Prospects for integrating food and feed production in Welayita
Sodo, Ethiopia
Adugna Tolera 1 and AN Said 2
1 Awassa College of Agriculture
Addis Ababa University
PO Box 5, Awassa, Ethiopia
2 International Livestock Centre for Africa (ILCA)
PO Box 5689, Addis Ababa, Ethiopia
ABSTRACT
Welayita Sodo is a densely populated and intensively cultivated mid-altitude area of Ethiopia.
Smallholder mixed farming is the dominant mode of production. The crop and livestock
subsystems are closely interdependent: livestock contribute to the cropping system in the form
of draught power, manure and sources of cash for purchase of crop production inputs, while
crop residues provide livestock feed, especially during the dry season.
Because of the high population density, land holdings per household are small. Inadequate
feed supply is the main constraint to livestock production. In order to optimise overall
productivity there is a need to integrate food and feed production. Introducing forage legumes
seems an acceptable approach: forage legumes will improve soil fertility, crop yields and
herbage quality, and make the system more sustainable. Hedgerows of multipurpose fodder
trees, productive backyard forages and undersowing or interplanting improved forages with
food or plantation crops will probably be the most successful forage development strategies in
this area. Research should look at temporal interactions between forage supply and form of
feeding and nutrient demand by animals to exploit opportunities for marketing animals and
their products.
RESUME
Perspectives de l'intégration des cultures vivrières et de la production des aliments du
bétail à Wolayita Sodo (Ethiopie)
Wolayita Sodo est une région d'altitude moyenne, densément peuplée, où l'agriculture
constitue une activité extrêmement répandue dominée par le système de la petite exploitation
mixte y est étroitement associé à l'agriculture: d'une part, le bétail fournit l'énergie de traction
et le fumier et assure des revenus monétaires servant à l'achat des intrants agricoles; de
l'autre, les résidus de récolte entrent dans l'alimentation des animaux, notamment au cours de
la saison sèche.
Compte tenu de la forte densité de population de cette région, les ménages doivent se
contenter de petits lopins de terres. Par ailleurs, les pénuries d'aliments du bétail constituent
ici le principal obstacle au développement de l'élevage. Pour optimiser la productivité de
l'ensemble du système, il convient de promouvoir l'intégration de la production des cultures
vivrières et des aliments du bétail. L'introduction de légumineuses fourragères dans ces
systèmes de la petite exploitation mixte semble constituer une stratégie viable. En effet celles-
ci permettront non seulement d'améliorer la fertilité des sols, les rendements des cultures et la
qualité du couvert herbacé, mais également de promouvoir la durabilité du système. La mise
en place de haies de ligneux fourragers d'usages multiples et de vergers d'embouche ainsi
que l'introduction d'espèces fourragères améliorées sous culture ou entre les rangées de
plantes vivrières ou pérennes constituent sans doute certaines des meilleures stratégies
possibles dans cette région. Enfin il convient, pour promouvoir la commercialisation du bétail
et des productions animales, d'effectuer une étude approfondie des interactions à long terme,
entre d'une part les disponibilités fourragères et de l'autre, le mode d'alimentation et les
besoins en éléments nutritifs des animaux.
INTRODUCTION
A diagnostic survey was carried out in 1988 to assess the feed resources and the animal
production situation in Welayita Sodo, Ethiopia. Data were collected by interviewing 102 heads
of households in the study area. In addition to the formal survey, supplementary information
was collected from the Ministry of Agriculture. The information reported in this paper is partly
based on this survey.
Welayita Sodo is located at 6° 49' N latitude and 39° 47' E longitude, about 400 km south-
west of Addis Ababa at an altitude of about 1900 m. The area has moderately drained, acidic
red soils (nitosols). The average monthly temperature in the area ranges between 11.9°C
(August) and 26.2°C (January) with a mean annual temperature of 18.9°C. Rainfall averages
1100 mm a year and is bimodal, with the short rains from February or March until April and the
long rains from June until September or October. The area is representative of medium
altitude, medium rainfall, acid soil areas of western Ethiopia, Kenya, Tanzania, Uganda,
Rwanda, Burundi and Cameroon. 
The population density at Welayita Sodo is 250 inhabitants per km², similar to that of other
densely populated areas in Ethiopia. Land holdings are very small, the average per household
being 0.96 ha. More than 65% of the total holding is cultivated, of which 68% is under food
crops. Cash crops occupy 32% while cultivated forages occupy only 0.04% of the cultivated
land.
The cropping system of the area is very intensive. Major food crops include, in order of
importance, maize, sweet potato, enset (false banana), teff (Eragrostis tef), haricot bean, taro,
sorghum, Irish potato, yam and cassava. Coffee is the major cash crop. Intercropping is a
common practice. Some farmers use chemical fertiliser, mainly diammonium phosphate, on
food crops such as teff and maize.
About 93% of the farmers are engaged in livestock production in addition to cropping. The
average livestock holding per household is 3.60 cattle, 0.74 sheep, 0.25 goats, 0.13 donkeys,
0.02 mules, 0.02 horses and 2.09 poultry. The predominant breed of cattle is the local zebu.
Animal manure is applied on garden crops; livestock also provide draught power for important
farm operations. Crop residues are mainly used as dry-season feed for livestock. It seems that
there is a strong positive interaction between crop and livestock production. This paper,
therefore, attempts to assess this complementarity and the prospects for further integration of
food and feed production in Welayita Sodo.
PREVAILING FEED RESOURCES IN WELAYITA SODO
Type and nutritional quality of available feed resources
Natural pasture is the main source of feed for most of the year, but because of population
pressure grazing land is limited to about 0.23 ha/household. Some 65% of the respondents do
not have access to common grazing land and so have to depend on their small individual
holdings. About 50% of the farmers buy-in additional grass from outside their farm.
Quantitatively, therefore, stubble grazing and crop residues also serve as important sources of
feed. Cereal crop residues (straws and stovers) are mostly stacked and fed to livestock during
the dry season when the quantity and quality of available fodder from natural pasture declines
drastically.
The use of concentrate feeds is very minimal or nonexistent. However, malted barley and
boiled maize are occasionally fed to lactating cows to stimulate milk production. Agro-
industrial byproducts such as meat and bone meal, oilseed cakes and flour mill byproducts
are not available in the area.
Two mineral-rich soils, locally known as Bole and Megadua are used as mineral supplements
for ruminants and equines, respectively.
Farmers also lop the leaves and branches of various trees and shrubs and feed them to their
animals during the dry season. They also collect herbaceous wild plants, mostly legumes, as
feed for lactating cows. Specimens of these plants were collected and identified at ILCA. They
include Acacia spp, Albizia spp, Buddleja polystachya, Clausena anisata, Combretum spp,
Cordia abyssinica, Dodonaea viscosa, Dracaena steudneri, Ehretia cymosa, Erythrina brucei,
Flacovirtia indica, Grewia spp, Ipomoea spp, Neonotonia wightii, Olea africana, Rubia
cordifolia, Schrebera alata, Triumfetta rhomboidea, Vernonia amygdalina, Zornia glochidiata
and Zornia setosa. Enset and cassava leaves and sweet potato vines are also fed to animals
during the dry season.
The chemical composition and in vitro dry-matter (DM) digestibility of some of these feeds
were determined. Crude-protein (CP) content ranged from 3.38% (sweet potato tuber) to
24.19% (cassava tops). Neutral detergent fibre (NDF) content ranged from 31.28% (Dracaena
steudneri) to 62.20% (sorghum leaves). Acid detergent fibre (ADF) content varied between
6.73% (sweet potato tuber) and 56.85% (Cordia abyssinica). Lignin content ranged from
1.35% (sweet potato tuber) to 27.86% (Cordia abyssinica). The ADF-ash content, which is
indicative of the silica content, ranged from 0% in some of the feeds to 17.13% in Combretum
sp. In vitro DM digestibility varied between 32.90 and 82.17% and was negatively correlated
with ADF (r = -0.71) and lignin (r = -0.60) contents. About 74% of the feeds commonly used
had in vitro DM digestibility values greater than 50%.
All the feeds analysed, with the exceptions of sweet potato tuber, Schrebera alata and
sorghum leaves, had relatively high CP content - above the 7.5% considered necessary for
optimum rumen function. Thus, most of them could be useful supplements to diets based on
crop residues and poor quality natural pastures, which have low protein content.
Most of the feeds analysed contained adequate amounts of all the required minerals, with the
exception of phosphorus and sodium. Bole and Megadua soils are rich in sodium, and could
be used as sources of the mineral, but are low in phosphorus. Moreover, the high calcium
content in the various feeds and mineral soils could aggravate the wide Ca:P ratio and
precipitate the problem of phosphorus deficiency. Attempts should be made to correct the
phosphorus levels of these feeds when they are used as dry-season supplements to diets
based on crop residues and dry pastures.
Availability of feed resources in Welayita district
Estimated quantities of feed resources
The total quantities of feed obtainable from crop residues in 1982/83 and 1983/84 are shown
in Table 1; there was drought in Welayita Sodo in 1983/84. Assuming 10% wastage, the
quantity potentially available for actual animal consumption was estimated as almost 166 000
and 107 000 t of DM in 1982/83 and 1983/84, respectively. Among the many types of crop
residue, maize stover and enset are quantitatively the most important, followed by teff,
sorghum, haricot bean and sweet potato.
In addition, natural sources provide more than 475 000 t of feed dry matter per year, about
73% of which is obtained from grazing and browsing of natural pasture (Table 2).
Table 1. Estimated quantities of feed dry matter obtainable from different residues in Welayita
district in 1982/83 and 1983/84 production years
Crop
Crop production (t) Conversion Crop residue production (t DM)
1982/83 1983/84 factor 1982/83 1983/84
Teff 10 810 9 934 1.5 16 215 14 901
Barley 804 343 1.5 1 206 515
Wheat 241 331 1.5 362 497
Maize 37 973 31 122 2.0 75 946 62 244
Sorghum 3 699 1 628 2.5 9 248 4 070
Horse bean 1 025 306 1.2 1 230 367
Chick pea 66 122 1.2 79 146
Haricot bean 4 156 2 791 1.2 4 987 3 349
Field pea 135 66 1.2 162 79
Potato 512 391 0.3 154 117
Sweet potato 8 027 9 366 0.3 2 408 2 810
Yam 458 592 0.3 137 178
Cassava 1 025 365 1.0 1 025 365
Coffee 2 302 1 931 0.4 921 772
Lentil 17 - 1.2 20 -
Enset a na na na 65 200 22 872
Banana a na na na 4 880 5 272
Total    184 180 118 554
a Crop production data not available. Crop residue production based on area planted (8150 and 2859 ha of
enset, and 610 and 659 ha of banana, in 1982/83 and 1983/84, respectively: MOA, 1984) and crop residue
yields of 8.0 t/ha per year for both crop (FAO, 1987)
Sources: Crop production data from MOA (1984). Conversion factors from FAO (1987)
Availability of feed dry matter relative to livestock population of the district
The livestock population of Welayita district is estimated at 468 288 tropical livestock units
(TLU) (MOA, 1984; Jahnke, 1982). The availability of feed DM was calculated to be 1.37 and
1.24 t DM/TLU per year, equivalent to 3.75 and 3.41 kg DM/TLU per day for the 1982/83 and
1983/84 production years, respectively. The maintenance requirement is estimated at 4.6 kg
DM/TLU per day (Kearl, 1982), so the available feed supply satisfied only about 78% of the
livestock feed requirement of the area. However, the animals are reared intensively; they are
kept in the yards within the household and younger animals or smallstock are kept inside the
family's house. Hence, the animals have very limited exposure to inclement weather and this
minimises their maintenance requirement. The animals also have constant access to
household wastes, which gives them a decided advantage with reduced movement resulting
in minimum energy loss. This, among other factors, is reflected in near acceptable animal
performance in spite of the estimated inadequate feed supply. On the other hand, the methods
used in estimating the quantities of feeds available may not have been accurate and therefore
there is a need to re-examine them.
Table 2. Feed dry matter obtainable from different land use types in Welayita district
Land use type Area (ha) DM yield (t/ha per year) Annual DM production (t)
Natural pasture 172 602 2.0 345 204
Aftermath 53 410 0.5 26 705
Fallow land 9 261 1.8 16 670
Forest 12 771 0.7 8 940
Woodland, bushland and shrubland 65 016 1.2 78 019
Total 475 538
Source: FAO (1987)
LIVESTOCK PRODUCTIVITY IN WELAYITA SODO
The productivity of the ruminant livestock in the area is summarised in Table 3. In general,
livestock productivity is low for all classes of animals. The relatively late age at maturity and
the extended parturition intervals reflect an environment in which animals are subjected to long
periods of nutritional stress. However, the estimated milk yield of cattle (about 500
kg/lactation, excluding milk consumed by the calf) can be considered acceptable for
unimproved local zebu cows and this could be attributed to the practice of hand-feeding with
various herbaceous and tree plants collected from the surroundings as well as with household
wastes.
Mortality rates could not be calculated from the data collected. However, farmers reporting
mortality of young animals cited drought and nutritional stress as the main causes of death.
MAJOR CONSTRAINTS TO LIVESTOCK PRODUCTION
Shortage of grazing land and inadequate feed supply are the major problems facing livestock
producers in the area. Most of the crop residues are used as livestock feed, but their supply is
seasonal and they are used in the traditional way, without any pretreatment and/or strategic
supplementation. The cultivation of fodder crops is limited due, among other reasons, to
shortage of land. The integration of food and fodder crops in Welayita Sodo is nil to minimal
due to inadequate research data and extension services and the lack of economic incentives
to the farmers. Generally, available feed resources are not sufficient for optimum animal
production. Lack of adequate financial resources and the absence of institutionalised credit
services for investment in livestock production are other constraints identified in the survey.
Table 3. Productivity data for ruminant livestock in Welayita Sodo, based on interview data
from 102 households
Performance parameters
Cattle Sheep Goats
Mean SD Mean SD Mean SD
Weaning age (months) 9.8 2.9 4.8 1.8 5.3 1.0
Age at sexual maturity (months) 52.3 10.2 11.4 4.6 11.0 6.6
Age at first purturition (months) 65.0 10.2 18.4 7.1 16.9 6.7
Parturition interval (months) 26.2 6.6 12.9 4.3 11.1 2.3
Daily milk yield (litres/head)
Maximum 2.4 0.9 - -
Minimum 1.0 0.5 - -
Mean 1.6 0.6 - -
Lactation length (months) 10.3 2.8 - -
PROSPECTS FOR INTEGRATING FOOD AND FEED PRODUCTION
The traditional complementarity of food and feeds in Welayita Sodo
Acute shortage of land and inadequate feed supply constrain animal output in Welayita Sodo
but the farmers in the area are amenable to supplementary feeding of their animals. Where
cash availability limits investment in livestock enterprises, as is the case in Welayita Sodo, the
incorporation of forage legume production with the crop subsystem may be an acceptable
approach.
Leguminous forages can contribute to better utilisation of cereal crop residues. Legumes are
rich in protein and other nutrients such as minerals and vitamins. Some forage legumes such
as Desmodium intortum, Macrotyloma axillare and Stylosanthes guianensis were shown in
initial agronomic screening trials to be suitable and productive; the trials were conducted by
the International Livestock Centre for Africa (ILCA), in collaboration with the Ministry of
Agriculture, in mid-altitude areas of Welayita Sodo and in the Rift Valley region of Ethiopia.
Some of the forage legumes were further evaluated in feeding and digestibility trials and were
found to be useful supplements to crop-residue-based diets (Table 4).
Table 4. Voluntary feed intake, nitrogen intention and body weight gain of lambs a basal diet
of maize stover supplemented with different levels of legume hays
Supplement g/head per
day
Dry matter intake (g/kg0.75 per
day) Nitrogen retention
(g/day)
Liveweight gain
(g/day)Maize
stover
Legume
hay
Total
Desmodium intortum
250 28.9a 26.8c 55.7b -0.19 7.3b
350 26.9b 35.0b 61.9a 1.51 24.9a
450 17.0c 44.7a 61.7a 1.41 30.9a
Stylosanthes guianensis
250 33.4a 26.0c 59.4b 0.82 15.0b
350 22.2b 36.6b 58.8b 1.61 13.9b
450 20.3c 44.4a 64.7a 1.37 30.3a
Within legume species, values in the same column followed by the same or no
letter do not differ significantly (P>0.05)
Leguminous fodders also make a very important contribution to the crop subsystem. In
systems with minimum fertiliser inputs legumes enhance crop yields by reducing the rate of
decline of soil fertility. Nitrogen fixed by the legumes will be available to food or plantation
crops grown by the smallholders during concurrent and/or subsequent seasons. For example,
the planting of Desmodium intortum cv Greenleaf under coffee increased coffee berry yields
by up to 19% on peasant holdings in Welayita Sodo (Lazier, 1987). Similarly, Haque (1990)
found that intercropping maize with Macrotyloma axillare produced significantly more total dry
matter than sole maize plots, although the intercropped plots yielded less than half as much
grain as maize in pure stand. From this finding, it was concluded that the production of large
amounts of high quality feed during the dry season by Macrotyloma axillare may be sufficient
to offset the loss in maize grain yield for farmers with mixed crop-livestock farms and
encourage them to adopt cereal-legume intercropping (Haque, 1990). But this tentative
assumption needs to be tested in the field against farmers' responses/reactions to low grain
yields vis-à-vis potential animal productivity and marketing in Welayita Sodo. Alley cropping
with leguminous multipurpose trees such as Sesbania sesban, Leucaena leucocephala and
Gliricidia spp offers alternatives to intercropped forage legumes.
Stratification of livestock production
The adoption of forage legume technologies would also assist in stratification of livestock
production. During the dry season there is an acute shortage of feed in the lower altitude
rangelands (mainly inhabited by pastoralists), resulting in substantial body weight loss and/or,
during severe droughts, even death of the animals. These losses could be partially averted by
selling some meat animals (steers and small ruminants) to highland farmers for fattening, but
this exercise would only be feasible if adequate feed can be made available in the highlands,
through improved forage production and if there are suitable marketing facilities.
In general, stratification of animal production is an advantageous strategy for both the lowland
pastoralists and the mixed crop-livestock farmers of the highlands. Studies on animal health
and on marketing are needed to assess the sustainability of this intervention. For the
pastoralists, this strategy will:
reduce feed problems for animals domiciled in the rangelands 
protect the rangelands from the likely slow land degradation by adjusting
stocking rates through increased offtake 
generate some cash income 
And the highland mixed farmers will benefit from:
increases in income, general welfare and cash input for crop production
enhanced soil fertility in the cropping areas from animal manure and from
the forage legumes
Temporal interaction between forage supply and demand
If farmers are to adopt a sustainable integrated crop-livestock production system there is a
need to develop packages that will fit forage production and conservation with the nutrient
demand curves by the animals to exploit market opportunities for livestock and livestock
products. Researchers need to work in a more problem-oriented and integrated approach to
overcome the isolation of disciplines like nutrition, forage agronomy and socio-economics.
Research should be extension-oriented, addressing the needs of the smallholder in Welayita
Sodo.
The four "I"s
Irrespective of the locality or the scale of operation, successful integration of food and feed
production subsystems will depend on:
institutional factors resulting from government agricultural policy 
infrastructure that is conducive to agriculture production and marketing, nationally,
regionally or internationally 
inputs to the production systems—credit and financial facilities 
incentives, interrelated to agricultural policy, infrastructure and ease with which inputs
are available.
In most cases farmers' incentives to increasing agricultural productivity are spontaneous and
self-triggered. Lamentably the natural wisdom of the peasantry in rationalising the
practicabilities of these issues, as they affect their well-being, is often assumed not to exist,
underplayed or neglected by policy makers, research workers and even by extension services.
We might be excused, sometimes, for assuming that, as researchers, we have better tools to
evaluate sustainability issues and other long-term interests of the farmers we are supposed to
help. That should not blind us to the fact that they, the farmers, also know what are their
priorities, in the short and medium term.
CONCLUSIONS
Our success in increasing agricultural production within the multiplicity of existing subsystems
in sub-Saharan Africa depends upon our ability to recognise and appreciate the
interrelationships of the four "I"s outlined. There is a need to develop these infrastructures at
the rural level and especially, in these days of gender issues, to lessen the load of women in
feeding their families. For Welayita Sodo, forage production technology will be more
acceptable if it fits into the existing production system, based on the available marketing
infrastructure and financial inputs. Forages should be integrated with cropping systems rather
than displace crops. Other forages could be produced from fence lines and hedges or in a
backyard forage bank. We advocate that research should be devoted to the temporal
interaction between forage supply and nutrient demand curves as they relate to a possible
exploitation of market opportunities of animals and their products in Welayita Sodo.
REFERENCES
FAO (Food and Agriculture Organization of the United Nations). 1987. Master land use plan,
Ethiopia range/livestock consultancy. Report prepared for the Government of the Peoples'
Democratic Republic of Ethiopia. Technical Report. AG/ETH/82/010. FAO, Rome, Italy.
Haque I. 1990. Effect of intercropping maize with Macrotyloma axillare on feed output and
grain yield. In: ILCA Annual Report 1989. ILCA (International Livestock Centre for Africa),
Addis Ababa, Ethiopia, pp. 86-87.
Jahnke H E. 1982. Livestock production systems and livestock development in tropical Africa.
Kieler Wissenshaftsverlag Vauk, Kiel, FR Germany. 253 pp.
Kearl L C. 1982. Nutrient requirements of ruminants in developing countries. International
Feedstuffs Institute, Utah Agricultural Experiment Station, Utah State University, Logan, Utah,
USA. 381 pp.
Lazier J R. 1987. The effect of D. intortum undersown in coffee on coffee yields. FNE
Newsletter 17:16-18. FNE (Forage Network in Ethiopia). ILCA (International Livestock Centre
for Africa), Addis Ababa, Ethiopia.
MOA (Ministry of Agriculture of Ethiopia). 1984. General agricultural survey. Preliminary
Report 1983/84 (1976 Ethiopian Calendar). Vol. II. Sidamo. Planning and Programming
Department, MOA, Addis Ababa, Ethiopia. 232 pp.