The study provides a comprehensive overview on future climate risks in Ethiopia’s agricultural sector, such as related to water availability, weather extremes and crop yields, derived from state-of-the-art impact models. It also features a spatial vulnerability assessment, which guides targeting of adaptation interventions. Based on the impact analysis, adaptation needs for Ethiopia were identified and suitable adaptation strategies selected and evaluated together with the Ethiopian government and local stakeholders, to provide recommendations for adaptation. The study and complimentary documents are available in the Downloads section.
While most adaptation strategies seek to minimise risks stemming from climate change, not all risks can be eliminated. Weather perils, such as droughts, storms or erratic precipitation represent so-called systemic risks that go beyond the farmers’ or communities’ coping ability. Thus, mechanisms are needed that distribute risk to avoid that certain groups or individuals are particularly affected and lose their livelihoods. One of such risk transfer solutions is crop insurance, which allows farmers to insure their crop yields against weather-induced losses. It is also a risk-specific adaptation strategy, which becomes irrelevant in the absence of weather and climate risks. While insurance usually is based on indemnity-assessment, with smallholder farmers this model is problematic due to the high transaction costs such an insurance scheme entails, e.g. for claim disbursements. Thus, a more suitable approach for smallholder farmers are weather index-based insurances (WII), a scheme that uses a weather index, such as temperature or precipitation to determine a payout. Alternative index-based insurance schemes can also be useful, such as area-yield index insurance.
Index insurance schemes for crops and livestock have been developed and tested in several pilot schemes in Ethiopia, but are not widely implemented yet. There is continued interest and engagement from within the country to further promote insurance solutions for the agricultural sector. For example, the Japan International Agency for Cooperation (JICA) recently launched a new “Index-based Crop Insurance Promotion (ICIP) project” in 2019 together with the Ministry of Agriculture and the Oromia Bureau of Agriculture and Natural Resources (OBoANR) (JICA, 2019). The programme is expected to cover 20,000 farmers in the Oromia region over the next five years.
During expert and stakeholder interviews conducted for this study, there was wide consensus that crop and livestock insurance has a low uptake in Ethiopia as of yet, and is regarded as having only limited upscaling potential in the country. Accordingly, interest in insurance among interviewees, survey and workshop participants was low, which may, however, be rather an expression of difficult implementation of insurance schemes than of general lack of interest, as the in-depth interviews revealed. Thus, there appears to be a need for further research on how to effectively operationalise insurance in Ethiopia and how to ensure better uptake and sustainability.
In sum, crop insurance can be considered an important adaptation strategy, because it can address risk, which cannot be mitigated in an economically sensible way with physical adaptation measures and acts as a safety net for farmers in times of extreme weather events. However, premium costs may not be affordable for farmers, requiring financial support in order to increase uptake.
Japan International Cooperation Agency (JICA). (2019). Index-based Insurance to benefit Small-holder Farmers through JICA Project. Retrieved June 13, 2019, from https://www.jica.go.jp/ ethiopia/english/office/topics/190419.html.
Fodder and feed improvement is regarded as a promising adaptation strategy that, according to interviewed experts, has high potential for upscaling in Ethiopia. Fodder and feed improvement is an umbrella term subsuming different strategies and technologies to improve nutritional quality, digestibility, quantity and availability of fodder and feed resources for livestock production. These include, for example, integration of pasture and forages into farm production, establishing fodder banks with improved forages and fodder trees, treatment of crop residues (e.g. with urea), silage and hay production, irrigation for production of off-season pasture and feed crops, improved grazing land resources management, increase of administering high-quality feed concentrate (Birhan & Adugna, 2014; Birhanu, Girma, & Puskur, 2017b).
To improve fodder and feed, a number of adaptation strategies have proven successful in Ethiopia, for example: Improved and high-yielding forage varieties, intercropping grasses and cereals with legumes, cultivation of irrigated fodder banks, natural pasture improvement through removing of invasive weeds, temporal zero-grazing and cut-and-carry feeding regimes on degraded pastures to restore and increase carrying capacity. Such strategies can boost livestock production, resilience and farmer income.
We conducted a cost-benefit analysis of one specific strategy to improve fodder for livestock in Ethiopia: irrigated Napier grass. Cultivating Napier grass and irrigating it was shown to be highly cost-effective. Importantly, the farmer’s investment in Napier grass will pay off after three years when the break-even point between net costs and net benefits is reached. It thus has a high and fast return to investment, as the following figure visualises.
While the examples above demonstrate the potential to increase livestock production even under a changing climate, the adoption of improved feed and fodder management strategies across the country so far remains rather low (Birhanu et al., 2017a), which was also confirmed by experts interviewed for this study. Yet, stakeholders and experts consistently expressed high interest in this measure throughout interviews, the survey and the workshop. In fact, due to the high importance and increasing policy awareness of this topic, experts interviewed see high upscaling potential for adoption of improved feed technologies in the country. Not all concrete strategies can be applied everywhere, irrigated Napier grass for instance may be difficult to implement in the lowland pastoralist and agro-pastoralist regions, but suitable strategies and improvements over the current practices can be found for all regions.
However, several issues need to be addressed to achieve this goal. Insufficient financial means, lack of market access and infrastructure and limited capacity among extension agents to provide adequate information and training are common constraints to technology adoption in smallholder farming that also apply to livestock production (Gebremedhin, Ahmed, & Ehui, 2003).
Table 4 lists some key potential co-benefits or maladaptive outcomes that may accrue with improved fodder and feed.
All in all, we conclude that looking at this specific adaptation strategy, action is much more profitable than inaction, although other improved fodder and feed strategies or factors, such as prices would change the results.
Birhan, M., & Adugna, T. (2014). Livestock feed resources assessment, constraints and improve-ment strategies in Ethiopia. Middle-East Journal For Scientific Research, 21(4), 616–622.
Birhanu, M. Y., Girma, A., & Puskur, R. (2017a). Determinants of success and intensity of livestock feed technologies use in Ethiopia: Evidence from a positive deviance perspective. Technological Forecasting and Social Change, 115, 15–25.
Birhanu, M. Y., Girma, A., & Puskur, R. (2017b). Determinants of success and intensity of livestock feed technologies use in Ethiopia: Evidence from a positive deviance perspective. Technological Forecasting and Social Change, 115, 15–25.
Gebremedhin, B., Ahmed, M., & Ehui, S. (2003). Determinants of adoption of improved forage technologies in crop-livestock mixed systems: Evidence from the highlands of Ethiopia. Retrieved from https://cgspace.cgiar.org/ handle/10568/27820.
Agroforestry is a complex field of interventions, comprising many different specific practices. In the climate risk analysis, we mainly considered the integration of trees in farming systems and tree-crop production. Agroforestry practices are considered as climate change adaptation for several reasons: Trees integrated in farming systems provide shade and thus lower temperature and enhance soil moisture, regulating the microclimate (Lasco, Delfino, & Espaldon, 2014). They generally save water, as they reduce evapotranspiration and improve soil fertility, for instance with falling leaves acting as mulch. Further, agroforestry systems can reduce pests and diseases. In terms of risk response, agroforestry systems are thus able to reduce risk from changing climatic conditions, such as rising temperatures and erratic precipitation. In addition, soil erosion can be lowered with targeted forestation, particularly on steep slopes.
An analysis of the effectiveness of agroforestry using a process-based crop model showed that agroforestry has the potential to stabilise maize yields in zones which are projected to experience yield losses under climate change (see Figure 5). 10 % or 20 % shade levels can reduce the losses projected (between 4-26 % in some zones), but would negatively affect yields in zones which are projected to benefit from climate change. However, those results are rather conservative, as they do not take into account the potential yield increases and other benefits of enhanced soil organic carbon due to agroforestry strategies, for instance.
The economic analysis showed that adapting maize production with agroforestry is very beneficial in comparison to the inaction scenario. Over time, it has a highly positive return on investment, with a benefit-cost ratio (BCR) of 5.1 and an internal rate of return (IRR) of 42.7 % (see Figure 6).
Further, agroforestry practices offer scope for many development co-benefits, which are described in Table 3.
All in all, agroforestry appears to be one of the most promising adaptation strategies considered in the climate risk analysis, thanks to the many co-benefits it can offer, when implemented wisely.
Lasco, R. D., Delfino, R. J. P., & Espaldon, M. L. O. (2014). Agroforestry systems: Helping small-holders adapt to climate risks while mitigating climate change. Wiley Interdisciplinary Reviews: Climate Change, 5(6), 825–833.
Improved crop management, such as using improved seeds, applying fertiliser and shifting the planting dates, has high transformative potential for increasing yields. We used process-based and machine learning techniques to evaluate the effectiveness of different crop management strategies for farmers to adapt to climate change. Our analysis found that increasing soil organic carbon in Ethiopia by 20% has positive effects on crop suitability for all crops (increasing suitability between 2-6%, depending on scenario and crop), especially for maize and wheat (see Figure 2). Enhancing organic carbon produces the greatest suitability increases under RCP8.5 for maize, teff and sorghum (ca. 5% suitability increase) and also has positive mitigation effects.However, shifting the growing season forward by four weeks will result in detrimental effects on suitability of the four crops (wheat, teff, maize and sorghum), with suitability losses of up to 10 % projected, and can thus not be recommended as an adaptation strategy.
Using the process-based crop model APSIM, we also evaluated the effect of increasing first basal and then top dressing NPK (Nitrogen, Phosphorous, Potassium) fertiliser on maize yield in Ethiopia for all zones (All) and for zones projected to experience yield losses (Loss). Fertiliser application among smallholders in Ethiopia is estimated to be of rather low intensity, some 30-40% of smallholder farmers apply fertiliser (Spielman, Mekonnen & Alemu, 2011). Applying fertiliser is one means for improving lower and more variable yields due to climate change impacts and can thus also be regarded as an adaptation strategy. Yet, increasing synthetic nitrogen fertiliser application will also lead to higher CO2 emissions, a thorough assessment of its usefulness for each specific case is thus needed. The results show that increasing basal fertiliser by 50% will increase yields by between 10 and 200% depending on the zone. At national level, an average increase of 56% for all modelled zones under current climatic conditions is projected (Figure 3).
We also conducted a cost-benefit analysis on one specific crop management strategy, namely a shift in cultivation from maize to sorghum, following the future suitability projections. The analysis shows that in comparison to the no adaptation scenario, the crop switch (adaptation scenario) will be economically beneficial from the year 2041 on (see Figure 4). From then on, the crop switch has a positive return on investment. The following figure shows this development of the net present value (NPV) from 2020 to 2050.
The late break-even point suggests that switching from maize to sorghum cannot be recommended in the near future, but rather in the medium term, once climate change impacts on the crop sector in Ethiopia further materialise.
In conclusion, improved crop management can generally be recommended for adaptation to climate change in Ethiopia, although some strategies require ex-ante evaluation. Shifting planting dates for instance is not necessarily beneficial, but applying more (organic) fertiliser and investing in improved seeds can generally bring about better yields and higher resilience.
Spielman, D. J., Mekonnen, D. K., and Alemu, D. (2011). Seed, Fertilizer and Agricultural Extension in Ethiopia. IFPRI, ESSP II Working Paper 20.
Irrigation can help smallholder farmers to compensate for the negative impacts of erratic and insufficient precipitation and significantly stabilise agricultural production (Woldemariam & Gecho, 2017). More specifically, it can raise agricultural production, allow for greater cropping intensity and crop diversity (i.e. higher-value crops), and lengthen agricultural seasons (Awulachew, 2010; Woldemariam & Gecho, 2017). Irrigation thus serves three main adaptive purposes: 1) Increasing yields by supplying water needed, 2) reducing risk due to a more constant water supply and 3) enabling multiple harvests and cultivation of high-value cash crops, as irrigation supplies water in the dry season.
Currently, irrigation is not wide-spread yet in Ethiopia (estimates range between 2-3 % of agricultural land, with water coming from Ethiopia’s ample surface water resources), with considerable potential to upscale its usage.Stakeholder consultations, interviews, the expert survey conducted and document analysis made clear that irrigation is a key adaptation priority in Ethiopia. The Ethiopian government is aiming to transform its agricultural sector from a subsistence mode to a market-oriented one. The potential for irrigation in Ethiopia is enormous, as it has ample surface water and groundwater resources on the one hand and land suitable for irrigation on the other hand (Woldemariam & Gecho, 2017). Twelve major river basins lie in Ethiopia, which form four main drainage systems. However, there is high spatial and temporal variability (FAO, 2005; Worqlul et al., 2015). According to various studies, there is sufficient water in Ethiopia to develop around 4.5 million hectares of agricultural land that could be irrigated through pump, gravity, pressure, underground water, water harvesting and other mechanisms (Makombe et al., 2011; Woldemariam & Gecho, 2017; Worqlul et al., 2017). The hydrological analysis in Chapter 2 also confirms this and projects ample water available for irrigation in the future.
A cost-benefit analysis of switching from rainfed maize production to irrigated maize production showed that adopting irrigation is beneficial for Ethiopian farmers. Over time, irrigation has a positive return on investment (see Figure 1).
Yet, irrigation requires a considerable investment and only becomes profitable after some years, depending on the type of irrigation system and the farm location. Institutional support is usually required and care has to be taken to avoid potential maladaptive outcomes from irrigation. Table 1 gives an overview of potential co-benefits and maladaptive outcomes from adaptation in Ethiopia.
Constraints to adaptation uptake in Ethiopia include weak institutional capacity and lack of physical infrastructures, such as pumps, conveyance structures and storage facilities, but also access to electricity in rural areas (Awulachew, 2010; FAO, 2015b; Worqlul et al., 2017).
Overall, irrigation is an important adaptation strategy in Ethiopia with large potential to transform the agricultural sector and increase yields. Government support and careful policy design is needed to make implementation succeed.
Note to the reader: This evaluation is only to be viewed as a careful model‐based and expert assessment, which can by no means replace a thorough analysis for specific project design and local implementation planning. It gives an indication of the overall feasibility and suitability of the selected adaptation strategies in Ethiopia. Actual selection of adaptation strategies, however, should always be based on specific needs and interests of local communities.
Awulachew, S. B. (2010). Irrigation potential in Ethiopia Constraints and opportunities for enhancing Irrigation potential in Ethiopia Constraints and opportunities for enhancing the system International Water Management Institute Teklu Erkossa and Regassa E . Namara. IWMI Research Report.
FAO (UN Food and Agriculture Organization), (2005). Irrigation in Africa in figures – AQUASTAT Survey 2005, 1–14.
FAO, (2015b): Analysis of price incentives for red sorghum in Ethiopia for the time period 2005-2012. Rome: FAO.
Makombe, G., Namara, R., Hagos, F., Awulachew, S. B., Ayana, M., & Bossio, D. (2011). A com-parative analysis of the technical efficiency of rain-fed and smallholder irrigation in Ethiopia. IWMI Working Papers (Vol. 143). https://doi.org/10.5337/2011.202.
Woldemariam, P., & Gecho, Y. (2017). Deter-minants of Small-Scale Irrigation Use: The Case of Boloso Sore District, Wolaita Zone, Southern Ethiopia. American Journal of Agri-culture and Forestry, 5(3), 49.
Worqlul, A. W., Collick, A. S., Rossiter, D. G., Langan, S., & Steenhuis, T. S. (2015). Assess-ment of surface water irrigation potential in the Ethiopian highlands: The Lake Tana Basin. Catena, 129, 76–85. https://doi.org/10.1016/ j.catena.2015.02.020.
Worqlul, A. W., Jeong, J., Dile, Y. T., Osorio, J., Schmitter, P., Gerik, T., … Clark, N. (2017). Assessing potential land suitable for surface irrigation using groundwater in Ethiopia. Applied Geography, 85, 1–13.
In response to increasing greenhouse gas (GHG) concentrations, air temperature over Ethiopia is projected to rise by 1.6 to 3.7 °C (very likely range) by 2080 relative to the year 1876, depending on the future GHG emissions scenario (Figure 2). Compared to pre-industrial levels, median climate model temperature increases over Ethiopia amount to approximately 1.5 °C in 2030, 1.8 °C in 2050 and 1.8 °C in 2080 under the low emissions scenario RCP2.6. Under the medium / high emissions scenario RCP6.0, median climate model temperature increases amount to 1.5 °C in 2030, 1.8 °C in 2050 and 2.4 °C in 2080.
Very hot days
In line with rising mean annual temperatures, the annual number of very hot days (days with daily maximum temperature above 35 °C) is projected to rise substantially and with high certainty, in particular over eastern Ethiopia (Figure 3). Under the medium / high emissions scenario RCP6.0, on average over all Ethiopia, the multi-model median projects 18 more very hot days per yearin 2030 than in 2000, 26 more in 2050 and 50 more in 2080. In some parts, especially in eastern Ethiopia, this amounts to about 200 days per year by 2080.
Future projections of precipitation are less certain than projections of temperature change due to high natural year-to-year variability (Figure 4). Out of the three climate models underlying this analysis, one model projects almost no change in mean annual precipitation over Ethiopia, while the other two models project an increase. Median model projections for RCP2.6 show almost no change in total precipitation per year until 2080, while median model projections for RCP6.0 show a precipitationincrease of 85 mm / year by 2080 compared to year 2000.
Heavy precipitation events
In response to global warming, extreme precipitation events are expected to become more intense in many parts of the world due to the increased water vapour holding capacity of a warmer atmosphere. At the same time, the number of days with heavy precipitation events is expected to increase. This tendency is also found in climate projections for Ethiopia (Figure 5), with climate models projecting a slight increase in the number of days with heavy precipitation events, from 7 days / year in 2000 to 8 days / year in 2080 under RCP2.6 and 9 days / year under RCP6.0 by 2080.
Soil moisture is an important indicator for drought conditions. In addition to soil parameters, it depends on both precipitation and evapotranspiration and therefore also on temperature as higher temperatures translate to higher potential evapotranspiration. Annual mean top 1-m soil moisture projections for Ethiopia show almost no change to a slight decrease for RCP2.6, while under RCP6.0, soil moisture is projected to slightly increase approaching a 1 % change by 2080 compared to the year 2000 (Figure 6). However, looking at the different models underlying this analysis, there is large year-to-year variability and modelling uncertainty, which makes it difficult to identify a clear trend.
Potential evapotranspiration is the amount of water that would be evaporated and transpired if sufficient water were available at and below land surface. Since warmer air can hold more water vapour, it is expected that global warming will increase potential evapotranspiration in most regions of the world. In line with this expectation, hydrology projections for Ethiopia indicate a stronger and more continuous rise of potential evapotranspiration under RCP6.0 than under RCP2.6 (Figure 7). Under RCP6.0, potential evapotranspiration is projected to increase by 2.0 % in 2030, 2.7 % in 2050 and 4.4 % in 2080 compared to year 2000 levels.
5 Changes are expressed relative to year 1876 temperature levels using the multi-model median temperature change from 1876 to 2000 as a proxy for the observed historical warming over that time period.
Smallholder farmers in the global South mostly use traditional crop varieties, which can be vulnerable to climate impacts such as droughts, floods or also diseases. In order to improve the resilience of crops to climatic shocks and to raise yields, improved crop varieties are bred from traditional varieties. The process of breeding is lengthy and costly, but once better varieties are released and used, they can substantially improve agricultural yields and resilience, depending on their specific characteristics. As such, breeding improved varieties is an institution-led approach, since it requires resources and time, which smallholder farmers most often do not have.
Improved crop varieties can offer a number of interesting development co-benefits, especially linked to increased agricultural production and income. However, they are expensive to develop: a multitude of factors will determine prices of the seeds, such as demand, scale of adoption, and – for farmers – potential government subsidies
A number of improved crop varieties already exist or are being developed in Ghana, for instance for maize, rice, cassava and cocoa, with sought-after properties being drought resistance, flood resistance and achievement of high yields. Maize appears to be the focus of breeding efforts in Ghana, as several publications (e.g. Alhassan, Salifu & Adebanji, 2016; Danso-Abbeam et al., 2017) and the number of active breeders for maize (10 out of 26) confirm (Mabaya et al., 2017).
An analysis conducted using the biophysical crop model APSIM shows that improved maize varieties indeed hold large potential for increasing yields under climate change (Figure 3). However, the size of the effect and sometimes even direction depends on the location, with the impact in different districts in Ghana differing considerably. Compared to no adaptation and other agronomic adaptation measures such as applying manure or delaying the sowing date, a carefully selected improved variety may indeed have a larger impact on increasing maize yields.
Another analysis conducted with machine-learning based crop suitability models shows how generally the effect of agronomic adaptation strategies varies considerably across crops. While maize and cassava appear to benefit from many agronomic adaptation strategies under climate change, for sorghum this is much more the case under the high-emission climate scenario (RCP8.5) and for groundnut, none of the adaptation strategies proposed has a positive effect.
Agronomic adaptation strategies and the use of improved seeds should thus be carefully evaluated for each specific region, crop and climate impact scenario. Nonetheless, improved crop varieties are seen as a transformative tool for buffering climate impacts in Ghanaian agriculture. Breeding is costly and time intensive, but where improved varieties already exist, they can contribute importantly to a higher agricultural output.
Alhassan, A., Salifu, H., & Adebanji, A. O., (2016). Discriminant analysis of farmers’ adoption of improved maize varieties in Wa Municipality, Upper West Region of Ghana. SpringerPlus, 5(1).
Danso-Abbeam, G., Bosiako, J. A., Ehiakpor, D. S., & Mabe, F. N., (2017). Adoption of improved maize variety among farm households in the northern region of Ghana. Cogent Economics and Finance, 5(1), 1–14.
Mabaya, E., Adzivor, S. Y., Wobil, J., & Mugoya, M., (2017). Ghana Brief 2017 – The African Seed Access Index, (December).
Rainwater harvesting (RWH) allows storing irrigation water for critical times in the growing period of otherwise rain-fed crops. It is also a promising practice for reducing costs for irrigation uptake as its installation is considerably cheaper than building groundwater irrigation infrastructure.
In Ghana, limited evidence on the use and potential of RWH for small-scale irrigation systems exists. Even though Ghana has abundant water resources for irrigation, its uptake is very low. RWH can provide a cost-efficient alternative to irrigation installment and is notably a strategy with usage potential at different scales. It is particularly well suited to be coupled with farm-level horticulture production, either for additional income from vegetable sales or satisfying changing demands in household consumption (for Sub-Saharan Africa as a whole, see e.g. OECD & FAO, 2016). This could also yield nutritional and thus health benefits, with vegetables enriching otherwise staple-based diets in Ghana. A combination with conservation agriculture techniques can be particularly useful, with further improvements in soil water storage capacity enhancing the water use efficiency of RWH.
In addition, RWH has the potential to deliver gender co-benefits. Since it is usually women who are in charge of fetching water and who engage in backyard vegetable farming, collecting rainwater could save women time, making time for other activities and enabling additional farming activities.
Rainwater harvesting and small-scale irrigation are a good alternative for action at smallholder level, with simple installation techniques proving less difficult to implement, maintain and refinance as compared to large-scale irrigation systems.
Thus, RWH is a promising adaptation strategy meeting local interest in Ghana. It can be implemented by farmers autonomously and decrease dependency on precipitation.
OECD/FAO, (2016). Agriculture in Sub-Saharan Africa: Prospects and challenges for the next decade, in OECD-FAO Agricultural Outlook 2016-2025, OECD Publishing, Paris. https://dx.doi.org/10.1787/agr_outlook-2016-5-en.
In areas short of precipitation, irrigation can be a key strategy to enable plant growth and increase yields. Water can be drawn from different sources, such as groundwater, surface water and in some countries even desalinated seawater, to enable a better growth of plants. Irrigation schemes at smaller scale can be initiated and implemented by farmers themselves, but for larger irrigation installations, technical agencies and extension officers play an important role.
As of yet, irrigation is not widely spread in Ghana, with only an estimated 1.6% of the area with a respective potential actually being irrigated (Mendes, Paglietti & Jackson, 2014), mostly for rice and horticulture cultivation (Namara et al., 2011). This limited uptake implies that further installation of irrigation schemes is possible and could increase Ghana’s agricultural production. However, this may incur high costs and technically challenging installation and maintenance. As a climate change adaptation strategy, the case for irrigation depends on the climate scenario, with the North of Ghana projected to see less rainfall under future climate change. However, as a measure to intensify agricultural production and enable multiple harvests, irrigation can also be considered useful today, where water is available for agricultural use.
Figure 2 shows the net value of maize production in the whole of Ghana under different irrigation scenarios, reflecting the varying costs that different irrigation techniques and installations incur. As can be seen from the comparison with the baseline scenario and the scenario under climate change, not all irrigation scenarios are able to offset the losses projected under climate change.
In sum, while irrigation has the potential to increase agricultural production in Ghana, it is also a costly strategy, often requiring institutional support for implementation and maintenance.
Mendes, D.M, Paglietti, L. & Jackson, D., (2014). Ghana: Irrigation market brief. Food and Agri-culture Organization of the United Nations. Rome.
Namara, R., Horowitz, L., Nyamadi, B. & Barry, B., (2011). Irrigation Development in Ghana: Past experiences, emerging opportunities, and future directions, International Food Policy Research Institute, GSSP Working Paper No. 27, Ghana Strategy Support Program, Accra.