. The methodology in determining feed quality will be
to select from one herd in each of the sub-samples two to four matched
sentinel animals and observe at regular intervals the composition of
vegetation being eaten. Samples will be collected for food analysis.
This method of grab sampling or simulated grazing has given good
representative results of the quality of ingested feed (Dicko et al,
1981).
Habitat utilisation. Habitat utilisation by sample households and
their herds will be monitored in a seasonal time-frame through
periodic recall enquiries on herd movement, grazing and watering
strategies and labour force. Behaviour recording will be done on the
same sub-sample as mentioned in animal nutrition to determine the 24-
hour activity profiles and grazing conditions of the diurnal cycle.
These habitat utilisation observations and measurements will be made
within the same sample framework as the productivity and nutrition
studies described above-
Epidemiological studies. Rigorous epidemiological data collection and
analysis system will be integrated with the studies described above
and will therefore have an ecological perspective which takes full
cognisance of environmental and socio-economic factors as determinants
of disease. Animal health monitoring will be done on:
- 451 -
1. the herd productivity groups;
2. geographically selected samples throughout the project
zone - retrospective studies will combine animal production data and
case-control methodology;
3. site-specific investigative field epidemiology where
problems occur, including water point and environmental contamination.
Livestock reproduction. Research will be directed to identify and
determine the relative significance of factors which cause such high
infertility and reproductive failure. Particular emphasis will be on
the non-nutritional causes of lowered reproductivity. By utilizing
retrospective herd studies and prospective cohort studies, a detailed
herd reproductive history will be obtained in an on-going manner with
the present and continuing herd reproductive status being assessed.
This data base will be used to determine reproductive indicators.
Determination of the infectious causes of infertility and reproductive
failure will also be done. This will be accomplished largely by the
epidemiological disease determination surveys.
The application of the techniques of early pregnancy
diagnosis will be examined in cattle, camels, sheep and goats. Inter
nal palpation and external ballotment will be performed on the various
species. This knowledge may be used to assist herders to make more
informed decisions concerning the sale and/or selective feed sup
plementation of their animals.
Field trials. These trials will be carried out in simple with/without
comparisons of matched paires of animals of similar characteristics
in order to reduce numbers and costs. They will be conducted in the
same production units as the field research, with the control animals
providing the data for the parameters of the existing system. Field
trials on individual vitamin or mineral inputs and certain combinations
will be carried out on 10 to 20 pairs of animals for each input.
Two classes of sheep and cattle (newborn animals and reproductive
females) will be tested for two years.
The ILP project will gradually test the technical and
economic feasibility, and the social acceptability, of limited
- 452 -
concentrated feeding (1 kg of cottonseed or 0.5 kg of groundnut cake)
to one or two lactating cows from each of 10 herds for two months in
the late dry/early wet season. Their performance will be compared
to 10 to 20 control animals of similar characteristics through bi
weekly milk recording (offtake) and measurements of calf growth;
the technical and economic coefficients will be estimated.
The ILP project will also establish the feasibility of calf
supplementation using locally produced groundnut cake/maizemeal and
a vitamin/mineral mixture. On the basis of differences in calf
mortality and daily gain, a cost-benefit ratio will be calculated in
order to make a decision on possible extension.
Socio-economic evaluation
Results from the field trials will be evaluated to determine whether
full implementation is warranted. Evaluations will be made of costs
and benefits, feasibility, sustainability, social acceptability and
environmental impact. Pilot interventions will be carried out at the
level of the total production unit and will comprise two categories:
1. Those interventions which need to be tested on a herding
unit (e.g. grazing management, protection from flies in the late rainy
season, vaccinations against contagious diseases) , and are not yet
sufficiently tested to establish economic viability.
2. Those interventions which have already proven their
economic viability, but for which it is necessary to establish their
social acceptability. In such cases, consumer demand for products -
those paid for by the herders themselves - will be the basis of
evaluation.
References :
Dicko, M.S., Lambourne, J.B., de Leeuw, P.N. and de Haan, C. 1981.
Voluntary intake and livestock productivity in the Sahel
zone of Mali. Proc. 32nd meeting EAAP., August 1981,
Zagreb.
Ferguson, D.S. 1976. A conceptual framework for the evaluation of
livestock production development projects and programs in
sub-Saharan Africa. CRED, Univ. of Michigan, Ann Harbor.
- 453 -
Glazier, D. 1983. Herding Dynamics. USAID/NRL, Niamey.
Goldschmidt, W. 1980. The failure of pastoral economic development
programs in Africa. In: Galaty, G.J., D. Aronson and P.C.
Salzman. The future of pastoral people. IDRC 17e, Ottawa.
Horowitz, M.N. 1979. The sociology of pastoralism and African
livestock projects. AID, Program Evaluation Discussion
Paper 6, Washington.
Klein, H.D. 1981. Contribution a l'estimation de la production sur
paturage sahelien au Niger. Rev. Elev. Med. Vet Pays Trop.
34 : 211-20.
Knight, J. NRL water point survey. USAID, Niamey.
Louis, S. 1982. Rapport technique sur la production animale dans la
zone pastorale du Niger-proposition d'amelioration. USAID/
NRL, Niamey.
Loutan, L. 1982. Health and nutrition in a group of WoDaaBe (Bororo)
herders in central Niger. USAID /NRL, Niamey.
De Leeuw, P.N. and Swift, J. 1983. Opportunites de developpement
pastoral. Economic Dev. Inst. / ILCA (In press).
Maliki, A.B. 1982. NGAYNAAKA: Herding according to the WoDaaBe.
USAID/NRL, Niamey.
Milligan, K. 1982a. Aerial survey of human, livestock and environ
mental conditions in a central region of the pastoral zone
of Niger. Final Report, USAID, Niamey.
Milligan, K. 1982b. Wet-season aerial survey of the human and live
stock populations and environmental conditions in a central
region of the pastoral zone in Niger. Final report, USAID,
Niamey.
Sollod, A.E. 1981. Patterns of disease in syvopastoral herds of
central Niger. USAID/NRL, Niamey.
Swift, J. 1982. NRL final report, first discussion draft. NRL rep.,
USAID, Niamey.
Wilson, R.T. and Wagenaar 1982. An introduction survey of livestock
population demography and reproductive performance in the
454 -
area of the Niger Range and Livestock Project. USAID/NRL,
Niamey.
Wylie, B., Senock, R. , Snyder, L., Roettgen, R.and Porter, B. 1982.
Range research and results. USAID/NRL, Niamey.
- 455 -
Proposition de développement pastoral en République du Niger
Résumé
La région-cible qui s'étend dans la zone pastorale de la République
2
du Niger a une superficie de 80 000 km . Sa population est de
quelque 170 000 éleveurs composés pour la majorité de Touaregs et
d'autres ethnies minoritaires, à savoir les Wodaabes et les Arabes.
On trouve dans la zone environ 330 000 bovins, 930 000 ovins et caprins
et 100 000 camélidés. La pluviométrie annuelle varie de 200 mm dans
le nord a 400 mm dans le sud et la plupart des précipitations se
produisent lors des deux a trois mois de saison des pluies entre
juillet et septembre.
La contrainte essentielle à laquelle les producteurs de la zone ont à
faire face a trait au caractère unimodal de la base des ressources
pastorales. Cela implique que la productivité du système est tribu
taire des deux à quatre mois de saisons des pluies au cours desquelles
la production d'éléments nutritifs est suffisante pour l'entretien de
la production en termes de croissance du troupeau. Il apparaît que le
manque de protéines soit le facteur limitant de saison sèche le plus
important pour les ruminants. La recherche a démontré que le sol
est très pauvre en cobalt, en cuivre et en sélénium. La distribution
inadéquate des points d'eau peut se traduire par une sous-utilisation
de certaines zones et par le surpâturage d'autres zones où l'eau est
abondante. La qualité de l'eau varie considérablement à l'intérieur
de la zone. Dans le passé, la maladie infectieuse qui constituait
l'obstacle le plus sérieux à la production animale était la peste
bovine. Toutefois, on ne dispose que de connaissances limitées sur
d'autres maladies animales graves dans la région et peu d'efforts ont
été déployés pour lutter systématiquement contre celles-ci. A cause
de ces contraintes, la production est insuffisante. Cette situation
se reflète dans la faiblesse des taux de croissance et de reproduction
ou dans le caractère tardif de la maturité.
La phase d'expérimentation et de vulgarisation du projet proposé fait
appel à l'évaluation technique et économique de stratégies proposées
dans des situations contrôlées, à des essais au niveau du troupeau, à des
- 456 -
stratégies dont l'efficacité a été prouvée dans des situations
contrôlées, et à la vulgarisation de stratégies confirmées au
niveau de l'association des éleveurs.
Au cours de l'exécution du projet proposé, les contraintes et les
limites relatives aux techniques disponibles de mise en valeur des
ressources en eau deviendront plus évidentes, alors que des innovations
plus efficaces seront identifiées.
Les innovations potentielles portent notamment sur :
- le renforcement de la margelle des puits;
- l'amélioration des harnais des animaux de trait;
- l'installation de panneaux de béton dans la couche aquifère
pour éviter l'effondrement de portions de celle-ci en vue
d'accroître le volume d'eau disponible à n'importe quel moment
et de faciliter la reconstitution de la nappe;
- la construction de couvercles de puits pour permettre la
fermeture temporaire des puits;
- la mise au point de méthodes de canalisation de l'eau en
direction de bassins d' abreuvement traditionnels sans réduire
le tirant d'eau;
- la mise au point de procédés de captage des eaux dans les zones
sableuses .
Les objectifs de la composante production et santé animales sont les
suivants :
- identification d'accroissements économiquement réalisables et
socialement acceptables de la production de bovins, d'ovins, de
camélidés et de caprins dans les systèmes de la zone pastorale,
par le biais de l'amélioration de la productivité par animal et
de l'efficacité de la production grâce à des programmes
vétérinaires destinés au troupeau;
- développement avec l'aide du Gouvernement du Niger des systèmes
de production et de santé animales, des associations d'éleveurs
et de centres vétérinaires notamment par:
- 457
la formation suivie d'un auxiliaire vétérinaire pour chaque
association d'éleveurs;
l'appui aux campagnes de vaccination annuelle du service de
l'élevage;
la construction de postes vétérinaires dans les zones pastorales
et l'installation de matériel de contention dans la zone;
l'introduction d'une approche à la recherche de terrain dans
le service de l'élevage.
- 458 -
Summary of Discussion Session 11.
Chairman: Dr Samson Chema (Kenya)
Discussion led by Dr Rhissa (Niger)
In response to a question by Dr Wilson on the importance of Vitamin A,
Dr de Haan said that Vit. A had been identified by the pastoralists
as a major constraint to night grazing in the late dry season. This
was subsequently confirmed by a consultant veterinarian who identified
between 10 and 20% of the herd in question with night-blindness. The
main cause was the very low proportion of browse trees in the region,
especially following drought. When asked by Dr Wilson why donkeys
had not been included in the studies described, Dr de Haan said that
donkeys were not an obvious constraint in the system, and therefore
did not command a high priority at the stage. Dr Akilu commented
that there were also problems of Vitamin A deficiency which had been
diagnosed and described by herdsmen and vets. As regards camels,
discussions were taking place with IEMVT to identify constraints,
the most important being parasites (helminthesises) and Salmonella
diseases.
Dr Suleiman observed that the discussion had concentrated on animal
management - why were not range monitoring studies, such as biomass
production, composition changes and vigour, not incorporated into the
plan for trend analysis described by Dr de Leeuw? Dr de Leeuw
explained that such parameters would indeed be monitored in Niger.
Ground truth would be collected on species composition, biomass growth
during the rainy season and during the dry season in specific areas
of the zone. These areas would be repeatedly visited during each
ground survey. This was part of an integrated space (satellite),
aircraft and ground inventory of grazing resources in time and space
for the zone .
Dr Diakite stressed the importance of quantity and quality of water,
and felt it was important to continue with the chemical and bacterio
logical analysis of water. A problem was that of tapping deep water
resources without spending too much money. Yet no development was
possible in these areas of Niger without water.
- 459
Résumé des débats de la onzième séance
Président: Dr Samson Chema (Kenya)
Débats dirigés par le Dr Rhissa (Niger)
En réponse à la question de M. Wilson sur l'importance de la vitamine
A, M. de Haan a déclaré que la vitamine A avait été identifiée par les
éleveurs comme la contrainte essentielle au pâturage de nuit à la fin
de la saison sèche. Cela avait été par la suite confirmée par un
vétérinaire consultant qui a identifié entre 10 et 20% des animaux
du troupeau en question frappés de cécité nocturne. La cause
essentielle en était la très faible proportion de ligneux dans la
région en particulier après la sécheresse. Lorsque M. Wilson a
demandé pourquoi les ânes n'avaient pas été inclus dans les études
décrites, M. de Haan a déclaré que les ânes ne constituaient pas une
contrainte évidente dans le système et que par conséquent on ne leur
avait pas accordé un rang élevé de priorité à ce niveau.
M. Aklilu a déclaré qu'il y avait également d'autres problèmes de
carence de vitamine A qui avaient été démontrés et décrits par les
éleveurs et par les vétérinaires. En ce qui concerne les chameaux,
des discussions étaient en cours avec l'iEMVT pour identifier les
contraintes, les plus importantes de celles-ci étant les maladies
parasitaires (l'helminthiase et la salmonellose).
M. Suleiman a fait remarquer que les débats avaient été axés sur la
gestion des animaux et s'est posé la question de savoir pourquoi les
études de suivi continu sur des thèmes tels que la production, la
composition, l'évolution et la vigueur de la biomasse des terrains
de parcours n'avaient pas été incorporées dans le plan utilisé pour
l'analyse de l'évolution décrite par M. de Leeuw. M. de Leeuw a
expliqué que de tels paramètres seraient étudiés au Niger. Des
données seraient recueillies au sol sur la composition des espèces et
la croissance de la biomasse pendant les saisons pluvieuse et sèche
dans des zones spécifiques de la région. Ces zones feraient l'objet
de plusieurs visites pendant chaque enquête au sol. Ces activités
s'inscrivaient dans le cadre d'un recensement aérien (par satellite
et par avion) et terrestre des pâturages de la zone échelonné dans le
temps et dans l'espace.
- 460 -
Le Dr Diakité a souligné l'importance de l'eau aux plans qualitatif
et quantitatif et a déclaré qu'il était important de poursuivre son
analyse chimique et bactériologique. L'un des problèmes à résoudre
était celui de l'exploitation des eaux souterraines sans dépenser
trop d'argent. Et pourtant, aucun type de développement n'était
possible dans ces régions du Niger sans l'eau.
- 461 -

STRENGTHS AND WEAKNESSES OF PSR

Summary of Discussion Session 12.
The strengths and weaknesses of pastoral systems research with
particular reference to ILCA's programmes.
Chairman: Mr Stephen Sandford (ILCA)
Discussion Panel: Mr Larry Ngutter (Kenya)
Dr Noumou Diakite (Mali)
Dr Jackson Kategile (IDRC, Kenya)
Dr Zakari Rhissa (Niger)
Prof. Saka Nuru (Nigeria)
Dr Assefa W. Giorgis (Ethiopia)
Mr Ngutter felt that the following should be ILCA's priorities in the
future:
1 . ILCA should continue its present efforts and programmes in
- micro-economic household level data collection;
- marketing studies;
- resource inventory, including livestock census;
- insistence on cost-effectiveness of proposed intervention
packages;
- aiming to influence traditional research from research
stations to the field in pastoral systems research, and
produce usable results at the earliest possible time.
2. ILCA should modify its present approaches to
- dovetail with national policies, objectives and strategies;
- integrate its work with national and international research,
policy, extension, training and other institutions;
- expand the scope and area coverage of its on-going programmes.
3. ILCA should explore new horizons in the areas of
- public policy formulation/analysis, particularly as policy
is affected by or affects research efforts;
- land issues in PSR, specifically
- 463 -
- impact of land individualisation on pastoralism;
- impact of cultivation in arid areas on pastoralism;
- ways and means of preserving land resources;
- drought prediction.
Dr Diakite referred to ILCA's work in Mali and said that ILCA's
co-operation there in the future could focus on :
- support for training;
- support for the initiation of extension methodology or pre-
extension work ;
- financial support for costs of offices, garages, accommodation,
etc.
ILCA could have an important role in the negotiations for obtaining
finance. ILCA's image was in general good. But there was a Fulani
proverb which said that 'the kid licks whoever licks the kid'.
ILCA's good image required that those responsible for ILCA keep
'licking' it in order to maintain it.
Dr Kategile said that in view of the fact that ILCA was not in a
position to cover all African countries directly in its PSR activities
due to financial constraints and that PSR results were applicable
within only one socio-economic situation, ILCA should adopt a
different strategy. He felt that ILCA could play a leading role in
PSR in Africa. This could be done by co-operating with other research
institutions in Africa. ILCA's roles could be:
1. popularisation of PSR among research institutions in Africa;
2. development of research methodology in PSR. Methodologies should
preferably be applicable to African institutions. Methods should
continue to be reviewed;
3. if methods were agreed upon, data collection and analysis could be
centralised where necessary;
4. ILCA should form cohesive PSR teams that can set an example;
5. the training of African research scientists in PSR;
6. to visit and advise PSR teams in Africa;
- 464 -
7. to collect and disseminate PSR activities in Africa - to foster
exchange of information;
8. to act as a source of germplasm, specifically for pastures;
9. to co-operate directly with PSR teams in African institutions.
Dr Kategile also questioned whether or not ILCA should get involved
into agroforestry and intensive livestock systems, rather than concen
trate solely on pastoral systems.
Dr Rhissa said that it was important that any collaboration between
ILCA and other organisations should pass through the correct
administrative channels. Once collaboration had been set up with
national institutions this should be open and sincere. Agreement must
first be reached on the objectives of the work and there should be
joint effective exploration of the existing potential in terms of
documentation, technical methods etc. Subsequently ILCA should
regularly inform the national institutions of the results of its
activities. ILCA should encourage the training of technical staff by
organising seminars on pertinent subjects. ILCA should also hold
meetings at the request of national governments.
Prof. Saka Nuru said that co-operation with national research and
development agencies was a top priority for ILCA. Such co-operation
involving universities, state, local and federal government etc. had
several advantages:
- pooling of technical expertise towards a common goal;
- the determination of national priorities relevant to national
needs ;
- use of local knowledge to shorten the descriptive/diagnostic
phase;
- financial benefits from national governments; and
- easier dissemination of results.
He advocated the siting of ILCA team offices in national research
institutions. Prof. Saka Nuru said he thought that ILCA's techniques
- 465 -
were in general good, but stressed that local pastoralists should be
actively involved in the planning and execution of the research work.
He felt that the failure of many national and international projects
had been due to the lack of such a grass-roots involvement. He thought
that ILCA could have achieved more if its research had been extended
beyond the 'case study area' concept and had involved local development
agencies to a greater extent.
- 466 -
Résumé des débats de la douzième seance
Forces et faiblesses de la recherche sur les systèmes pastoraux
à la lumière des programmes du CIPEA.
Président: M. Stephen Sandford (CIPEA)
Participants: M. Larry Ngutter (Kenya)
Dr Nounou Diakité (Mali)
M. Jackson Kategile (CRDI, Kenya)
Dr Zakari Rhis6a(Niger)
Prof. Saka Nuru (Nigeria)
M. Assefa W. Giorgis (Ethiopie)
M. Ngutter a déclaré qu'il estimait que les futures priorités du
CIPEA devraient se présenter comme suit :
1. Le CIPEA devrait poursuivre ses efforts et programmes actuels dans:
- la collecte des données micro-économiques au niveau du ménage;
- les études de marché;
- l'inventaire des ressources, y compris le recensement du
bétail;
- la proposition de programmes d'intervention économiquement
efficaces et susceptibles d'inciter la recherche traditionnelle
sur les systèmes pastoraux à quitter la station de recherche
pour aller sur le terrain ainsi que la production de résultats
utilisables, le plus tôt possible.
2. Le CIPEA devrait modifier sa philosophie actuelle pour :
- harmoniser ses politiques, objectifs et stratégies avec ceux
des pays africains;
- intégrer ses travaux à ceux des institutions nationales et
internationales de recherche, de planification, de vulgarisa
tion, de formation et autres;
- élargir ses programmes en cours et ses zones d'étude.
3. Le CIPEA devrait explorer de nouvelles perspectives dans les
domaines suivants:
- 467 -
- formulation/analyse de politiques gouvernementales,
notamment lorsque les activités de recherche influent sur
la politique ou vice-versa; v
- questions foncières dans la RSP et notamment:
- l'impact de l'individualisation de la propriété foncière
sur le pastoralisme;
- l'impact des cultures des zones arides sur le pasto
ralisme;
- les voies et moyens à mettre en oeuvre pour conserver
les ressources foncières;
- la prévision des sécheresses.
Parlant des travaux du CIPEA au Mali, le Dr Diakité a déclaré que la
coopération du CIPEA avec son pays devrait à l'avenir être axée sur:
- l'assistance en matière de formation;
- l'assistance en matière de mise au point de méthodologies
de vulgarisation ou de pré-vulgarisation;
- l'assistance financière par la prise en charge des coûts de
bureaux, de garages, de logements etc... Le CIPEA pourrait
jouer un rôle important dans les négociations en vue de
l'obtention de financements. L'image du CIPEA était en
général bonne. Mais il y a un proverbe peul qui dit que
le "petit lèche celui qui le lèche". Pour les besoins de
la cause, il fallait que les responsables du CIPEA continuent
à lécher le Centre.
M. Kategile a déclaré qu'étant donné que le CIPEA n'était pas en
mesure de couvrir directement tous les pays africains dans ses
activités de RSP, notamment en raison des restrictions budgétaires,
et que les résultats de la RSP n'étaient applicables que dans un
contexte socio-économique seulement, le CIPEA devrait adopter une
stratégie différente. Il estimait que le CIPEA pourrait jouer un
rôle de premier plan dans la RSP en Afrique. Cela pouvait se faire
grâce à la coopération avec d'autres institutions de recherche en
Afrique. Le CIPEA pourrait avoir pour rôle :
- 468 -
1 . La vulgarisation de la RSP au niveau des institutions de recherche
de l'Afrique.
2. La mise au point de méthodologies de recherche dans le domaine de
la RSP. Les méthodologies devraient de préférence être utilisables
par les institutions africaines, et les méthodes devraient faire
l'objet d'une révision permanente.
3. La centralisation de la collecte et de l'analyse des données là
où cela s'avérera nécessaire, si des méthodes sont conjointement
adoptées.
A. La formation par le CIPEA d'équipes homogènes de RSP susceptibles
de servir de modèles.
5. La formation de chercheurs africains spécialisés dans la RSP.
6. La fourniture de services consultatifs aux équipes de RSP en
Afrique, dans le cadre de visites périodiques.
7. La collecte et la diffusion de données sur la RSP en Afrique et
l'intensification des échanges d'informations.
8. La fourniture de matériel génétique, notamment pour les pâturages.
9. La coopération directe avec les équipes de RSP dans les institutions
africaines. M. Kategile s'est également demandé si oui ou non le
CIPEA devait toucher à 1' agro-foresterie et au système d'élevage
intensif au lieu de se limiter aux systèmes pastoraux.
Le Dr Rhissa a déclaré qu'il était important que toute collaboration
entre le CIPEA et d'autres organisations passe par les circuits
administratifs appropriés. Dès la mise en place des mécanismes de
coopération avec les institutions nationales, la collaboration devrait
être franche et sincère. Un accord doit d'abord être conclu sur les
objectifs des travaux et il devrait y avoir une exploration commune
du potentiel existant en termes de documentation, de méthodes,
- 469 -
techniques, etc. Par la suite, le CIPEA devra régulièrement informer
les institutions nationales des résultats de ses activités. Le CIPEA
devra encourager la formation du personnel technique en organisant des
séminaires sur des sujets pertinents. Le CIPEA devrait également tenir
des réunions à la demande des gouvernements nationaux.
Le Prof. Saka Nuru a déclaré que la coopération avec les organismes
nationaux de recherche et de développement était une priorité
essentielle pour le CIPEA. Une telle coopération avec les universités,
les administrations nationales, municipales et fédérales comportait
plusieurs avantages :
- la mise en commun des connaissances techniques vers la
réalisation d'un objectif commun;
- la détermination de priorités nationales conformes aux
besoins nationaux;
- l'utilisation des connaissances locales pour raccourcir la
phase diagnostic/description;
- des gains financiers pour les gouvernements; et
- une diffusion plus facile des résultats.
Il a préconisé l'installation des bureaux des équipes du CIPEA au sein
des institutions nationales de recherche. Le Prof. Saka Nuru a
déclaré qu'il pensait que les techniques du CIPEA étaient généralement
bonnes. Mais il a souligné que les éleveurs locaux devraient
participer plus activement a la planification et à l'exécution des
travaux de recherche. Il a déclaré que l'échec de plusieurs projets
nationaux et internationaux était dû à la non-participation de ces
éleveurs à ces entreprises. Il estimait que le CIPEA aurait pu
avoir des résultats plus probants si sa recherche avait été poursuivie
au-delà de sa zone d'étude et s'il avait collaboré de manière plus
suivie avec les organismes locaux de développement.
- 470 -
FINAL DISCUSSION ON PSR

Summary of Discussion Session 13.
Pastoral Systems Research
Chairman: Mr Stephen Sandford (ILCA)
Dr Chema suggested that in order to sell the systems approach further,
ILCA might make a systems film, describing the kind of personnel
involved in such work and what methodologies they follow. Sets of
slides might also be helpful in supplementing specific areas of
research. He also cautioned that ILCA's Programme Committee would
need to consider national commitments to joint activities when consi
dering changes in ILCA's in-country research programmes.
Dr de Haan pointed out to the meeting that there was no pastoral
systems research methodology available when ILCA started its operations.
He felt that ILCA was getting near to developing that capability now.
He cited the examples of the joint projects under negotiation in
Mali and in the Cameroons, both of which would be housed with national
programmes.
Dr El Kaluba stated that in any livestock production system there were
two factors, namely genetic material and the environment, that
influenced the productivity and efficiency of the system. He felt that
most of the presentations in the workshop had dealt extensively with
environmental effects, but had paid little attention to genetic
material. He felt there had been a tendency by most development
agencies to overlook the importance of local livestock breeds in
favour of exotic breeds, often very unsuited to African conditions.
ILCA should consider the importance of the local breeds in the areas
in which it is working.
Dr Zulberti noted that the comments of participants, particularly
those of national representatives, indicated the broad acceptance of
the concepts of pastoral systems research. As Prof. Saka Nuru had put
it, this represented a major shift from the traditional approach to
research. However, Dr Zulberti felt that some of the workshop
presentations did not follow the lines of PSR too closely. But he
said that a workshop such as this one could not have taken place 10
years ago. The systems research concept was not a crazy idea any more.
- 471 -
He proposed that the philosophy of systems research imposed the
following conditions on those pursuing such an approach:
- a commitment to development;
- a commitment to recognise one's mistakes, from which one can
learn;
- a commitment to a well-defined purpose;
- a commitment of ILCA professionals to work with local
professionals and vice versa;
- a commitment to multi-disciplinarity ;
- a commitment to work with pastoralists ;
- a commitment to be useful to pastoralists, to generate action
programmes to serve pastoralists as mentioned in de Haan's
paper.
Dr Abel referred to the adverse effects on pastoralists of rural-
urban and arable-pastoral terms of trade, and thus felt that if
technical research was to have any relevance it must incorporate
policy analysis into a systems approach. He also thought that a
historical analysis of the origins of the present pastoral problems
was lacking from the systems framework of the workshop. It was
necessary to search for historical explanations of why, how and when
people once politically and economically strong and ecologically
secure had become dependant, impoverished and devastated by droughts.
Dr Grandin agreed with Mr Sandford that systems analysis was a
continuing process. She felt that ILCA needed to develop a data base
from their research which could be made available to other African
researchers. The raw data should be made available in this way.
Dr Barry took the opportunity to thank ILCA's Director General and
his colleagues for having made the participants' stay possible. The
problems of livestock production formed a major part of the preoccu
pations of he and his colleagues from the Sahel. The desire to
improve the income of pastoralists and to maintain the Sahel countries
as exporters of livestock and meat had shown itself soon after
- 472 -
independence. This was the reason the economic community of meat and
livestock was set up in 1971 as part of the development programme in
the member countries concerned with this agreement. For the same
reasons the member countries of CILSS had set up a unit responsible for
the problems of livestock production following the great drought. In
the Sahel sub-region there were several organisations such as CEBV,
CILSS, CEAO and ECOWAS working in a co-ordinated way. The information
provided during the workshop would certainly complement the data
already possessed by the countries of the Sahel. However, the problem
was to succeed in implementing concrete measures to improve the lives
of the herdsmen and their communities. These measures should include
all types of production system. Research should not be carried out
for research's sake. There was not much time to spare - development
had to be rapid. In this respect ILCA had a very important role to
play.
,•; -,i •*i\Mi 'y
- 473 -
Résumé des débats de J.a treizième séance sur la recherche sur les
systèmes pastoraux
Président: M. Stephen Sandford (CIPEA)
Le Dr Chema a suggéré que pour mieux promouvoir l'approche par système,
le CIPEA pourrait faire un film sur le système décrivant le type
d'agents participant à de tels travaux et les méthodologies appliquées.
Des séries de diapositives pourraient également être utilisées pour
expliciter certains domaines spécifiques de recherche. Il a également
souligné qu'en examinant les modifications à apporter aux programmes
de recherches zonaux du CIPEA, le Comité de programme du CIPEA devrait
envisager la participation du Centre à des activités nationales dans
un cadre coopératif.
M. de Haan a signalé aux participants que lorsque le CIPEA démarrait
ses activités, il n'y avait pas de méthodologie de recherche sur les
systèmes pastoraux disponible. Il estimait que le CIPEA était sur le
point d'acquérir une telle méthodologie. Il a cité l'exemple de
projets en cours de négociation au Mali et au Cameroun qui auront
pour siège des programmes nationaux.
Le Dr El Kaluba a déclaré que dans tout système de production animale,
il y avait deux facteurs, notamment le matériel génétique et l'environ
nement qui influençaient la productivité et l'efficacité du système.
Il pensait que la plupart des documents présentés au cours du séminaire
avaient traité de manière extensive des effets de l'environnement,
mais avaient négligé le matériel génétique. Il pensait que la plupart
des organismes de développement avaient tendance à négliger les races
locales et à privilégier les races exotiques souvent mal adaptées aux
conditions de l'Afrique. Le CIPEA devait tenir compte des races
locales dans les zones dans lesquelles il travaillait.
M. Zulberti a noté que les observations des participants, particulière
ment celles des représentants des pays, indiquaient qu'ils acceptaient
en général les concepts de la recherche sur les systèmes pastoraux.
Comme l'avait souligné le Prof. Saka Nuru,cela représentait une
évolution majeure par rapport à l'approche traditionnelle en matière
de recherche. Toutefois, M. Zulberti estimait que certains des exposés
- 474 -
ne suivaient pas de très près les lignes de la RSP. Mais il a
déclaré qu'un séminaire tel que celui-ci ne pouvait pas avoir lieu
dix ans auparavant. Le concept de la recherche sur les systèmes
n'était plus une idée absurde. Il a soutenu que la philosophie de la
recherche sur les systèmes imposait les contraintes suivantes à ceux
qui l'adoptaient:
- une volonté de développement;
- l'engagement de reconnaître ses erreurs, démarche qui permet
d'aller de l'avant;
- l'engagement de bien définir les objectifs;
- l'engagement des scientifiques du CIPEA à travailler avec les
scientifiques locaux et vice-versa;
- l'engagement à travailler dans un cadre multidisciplinaire;
- l'engagement à travailler avec les éleveurs;
- l'engagement à servir les éleveurs et à mettre au point des
programmes d'action au service des éleveurs comme 1 a
mentionné M. de Haan dans son étude.
Le Dr Abel a fait allusion aux effets négatifs de la détérioration
des termes de l'échange rural/urbain et agricole/pastoral sur l'éleveur
et a estimé que pour être applicable, la recherche technique doit
incorporer l'analyse des politiques dans l'approche par système. Il
pensait également que dans le cadre du séminaire, on avait omis de
procéder à l'analyse historique des origines des problèmes
pastoraux. Il était nécessaire de rechercher les explications
historiques qui permettraient de savoir pourquoi, comment et quand
des peuples jadis politiquement et économiquement forts, vivant dans
un cadre écologique sûr, ont été appauvris et affaiblis par la
sécheresse. Mlle Grandin a reconnu avec M. Sandford que l'analyse
des systèmes était un processus continu. Elle pensait que le CIPEA
devait développer une base de données à partir de sa recherche, base
qui pourrait être mise a la disposition des autres chercheurs africains.
Les données brutes devraient être mises à la disposition des
utilisateurs éventuels de cette manière.
- 475 -
Le Dr Barry a saisi cette occasion pour remercier le Directeur général
du CIPEA et ses collègues pour avoir rendu possible le séjour des
participants à Addis. Les problèmes de production animale consti
tuaient une partie importante de ses préoccupations et de celles de
ses collègues du Sahel. Le désir d'améliorer le revenu des éleveurs
et de permettre aux pays du Sahel de continuer à exporter du bétail et
de la viande s'est manifesté dès le lendemain des indépendances.
C'était là la justification de la Communauté économique du bétail et
de la viande qui a été mise sur pied en 1971 en tant que partie
intégrante des programmes de développement des pays membres signataires
de cet accord. Pour les mêmes raisons, les pays membres de CILSS
avaient mis en place un service chargé des problèmes de production
animale à la suite de la grande sécheresse. Dans la sous-région du
Sahel, il y avait plusieurs organisations telles que la CEBV, le
CILSS, la CEAO, et la CEDEAO qui travaillaient de concert. Les
informations fournies au cours du séminaire compléteraient certainement
les données déjà disponibles dans les pays du Sahel. Toutefois, le
problème était de réussir à mettre en oeuvre des mesures concrètes
pour améliorer la qualité de la vie du pasteur et de sa collectivité.
Ces mesures devraient porter sur tous les types de systèmes de
production. La recherche ne devrait pas être entreprise simplement
pour l'amour de la recherche. Il n'y avait pas de temps à perdre. Le
développement devait être rapide. A cet égard, le CIPEA avait un rôle
très important à jouer.
- 476 -
LIST OF WORKSHOP PARTICIPANTS

Workshop participants
BOTSWANA Mr Mbakiso Godfrey Lebani
Senior Animal Production Officer
Ministry of Agriculture
Private Bag 003
Gaborone
ENGLAND Dr Barry Nestel
38 Hatchlands Road
Redhill
Surrey
ETHIOPIA Dr Befekadu Degefe
Institute of Development Research
P 0 Box 1176
Addis Ababa
ETHIOPIA Dr Assefa W Giorgis
Head
Animal Resources Development Dept.
Ministry of Agriculture
P 0 Box 1052
Addis Ababa
KENYA Dr Samson Chema
Director of Livestock Research
Ministry of Livestock Development
P 0 Box 63228
Nairobi
KENYA Mr Larry Ngutter
Head
Development Planning Division
Ministry of Livestock Development
P 0 Box 63228
Nairobi
MALI Dr Noumou Diakite
Directeur General
ODEM
Mofti
- 477 -
MALI Dr Gagny Timbo
Chef Division de Projets et Programmes
Direction National de l'Elevage
Bamako
NIGER Dr Habou Akilu
Directeur
Projet de l'Elevage Zinder-Est
Zinder
NIGER Dr Zakari Rhissa
Coord inateur
Projet de Developpement Pastoral
Niamey
NIGERIA Professor Saka Nuru
Director
National Animal Production
Research Institute (NAPRI)
P M B 1096 - Shika
Zaria
NORWAY Professor Gunnar Sorbo
Department of Social Anthropology
N-5014 Bergen University
Bergen
SENEGAL Dr Ibrahim Sory Gueye
Directeur General
SODESP
B P 10282
Dakar
ZAMBIA Mr Eric Kuluba
Mochipaga Regional Research Station
P 0 Box 90
Choma
CILSS Dr Sitta Barry
Directeur de Projets et Programmes
Comite Permanent Interetats de Lutte
Contre la Secheresse dans le Sahel
Ouagadougou
Haute Volta
- 478 -
HIID Mr Carlos Zulberti
Harvard Institute for International
Development
P 0 Box 47098
Nairobi
Kenya
ICARDA Dr Euan Thomson
International Centre for Agricultural
Research in the Dry Areas
P 0 Box 5466
Aleppo
Syria
IDRC Dr Jackson A Kategile
International Development Research
Centre
Nairobi
Kenya
ILCA Participants
ILCA Ethiopia Ato Addis Anteneh
Dr Jean-Claude Bille
Dr Peter Brumby
Mr Noel Cossins
Dr Maimouna S Dicko
Dr Jim Lamboume
Dr Kevin Milligan (Consultant)
Ato Negussie Tilahun
Mr Stephen Sandford
ILCA
P 0 Box 5689
Addis Ababa
ILCA Botswana Mr Nick Abel
ILCA
P 0 Box 20604
Gaborone
ILCA Kenya Dr Peter de Leeuw
Dr Barbara E Grandin
Mr Patterson Semenye
Dr Solomon Bekure
ILCA
P 0 Box 46847
Nairobi
- 479 -
ILCA Mali Dr Pierre Hiemaux
Dr Trevor Wilson
ILCA
B P 60
Bamako
ILCA Nigeria Dr Christine Okali
ILCA
P M B 5320
Ibadan
Dr Ralph von Kaufmann
Al Haji Habibu Sulaiman
ILCA
P M B 2248
Kaduna
 
!