Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=tags20
International Journal of Agricultural Sustainability
ISSN: 1473-5903 (Print) 1747-762X (Online) Journal homepage: http://www.tandfonline.com/loi/tags20
Effectiveness of agricultural water management
technologies on rainfed cereals crop yield and
runoff in semi-arid catchment: a meta-analysis
M. S. Magombeyi, A. E. Taigbenu & J. Barron
To cite this article: M. S. Magombeyi, A. E. Taigbenu & J. Barron (2018): Effectiveness of
agricultural water management technologies on rainfed cereals crop yield and runoff in semi-
arid catchment: a meta-analysis, International Journal of Agricultural Sustainability, DOI:
10.1080/14735903.2018.1523828
To link to this article:  https://doi.org/10.1080/14735903.2018.1523828
View supplementary material 
Published online: 26 Sep 2018.
Submit your article to this journal 
View Crossmark data
Effectiveness of agricultural water management technologies on rainfed
cereals crop yield and runoff in semi-arid catchment: a meta-analysis
M. S. Magombeyia,b, A. E. Taigbenua and J. Barronb,c
aSchool of Civil and Environmental Engineering, Witwatersrand University, Johannesburg, South Africa; bInternational Water
Management Institute, Colombo Sri Lanka; cDepartment of Soil and Environment, Swedish University of Agricultural Sciences
(SLU), Uppsala, Sweden
ABSTRACT
Multiple agricultural water management (AWM) technologies are being promoted
worldwide in rainfed agro-ecological production systems, such as the Limpopo
River Basin, to close the yield gap, enhance food security and reduce poverty, but
evidences on yield gains and environmental impacts are varied. This paper
conducts a review of the performance of AWM technologies against conventional
farmer practices to produce adequate evidence on cereal yield and field runoff
changes. With the interrogation of literature from 1980 to 2013 using seven AWM
groupings, enough evidence was found that AWM technologies can deliver
substantial benefits of increased crop yield and water productivity with reduced
environmental impacts. Using random effects model, the standardized mean
difference (SMD) of yield between AWM and control was 0.27, while SMD of water
productivity was 0.46, indicating the effectiveness of the technologies (SMD > 0).
Subgroup analyses showed greatest yield responses on silty-clay-loam, clay-loam
and sandy soils compared to clay and loam-sandy soils, and higher yield increase
under low rainfall regime (200–500 mm) than under high rainfall regime (500–800
mm). Large yield change variations for different AWM technologies present a huge
opportunity for meeting the existing yield gaps and enhancing coping capacity in
dry years and under climate change.
KEYWORDS
Agricultural water
management; climate smart
agriculture; Limpopo; rainfed;
smallholder farmers;
sustainable intensification;
systematic review; yield gap
1. Introduction
Currently, Africa is experiencing an annual decrease in
agricultural production of 3% due to soil erosion, land
and environmental degradation (FAO, 2009; Owenya
et al., 2012), whereas an annual increase of 6% was
required to achieve the Millennium Development
Goals by 2015 and now the Sustainable Development
Goals set for 2030 (Bayala et al., 2012; United Nations,
2015). Further inherent risks to production are associ-
ated with seasonal variability of the amount and distri-
bution of rainfall (Cooper et al., 2008; Magombeyi,
Taigbenu, & Barron, 2016) and climate change.
Climate change projections indicate a decrease in
rainfall (5–15% per century) and hotter climate (temp-
erature rise of 2–5°C by 2050) in southern Africa,
where the Limpopo River Basin (LRB) is located
(IPCC, 2012; World Bank, 2012). A combination of
increased temperatures with increased rainfall varia-
bility due to climate change (Lobell et al., 2008) is
likely to erode production capacity by about 50%
(Müller, Cramer, Hare, & Lotze-Campen, 2011; Schlen-
ker & Lobell, 2010) and constrain beneficial invest-
ment. Hence, climate change will exacerbate the
current sub-optimal yield levels of 0.50–1.00 t ha−1
from smallholder farming in Sub-Saharan Africa (SSA)
(FAO and DWFI., 2015; Rockström et al., 2009; van Itter-
sum et al., 2016), well below the attainable yields of
6.00–12.00 t ha−1, typical under commercial farming
(Tittonell & Giller, 2013; van Ittersum et al., 2013).
The yield gap is particularly large for key staples
such as maize, sorghum andmillet in rainfed agro-eco-
logical production systems (Guilpart, Grassini, Sadras,
© 2018 Informa UK Limited, trading as Taylor & Francis Group
CONTACT M. S. Magombeyi manumagomb@yahoo.com
Supplemental data for this article can be accessed https://doi.org/10.1080/14735903.2018.1523828.
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY
https://doi.org/10.1080/14735903.2018.1523828
Timsina, & Cassman, 2017). Hence, sustainable inten-
sification of agriculture to increasing crop yields,
enhancing efficient use of rainfall and preserving the
quality of the soil, is paramount not only for small-
holder farmers in SSA but also at national and global
levels (Alemaw & Simalenga, 2015; Carberry et al.,
2013; Milgroom & Giller, 2013; Zhou et al., 2010).
However, the required changes to management prac-
tices to close yield gaps vary considerably by region
and production intensity levels (Mueller et al., 2012;
Tilman, Balzer, Hill, & Befort, 2011).
Agricultural research and development innovations
in SSA, including LRB have for decades focused on
smallholder rainfed farming systems with the aim of
increasing the value and productivity of land, water,
labour and capital so as to achieve both local and
national food security (Benin, Nin Pratt, Wood, &
Guo, 2011; CAWM, 2007; Magombeyi et al., 2016).
The smallholder farming system mainly produces
crops for subsistence purposes and is characterized
by low inputs and technology uses. The rainfall distri-
bution in the LRB also presents a huge production risk
as it is highly variable (20–45%) in space and time,
with a basin annual mean of 530 mm (Mupangwa,
2009; Sullivan & Sibanda, 2010). Furthermore, the
basin is affected by recurring droughts that occur at
least once in every four years (Meinke et al., 2006;
Twomlow et al., 2009) and lengthy dry spells (Magom-
beyi & Taigbenu, 2008; Mupangwa, 2009), which often
cause crop failures and livestock deaths, thereby nega-
tively affecting smallholder rural livelihoods.
To reduce the risks of production capacity, food
insecurity and livelihoods in SSA, governments and
the private sector have encouraged the implemen-
tation of improved AWM technologies in smallholder
farming systems (Jat, Wani, & Sahrawat, 2012). These
AWM technologies are applicable across diverse geo-
graphical, agro-ecological zones, soil types, plot
sizes, and crops throughout Africa (IIRR and ACT,
2005; Vohland & Barry, 2009). AWM constitutes a set
of key technologies which are aligned with the
pillars of agriculture, market access, water manage-
ment and the environment that are championed by
the New Partnership for Africa’s Development
(NEPAD), and the Comprehensive Africa Agricultural
Development Program (CAADP) (Owenya et al.,
2012). AWM is therefore well positioned in the devel-
opment agendas to achieving multiple objectives of
climate change adaptation and mitigation as well as
poverty alleviation and agro-ecosystem biodiversity
conservation (Milder, Majanen, & Scherr, 2011).
Despite most studies in Africa generally reporting
positive effects of AWM technologies on soil fertility
and crop productivity (e.g. Bayala et al., 2012;
IMAWESA, 2009; Munamati & Nyagumbo, 2010;
Owenya et al., 2012), crop yields, rainfall partitioning
and soil conservation in smallholder farming system
have consistently remained below those achieved by
large-scale producers in the same agro-ecological set-
tings (Munamati & Nyagumbo, 2010; Mupangwa,
Twomlow, & Walker, 2008). These results suggest
that the conditions for optimum performance of the
AWM technologies are yet to be fully context
defined (Bulcock & Jewitt, 2013). Hence, blanket state-
ments about the performance of AWM technologies
are often inappropriate and misplaced because of
the interplay of many climatic, environmental, policy,
institutional and farming factors which impact on
their adoption (Farooq, Flower, Jabran, Wahid, & Siddi-
que, 2011). Furthermore, the complexity, time-bound
and site-specific interactions between the com-
ponents of AWM technology on yield performance
requires long-term experiments for better under-
standing (Rusinamhodzi et al., 2011). By applying a
meta-analysis, valuable synthesized information on
the processes that make for the success or failure of
AWM technology can be obtained and used to
empower farmers on a suite of technologies appropri-
ate for their context (Bayala et al., 2012; Marongwe
et al., 2011). Hence, this paper is aimed at identifying
and quantifying opportunities for AWM technologies
to increase cereal crop yield and reduce field runoff,
compared to conventional/traditional farmer soil and
water management practices in arid and semi-arid
areas across diverse experimental set up (on-farm or
on-station), and agro-ecological conditions in the LRB.
2. Materials and methods
2.1. Study area
The case studies for this review were from the LRB
(Figure 1), located in southern Africa between lati-
tudes 20°S and 26°S and longitudes 25°E and 35°E,
with a total land area of 430,000 km2 (LBFP, 2010).
The proportions of area in each country are Botswana
(20%), Mozambique (21%), South Africa (44%) and
Zimbabwe (15%). The altitude of the area ranges
between 11 and 2330 m (mean 796 m). According to
Magombeyi et al. (2016), the average poverty levels,
based on US$1.25/capita/day, were 20% for Botswana
(2009/2010), 56% for South Africa (2010), 68% for
2 M. S. MAGOMBEYI ET AL.
Mozambique (2008/2009) and 69% for Zimbabwe
(2011). It is projected that water stress to absolute
water scarcity as a result of both natural and human-
made phenomenon will be experienced in the basin
by 2025 (Sulser et al., 2009). Water stress for an area
is when annual water supplies drop below 1700 m3
per person, while absolute water scarcity is when
the water supplies drop below 500 m3 per person
(FAO, 2007).
2.1.1. Climate
According to Köppen-Geiger’s five major Climate
Classification system (A–E) most of the LRB falls in cat-
egory B (potential evaporation and transpiration
exceed precipitation, and rainfall is relatively low),
with a small eastern portion of the basin in category
C (warm and humid summers, with mild winters)
(Hatfield Consultants, 2008). The basin rainfall is uni-
modal with the wet season from October/November
to March/April, followed by a long dry spell in the
winter (Willcocks & Twomlow, 1993). The annual
rainfall ranges per country in the Limpopo Basin are:
Botswana, 250–555 mm (mean 425 mm); Mozambi-
que, 355–865 mm (mean 535 mm); South Africa,
290–1050 mm (mean 590 mm); and Zimbabwe, 300–
635 mm (mean 465 mm), while the basin mean is
530 mm (CGIAR, 2003). Generally, rainfall should
exceed 20–30 mm in a single event to trigger runoff,
due to the basin’s flat terrain, high temperatures and
low humidity (LBPTC, 2010). While droughts are a
common occurrence – 1980, 1982–1983, 1987, 1992–
1993, 1994–1995, 1999, 2002–2004, 2005 and 2006–
2007 (DEWFORA, 2011; SARDC, 2002), floods with
devastating consequences from heavy rains do some-
times occur over Mozambique, South Africa and Zim-
babwe as in the 2013/2014 and 2016/2017
(Shewmake, 2008).
Temperatures follow a seasonal variation and alti-
tude levels, with the coolest months (0°C at night)
experienced in winter (June–August) and the highest
temperatures (above 40°C) experienced in early
summer (late November to early December). With
Figure 1. Reviewed case studies and the agro-ecological zones in the Limpopo Basin. Adapted after IIASA/FAO (2012).
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 3
the predicted average increase in temperature of 0.7°C
in the past 100 years, future regional air masses, basin
climate and crop production and productivity are
likely to be impacted (IPCC 2007). Annual evaporation
rates are between 1600 and 1700 mm yr−1 in the
cooler mountainous regions in the south-eastern
part of the basin, with the highest values of 2600–
3100 mm yr−1 in its warmer western and central
regions.
2.1.2. Soils
The two dominant soil types in the basin are older soils
that were formed from deep weathering of parent
material during a time of higher temperatures and
rainfall, such as the soils of highveld plateaus of
South Africa and Zimbabwe, and considerably
younger, shallower sediments from more recent ero-
sional or depositional activities under drier climates,
such as soils of the lowveld and coastal plains of
Mozambique (Limpopo River Awareness Kit, 2011).
Extreme soil degradation was noted in three areas in
Limpopo Province of South Africa, corresponding
with densely populated communal areas (former
homelands of Venda) and Lebowa (south of
Polokwane).
2.2. Farming systems for cereal production in
the basin
The prevailing crop farming systems (subsistence,
semi-commercial or emerging and commercial) in
the LRB reflect its cultural, socio-economic conditions,
agro-ecological potential and agricultural policies.
2.2.1. Conventional farming system
A mix of farming systems exists under conventional
farming practice. However, greater attention is paid
to smallholder or subsistence farming system,
which engages the largest proportion of farmers
in the basin. The smallholder, subsistence and con-
ventional farming systems are primarily low-input-
productivity systems, characterized by low level of
management (e.g. water and nutrients) and inten-
sive natural resources utilization that result in irre-
versible land degradation (Feed the Future, 2013;
Munamati, 2009). Land preparation involves
ploughing to a depth of about 0.10–0.20 m using
animal or mechanical power or hoeing and planting
on the levelled field (e.g. conventional tillage).
Ploughing destroys soil structure and weakens soil
aggregation by exposing soil organic carbon to
microbial oxidation.
2.2.2. Crop production under subsistence
farming system
The subsistence agriculture is typically a low-input–
output system adopted by local communities to mini-
mize risks from climate variability. Sorghum, millet,
groundnuts, beans/pulses and oilseeds such as
sunflower tend to perform better than maize, which
is the most widely grown crop (FAO, 2004; ReSAKSS,
2017). Average grain yields of maize in the traditional
(communal) farming system are of the order of
0.25 t ha−1 in Botswana, 0.80 t ha−1 in Mozambique
and Zimbabwe, and 0.70 t ha−1 in South Africa, while
the basin average yields for maize, grain sorghum
and groundnuts are about 0.64, 0.60 and 0.40 t ha−1,
respectively (FAO, 2004). In contrast, large-scale
rainfed commercial farming in the basin produces
maize yields between 3.00 and 8.00 t ha-1 (BFAP,
2014; LBFP, 2010).
The area cultivated per family is about 4.00 ha in
Botswana, 1.50–3.00 ha in Mozambique and Zim-
babwe, and about 0.75 ha in South Africa. Late and
poor land preparations are a common feature of the
basin due to limited access to draught power,
thereby making less labour intensive AWM technol-
ogies more attractive to farmers (Mupangwa, Dimes,
Walker, & Twomlow, 2011; Rockström et al., 2009; Rusi-
namhodzi et al., 2011; Schlenker & Lobell, 2010). The
use of improved seed for genetic diversity is limited,
with approximately 90% of the seed obtained from
previous harvest or local sources (Feed the Future,
2013; Netnou-Nkoana, Jaftha, Dibiloane, & Eloff,
2015; Progressio, 2009; The African Centre for Biodi-
versity, 2016), with an exception of Zimbabwe with
well-developed seed distribution facilities (Friis-
Hansen, 1992).
2.3. Benefits of AWM technologies
The benefits associated with AWM technologies
include higher plant-water availability, redistribution
of labour for land preparation to dry season, higher
productivity and income and reduced vulnerability
to erratic rainfall distribution and droughts (IPCC,
2007; Owenya et al., 2012). Furthermore, there is
potential for enhanced sequestration of carbon,
when combining AWM with cover crops and mulch
practises and reduction in runoff soil erosion, which
4 M. S. MAGOMBEYI ET AL.
is a major cause of soil degradation in semi-arid
(Oyedele & Aina, 2006).
2.4. Data search, collection and analysis
2.4.1. Literature Search
A comprehensive literature search provided cases of
impacts of AWM technologies, obtained from peer-
reviewed journals and non-peer reviewed conference
proceedings, theses and project reports in the LRB
from 1980 to 2013 (Figure 2). The sources searched
for cases through the end of April 2013 were: AGRIS,
CABDirect, ProQuest, EconLit, Text WebPublisher
from INMAGIC, Science direct, Database of Abstracts
of Reviews of Effects (DARE) and other databases
(e.g. Scopus, ISI Web of Knowledge and Google
Scholar).
2.4.2. Search strategy
The literature search used the following keywords and
their combinations: maize, sorghum, millet, rainfed,
reduced tillage, no-tillage, ripper, ridges, tied ridges,
conservation agriculture, minimum tillage, zero tillage,
infiltration pits, fanya juu, planting basins, bunds, ter-
races, contours, inter-cropping, agroforestry, rotation,
grain yield, runoff, nutrient and water harvesting.
Other words used were storage, retention, water
storage, supplemental irrigation, irrigation, crop resi-
dues, mulch, residue, organic, inorganic, manure, fertili-
zer, Africa, Sub-Sahara Africa, Botswana, Mozambique,
South Africa, Zimbabwe, Limpopo, Olifants, Mzing-
wane, Gwanda, Notwane, Chokwe and Xai Xai.
2.4.3. Eligibility criteria
All the identified studies were screened for relevance
first on title, then on abstract and full-text paper,
and further screening of the full papers that satisfied
all the specified eligibility checks. The schematic over-
view of the decision tree (Figure 2) shows the selection
process of the case studies used in the meta-analysis.
Both published and unpublished studies were
included in the meta-analysis to avoid publication
bias, as studies with significant results tend to be pub-
lished more often than those without significant
results (Wang, 2009).
The study’s empirical findings, which include depen-
dent variables of cereal crop yield, water productivity
and field runoff, were converted to an effect size. An
effect size represents the numerical way of expressing
the strength or magnitude of a reported relationship
of various technologies (Higgins, Altman, & Sterne,
2011). An effect size near zero indicates homogenous
result from control and experimental groups, while a
high effect size indicates an effective technology. An
effect size below zero indicates that the technology
had a negative or reverse effect compared to the
control group (Higgins et al., 2011).
2.5. Data synthesis and subgroup analyses
Meta-analysis using Review Manager (RevMan) 5.3
software (The Cochrane Collaboration, 2014) was
used to determine the overall effect size of standar-
dized mean difference (SMD) and to impose a 95%
confidence limits on the means. SMD is the difference
in mean effects from the AWM experimental treat-
ment and control groups in relation to the pooled
standard deviation of participants’ outcomes. SMD is
a summary statistic used when the studies in a
meta-analysis assess the same outcome (e.g. yield)
but measure it in different ways. SMD is not tied to
any specific unit of measurement, but SMD > 0
means the intervention is effective. The rule of
thumb used for SMD effect size from Schünemann
et al. (2011) was: SMD < 0.4, small effect, SMD 0.4–
0.7, moderate and SMD > 0.70, large effect.
The median and 25th and 75th percentiles were
used for synthesizing the field runoff, as the mean is
sensitive to extreme values. The random-effects
model (Schünemann et al., 2011) was used because
its estimate assumes there is a distribution of effects
which, with its confidence interval (CI), addresses the
question ‘what is the best estimate of the average
effect?’ In contrast, the fixed effect model addresses
the question ‘what is the best (single) estimate of
the effect? A funnel plot, which is a scatter plot of
the effect estimate (SMD) in relation to the size or pre-
cision of each study, standard error (SE), was used for
the assessment of risk of bias that may affect the
cumulative evidence, such as publication bias and
selective reporting within studies.
Factors such as soil texture, location and climate
have well-known effects on grain yield and field
runoff and may play a role in the observed yield pat-
terns. Factors used as covariates for the response of
crop grain yield to AWM technologies included:
long-term rainfall, altitude, soil texture and field slope.
The subgroup analyses examined whether the
subgroups reduce any heterogeneity in each AWM
main group. Subgroups analyses of low (200–
500 mm) and medium (500–800 mm) seasonal rainfall
regimes, manure, nitrogen fertilizer input, categorized
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5
as low (<35 kg ha−1) and high (36–300 kg ha−1) and
soil texture of clayey soil, sandy soil and loamy soil
were considered.
2.6. Interpretation of results
Results from both individual references and meta-ana-
lyses are reported with a point estimate together with
an associated 95% CI. The point estimate is the best
guess of the magnitude and direction of the AWM
experimental intervention’s effect compared with
the control one.
The p-value used in this study relates to the
summary effect in a meta-analysis and is from a Z-
test of the null hypothesis that there is no effect of
the technology. In this paper the Z-test is reported,
Figure 2. Schematic overview of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) process flow diagram for the
search result modified from Moher, Liberati, Tetzlaff, and Altman (2009).
6 M. S. MAGOMBEYI ET AL.
where p < 0.05 indicate statistical significant results
(Higgins & Green, 2008; Rücker, Schwarzer, Carpenter,
& Schumacher, 2008).
Consistence of results was measured by I2 statistic,
which is the percentage of observed total variation
across studies as a result of real heterogeneity rather
than chance or sampling error. Negative values of I2
are set to zero, with a value of 0% indicating no observed
heterogeneity. The following rule-of thumb was used to
classify I2 statistic (Cohen, 1988; Higgins & Green, 2008):
<40% is low; 30–60% is moderate; 50–90% is substantial
and 75–100% is considerable heterogeneity.
2.7. Rainfall classification
Partly based on FAO (2007) guidelines, long-term
mean annual and season rainfall were categorized
into four classes as very low (<200 mm), low (200–
500 mm), medium (500–800 mm) and high
(>800 mm). Each season was analysed as a separate
case due to the differences in rainfall amount and dis-
tribution experienced during each growing season
over several years considered in the study.
2.8. Classification of AWM technologies
The AWM technologies were classified into seven
groups from consultation with other researchers and
extension officers in the basin and partly based on
Bayala et al. (2012), FAO (2009) and IPCC (2007). These
groups are: Group A – Reduced tillage (minimum or
zero tillage, ripper), Group B – in situ water retention
(tied ridges, planting basins, bunds, terraces, contours),
Group C – Evaporation suppressants (crop residues,
mulching), Group D – Nutrient only (organic – manure
and inorganic – fertilizer), Group E – Water harvesting
with storage (full or supplemental irrigation), Group
F – Cropping system and Agroforestry (trees and
crops: crop rotations and intercropping) and Group
G – Combination of two or more technologies
(ripper or planting basin or ridges combined with fer-
tility treatment (fertilizer and/ or manure), planting
basins or crop rotation with mulch) (see Table 1).
The conventional farming system is taken as the
control against which the experimental AWM technol-
ogy groups are compared.
2.9. Risk of crop yield to AWM technologies
The relative frequency of positive or negative effects
on crop yield for each soil texture was estimated.
The level of performance of each AWM technology
was assessed by the cumulative probability distri-
butions of first and second order stochastic domi-
nance of yield. An alternative to conventional
practice dominates if it has a smaller area under the
distribution plot at every outcome level (Uaiene,
2004).
2.10. Assessment of certainty of the evidence
from reviewed studies
The GRADE (Grading of Recommendations Assess-
ment, Development and Evaluation) approach, using
the GRADEpro GDT software (Schünemann, 2008)
was used to assess the strength of evidence for or
against AWM technology and develop recommen-
dations on the use of AWM technologies (Andrews
et al., 2013; Schünemann et al., 2011). GRADE is a
well-developed formal process to rate the quality of
scientific evidence in systematic reviews and to
develop evidence-based recommendations.
2.11. Assumptions of the study
The assumptions of the study are:
. Plant population across technologies may vary but
remain the same within each case study
. The management of crops was assumed to be
similar in the experiment and control plots.
. Crop species, variety and treatment effects were
similar, i.e. in each study the same crop species or
varieties were used in the experiment and control
groups.
A limitation of the current approach is that field
soil fertility and management prior to the initiation
of the experiment or observation, which could
affect performance of experimental trials in the sub-
sequent year, was not revealed in most of the case
studies.
3. Results and discussions
3.1. Descriptive statistics of AWM cases
Ninety-seven (97) references from 1980 to 2013 were
identified. They consisted of cases from on-station
(37%), on-farm (58%) and modelling (5%). A total of
1430 AWM paired cases (experimental and control)
stemmed from the 97 references and 50.1% were
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 7
peer-reviewed studies, 34.6% project reports, 8.5%
conference papers and 6.8% were theses (Figure 3).
The highest number of cases were from Group G –
combination of two or more AWM technologies. The
proportion of cases based on publication period
were 1980–1990 (0.2%), 1991–2000 (11.8%), 2001–
2010 (55.1%) and 2011–2013 (32.9%). The references
of the reviewed cases for the different AWM groups
and the highest number of consecutive years of imple-
menting the AWM technology or modelling are pre-
sented in supplementary material.
The number of case studies for each crop type
were: maize (1085: 76%), sorghum (208: 15%) and
millet (38: 2.7%), while cowpea (58: 4%) and ground-
nut (41: 2.9%) were used in some of the intercropping
cases. The proportion of paired cases based on
Table 1. Current farm practices and AWM Technologies.
AWM technology
class Group
No. of
paired
cases Water soil nutrient feature Typical names Management feature
Reduced tillage A 83 Increase infiltration/soil carbon Minimum or zero tillage;
ripping (no soil inversion)
Reduced and energy-efficient
manual, mechanized and animal
draft power
In-situ water
retention
B 284 Slow down and trap runoff and
enhance infiltration rates,
guard against erosion, and
help keep soil fertility in place
Zai pit/ planting basins,
planting/infiltration pit,
fanyaa juu, tied ridges,
bunds, terraces, contour
ridges
Manual, mechanized and animal
draft power for creating and
managing soil structures
Evaporation
suppress
C 115 Reduce runoff, increase
infiltration, reduce
evaporation, promote
decompose to provide
humus, protect soil from rain
and wind erosion. May cause
water logging under
continuous rainfall
Organic and inorganic
materials, e.g. wild grass,
crop residues or tree
biomass, either leguminous
or not), plastic sheeting, rock
Residues/mulch must provide at
least 30% soil cover (Unger,
Stewart, Parr, & Singh, 1991);
conflict between livestock feed
and mulch from crop residues
Nutrient only D 263 Improve soil nutrient, and
water holding and cation
exchange capacities. Crop
response to nutrients
depends primarily on
available moisture and is
poor under low seasonal
rainfall
Organic (manure) and
inorganic nutrient
components
Collect/ buy fertilizer or manure.
Station nutrient placement
method increases nutrient use
efficiency under smallholder
farming
Water harvesting
with storage
E 73 Improve water supply for the
plant, livestock and domestic
use. Can bridge and
smoothen seasonal water
supply variability
Full or supplemental irrigation Manual and mechanical capture of
rainfall in surface and
subsurface reservoirs, soil
profile, and groundwater
aquifers for irrigation
application
Cropping system
and
agroforestry
F 237 Improve soil fertility, regulate
nutrient cycling, reduce soil
erosion, increase labour
productivity and reduce the
risk of complete crop failure
due to water stress and pest
and disease
Trees and crops: crop rotations
& intercropping e.g.
Sorghum-cowpea, millet-
cowpea, maize-cowpea, and
sorghum-groundnuts
associations
The practise of intercropping of
annual and perennial crops with
trees and/or bushes with
beneficial use for crop-livestock
farming systems and infertile
soils prone to erosion (Van
Duivenboodew et al., 2000)
Combination of
two or more
interventions
G 445 Promote infiltration, conserve
moisture, enhance soil micro
flora and fauna in soil
Planting basin + fertility
treatment, reduced tillage +
mulch (conservation
agriculture)
Manual, mechanized and animal
draft power
Farmer/
conventional
practise
1485 Low input requirement,
promotes infiltration,
conserve moisture in soil,
organic matter mineralization
and destroy soil micro flora
and fauna, resulting in poor
soil structure and enhanced
sheet and wind erosion
Planting on flat ploughed
surface or on-line planting;
no use or minimal use of
nutrients
Manual, mechanized and animal
draft power
8 M. S. MAGOMBEYI ET AL.
duration of study were 1 year (17%: 256), 2 years (22%:
332), 3 years (27%: 404), 4 years (29%: 435), 5 years
(3.5%: 53) and above 5 years (1.3%: 11), indicating
that more than 80% of the paired cases had been
tested for more than two years. The proportion of
on-farm cases were highest with 870 cases (57%), fol-
lowed by on-station with 555 cases (37%) and model-
ling with 70 cases (5%).
3.2. Yield response for different AWM
technology groups
The overall value of SMD for yield was 0.27with a 95% CI
of 0.18–0.35, indicates an opportunity to enhance crop
production with AWM technologies in the basin
(Figure 4). The efficacy of AWM technologies was
highest for water harvesting with storage – group E
(SMD = 0.53), followed by in situ water retention –
group B (SMD = 0.38) and a combination of two or
more AWM technologies – group G (SMD= 0.31). This
result indicates the critical aspect of securing water for
improved crop yield and production in the Limpopo
agro-ecological landscapes. The widest variation in
yield observed for group E is indicative of the huge
potential of yield gains that can be realized from this
technology. Evaporation suppressants alone (group C)
had the least effect on the improvement of crop yield
(SMD = 0.07). Negative yield impacts were observed
for groups A, C and F (Figure 4), indicating unstable per-
formance of some AWM technologies because of the
complex relationship between farm management
practices and the soil–rainfall–crop system. The
median yield gains (t ha−1) and their marginal
changes in relation to conventional practise confirm
the performance ranking of AWM groups (Table 2).
Overall, yield performance comparison showed
that technologies implemented on-station (511
cases) performed better than those from on-farm
(758 cases). The median yield gains from on-station
were 63% (control of 1.23 t ha−1 and experiment of
2 t ha−1), while yield gains from on-farm were 29%,
with control of 1 t ha−1and experiment of
1.29 t ha−1. Water productivity improvement was
also higher under technologies implemented on-
station. This result showed reduced crop yield and
productivity performance due to poor replication of
on-station conditions at the farm level.
The low value of I2 (Figure 4) suggest low hetero-
geneity among the seven AWM groups, while the
p-value indicates statistically insignificant results
(Higgins & Green, 2008; Rücker et al., 2008).
However, Rücker et al. (2008) argue that I2 is gener-
ally limited in assessing clinically relevant heterogen-
eity and it is good practice to also look at efficacy
and covariate sizes.
The yield increase from AWM technologies com-
pared to the control (figures given in brackets) and
rainfall regimes are presented in Table 2. Median
yield increase ranged 9–95% (mean 34%), with an
absolute median yield range of 1.40–2.20 t ha−1 (not
shown). Table 2 indicates that although group E and
group A produced the highest absolute median
Figure 3. Paired case studies identified for each AWM technology group.
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 9
yields of 2.20 t ha−1 (control 1.00 t ha−1) and
1.78 t ha−1 (control yield 1.58 t ha−1), respectively,
the range of yield gains of the latter from low to
medium rainfall regimes is marginal compared to
the former. This result indicates that group E technol-
ogy achieves higher yield gains when annual rainfall is
low, and this is particularly significant in the semi-arid
environment of the basin. The nutrient only (D) and in-
situ water retention (B) groups seemed to perform
equally in both low and medium rainfall regimes.
The funnel plot of standard error (SE) against the
SMD in yield is presented in Figure 5, and it indicates
a symmetric scatter for the seven AWM groups about
the overall SMD of 0.27. This result suggests consider-
able strength of evidence that the number of cases
included for the analysis of the crop yield is comprehen-
sive enough to ensure less chance of bias and between
group heterogeneity (Schünemann et al., 2011).
3.3. Water productivity for different AWM
groups
The total water (rainfall and rainfall plus supplemental
irrigation) crop productivity variation for the seven
Figure 4. Synthesized grain yield (t·ha-1) change for 7 AWM technology groups in the Limpopo Basin. The diamond symbol represents the mean,
the horizontal axis of the diamond represent the 95% CI.
Table 2. Summary of AWM technologies and expected yields increase across all crops and rainfall regimes.
AWM technology groups
Median yield
(control) (t ha−1)
Median yield
increase (%)
Median yield increase per rainfall
regime 200–500 mm (501–800 mm) (%)
Water harvesting with storage: E (n = 58) 2.2 (1.0) 95 28 (129)
Cropping system and agroforestry: F (n = 167) 1.42 (0.95) 34 50 (11)
In-situ water retention: B (n = 190) 1.44 (1.11) 33 32 (35)
Combination of two or more interventions: G (n = 428) 1.59 (1.06) 25 42 (18)
Nutrient only: D (n = 247) 1.36 (1.1) 31 21 (23)
Reduced tillage: A (n = 83) 1.78 (1.5) 12 7 (18)
Evaporation suppressants: C (n = 115) 1.4 (1.29) 9 16 (30)
10 M. S. MAGOMBEYI ET AL.
AWM groups is presented in Figure 6. The best per-
forming AWM technology in terms of water pro-
ductivity is nutrient only – group D (SMD = 0.9)
followed by in situ water retention – group B (SMD
= 0.71) and a combination of two or more technol-
ogies – group G (SMD = 0.29). Group G showed the
largest variations, indicating a wide opportunity to
increase productivity by this technology. The evapor-
ation suppressants (Group C) had too few water pro-
ductivity data to do any statistical analysis, and
water harvesting with storage – group E (SMD =
0.21) resulted in marginal enhancement of water pro-
ductivity. It can be conjectured that there could be
untapped water productivity potential with some
AWM technologies that showed large water pro-
ductivity variations, while others could be already per-
forming in their optimum range in the basin. The
overall water productivity efficacy of SMD = 0.46 indi-
cates that AWM technologies (experiment) are more
effective than control treatment.
The funnel plot of SE against the SMD from
water productivity is presented in Figure 7 for the
six AWM groups (Figure 6) which had sufficient
data for the analysis. It is observed that there is
asymmetry in the scatter for these AWM technol-
ogies about the overall SMD of 0.46, reflecting a
Figure 5. Funnel plot from the seven AWM intervention groups on
cereal crop yield.
Figure 6. Water productivity (kg mm−1) variations for the seven AWM technology groups. The diamond symbol represents the summary effect
measure, the vertical axis of the diamond represents the point estimate and the horizontal axis of the diamond represents the 95% CI.
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 11
reduced strength of evidence that the cases
included for the analysis are not comprehensive
enough, thereby warranting additional studies in
the future to support the efficacy of AWM technol-
ogies on crop water productivity.
3.4. Yield responses from AWM groups under
different rainfall regimes
The performances of various sub-groupings of AWM
technologies in low and medium rainfall regimes are
presented in Figure 8. The I2 (Figure 8) showed sub-
stantial technological heterogeneity, while the p-
value indicated statistically significant results
(Higgins & Green, 2008; Rücker et al., 2008). The sub-
group of crop rotation plus mulch from group G per-
formed best in both low (SMD = 2.01) and medium
(SMD = 3.7) rainfall regimes, followed by planting
basin plus fertilizer. This suggests the importance of
crop rotation in providing a balanced soil nutrient
for the following crop unlike application of inorganic
fertilizer. There are no consistent trends in the per-
formances of other AWM sub-groups from the low
to medium rainfall regimes. This result indicates that
the yield potentials for some AWM technologies
could be substantial (SMD = 5.1) and marginal (SMD
= –1) for others, depending on the complex soil–rain-
fall–crop system that has to be understood by
farmers/researchers in applying these technologies.
There is marginal increase in effectiveness of AWM
technologies to improving yields in low rainfall
regime (SMD = 0.51) compared to the medium rainfall
regime with an overall SMD value of 0.39. These results
concur with Rusinamhodzi et al. (2011) who reported
significant yield increases under reduced tillage for a
lower rainfall regime of less than 600 mm compared
to a higher rainfall regime of 600–1000 mm. Hussain,
Olson, and Ebelhar (1999) also reported yield
decreases of 5–20% in wet years and 10–100%
increases in relatively dry years under conservation
agriculture compared to conventional tillage practices.
To strengthen the evidence of the influence of rainfall
in AWM technologies, intra-season or crop season
rainfall distribution should be collected in future
studies.
3.5. Yield performance of individual
technologies within groups D and F
The analysis breaks down the average performance
of an AWM group to an individual technology
within that group. A comparison of efficiency of fer-
tilizer alone with SMD of 0.20 in yield to intercrop-
ping with leguminous crops with SMD of 0.03,
indicates that intercropping has little effect on yield
(Table 3). The efficiency of organic manure only
was better (SMD = 0.30) compared to different
levels of application of inorganic fertilizers which pro-
duced values of SMD between 0.18 and 0.21. Micro-
dosing of inorganic fertilizers that did not exceed
35 kg ha−1 performed better (SMD = 0.21) than
large quantities of fertilizer of 36–300 kg ha−1 with
SMD of 0.18. This result underscores the significance
of adequate soil-water availability in order to realize
the efficiency of fertilizer on crop yield. The results
in Table 3 also indicate a considerable increase in
yield when the frequency of weeding per growing
season is increased, as weeds compete with crops
for nutrients and water.
The effect of evaporation suppressants (Group C)
on yield showed that large quantities of mulch of
4.00–10.00 t ha−1 (125 cases) enhance yield (SMD =
0.15) compared to less mulch of between 0.50 and
3.00 t ha−1 (102 cases), which gave a value of SMD
of 0.02. For water harvesting with storage (Group E)
based on 170 cases, the efficacy of supplemental irri-
gation was highest (SMD = 1.17) compared to full
sprinkler (SMD = 0.71) and full furrow (SMD = 0.36)
irrigation types. This result suggests that it is not
only the irrigation infrastructure that is important
Figure 7. Funnel plot from the six AWM intervention groups on water
productivity.
12 M. S. MAGOMBEYI ET AL.
Figure 8. Standardized yield difference for all AWM technologies in different rainfall regimes.
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 13
but the availability and effective management of
soil–water during critical long dry spells of the crop-
ping season.
3.6. Yield responses in different soil textures
AWM technologies have the highest likelihood of pro-
ducing good yields in silty clay loam soils which are
part of the cultivated basin area of 2,20,000 km2
(53% of basin area) as reported in Limpopo River
Awareness Kit (2011) than in other soils (e.g. clay
loam, loamy sandy and sandy) with clay soils present-
ing the greatest challenge to these technologies. Silty
clay loam soils cover an approximate area of
46,000 km2 (11%) of the basin area (Bangira & Manye-
vere, 2009) and present great potential for expansion
and intensification of crop production to the areas
that are currently uncultivated. These results agree
with Rusinamhodzi et al. (2011) who reported mostly
negative yield changes for clay soils but positive
yield changes in both loam and sandy soils when con-
ventional tillage and reduced tillage plus mulch were
compared. The yield change frequency histogram for
different soil textures for the combined AWM technol-
ogy, group G is shown in Figure 9. For sandy soil, the
modal frequency of 25% is observed for the yield
change classes of 0–25% and – (0–25%), while for
sandy loam, the modal frequency of 19% is for yield
change classes of 26–50%, 0–25% and – (0–25%).
For clay, the modal frequency of 37% is for yield
change class of – (0–25%) and for silty clay loam,
the modal frequency of 97% is for yield change class
of >75%.
3.7. Effect of site potential on yield responses
The SMD for yield changes for all AWM technologies in
relation to crop production potential at control sites is
presented in Figure 10. The I2 (Figure 10) shows mod-
erate heterogeneity among the AWM technologies,
while the p-value indicates statistically significant
results (Rücker et al., 2008). Using the mean yield cat-
egories at control sites of low (<0.50 t ha−1), medium
(0.50–2.00 t ha−1) and high (>2.00 t ha−1), there is
greater yield increase at low control yield sites than
at sites with medium and high control yields. The
high yield gains from AWM technologies at low yield
control sites indicate that AWM technologies increase
yields in large areas of low-yielding environments
occupied by smallholder farmers in the basin. For
maize crop, the control produced low category
mean yield of 0.35 t ha−1, while AWM technologies
produced a mean yield of 0.75 t ha−1. For the
medium yield category, the control was 1.10 t ha−1,
while for AWM technologies it was 1.64 t ha−1, and
in the high yield category, the control was
3.30 t ha−1, while for AWM technologies it was
3.2 t ha−1. Similarly for sorghum, in the low, medium
and high categories, the control was 0.30, 1.21,
2.78 t ha−1, respectively, while for AWM technologies
it was 0.45, 1.52 and 2.77 t ha−1, respectively. For
millet, reviewed cases were only obtained for the
low and medium categories which for the control
was 0.31 and 0.60 t ha−1, respectively, while for
AWM technologies it was 0.60 and 0.80 t ha−1,
respectively. There are generally little benefits from
AWM technologies when conventional practice
produced high yields of greater than 2.00 t ha−1
(SMD = –0.05).
3.8. Risk to yield response for different AWM
technologies
Essentially, all AWM technologies provide a 90%
chance to increasing yields in the 1.00–3.00 t ha−1
yield range. Only AWM technology groups of
Table 3. SMD effect size for individual AWM technologies from groups D and F.
Subgroup Type of AWM Number of cases Effect Estimate SMD [95% CI]
Nutrient only (Group D) 6 818 0.21 [0.07, 0.35]
Fertilizer only 1 342 0.20 [−0.01, 0.41]
Fertilizer less or equal to 35kg·ha-1 1 212 0.21 [−0.06, 0.48]
Fertilizer 36 – 300 kg·ha-1 1 126 0.18 [−0.17, 0.53]
Manure only 1 62 0.30 [−0.21, 0.80]
Weeding once per season 1 38 0.01 [−0.62, 0.65]
Weeding 2–3 times per season 1 38 0.50 [−0.15, 1.14]
Cropping system (Group F) 2 112 0.14 [−0.23, 0.51]
Intercropping 1 52 0.03 [−0.52, 0.57]
Improved seed variety 1 60 0.23 [−0.27, 0.74]
Note: CI is confidence interval; SMD: <0.4 represents a small effect, 0.4–0.7 moderate and >0.70 a large effect.
14 M. S. MAGOMBEYI ET AL.
intercropping and agroforestry (F) and water harvest-
ing with storage (E) provide higher reliability of yield
gains than conventional practice for all yield levels.
These results are deduced from the plot of the cumu-
lative probability distribution for the seven AWM tech-
nology groups and the controls for all 1430 paired
cases presented in Figure 11. The probability of
obtaining grain yields lower or equal to the control
constitutes a risk to farmers if the performance of a
technology, on average, across a wide range of con-
ditions, is used to give recommendations. The curve
of water harvesting with storage (Group E) is farthest
to the right of the control cumulative distribution
curve and, therefore, represents the most resilient
AWM technology that exposes the least risk of yield
failure to farmers (Figure 11). However, it is important
to note that all the AWM technologies reduce yield
failure risk to some extent for yield ranges of 0.50–
2.50 t ha−1.
3.9. Impacts of AWM technologies on runoff
and sediment loss
The environmental impacts of AWM technologies
from groups A (Reduced tillage), B (In situ water reten-
tion), and G (a combination of two or more technol-
ogies) are represented by the median sediment loss
reduction of 78% (65 cases) and runoff reduction of
62% (51 cases), compared to conventional farmer
practise (Figure 12). The influences of rainfall regime
Figure 9. Yield change frequencies in different soil textures for combined AWM technology, group G.
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 15
Figure 10. Yield changes under all AWM intervention groups for control sites with different yield potentials. The diamond symbol represents the
summary effect measure, the vertical axis of the diamond represents the point estimate and the horizontal axis of the diamond represents the 95% CI.
16 M. S. MAGOMBEYI ET AL.
on the environmental impacts of these three AWM
technologies are presented in Figure 13. The median
runoff reduction is higher with higher rainfall,
whereas the median sediment reduction is about the
same for both rainfall regimes, suggesting a limit to
the effectiveness of sediment reduction.
3.10. Summary of findings
The results presented in Table 4 from the GRADEpro
GDT software show high strength of evidence that
the SMD of yield between conventional and AWM
technologies is 0.27, while there is moderate
strength of evidence to support water productivity,
with SMD of 0.46. This confirms the results of the
funnel plots for yield and water productivity pre-
sented earlier.
3.11. Discussions
The results presented from 97 references with 1430
paired cases of AWM technologies in the Limpopo
Basin provide some indication of their relative per-
formances and major consistent trends without any
attempt to explain each individual variation. Of signifi-
cance to smallholder farmers is the improvement in
production, measured by crop yield and water pro-
ductivity, and environmental protection, assessed by
reduced sediment and field runoff losses.
The AWM technologies can enhance regional and
local food security through adaptation to expected
climate change and maximization of crop water pro-
ductivity. An overall assessment of the seven AWM
technologies, based on crop yield and water pro-
ductivity, showed that water harvesting with storage
(E) and in situ water retention (B) out-perform the
others, although intercropping and agroforestry (F)
and a combination of two or more AWM technologies
(G) gave consistently good yield results because of the
synergy effect. However, in spite of the synergy, G per-
formed lower than in situwater retention (B) and nutri-
ent management (D) (Figure 6). The synergy in this
group G is masked by other low yielding AWM tech-
nology combinations found in this group such as
ripper plus mulch or crop rotation plus mulch. The
reduced tillage improved median yield by 0.24 t ha−1
and in situ water retention by 0.33 t ha−1 when com-
pared with the control (Table 2). This result is
Figure 11. Cumulative distributions of yield under seven AWM technology groups and control across all crops.
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 17
consistent with Mafongoya et al. (2016) who reported
yield increases from direct seeding, rip-line seeding
and seeding into planting basins by 0.445, 0.258 and
0.241 t ha−1, respectively.
Water harvesting with storage (E) that resulted in
median yield increases of 34–95% provides the least
risks to farmers in terms of coping with the inherent
rainfall variability of the LRB. Some limitations of not
Figure 12. Change in runoff and sediment generated for the combined AWM technology group. The diamond symbol represents the mean, the
horizontal line in the box the median, the upper and lower boundaries of the box are the 25th and 75th percentiles, and the upper and lower
ends of the extended lines represent the minimum and maximum values of the data.
Figure 13. Sediment and runoff reduction for combined AWM technology group in different rainfall regimes. The diamond symbol represents
the mean, the horizontal line in the box the median, the upper and lower boundaries of the box are the 25th and 75th percentiles, and the upper
and lower ends of the extended lines represent the minimum and maximum values of the data.
18 M. S. MAGOMBEYI ET AL.
Table 4. Summary of findings.
AWM technologies versus farmer practice for increasing crop yield and water productivity in semi-arid areas
Patient or population: Subsistence farming with increasing crop yield and water productivity in semi-arid areas
Settings: Arid and semi-arid subsistence farming
Intervention: Improved AWM technologies versus farmer practice
Outcomes Illustrative comparative risks* (95% CI) Relative
effect
(95% CI)
No of
Participants
(AWM
Groups)
Quality of the
evidence
(GRADE)
Comments
Assumed risk Corresponding risk
Control Improved agricultural water management technologies versus
farmer practice
Cereal crop yield SMD for yield in all technology groups was
0.27 indicating AWM had small effect on yield
2517
(7 groups)
⊕⊕⊕⊕
high
AWM technologies are recommended
Water productivity SMD for water productivity in all technology groups was 0.46
indicating AWM had moderate effect on increasing water
productivity
242
(7 groups)
⊕⊕⊕⊝
moderate
113 cases and 129 controls – control
studies. AWM are recommended
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison
group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; SMD: < 0.4 represents a small effect, 0.4–0.7 moderate and > 0.70 a large effect (Schünemann et al. (2011).
GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.
IN
TERN
A
TIO
N
A
L
JO
U
RN
A
L
O
F
A
G
RIC
U
LTU
RA
L
SU
STA
IN
A
BILITY
19
having actual seasonal rainfall data nor rainfall distri-
bution to correlate to yield and water productivity,
and quite low number of cases that measured
erosion (67 cases) and runoff (51 cases) were con-
sidered to increase uncertainty of the results. This
lack of data suggests the need for improved field
monitoring to capture these aspects.
Understanding the bio-physical and socio-econ-
omic contexts of smallholder farmers is crucial in iden-
tifying the most appropriate suite of AWM
technologies, as the results indicate that every tech-
nology has relative chances of success and failure
depending on soils, rainfall regime, nutrient, water
and farmer management practices. For instance, the
decrease in median yields for cropping system and
agroforestry (F) and combination of two or more tech-
nologies (G) with an increase in rainfall regime implies
that farmers have to pay closer attention to the man-
agement of soil-water at higher rainfall to avoid water-
logging, crusting and leaching of nutrients and
pesticides. Similarly, the challenges posed by clay
soils imply that smallholder farmers need better man-
agement strategies to improve production and pro-
ductivity. Hence, the onus is on researchers to make
additional effort in understanding when and where
AWM technologies work.
The assessment showed large yield variations with
different AWM technologies, indicating that higher
yield gains, which are attainable, present huge
promise for meeting the substantial yield gaps that
currently exist in conventional farming practice. It is
expected that with further technology development,
continued on-farm experiments of different technol-
ogies and appropriate knowledge transfer mechan-
isms and support including enabling market-policy
conditions, these technologies can readily be
adopted by smallholder farmers in the basin.
The reductions of runoff and sediment loss are
important in partitioning the total precipitation,
maintaining soil stability, and conserving the inherent
and applied nutrients in agricultural fields in semi-
arid areas (Mzezewa and van Rensburg 2011). Further-
more, AWM technologies partition rainfall in a way
that increases soil–water infiltration and reduces evap-
oration rate to enhance crop soil–water availability to
bridge the intra-seasonal dry spells for increased yields
and water productivity (Moroke, Dikinya, & Patrick,
2009). Field runoff losses above 50% of the rainfall
from untilled bare lands, which constitute an erosion
hazard, have been reported in the basin (Mupangwa,
Twomlow, & Walker, 2012), while comparable runoff
losses up to 46% have been reported in Ethiopia
(Zere, Hensley, & Van Huyssteen, 2006) and 25–30%
under conventional tillage in SSA (Rockström, 2000).
High soil losses ranging from 10.00–50.00 t ha−1 y−1 in
both low and high rainfall zones have resulted in low
productivity in over 25% of the smallholder areas in
Zimbabwe (Nyamadzawo, Nyamugafata, Wuta, Nya-
mangara, & Chikowo, 2012; Vogel, 1992; Whitlow &
Campbell, 1989). The added benefit of AWM technol-
ogies in climatic adaptation through the carbon seques-
tration requires further investigation.
We suggest that better institutional capacity and
better context-specific advice to farmers could
enable successful AWM up-scaling. Renewed efforts
and commitment by CAADP (2017) and African
Union (2014) through the Malabo Declaration have
addressed the policy issues. However, the constraint
to farmers to accessing affordable technologies that
suit their context despite climatic variability still
persist. This was also discussed in Bulcock and
Jewitt (2013) who noted the need for a new set of
guidelines that are broader than the current one
for the uptake of water harvesting. There is a
general weakness in documentation and communi-
cation of the experiences and lessons learned
under AWM technologies in the basin. Hence, this
paper contributes to share knowledge and best
practices on AWM technologies, their capabilities
and limitations under diverse conditions.
There is a need to support farmers with appropriate
AWM technologies based on soil and rainfall regimes.
Support to accessing appropriate credit facilities and
micro-finance to invest and reduce economic risk in
view of the prevailing climatic risk and variability is
also needed. This support may include contract
farming, crop insurance and farmers participating in
inputs and/or outputs markets through farmer organ-
izations to reduce transaction costs (Markelova,
Meinzen-Dick, Hellin, & Dohrn, 2009; Nyagumbo, Mut-
samba, Barrett, Dengu, & Thierfelder, 2012). Further-
more, provision of advice on seasonal rainfall regime
forecast to provide early-warning systems to support
implementation of appropriate AWM technologies
such as planning for fertilizer and supplemental irriga-
tion application (Van Duivenboodew, Paln, Studer,
Bielders, & Beukes, 2000) is required.
4. Conclusions
This meta-analysis has captured for the first time the
diversity of environments in which AWM technologies
20 M. S. MAGOMBEYI ET AL.
among smallholder farmers have taken place in the
LRB, and helped to identify the potential biophysical
zones where studied technologies are being
researched and applied. The meta-analysis showed
that the environment played an important role in
determining the relative agricultural crop yield pro-
duction level for AWM technologies. Overall there is
moderate to high evidence that AWM technologies
can deliver substantial benefits of climate-smart agri-
culture in terms of increased crop yield and water pro-
ductivity to smallholder farmers, whilst attaining
desired environmental impacts through retention of
sediment and runoff. Yield stability analysis showed
that under prolonged drought or very high rainfall
conditions, no AWM technology, except water har-
vesting with storage (Group E) can offset the effects
of these extreme conditions. Furthermore, no single
AWM technology fits all circumstances to achieve sus-
tainable smallholder agricultural production. Evidence
suggests that a combination of two or more AWM
technologies (G) and intercropping and agroforestry
(F) demonstrate high yield opportunity in low rainfall
regimes. The efficiency of organic manure only was
better (SMD = 0.3) compared to different application
levels of inorganic fertilizers which produced values
of SMD between 0.18 and 0.21. Micro-dosing with
inorganic fertilizers less than 35 kg ha−1 performed
better than large quantities of fertilizer of 36–
300 kg ha−1, suggesting the need for adequate soil-
water availability in order to realize the efficacy of fer-
tilizer on crop yield.
Variations in reported yields within each AWM
group suggest that there are still potential opportu-
nities to increase yields beyond average values for
enhanced food security and income generation.
Attempts to replicate AWM successes at specific
sites should be earnestly pursued and appropriately
targeted to climatic and edaphic conditions with
adequate inputs (fertilizers, seeds, and herbicides)
and correct timing of farming operations for best
results. Success of AWM technologies depends on
the transformation of conventional practices
through social learning and participatory approaches
between farmers, donors, researchers, and prac-
titioners in the public and private sectors. Given the
negative implications of climate change on agricul-
tural production in SSA including the Limpopo
Basin (IPCC, 2014), AWM technologies can be
implemented as a viable strategy to build resilience
and food security for people living in high rainfall
variability areas.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This work was supported by The CGIAR Challenge Program on
Water and Food (CPWF) and and the CG Research Program
’Water, Land and Ecosystems’ (WLE) and CGIAR Fund Donors:
[Grant Number CPWF: 2010-2013].
References
The African Centre for Biodiversity. (2016). Mapping farmer
seed varieties in Manica, Mozambique: Report on initial
investigations into agricultural biodiversity. The African
Centre for Biodiversity. Johannesburg, South Africa. www.
acbio.org.za.
African Union. (2014). Malabo Declaration on Africa Accelerated
Agricultural Growth and Transformation (3AGT) for shared
prosperity and improved livelihoods Doc. Assembly/AU/2
(XXIII). Addis Ababa, Ethiopia.
Alemaw, B. F., & Simalenga, T. (2015). Climate change impacts
and adaptation in rainfed farming systems: Amodeling frame-
work for scaling-Out climate smart agriculture in Sub-Saharan
Africa. American Journal of Climate Change, 4, 313–329. doi:10.
4236/ajcc.2015.44025.
Andrews, J. C., Schünemann, H. J., Oxman, A. D., Pottie, K.,
Meerpohl, J. J., Coello, P. A.,… Guyatt, G. (2013). GRADE guide-
lines: 15. Going from evidence to recommendation-determi-
nants of a recommendation’s direction and strength. Journal
of Clinical Epidemiology, 66, 726–735.
Bangira, C., & Manyevere, A. (2009). WaterNet Working Paper 8:
Baseline Report on The Soils of the Limpopo River Basin a con-
tribution to the Challenge Program on Water and Food
Project 17 “Integrated Water Resource Management for
Improved Rural Livelihoods: Managing risk, mitigating drought
and improving water productivity in the water scarce Limpopo
Basin”. Harare, Zimbabwe. www.waternetonline.org.
Bayala, J., Sileshi, G. W., Coe, R., Kalinganire, A., Tchoundjeu, Z.,
Sinclair, F., & Garrity, D. (2012). Cereal yield response to con-
servation agriculture practices in drylands of West Africa: A
quantitative synthesis. J. Arid Environ, 78, 13–25.
Benin, S., Nin Pratt, A., Wood, S., & Guo, Z. (2011). Trends and
Spatial Patterns in Agricultural Productivity in Africa, 1961–
2010. ReSAKSS Annual Trends and Outlook Report 2011.
International Food Policy Research Institute (IFPRI).
Bulcock, L. M., & Jewitt, G. P. W. (2013). Key physical characteristics
used to assess water harvesting suitability. Physics and Chemistry
of the Earth, 66, 89–100. doi:10.1016/j.pce.2013.09.005.
Bureau for Food and Agricultural Policy (BFAP). (2014).Maize trust
progress report & results the BFAP farming systems program.
South Africa: Pretoria. http://www.maizetrust.co.za/upload/
WEBSITE/Market%20Information/2014/20160211BFAP%
20Progress%20Report%20and%20Results%202014.pdf.
Carberry, P. S., Liang, W.-L., Twomlow, S., Holzworth, D. P., Dimes,
J. P., McClelland, T.,… Keating, B. A. (2013). Scope for
improved eco-efficiency varies among diverse cropping
systems. PNAS, 110(21), 8381–8386.
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 21
The Cochrane Collaboration. (2014). Review manager (RevMan)
[computer program]. version 5.3. Copenhagen: The Nordic
Cochrane Centre.
Cohen, J. (1988). Statistical power analysis in the behavioural sciences
(2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Comprehensive Africa Agriculture Development Programme
(CAADP). (2017). 13th Comprehensive Africa Agriculture
Development Programme (CAADP) Partnership Platform
themed, ‘Strengthening Mutual Accountability to Achieve
CAADP/Malabo Goals and Targets”. Kampala, Uganda. www.
au.int.
Comprehensive Assessment of Water Management in
Agriculture (CAWMA). (2007). Water for Food, Water for Life:
A Comprehensive Assessment of Water Management in
Agriculture. London: Earthscan, and Colombo: International
Water Management Institute.
Consultative Group on International Agricultural Research
(CGIAR). (2003). Limpopo Basin Profile. Accessed at: http://
www.arc.agric.za/limpopo/pdf/basin_profile_intro.pdf.
Accessed on February 2, 2010.
Cooper, P. J. M., Dimes, J., Rao, K. P. C., Shapiro, B., Shiferaw, B., &
Twomlow, S. (2008). Coping better with current climatic
variability in the rain-fed farming systems of sub-Saharan
Africa: Anessentialfirst step inadapting to future climate change?
Agricultural, Ecosystems and Environment, 126, 24–35.
DEWFORA. (2011). Drought Early Warning and Forecasting to
strengthen preparedness and adaptation to droughts in
Africa (DEWFORA). A 7th Framework Programme
Collaborative Research Project. Work Package 2 Assessing
existing drought monitoring and forecasting capacities, miti-
gation and adaptation practices in Africa.
FAO and DWFI. (2015). Yield gap analysis of field crops –Methods
and case studies, by Sadras, V.O., Cassman, K.G.G., Grassini, P.,
Hall, A.J., Bastiaanssen, W.G.M., Laborte, A.G., Milne, A.E.,
Sileshi, G., Steduto, P. FAO Water Reports No. 41, Rome,
Italy. http://www.fao.org/3/a-i4695e.pdf.
Farooq, M., Flower, K. C., Jabran, K., Wahid, A., & Siddique, K. H. M.
(2011). Crop yield and weed management in rainfed conser-
vation agriculture. Soil & Tillage Research, 117, 172–183.
Feed the Future. (2013). Smallholder Farmers in Mozambique
Reap the Benefits of Conservation Agriculture. US
Government’s Global Hunger and Food security Initiative.
Feed the Future Newsletter, November 22, 2013. https://
feedthefuture.gov/article/smallholder-farmers-mozambique-
reap-benefits-conservation-agriculture.
Food and Agriculture Organisation (FAO). (2004). Drought Impact
Mitigation and Prevention in the Limpopo River Basin: a situ-
ation analysis. Land and Water Discussion Paper 4. Rome,
Food and Agriculture Organization of the United Nations.
Food and Agriculture Organization (FAO). (2007). Adaptation to
climate change in agriculture, forestry, and fisheries:
Perspective, framework and priorities. Rome, Italy: Food and
Agriculture Organization of the United Nations.
Food and Agriculture Organization (FAO). (2009). Scaling-up
Conservation Agriculture in Africa: Strategy and Approaches.
The FAO Sub-regional Office for Eastern Africa. Thiombiano,
L and Meshack, M (Edits). Addis Ababa, Ethiopia.
Friis-Hansen, E. (1992). The performance of the seed sector in
Zimbabwe: an analysis of the influence of organisational
structure. Working paper 66. ISBN 0 85003 1842, Overseas
Development Institute, London.
Guilpart, N., Grassini, P., Sadras, V. O., Timsina, J., & Cassman, K. G.
(2017). Estimating yield gaps at the cropping system level.
Field Crops Research, 206, 21–32.
Hatfield Consultants. (2008). Regional Aquatics Monitoring
Program (RAMP). Accessed at: http://www.ramp-alberta.org/.
Accessed on: March 14, 2014.
Higgins, J. P. T., Altman, D. G., & Sterne, J. A. C. (2011). Chapter 8:
Assessing risk of bias in included studies. In J. P. T. Higgins, & S.
Green (Eds.), Cochrane handbook for systematic reviews of
interventions version 5.1.0 (updated march 2011). www.
cochrane-handbook.org.
Higgins, J. P., & Green, S. (2008). Cochrane handbook for systema-
tic reviews of interventions version 5.0.1. 2008. London: The
Cochrane Collaboration. http://www.cochrane-handbook.org
Hussain, I., Olson, K. R., & Ebelhar, S. A. (1999). Impacts of tillage
and no-till on production of maize and soybean on an eroded
illinois silt loam soil. Soil & Tillage Research, 52, 37–49.
IIASA/FAO. (2012). Global Agro-ecological Zones (GAEZ v3.0).
IIASA, Laxenburg, Austria and FAO, Rome, Italy.
Improved Management of Agricultural Water in Eastern and
Southern Africa (IMAWESA). (2009). Water for Agriculture is
Bankable: Experiences from Eastern and Southern Africa.
Conference on: Enhancing the Productivity of High Value
Crops and Income Generation with Small-Scale Irrigation
Technologies. March 31, 2009. KARI, Nairobi, Kenya.
Intergovernmental Panel on Climate Change (IPCC). (2007). State
of the science. Contribution of working group I to the fourth
assessment report of the intergovernmental panel on climate
change. Cambridge: Cambridge University Press.
Inter-governmental Panel on Climate Change (IPCC). (2012).
Summary for Policy-makers. In: Managing the Risks of
Extreme Events and Disasters to Advance Climate Change
Adaptation [Field CB, Barros V, Stocker TF, Qin D, Dokken
DJ, Ebi KL, Mastrandrea MD, Mach KJ, Plattner G-K, Allen SK,
Tignor M and Midgley PM (eds.)]. A Special Report of
Working Groups I and II of the Intergovernmental Panel on
Climate Change. Cambridge University Press, Cambridge,
UK, and New York, NY, USA, pp. 3–21.
Inter-governmental Panel on Climate Change (IPCC). (2014).
Chapter 22: Africa. Climate change 2014: Impacts, adaptation
and vulnerability, contribution of working group II to the fifth
assessment report of the intergovernmental panel on climate
change. Cambridge: Cambridge University Press.
International Institute of Rural Reconstruction and African
Conservation Tillage Network (IIRR and ACT). (2005).
Conservation agriculture: A manual for farmers and extension
workers in Africa. International Institute of Rural
Reconstruction (IIRR), Nairobi; African Conservation Tillage
Network (ACT), Harare. ISBN 9966-9705-9-2.
Jat, R. A., Wani, S. P., & Sahrawat, K. L. (2012). Chapter Four –
Conservation Agriculture in the Semi-Arid Tropics: Prospects
and Problems. In: Advances in Agronomy. Advances in
Agronomy, 117. Academic Press, pp. 191–273. ISBN 978-970-
12-394278-4.
Limpopo Basin Focal Project (LBFP). (2010). Poverty in the
Limpopo Basin. Report submitted to the Challenge Program
on Water and Food, phase 1.
Limpopo Basin Permanent Technical Committee (LBPTC). (2010).
Joint Limpopo River Basin Study – Scoping Phase Final Report
– Main Report January 2010. Prepared by BIGCON
Consortium. Mozambique.
22 M. S. MAGOMBEYI ET AL.
Limpopo River Awareness Kit. (2011). National policies and laws,
Mozambique. Website: www.limpoporak.org/.
Lobell, D. B., Burke, M. B., Tebaldi, C., Mastrandrea, M. D., Falcon,
W. P., & Naylor, R. L. (2008). Prioritizing climate change adap-
tation needs for food security in 2030. Science, 319, 607–610.
Mafongoya, P., Rusinamhodzi, L., Siziba, S., Thierfelder, C., Mvumi,
B. M., Nhau, B.,… Chivenge, P. (2016). Maize productivity and
profitability in conservation agriculture systems across agro-
ecological regions in Zimbabwe: A review of knowledge and
practice. Agriculture, Ecosystems and Environment, 220, 211–
225. ISSN 0167-8809.
Magombeyi, M. S., & Taigbenu, A. E. (2008). Crop yield risk analy-
sis and mitigation of smallholder farmers at Quaternary catch-
ment level: Case study of B72A in Olifants river basin, South
Africa. Physics and Chemistry of the Earth, 33, 744–756.
Magombeyi, M. S., Taigbenu, A. E., & Barron, J. (2016). Rural food
insecurity and poverty mappings and their linkage with water
resources in the Limpopo river basin. Physics and Chemistry of
the Earth, 92, 20–33.
Markelova, H., Meinzen-Dick, R., Hellin, J., & Dohrn, S. (2009).
Collective action for smallholder market access. Food Policy,
34, 1–7.
Marongwe, L. S., Kwazira, K., Jenrich, M., Thierfelder, C., Kassam,
A., & Friedrich, T. (2011). An African success: The case of con-
servation agriculture in Zimbabwe. International Journal of
Agricultural Sustainability, 9(1), 153–161.
Meinke, H., Nelson, R., Kokic, P., Stone, R., Selvaraju, R., &
Baethgen, W. (2006). Actionable climate knowledge: From
analysis to synthesis. Climate Research, 1(33), 101–110.
doi:10.3354/cr033101.
Milder, J. C., Majanen, T., & Scherr, S. J. (2011). Performance and
potential of conservation agriculture for climate change adap-
tation and mitigation in Sub-Saharan Africa. Eco-agriculture
Discussion Paper no. 6. Washington, DC: Eco-agriculture
Partners.
Milgroom, J., & Giller, E. (2013). Courting the rain: Rethinking sea-
sonality and adaptation to recurrent drought in semi-arid
Southern Africa. Agricultural Systems, 118, 91–104.
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). The
PRISMA group. Preferred reporting items for systematic
reviews and meta-analyses: The PRISMA statement. PLoS
Medicine, 6(7), 1–6. doi:10.1371/journal.pmed.1000097
Moroke, T. S., Dikinya, O., & Patrick, C. (2009). Comparative assess-
ment of water infiltration of soils under different tillage
systems in eastern Botswana. Physics and Chemistry of the
Earth, 34, 316–323.
Mueller, N. D., Gerber, J. S., Johnston, M., Ray, D. K., Ramankutty,
N., & Foley, J. A. (2012). Closing yield gaps through nutrient
and water management. Nature. doi.org/10.1038/
nature11420.
Müller, C., Cramer, W., Hare, W. L., & Lotze-Campen, H. (2011).
Climate change risks for African agriculture. PNAS, 108
(11), 4313–4315. www.pnas.org/cgi/doi/10.1073/pnas.1015
078108.
Munamati, M. (2009). Literature study on practical Agriculture
Water Management interventions in Limpopo Basin. Report
to the Limpopo Basin Focal Project (LBFP), December 2009.
Munamati, M., & Nyagumbo, I. (2010). In situ rainwater harvesting
using dead level contours in semi-arid southern Zimbabwe:
Insights on the role of socio-economic factors on performance
and effectiveness in gwanda district. Physics and Chemistry of
the Earth, 35, 699–705.
Mupangwa, W. (2009). Water and nitrogen management for risk
mitigation in semi-arid cropping systems. PhD Thesis, 351pp.
University of the Free State, Republic of South Africa.
Mupangwa, W., Dimes, J., Walker, S., & Twomlow, S. (2011).
Measuring and simulating maize (Zea mays L.) yield responses
to reduced tillage and mulching under semi-arid conditions.
Agr. Sci, 2(3), 167–174.
Mupangwa, W., Twomlow, S., & Walker, S. (2008). The influence of
conservation tillage methods on soil water regimes in semi-
arid southern Zimbabwe. Physics and Chemistry of the Earth,
33, 762–767.
Mupangwa, W., Twomlow, S., & Walker, S. (2012). Dead level con-
tours and infiltration pits for risk mitigation in smallholder
cropping systems of southern Zimbabwe. Physics and
Chemistry of the Earth Parts A/B/C, 47–48, 166–172.
Mzezewa, J., & van Rensburg, L. D. (2011). Effects of tillage on
runoff from bare clayey soil on a semi-arid ecotope in the
Limpopo Province of South Africa. Water SA, 37(2), 165–172.
Netnou-Nkoana, N. C., Jaftha, J. B., Dibiloane, M. A., & Eloff, J.
(2015). Understanding of the farmers’ privilege concept by
smallholder farmers in South Africa. South African Journal of
Science, 111(1/2), 1–5. doi:10.17159/sajs.2015/2013-0344
Nyagumbo, I., Mutsamba, E., Barrett, T., Dengu, E., & Thierfelder, C.
(2012). Prospects for up-scaling conservation agriculture in
Zimbabwe using animal traction mechanization technologies.
Harare: University of Zimbabwe.
Nyamadzawo, G., Nyamugafata, P., Wuta, M., Nyamangara, J., &
Chikowo, R. (2012). Infiltration and runoff losses under fallow-
ing and conservation agriculture practices on contrasting
soils, Zimbabwe. Water SA, 38, 233–240.
Owenya, M., Mariki, W., Stewart, A., Friedrich, T., Kienzle, J.,
Kassam, A.,…Mkomwa, S. (2012). Conservation Agriculture
and Sustainable Crop Intensification in Karatu District,
Tanzania. Integrated Crop Management Vol.15–2012. FAO,
Rome, Italy.
Oyedele, D. J., & Aina, P. O. (2006). Response of soil properties and
maize yield to simulated erosion by artificial topsoil removal.
Plant and Soil, 284(1–2), 375–384. doi:10.1007/s11104-11006-
0041-0040.
Progressio. (2009). Seed saving and climate change in Zimbabwe.
London: Progressio. www.progressio.org.uk
ReSAKSS. (2017). Yield, maize, Southern Africa. Regional Strategic
Analysis and Knowledge Support System. http://www.resakss.
org/node/11. Retrieved from 14/6/2017.
Rockström, J. (2000). Water resources management in small-
holder farms in eastern and Southern Africa: An overview.
Phys. Chem. Earth (B, 25(3), 275–283.
Rockström, J., Kaumbutho, P., Mwalley, J., Nzabi, A. W., Temesgen,
M., Mawenya, L.,… Damgaard-Larsen, S. (2009). Conservation
farming strategies in east and Southern Africa: Yields and rain
water productivity from on-farm action research. Soil Till. Res,
103, 23–32.
Rücker, G., Schwarzer, G., Carpenter, J. R., & Schumacher, M.
(2008). Undue reliance on I2 in assessing heterogeneity may
mislead. BMC Medical Research Methodology, 8(79). doi:10.
1186/1471-2288-8-79.
Rusinamhodzi, L., Corbeels, M., van Wijk, M. T., Rufino, M. C.,
Nyamangara, J., & Giller, K. E. (2011). A meta-analysis of
INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 23
long-term effects of conservation agriculture on maize grain
yield under rain-fed conditions. Agron. Sust. Dev, 31, 657–673.
Schlenker, W., & Lobell, D. B. (2010). Robust negative impacts of
climate change on African agriculture. Environmental Research
and Letters, 5(1), 1–8. doi:10.1088/1748-9326/5/1/014010
Schünemann, H. (2008). Criteria for applying or using GRADE.
Reviewed and approved by GRADE Working Group on
March 24, 2016. http://www.gradeworkinggroup.org/docs/
Criteria_for_using_GRADE_2016-2004-05.pdf.
Schünemann, H. J., Oxman, A. D., Vist, G. E., Higgins, J. P. T., Deeks,
J. J., Glasziou, P., & Guyatt, G. H. (2011). Chapter 12:
Interpreting results and drawing conclusions. In J. P. T.
Higgins, & S. Green (Eds.), Cochrane handbook for systematic
reviews of interventions. Version 5.1.0 [updated March 2011].
London: The Cochrane Collaboration. Retrieved from www.
cochranehandbook.org
Shewmake, S. (2008). Vulnerability and the impact of climate change
in South Africas Limpopo River Basin. IFPRI Discussion Paper 804.
Washington, D.C. (USA): IFPRI. http://www.ifpri.org/sites/default/
files/publications/ifpridp00804.pdf.
Southern African Research and Documentation Centre (SARDC).
(2002). Limpopo River Basin Factsheet, Southern African
Research and Documentation Centre (SARDC), Harare,
Zimbabwe. Accessed at: http://www.sardc.net/IMERCSA/
Limpopo/pdf/Limpopo1.pdf. Accessed on: February 2, 2010.
Sullivan, A., & Sibanda, L. M. (2010). Vulnerable populations, unre-
liable water and low water productivity: A role for institutions
in the Limpopo Basin’.Water International, 35(5), 545–572. doi:
10.1080/02508060.2010.510590. URL: doi: 10.1080/02508060
.2010.510590.
Sulser, T. B., Ringler, C., Zhu, T., Msangi, S., Bryan, E., & Rosegrant,
M. W. (2009). Green and blue water accounting in the Limpopo
and Nile basins: Implications for food and agricultural policy.
IFPRI Discussion Paper No. 907. Washington, DC:
International Food Policy Research Institute.
Tilman, D., Balzer, C., Hill, J., & Befort, B. L. (2011). Global food
demand and the sustainable intensification of agriculture.
PNAS, 108(50), 20260–20264.
Tittonell, P., & Giller, K. E. (2013). When yield gaps are poverty traps:
The paradigm of ecological intensification in African small-
holder agriculture. Field Crops Research, 143, 76–90.
Twomlow, S., Rohrbach, D., Dimes, J., Rusike, J., Mupangwa, W.,
Ncube, B.,…Maphosa, P. (2009). Micro-dosing as a pathway to
Africa’s green revolution: Evidence from broad-scale on-farm
trials. Nutrient Cycling in Agroecosystems, 88(1), 3–15.
Uaiene, R. N. (2004). Maize and sorghum technologies and the
effects of marketing strategies on farmers’ income in
Mozambique. MSc Thesis. Purdue University. West Lafayette,
United States.
Unger, P. W., Stewart, B. A., Parr, J. F., & Singh, R. P. (1991). Crop
residue management and tillage methods for conserving
soil and water in semi-arid regions. Soil & Tillage Research,
20, 219–240.
United Nations. (2015). Transforming our world: the 2030 Agenda
for Sustainable Development. Resolution adopted by the
General Assembly on 25 September 2015. http://www.un.
org/ga/search/view_doc.asp?symbol=A/RES/70/1&Lang=E.
Van Duivenboodew, N., Paln, M., Studer, C., Bielders, C. L., &
Beukes, D. I. (2000). Cropping systems and crop complemen-
tarity in dryland agriculture to increase soil water use
efficiency: A review. Netherlands Journal of Agricultural
Science, 48, 213–236.
van Ittersum, M. K., Cassman, K. G., Grassini, P., Wolf, J., Tittonell,
P., & Hochman, Z. (2013). Yield gap analysis with local to
global relevance – a review. Field Crops Research, 143, 4–17.
van Ittersum, M. K., van Bussel, L. G. J., Wolf, J., Grassini, P., van
Wart, J., Guilpart, N.,… Cassman, K. G. (2016). Proceedings of
the National Academy of Sciences of the United States of
America (PNAS), 113(52), 14964–14969. doi:10.1073/pnas.
1610359113.
Vogel, H. (1992). Effects of conservation tillage on sheet erosion
from sandy soils at two experimental sites in Zimbabwe.
Applied Geography, 12, 229–242.
Vohland, K., & Barry, B. (2009). A review of in situ rainwater har-
vesting (RWH) practices modifying landscape functions in
African drylands. Agriculture, Ecosystems & Environment, 131,
119–127.
Wang, V. C. X. (2009). Handbook of research on E-learning appli-
cations for career and technical education: Technologies for
vocational training, volume II. Information Science Reference,
Hershey, New York. http://www.igi-global.com/reference.
Whitlow, R., & Campbell, B. (1989). Factors influencing erosion in
Zimbabwe: A statistical analysis. Environmental Management,
29, 17–29.
Willcocks, T. J., & Twomlow, S. J. (1993). A review of tillage
methods and soil and water conservation in Southern
Africa. Soil &Tillage Research, 27, 73–94.
Word Bank. (2012). Turn Down The heat: why a 4°C warmer world
must be avoided. A Report for the World Bank by the Potsdam
Institute for Climate Impact Research and Climate Analytics.
Washington DC, USA. Retrieved from http://climatechange.
worldbank.org.
Zere, T. B., Hensley, M., & Van Huyssteen, C. W. (2006). Evaluating
maize production potential of two similar semi-arid ecotopes
in the free state province. South African Journal of Plant and
Soil, 23, 157–162.
Zhou, P., Gwimbi, P., Maure, G. A., Johnston, P., Kanyanga, J. K.,
Mugabe, F. T.,…Nelson, G. (2010). Assessing the
Vulnerability of Agriculture to Climate Change in Southern
Africa. Synthesis Report. Food, Agriculture and Natural
Resources Policy Analysis Network (FANRPAN). Pretoria,
South Africa.
24 M. S. MAGOMBEYI ET AL.