Peden, D., Tadesse, G., Haileslassie, A.,. 2009. Livestock water productivity: implications for sub-Saharan Africa. The Range land Journal 31

CSIRO PUBLISHING www.publish.csiro.au/journals/trj The Rangeland Journal, 2009, 31, 187–193 Livestock water productivity: implications for sub-Saharan Africa D. Peden A,B, G. Taddesse A and A. Haileslassie A A International Livestock Research Institute, PO Box 5689, Addis Ababa, Ethiopia. Corresponding author. Email: d.peden@cgiar.org B Abstract. Water is essential for agriculture including livestock. Given increasing global concern that access to agricultural water will constrain food production and that livestock production uses and degrades too much water, there is compelling need for better understanding of the nature of livestock–water interactions. Inappropriate animal management along with poor cropping practices often contributes to widespread and severe depletion, degradation and contamination of water. In developed countries, diverse environmental organisations increasingly voice concerns that animal production is a major cause of land and water degradation. Thus, they call for reduced animal production. Such views generally fail to consider their context, applicability and implications for developing countries. Two global research programs, the CGIAR ‘Comprehensive Assessment of Water Management and Agriculture’ and ‘Challenge Program on Water and Food’ have undertaken studies of the development, management and conservation of agricultural water in developing countries. Drawing on these programs, this paper describes a framework to systematically identify key livestock–water interactions and suggests strategies for improving livestock and water management especially in the mixed crop–livestock production systems of sub-Saharan Africa. In contrast to developed country experience, this research suggests that currently livestock water productivity compares favourably with crop water productivity in Africa. Yet, great opportunities remain to further reduce domestic animals’ use of water in the continent. Integrating livestock and water planning, development and management has the potential to help reduce poverty, increase food production and reduce pressure on the environment including scarce water resources. Four strategies involving technology, policy and institutional interventions can help achieve this. They are choosing feeds that require relatively little water, conserving water resources through better animal and land management, applying well known tools from the animal sciences to increase animal production, and strategic temporal and spatial provisioning of drinking water. Achieving integrated livestock–water development will require new ways of thinking about, and managing, water by water- and animal-science professionals. Additional keywords: animal production, feed, land degradation, crop–livestock systems, water accounting, water conservation, sustainability. Introduction The human population of ~6.5 billion people requires ~7130 km3 of water annually to produce their food (Molden 2007), amounting to ~70% of global blue water withdrawals. Given current demographic trends and predicted dietary changes, water use will increase to 12 000–13 500 km3/year requiring annual increase in agricultural water demand of ~100–130 km3 in the next few decades. This increase is about three times the volume of water supplied to Egypt through the Aswan High Dam (Molden 2007) and does not take into account additional demands for possible production of biofuel or for sequestering carbon in response to global demand to combat human induced climate change. It also does not accommodate requirements for maintaining ecosystems services and environmental health. The world community must find ways to meet current and future food requirements without use of additional water resources. Aggravating the continuing population growth’s impact on food demand is the trend towards urbanisation and increasing Ó Australian Rangeland Society 2009 discretionary expenditures leading to increased consumption of animal source foods. Global consumption of meat and milk is growing at 2.1% and 1.7%, respectively, and are higher than human population growth rates. While animal source foods are very important for satisfying human nutritional needs, in excess of those needs, they are inefficient sources of energy if plant-based alternatives are available. Thus in some cases, reduced animal production may be warranted. Nevertheless, accommodating the world’s livestock will provide unique opportunities for improving the productivity of agricultural water resources. Steinfeld et al. (2006) highlight many negative impacts that livestock production generates in terms of excessive depletion of scarce water resources, adverse environmental consequences including contamination and sedimentation of water bodies, loss of aquatic and riparian biodiversity and competition for water required by people and ecosystems. Their critique of the livestock sector is most relevant to developed country production systems and intensified agricultural production in developing countries. 10.1071/RJ09002 1036-9872/09/020187 188 The Rangeland Journal These authors call for increased water productivity of livestock as an essential means to free up water for the natural environment and other users. Two linked global research processes have recently initiated transformed thinking about water use in developing country agriculture including means to increase agricultural water productivity. The Comprehensive Assessment of Water Management in Agriculture (CA 2008) undertook a critical review of the benefits, costs and impacts of 50 years of water development, the water management challenges communities face today, and the solutions people have developed around the world (Molden 2007). The CGIAR Challenge Program on Water and Food (CPWF 2008) builds on knowledge generated by the CA and undertakes new research to contribute to increasing agricultural water productivity, leaving more water for other users and the environment. Anticipated benefits include achieving food security, poverty alleviation through increased sustainable livelihoods in rural and peri-urban areas, improved human health through better nutrition, lower agriculture-related pollution, reduced water-related diseases, and environmental security through improved water quality as well as maintenance of water related ecosystems and biodiversity. The CA and CPWF processes revealed that some generalisations about excessive use of water by livestock (e.g. Goodland and Pimental 2000; Nierenberg 2005) are generally not applicable to developing country contexts and that past investments in agricultural water development in subSaharan Africa resulted in low investment returns and were environmentally and economically unsustainable (Peden et al. 2005). The result has been inefficient and ineffective use and enhanced degradation of water and land resources. However, great opportunities exist to increase water productivity of both crops and livestock especially in rainfed areas of sub-Saharan Africa (Molden 2007). Peden et al. (2007) concluded that greater integration of livestock and water development in sub-Saharan Africa could: reduce poverty, increase food production and reduce pressure on scarce water resources and the environment, reduce water used for current levels of livestock production by at least 50%, and that a water-accounting framework can help unravel the complexity of livestock–water interactions making progress in their management more feasible. This paper outlines one promising framework for assessing livestock water productivity (LWP) and describes how renewable fresh water is channelled through livestock production systems, how and where water is lost from them, and what management strategies might help in increasing LWP. This work focuses on sub-Saharan Africa, especially in the Nile River Basin, but many of the lessons learned are globally applicable. Other case study papers in this special edition of The Rangeland Journal describe various methodological details for assessing LWP (Cook et al. 2009; Descheemaeker et al. 2009; Gebreselassie et al. 2009; Haileslassie et al. 2009). * * Livestock water productivity Agricultural water productivity (WP) is defined as the ratio of the net benefits from crop, forestry, fishery, and livestock to the D. Peden et al. amount of water required to produce those benefits (Molden et al. 2007). Water productivity is the quantity and quality of food, income, livelihoods, environmental services produced per unit of water used in an agroecosystems. In this paper, we focus on livestock water productivity (LWP). Livestock water productivity is the ratio of the net beneficial livestock-related products and services to the water depleted in producing them and it might also be called the marginal water productivity of livestock. Livestock water productivity is a systems concept based on water accounting principles (Molden and Sakthivadivel 1999) that is applicable to diverse agricultural production systems in sub-Saharan Africa (Fig. 1) and to scales ranging from household to river basin levels. As a finite resource, water can only be used if it is available to users in an agroecosystem – in this case, livestock and livestock keepers. Water can be stored within the system as blue or green water. Blue water consists of lakes, rivers, ponds and reservoirs. Green water (Falkenmark et al. 1998) consists of soil moisture and water locked up in the tissues of plants, animals and microorganisms. Water enters a system in the form of rainfall and surface and subsurface inflow. Water depletion or loss from the system includes transpiration, evaporation, and downstream discharge. Sustainable water management requires long-term in-flow and depletion to be in balance preferably with sufficient storage of blue and green water to offset short-term scarcity to droughts. Transpiration is an essential loss of water necessitated by ecological and agricultural needs to sustain photosynthesis and plant growth. Depleted water has little or no value to the system and cannot readily be recovered notwithstanding some value derived from evaporative cooling within the system and requirements of downstream users outside of the system. Thus, transpiration is preferable to other forms of depletions from the perspective of the system losing water. Estimating livestockrelated water inflow, depletion and storage is a primary requirement of assessing water balance and LWP typically measured in units such as kg/m3 or US$/m3. Domestic animals are multi-purpose assets providing meat, milk, blood, manure, hides, farm power, a preferred means of securing household wealth, and cultural values, particularly in developing countries. Net beneficial livestock-related products and services include all of these, and they may be derived from multiple species and breeds of animals. Assessing LWP requires estimates of all major goods and services provided by animals. Because quantities of these beneficial outputs are not directly comparable, they are often converted to monetary units to obtain total value of animal production. Huge variations in estimated water used for feed confuse efforts to use this resource more productively and sustainably. For example, Peden et al. (2007) noted that water productivity estimates in the literature varied 80 fold from the least to most efficient feed plants. This variation is often largely an artefact of divergent concepts and methodologies. For example, consider irrigated forage sorghum from Sudan and irrigated alfalfa from Sudan (Saeed and El-Nadi 1998) in which the authors estimate water use on the basis of ET during the growing season at the field level. In contrast, Sala et al. (1988) used annual rainfall to estimate ‘rain use efficiency’ of Wyoming rangelands at a landscape level implying that the rainfall and evaporation during the non-growing season is also an input into plant biomass Livestock water productivity in sub-Saharan Africa The Rangeland Journal 189 Fig. 1. A framework for assessing livestock water productivity helps identify options for reducing water depletion, increasing livestock production, and enhancing ecosystems services associated with animal keeping [Source: Updated from Peden et al. (2007) and the modified image reproduced with permission from the International Water Management Institute (www.iwmi.org) and Earthscan (www.earthscan.co.uk)]. production. Sinclair et al. (1984) noted years ago that water use efficiency is a problematic term referring to multiple and confusing concepts where production has been defined in terms of carbon dioxide uptake, grain yield, or total plant yield often implying above ground plant production only. Water use estimates have been based on water use ranging from a few minutes to as much as year including non-growing seasons. Landscapes with diverse plant species and plant structures further complicate the challenge of estimating water productivity. Estimating the feed component of LWP will require rigorous standardisation of the working definition and measurement of production and water depletion. Without water loss through transpiration, plants cannot grow. Because disaggregating evaporation and transpiration is difficult in practice, it is rarely done. Most published research combined estimates of transpiration (T) and evaporation (E) into one index, evapotranspiration (ET), for the purpose estimating water depletion in agriculture (e.g. Renault and Wallendar 2000). However, shifting water depletion from evaporation and discharge to transpiration (Keller and Seckler 2005) and 190 D. Peden et al. The Rangeland Journal increasing the value of animal products and services (Peden et al. 2007) are the key means to increase WP and consequently LWP. Thus, the ratio of T to E is important. A high T to E ratio is indicative of probable higher agricultural water productivity than a lower ratio because T represents water used to enable plant growth while E represents non-productive water loss. For example, landscapes with little green vegetation as indicated by the leaf area index (LAI) and much bare ground will lose much water and have relatively low levels of plant production. In contrast, land with a high LAI and little bare ground, will have relatively higher plant production if plant species composition and environment and water inflow are held constant. One consequence of failure to distinguish T from E is that a key opportunity to increase WP is lost and non productive losses of water through evaporation can high. There are four basic livestock keeping strategies that can help increase LWP (Fig. 1). These are optimal feed sourcing, enhancing animal productivity, conserving water resources, strategic spatial and temporal provisioning of drinking water to livestock. These strategies are not merely technology interventions. All also require a mix of management practices, institutional arrangements, and policies for success (Amede et al. 2009). Feed sourcing Provision of feed is a major livelihood challenge and high labour and farm input cost for farmers in sub-Saharan Africa. In terms of volume, the major water requirement for livestock production is often that which is needed to produce animal feed. One key strategy for increasing LWP lies in selecting feed sources that use relatively little water or that use water that has little value for other human needs for support of ecosystem services. Water for feed is a concept that is closely linked to ‘water for vegetation’ where vegetation may or may not be consumed by animals as feed. The influence of livestock on vegetation and the management of vegetation for production of feed are discussed separately in the section ‘Conserving water resources’. In this section, we look at feed sources independent of the sustainability and state of vegetative degradation of the production systems involved. Widely prevailing views that livestock are wasteful users of water resources are largely predicated on the assumption that cattle mostly consume feed grains for which cropland and by implication all irrigation and rainwater used are allocated directly and purposely for animal meat and milk production (SIWI, IFPRI, IUCN and IWMI 2005). This view has some legitimacy for developed countries, but even there, livestock producers are responding to market pressure and beef cattle are spending less time being fed grain-based rations in feed lots. In sub-Saharan Africa, few cattle are fed grain. Rather, they depend on extensive grazing and crop residues and byproducts. Since crops will be grown for grain with or without the presence of livestock, any residues and byproducts generated through crop production to meet human nutritional needs can be made available to livestock without any additional water cost. Furthermore, grazing on vegetation that is not used for other human needs also constitutes a low water cost source for feed unless that water is vital for maintaining ecosystem structure and functioning. In the extreme, importing feed enables animal production by externalising the water cost of feed. In essence, virtual water supports animal production by transferring internal demands for water to external production systems often where the cost of water may be lower. In extensive production systems, animals’ dry matter digestibility is variable (e.g. Mpairwe 1998). If dry matter is ~50% digestible, ~50% of feed intake emerges from the animal as manure. In sub-Saharan Africa, manure is a highly valued resource that is widely used for replenishing soil fertility, domestic fuel and as a construction material for housing. Thus, approximately half of the water used to produce feed for livestock is actually used to provide additional value. Taking this into account can double estimates of LWP by either reducing the water depleted attributable to feed production or increasing the benefits derived from livestock keeping. Oxen provide a vital role in crop production in some areas of sub-Saharan Africa and Asia including the Blue Nile highlands. Without oxen, few crops would be produced. Where animals are used to cultivate fields and thresh grains, their use of water can be legitimately added to the water cost of producing crops. Equines are also important especially for transporting farm products to markets. For this reason, water used for feed to enable farmers’ access to animal-based farm power is an essential input into agricultural crop production and meat and other products derived from these animals are a secondary priority. Enhancing animal productivity Historically, animal sciences research has emphasised increasing animal production with much of it focused on single outputs especially meat and milk. Since most of this research was done in developed countries, relatively little emphasis was given to developing country demands for multiple animal products including farm power, hides, manure, livestock as a means of storing wealth and cultural values. Yet, the value of animals is the sum of these offset by any negativity. This is especially important in developing countries, and almost none of this research addressed the issue of total water use for production of multiple animal products and services. No matter how much or how little water plants transpire to produce feed, LWP will be low if that feed is not used efficiently for production of beneficial animal products and services. Livestock require energy for maintenance. The basal metabolic rate (BMR) is the intracellular energy consumption of a fasting animal in a thermoneutral environment. In 1932, Kleiber (1975) demonstrated that BMR of mammals ranging in size from mice to elephants is proportional to their liveweight0.75. One cow weighing 250 kg liveweight ( TLU) will require 18.4 MJ/day for BMR alone. Because of principles underlying Kleiber’s 3/4 law, five goats weighing 50 kg each will require a total of ~28.7 MJ/day for BMR. This feed energy requirement is a fixed cost before animals produce any goods and service. Other maintenance costs amount to ~27.6 MJ/TLU/day (FernandezRivera, pers. comm.) include thermoregulation, limited activity associated with feeding and watering, and the energy cost of maintaining rumen function. These data are based on unpublished synthesis of many studies and reports emerging from ILRI’s feeding trial research in sub-Saharan Africa. Thus, total energy Livestock water productivity in sub-Saharan Africa needs are ~46 MJ/TLU/day. Based on typical feeds available in Africa, maintenance of cattle requires ~450 m3/TLU/year of water for feed production to meet these energy needs. This is a fixed water cost associated with animal maintenance even if the animal is not growing, working or reproducing. To reduce the water cost of animal production requires two interacting steps: 1. Increasing the ratio of feed energy in producing products and services to that used in maintenance. Some maintenance costs can be reduced by husbandry that minimises stress on the animal associated with factors such as excessive trekking to watering sites, provision of shelter that reduces exposure to extremes of heat and cold, and reduction of morbidity and mortality due to disease and other threats. Energy not used for maintenance is available for production. For example, Muli (2000) and Staal et al. (2001) showed that providing on-site drinking water to livestock reduces stress and energy costs associated with drinking enabling substantive increases in dairy production. Where practical, night grazing can also reduce heat stress on animals enabling greater production. Traction power from oxen is a vital input for crop production in Ethiopia. Oxen are used for only short periods each year. However, oxen along with the larger bovine herd structure must be maintained throughout the year, and maintenance energy costs for the herd will be relatively high. 2. Increasing the efficiency with which feed energy is used by domestic animals. One of the axioms of animal sciences has been promotion of strategies that increase animals’ daily feed intake. Although this strategy may help increase the ratio of feed energy for production to feed energy for maintenance, it may not increase feed conversion efficiency for that production. Basarab (2003) explains that there is potential for selecting and breeding cattle that have higher feed conversion efficiency and not just higher rates of intake. Selecting appropriate species and breeds may also help reduce maintenance energy costs associated with vulnerability to heat, drought and diseases. The LWP concept adopts monetary indices as a means to weight and aggregate diverse animal products and services. It follows that market conditions will affect the ‘conversion of water to beneficial animal outputs’. Consequently, LWP may be higher when livestock keepers have good access to markets, have disease-free quality products, and can add value at the farm gate such as by converting liquid milk to butter or cheese. Not only do livestock produce diverse animal source foods such as milk and meat, these come from diverse animal species and each contains diverse nutrients of importance to human nutrition. For some nutrients (e.g. vitamin B12) in some environments, there may be no practical alternative food source. Thus, reliance only on market value of a commodity such as meat may be misleading. In this paper, we are focusing the discussion on ‘productivity’ and not directly on ‘production’. In this case, ‘productivity’ is defined on the basis of water depletion associated with the generation of beneficial livestock products and services. It is entirely possible that systems with high productivity or efficiency have unacceptably low levels of production. It is also true that focusing on systems efficiency or productivity may overlook agroecosystem resilience, a key requirement in African mixed crop–livestock and pastoral systems. The Rangeland Journal 191 Conserving water resources Conserving water is a key strategy for increasing LWP. From a water accounting perspective, this means managing water in ways that increase transpiration and thus growth of forages, and decrease depletion especially in the form of evaporation and uncommitted discharge. It also means encouraging infiltration that recharges soil moisture and ground water making water more available for plant growth during dry seasons. The primary challenge to conserving agricultural water is maintaining high levels of vegetative ground cover and, under this canopy, soil that allows high rates of infiltration and water holding capacity. In grazing areas, this means limiting animal stocking rates to a level that allows moderate production and avoids heavy grazing intensity the removes excessive ground cover or shifts the plant species composition from palatable to unpalatable types. Providing drinking water Without drinking water, animals die. Stress caused by a shortage of drinking water reduces animal production. The process of drinking takes place within a system and thus water drunk is not lost from it. After water has been consumed, it can be lost from the animal as faecal moisture or urine, in which case it is deposited on the soil–vegetation complex from which it may infiltrate or evaporate. A very small amount may be lost as evaporation from pulmonary and other animal tissues, but this depletion is not part of the process of drinking. Nevertheless, drinking water must be of high quality and available in small but adequate quantities. Although the cost of providing a unit of drinking water may also be high, the amount of water drunk is less than two percent of that needed to produce feed. The amount of water livestock drink depends on many factors such as species, breed, ambient temperature, water quality, levels and water content of feed, animal activity, pregnancy and lactation (King 1983; Pallas 1986; Seleshi et al. 2003). Water loss through urine and faeces must also be replaced through drinking or with the water content of feed. Water drunk per kg of feed intake ranges from ~3.6 to 8.5 L at ambient temperature below ~158C to 278C, respectively. Lactating cows drink more and high producing cows can drink as much as 85 L. Water deprivation reduces feed intake and hence, weight gains and milk production. Particularly in mixed crop–livestock systems of sub-Saharan Africa, piped water delivered to the farm combined with zero grazing will increase production and probably LWP (Muli 2000). Perhaps the most important contribution of providing drinking water is the opportunity to more optimally distribute livestock, especially cattle, on pasture land enabling them to make more effective use of forages without overgrazing them. For example, studies in Wyoming demonstrated in one case that 77% of grazing took place within 366 m of watering points and that 65% of available pasture was more than 730 m from water (Gerrish and Davis 1999). In dryland areas of Africa, many factors force livestock to travel much farther for drinking and daily watering is often not possible. One consequence is that related stress from excessive walking, heat and limited feed supplies limit both animal production and LWP. For example, Faki et al. (2008) concluded that in the vast Central Belt of Sudan, feed supplies are ‘low near watering points because high animal concentration has degraded the nearby pastures’. Many water and feed sources are 192 D. Peden et al. The Rangeland Journal often seasonal forcing pastoralists to adopt migratory grazing patterns. In most sub-Saharan Africa grazing lands, livestock watering points are inadequate in number and sub-optimally distributed in terms of providing access to feed. Poor management of watering points and adjacent pastures results in contaminated and siltation of water bodies, and excessive loss of vegetative biodiversity and biomass. Livestock keepers thus suffer from lack of both quality feed and water. Implications of livestock water productivity for sub-Saharan Africa Livestock keeping is an important livelihood strategy for poor farmers in pastoral, agro-pastoral and mixed crop–livestock systems of sub-Saharan Africa. Contrary to common views in developing countries, livestock use of water is not wasteful compared to crop production but significant increases in LWP are possible and feasible. A water productivity framework taking a water accounting approach suggests that sourcing of feeds derived from plants that require and deplete relatively little water can help increase LWP. Equally important is the need to enhance animal production by reducing morbidity, mortality and stress risks to which herds are exposed. Integrating water conservation into animal production practices helps reduce excessive run-off, downstream flooding, soil loss, and contamination of water supplies and helps increase transpiration that drives plant production making biomass more available for food, feed and environmental services. One key is to ensure that livestock feeding and trekking behaviour is managed in ways that maintain vegetative cover and biomass to prevent gulleying that accelerates soil loss and run-off. Finally provision of drinking water in areas having surplus feed and pasture while limiting animal access to existing and newly developed pastures, water bodies and their adjacent riparian habitats can increase LWP and environmental health at landscape scales. These strategies are not simply technical in nature. They all may be gendered in imposing different costs and delivering benefits differently on children, women, men and marginalised socio-economic groups. They all require enabling policies or community-based management of common property pasture and water resources, details of which are reported elsewhere (Molden 2007; Peden et al. 2007; Amede et al. 2009). Because the livestock–crop-water nexus falls across multiple ministries in most national governments, there is need for coherent policies that amplify the potential for multi-sector benefits for farmers and herders. Often these four strategies need to be applied simultaneously. For example, Faki et al. (2008) suggest that across the complex of production systems that make up the Central Belt of Sudan ‘improved natural resources legislation, institutional arrangements, marketing of livestock products, and veterinary care combined with efforts optimally to expand watering sites while limiting animal densities near them can help increase LWP’. In brief, increasing LWP will likely require a site-specific set of multiple strategies and not just one alone. Achieving increased LWP requires integrated investments in livestock and agricultural water development and management. Apart from provision of drinking water in dominantly grazing areas, such integration has been lacking. Evidence suggests that integrating water management for livestock and crops can increase sustainability and investment returns. Encouraging collaboration among research and development agencies working in the livestock, land, water and crop sectors is the first step. Success will also depend on engaging farmers, herders, and local communities especially in adopting intervention options and in collective management that can sustainably increase water productivity of crops and livestock especially those dependent on access to common property land and water resources. Livestock water productivity is a new and evolving concept and several questions still need to be addressed. To date, we have only considered water depleted in terms of volume (m3) lost, but the price of water may be more meaningful. For example, rainwater in rainfed arid rangelands may have lower value than irrigation water because the former can only be used in agriculture for livestock but the latter can support diverse crop and animal production options. Additionally, we have restricted the discussion to the monetary benefits derived from animal products and services. There may be need for analyses that focus on the water productivity of specific products such as protein if the development goal is non-economic. The importance of the LWP concept lies not merely in estimating LWP numerically. Spatial and temporal trends and patterns in LWP may help natural resource managers target interventions, locations and scales of analysis where investments in agricultural water will give the greatest sustainable returns on investments. One question frequently raised about the LWP concept relates to its relevance in dry land areas where crops cannot be grown. The argument is that LWP does not matter if water availability is too low or water access for crops too costly. Our experience suggests that even in extensive semi-arid grazing systems, there are huge opportunities for increasing LWP and that by using available rainwater more productively, options exist to either increase animal production or rehabilitate degraded landscapes. Future development of LWP concepts needs to give priority to assessments that separate evaporation and transpiration to allow for identification of technical, management and policy options that increase transpiration and infiltration and reduces evaporation and excessive discharge. Finally, a major challenge in mixed crop–livestock systems is to integrate efforts to increase both crop and livestock water productivity into one coherent approach to enhance overall agricultural water productivity while ensuring adequate water remains available for the support of environmental services. In conclusion, research on livestock water productivity in the Nile River Basin suggests that at least in sub-Saharan Africa, increasing LWP can make an important contribution to increasing overall agricultural water productivity that is deemed essential to sustainably meeting future human food security needs and reducing poverty (Molden et al. 2007). Acknowledgements This paper presents findings from PN37, ‘Nile Basin Livestock Water Productivity’, a project of the CGIAR Challenge Program on Water and Food in collaboration with the CGIAR Comprehensive Assessment of Water Management in Agriculture. We acknowledge with gratitude the scientific advice and encouragement provided by David Molden, Deborah Bossio and Shirley Tarawali. Livestock water productivity in sub-Saharan Africa The Rangeland Journal References Amede, T., Geheb, K., and Douthwaite, B. (2009). Enabling the uptake of livestock–water productivity interventions in the crop– livestock systems of sub-Saharan Africa. The Rangeland Journal 31, 223–230. Basarab, J. (2003). Feed efficiency in cattle. (Canada Alberta Beef Industry Development Fund: Calgary.) Available at: https://mail.une.edu.au/ lists/archives/beef-crc-technet/2003-November/000016.html (verified fi 20 April 2009). CA (2008). Comprehensive Assessment of Water Management in Agriculture. Available at: www.iwmi.org\assessment (verified 20 April 2009). Cook, S. E., Andersson, M. S., and Fisher, M. J. (2009). Assessing the importance of livestock water use in basins. The Rangeland Journal 31, 195–205. CPWF (2008). CGIAR Challenge Program for Water and Food. Available at: www.waterandfood.org (verified 20 April 2009). Descheemaeker, K., Raes, D., Nyssen, J., Poessen, M., Haile, M., and Deckers, J. (2009). Changes in water flows and water productivity upon vegetation regeneration on degraded hillslopes in northern Ethiopia: a water balance modelling exercise. The Rangeland Journal 31, 237–249. Faki, H., El-Dukheri, I., Mekki, M., and Peden, D. (2008). Opportunities for increasing livestock water productivity in Sudan. In: ‘Fighting poverty through sustainable water use. 2nd International Forum on Water and Food. Vol. II’. (Eds E. Humphreys, R. S. Bayot, M. van Brakel, F. Gichuki, M. Svedsen, P. Wester, A. Huber-Lee, S. Cook, B. Douthwaite, C. T. Hoanh, N. Johnson, S. Nguyen-Khoa, A. Vidal, I. MacIntyre and R. MacIntyre.) pp. 57–60. (Challenge Program on Water and Food: Colombo.) Falkenmark, M., Klohn, W., Lundqvist, J., Postel, S., Rockstrom, J., Seckler, D., Shuval, H., and Wallace, J. (1998). Water scarcity as a key factor behind global food insecurity: round table discussion. Ambio 27, 148–154. Gebreselassie, S., Peden, D., Haileslassie, A., and Mpairwe, D. (2009). Factors affecting livestock water productivity: animal scale analysis using previous cattle feeding trials in Ethiopia. The Rangeland Journal 31, 251–258. Gerrish, J., and Davis, M. (1999). Water availability and distribution. In: ‘Missouri Grazing Manual’. (Eds J. Gerrish and C. Roberts.) pp. 81–88. (University of Missouri Extension: Columbia.) Goodland, R., and Pimental, D. (2000). Environmental sustainability and integrity in natural resources systems. In: ‘Ecological Integrity’. (Eds D. Pimentel, L. Westra and R. Noss.) pp. 121–138. (Island Press: Washington, DC.) Haileslassie, A., Peden, D., Gebreselassie, S., Amede, T., Wagnew, A., and Taddesse, G. (2009). Livestock water productivity in the Blue Nile Basin: assessment of farm scale heterogeneity. The Rangeland Journal 31, 213–222. Keller, A., and Seckler, D. (2005). Limits to the productivity of water in crop production. In: ‘California Water Plan Update 2005. Vol. 4’. (California Department of Water Resources: Sacramento.) King, J. (1983). ‘Livestock Water Needs in Pastoral Africa in Relation to Climate Change and Forage.’ ILCA Research Report 7. (ILCA: Addis Ababa.) Kleiber, M. (1975). ‘The Fire of Life: An Introduction to Animal Energetics.’ (Robert E. Krieger Publishing: New York.) Molden, D. (2007). ‘Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture.’ (Earthscan: London, and International Water Management Institute: Colombo.) 193 Molden, D., Oweis, T. Y., Steduto, P., Kijne, J. W., Hanjara, M. A., and Bindraban, P. S. (2007). Pathways for increasing agricultural water productivity. In: ‘Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture’. (Ed. D. Molden.) pp. 279–310. (Earthscan: London, and International Water Management Institute: Colombo.) Molden, D., and Sakthivadivel, R. (1999). Water accounting to assess uses and productivity of water. International Journal of Water Resources Development 15, 55–71. doi: 10.1080/07900629948934 Mpairwe, D. (1998). Integration of forage legumes with cereal crops for improved grain yield, forage production and utilization for smallholder dairy production systems. Ph.D. Thesis, Makerere University, Kampala, Uganda. Muli, A. (2000). Factors affecting amount of water offered to dairy cattle in Kiambu District and their effects on productivity. B.Sc. Thesis, University of Nairobi (Range Management), Kenya. Nierenberg, D. (2005). ‘Happier Meals: Rethinking the Global Meat Industry.’ Worldwatch Paper #171. (Worldwatch Institute: Washington, DC.) Pallas, P. (1986). ‘Water for Animals.’ (FAO: Rome.) Peden, D., Freeman, A., Astatke, A., and Notenbaert, A. (2005). ‘Investment Options for Integrated Water-Livestock–Crop Production in sub-Saharan Africa.’ ILRI Research Paper No. 1. (ILRI: Nairobi.) Peden, D., Tadesse, G., and Misra, A. K. (2007). Water and livestock for human development. In: ‘Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture’. (Ed. D. Molden.) pp. 485–514. (Earthscan: London and International Water Management Institute: Colombo.) Renault, D., and Wallendar, W. (2000). Nutritional water productivity and diets. Agricultural Water Management 45, 275–296. doi: 10.1016/ S0378-3774(99)00107-9 Saeed, I., and El-Nadi, A. (1998). Forage sorghum yield and water use efficiency under variable irrigation. Irrigation Science 18, 67–71. doi: 10.1007/s002710050046 Sala, O., Parton, W., Joyce, A., and Lauenroth, W. (1988). Primary production of the Central Grasslands of the United States. Ecology 69, 40–45. doi: 10.2307/1943158 Seleshi, Z., Tegegne, A., and Tsadik, T. (2003). Water resources for livestock in Ethiopia: implications for research and development. In: ‘Integrated water and land management research and capacity building priorities for Ethiopia. Proceedings of a MoWR/EARO/IWMI/ILRI International Workshop’. (Eds P. McCornick, A. Kamara and G. Tadesse.) pp. 66–79. (ILRI: Addis Ababa.) Sinclair, T. R., Tanner, C. B., and Bennett, J. M. (1984). Water-use efficiency in crop production. Bioscience 34, 36–40. doi: 10.2307/1309424 SIWI, IFPRI, IUCN and IWMI (2005). ‘Let it Reign: The New Water Paradigm for Global Food Security.’ Final report to CSD-13. (Stockholm International Water Institute: Stockholm.) Staal, S., Owango, M., Muriuki, G., Lukuyu, B., Musembi, F., Bwana, O., Muriuki, K., Gichungu, G., Omore, A., Kenyanjui, B., Njubi, D., Baltenweck, I., and Thorpe, W. (2001). ‘Dairy Systems Characterization of the Greater Nairobi Milk-shed.’ SDP Research Report. (Kenya Ministry of Agriculture and Rural Development, Kenya Agricultural Research Institute, and International Livestock Research Institute: Nairobi.) Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., and de Hann, C. (2006). ‘Livestock’s Long Shadow: Environmental Issues and Options.’ (FAO: Rome.) Manuscript received 14 January 2009; accepted 8 April 2009 http://www.publish.csiro.au/journals/trj