Assuring Water for Food and Environmental Security

David Molden and Frank Rijsberman International Water Management Institute1 We need a Blue Revolution in agriculture that focuses on increasing productivity per unit of water more crop per drop. These are words that might be heard within IWMI corridors from time to time, but in this case the words come from Mr. Kofi Annan, Secretary General of the United Nations, Report to the Millennium Conference, October, 2000 in reference to the need to solve the water crisis we are in.

In this presentation I will explore the nature of the world water crisis, the role of agriculture, and the potential to resolve this crisis.

What is the nature of the crisis?

We are all quite aware of issues of dried up and polluted rivers, of endangered aquatic species, of accumulation of agricultural chemicals in natural ecosystems2. In many areas, salinization and groundwater decline are symptoms of the crisis3. In the wake of economic development, some of the worlds great rivers do not reach the sea. Cotton uses nearly the entire flow of The Amu Darya and Syr Darya in Central Asia. The Yellow River did not reach the sea for 7 months in 1997. Similarly, very little Nile, Indus, or Colorado River water reaches the sea.

Prepared for the CGIARs Mid Term Meeting in Durban, South Africa, May 24, 2001. The authors are Senior Researcher and Director General of IWMI. 2 The current situation is already one characterized by dried-up dried up and polluted rivers, lakes, and groundwater resources. Species are disappearing. World wide, 20% of freshwater fish are vulnerable, endangered or extinct; 20% of insects have aquatic larval stages; and 57% of freshwater dolphins are endangered. Rapidly growing cities, burgeoning industries, and rising use of chemicals in agriculture have undermined the quality of many rivers, lakes, and aquifers. Agricultural chemicals used to boost agricultural productivity slowly accumulate in aquifers and natural ecosystems. Groundwater, the preferred source of drinking water, is extremely difficult to clean. 3 Sandra Postels Pillars of Sand (1999)

In spite of development of water resources intended at food production, malnutrition persists, mostly in South Asia and Sub-Saharan Africa4. Much of this is in regions dubbed economically water scarce5, meaning that while there is water available in nature, sometimes abundantly, it has not been developed for human use. Small farmers and the poor are particularly disadvantaged and can face acute water scarcity. They do not have access to water to satisfy their needs for either food security or sustainable livelihoods. The agriculture community see continued growth of irrigation as an imperative to achieve the goals adopted by the international community to reduce hunger and poverty. Under a base scenario that included optimistic assumptions on productivity growth and efficiency, IWMI estimated that 29% more irrigated land would be required by the year 2025, and because of gains in productivity and more efficient water use, the increase in diversions to agriculture would be 17%6. FAO and Igor Shiklomonov of the Russian State Hydraulics Institute had similar results. Citing similar international commitments to maintain and improve environmental quality and biodiversity, many in the environmental community see it as imperative that water withdrawn for agriculture is reduced, not increased. Taken from the perspective of sustainable use, a colleague Joe Alcamo of Kassel University projected an 8% decrease in the amount of water that should be diverted to irrigation. The difference between the 17% increase and 8% decrease is on the order of 625 km3 of water close to the 800 km3 of water that is presently used globally for urban and industrial use7. The crisis and conflict is not one of cities versus agriculture. Cities withdraw only a small portion of water. Plus the value of use in cities is so much higher that they naturally should have first priority. While there can certainly be sharp conflicts locally, and agriculture is indeed displaced by urban and industrial needs, there is no real competition. It is more like a lost battle: water for urban areas wins hands down over water for agriculture. On the whole, the conflict, or the need to find harmony or balance, is between uses of water in agriculture and uses of water in nature. From this perspective, how much irrigation do we really need is one of the burning questions of our times. How we resolve the world water crisis very much depends on how well water is managed in agriculture.

In 1997, 790 million people in developing countries remain food insecure, 60 percent of whom live in South Asia and Sub-Saharan Africa. In Sub-Saharan Africa, the number of food insecure people has risen from 125 to 186 million people over the last from 1980 to 1997 (from Marc Cohen and Per Pinstrup Andersen prepared for the CGIAR based on FAO data). 5 From IWMI Water Scarcity Studies, IWMI (2000).
6

FAO and estimated a 34% increase in irrigated area, and a 12% increase in irrigation diversions, and similarly Shiklomanov projected a 27% increase in irrigated diversions. 7 To get an idea of the magnitude, Egypts High Aswan Dam releases about 55 km3, so the difference is equivalent to more than 10 High Aswan Dams annual supply of water.

It is often stated that irrigation uses 70% of all water withdrawn, and in some countries this number reaches 90%8. In fact, irrigation withdrawals are on the order of 2500 km3 , which is approximately 6% of the worlds renewable resources. The other 94% of the renewable resources are used to are used to support crop cultivation and terrestrial, aquatic and coastal ecosystems. Seen from another perspective, IWMI estimates that of evaporation from earth surfaces, lands supporting crop-based agriculture evaporate 20% about 15% of which is from lands supporting rainfed agriculture and 5% by irrigated lands9. Certainly, we think that managing water in agriculture should not exclusively focus on improving the efficiency of the 2,500 km3 diverted to irrigation, but must include improving the productivity of the 16,000 km3 used in rainfed agriculture as well. Productivity of Water in Agriculture A common perception is that increasing efficiency in agriculture is the solution to the water crisis. Technically defined, efficiency tells us how much diverted water reaches the crops, and how much is wasted down the drain. Unfortunately, this is a widespread misperception as I will illustrate. The real wastage comes from not being as highly productive as possible with the water that is currently consumed (not wasted down the drain) in agriculture. The Chistian Irrigated area is located in Pakistans Punjab with a landscape heavily dominated by agriculture. To get an idea of how efficiently water was used, IWMI performed a water accounting exercise10. During the 1993/94 agricultural year, 740 million cubic meters (MCM) of water entered the area11 from irrigation deliveries, rain and groundwater. Human use, dominated by crop agriculture, consumed 90% of the supplies, evidently quite efficient. From this larger, basin perspective, farmers are very effective in converting water into crop production. But, groundwater was mined during the year, and in this area very little water was available for environmental purposes such as flushing salts, or for ecosystem sustenance. Farmers as a group are, if anything, too efficient! Certainly increasing the efficiency, and leaving even less for other uses, is not recommended.

From 1900 to 1995, withdrawals for human use have increased from 600 km3/year to 3,800 km3/yr. Agricultural withdrawals are on the order of 2500 km3/year in many developing countries this is over 90% of all water withdrawn for human uses. From another perspective, of the 100,000 km3 per year reaching the earths surface, only 40% or 40,000 km3 are considered renewable water resources because they contribute to river runoffs and groundwater storage. Of this amount, some 10% or 3,800 km3 is diverted from its natural courses, of which 2,500 km3, 7% is withdrawn for irrigation (based on Shiklomonov, 1998). 9 This preliminary estimate by IWMI was done by overlaying World Water and Climate Atlas grids on the USGS land cover data set. 10 Based one Molden et al (2001) 11 504 MCM from irrigation diversions, 143 MCM as rain, and 73 MCM as net groundwater abstraction. Crop evapotranspiration was 595 MCM, while evaporation from cities was about 50 MCM.

r o p s

S u r fa c e

B E N E F I C IA L

ity

u s e s

V e g e ta t io n

R a in

S to ra g e

h is h tia n

O e E n v ir o c o m m

th e v a p n m itm

r o r a tio n e n ta l e n ts

While efficiency is very high, productivity is very low. Wheat yields are on the order of only 2 tons per hectare, while rice yields are on the order of 1.4 tons per hectare. In terms of kilograms and dollars per cubic meter, water productivity is on the low end of the spectrum when comparing to other systems worldwide12. For wheat, water productivity is on the order of 0.6 kg/m3 compared to a range of about 0.5 go 1.5 kg/m3.

Figure 1. Water accounting for the Chistian sub-division, Pakistan. Where water is limiting, there is a clear need to shift from an exclusive focus on productivity of land resources, yield in tons per hectare, to a view that focuses on productivity of water resources, tons per cubic meter, and in a broad sense, overall benefits derived from water used. Productivity of Water How will it help?

12

For wheat this converts to 0.6 kg/m3 of water. We have found a range of water productivity of wheat from 0.6 to about 1.5 kg/m3 worldwide. The gross value of production for the rice-wheat cropping system per cubic meter of evapotranspiration is on the order of US$0.07, at the low end of the spectrum (Sakthivadivel et al, 1999). For 40 systems, IWMI calculated a range of water productivity calculated in this way from 0.05 to about 0.80 $US per cubic meter.

Why is getting more crop per drop so important? The answer is simple growing more food with less water alleviates scarcity, contributes to achieving food security, and puts less strain on nature. Reducing water withdrawn by agriculture contributes by freeing up more water for nature, for drinking, and industrial uses. Can this be done and still provide food security and improved rural livelihoods. Here are the results of a global calculation using the IWMIs Podium Model13. In this scenario14, there is a moderate expansion of 3% of the harvested area, and 10% of irrigated area. But we have actually required withdrawals by irrigation to decrease by about 10%. The only way that enough food can be grown is by increases in water productivity on rainfed and irrigated land. For the period of 2000 to 2025, we have estimated that an annual growth rate of about 1.8% or roughly a 60% percent increase for the period, on irrigated land, and 1.0%, or a 30% increase on rainfed land in water productivity would be required (see Table 1)15. This marked change in water productivity from business as usual scenarios is the challenge.

Table 1. Water productivity and yield growth rates for a scenario meeting goals of food and environmental security. Irrigated Rainfed Recent Annual Growth Rates (%) in Yield 1.0% 0.5% Business as Usual Scenarios - Growth in Yield 1.0% 0.5% - Growth in Water Productivity 0.6% 0.5% - Growth in Water Productivity (25 years) 20% 15% Food and Environmental Security Scenario - Growth in Yield 1.3% 1.0% - Growth in Water Productivity 1.8% 1.2% - Growth in Water Productivity (25 years) 60% 35%

The billion dollar question is, of course, whether such an increase is feasible! Increases on rainfed land can be achieved by several means: improved varieties, better nutrient management, improved soil-water management practices, and by introducing supplemental irrigation to fill in the water gaps. It is estimated that in arid areas, 50% of rainfall evaporates back to the atmosphere without contributing to crop productivity16.
13 14

Seckler et al, 2000, and http://www.iwmi.org Population grows as per the UN Medium population growth forecast to 7.8 million, and the calorie level is assumed to increase from a present per capita value of 2700 to 3000. 15 For irrigation, water productivity was calculated as kg per cubic meter of water withdrawn. On irrigated land, we calculated the growth in terms of kg per unit of evapotranspiration. 16 See Rockstrom, 1999 for a discussion.

0.7

0.6

0.5

0.4
US$/m3

0.3

0.2

0.1

Capturing this water before it evaporates, through improved crop properties such as fast growing roots, or improved tillage practices seems to offer potential. Drought tolerant crops, while not necessarily lifting the yield ceiling, can improve water productivity. And supplying a small amount of water at a time of stress can greatly contribute to productivity. In many areas, potential productivity is not realized and this is in part due to poor irrigation management. Considering the productivity of water in more than 40 irrigation systems worldwide, Sakthivadivel et al (1999) demonstrated a 10-fold difference in the gross value of output per unit of water consumed by evapotranspiration (Figure 2). Some of this difference is due to the price of grain versus high valued crops, and certainly not all agriculture can be devoted to high valued crops. But even among grain producing areas, the differences are large. Figure 2. Water Productivity values in terms of standardized gross value of output per unit of evapotranspiration (Sakthivadivel, 1999). In many places, real water savings is an important mechanism to increase the productivity of water. In China, and many other places of the world, water is moving out of agriculture. In China, Wuhan University, ACIAR, CSIRO, IRRI, IWMI and other Chinese partners are carefully looking at practices that improve the productivity of water in rice areas, allowing sustained production, yet freeing up water for other uses. The Zhang He reservoir, situated in the Yangtze River basin, was constructed primarily for irrigated agriculture. Over time reservoir water also met increasing demands from higher valued urban and industrial water uses. Water managers farmers, irrigation service

providers, and water resource managers - were able to shift water out of agriculture to meet these other needs (Figure 3). Production levels remained stable over the time period in spite of this massive shift of water out of agriculture (Table 2). The increase in water productivity can only partly be explained by yield growth which nearly doubled over a thirty year period. This is compared to a near trebling of water productivity attributed to the Zhang He supply. Growing more rice with less water improving the productivity of water was made possible through on-farm water saving irrigation practices, ample recycling through the melons-on-the-vine system of reservoirs, pricing water, and strong institutions to back these approaches. (Hong et al, 2001 and IWMI annual report 2000).
Zhanghe Irrigation System, Hubei, China Annual Water Allocations for Irrigation and Other Beneficial Uses 1974-1998 ( Five Year Moving Averages )
700

600

D ESTIC, IN OM DUSTRIAL AN D H YDROPOW ER IRRIGATION

500 MILLION CUBIC MERERS

400

300

200

100

0 1976 1979 1982 1985 YEARS 1988 1991 1994

Figure 3. Annual deliveries to irrigation and other uses at Zhang He Irrigation District, China. Table 2. Changes in land and water productivity in Zhanghe irrigation district 1966-1998

Period

Annual irrigated area (103 ha) 139 135 118

Rice crop production (103 tons) 561 905 920

Rice yield (T/ha)

Rice water productivity (kg/m3 water supply) 0.65 1.17 2.24

1966-78 1979-88 1989-98

4.04 6.72 7.80

Some Riders on the Challenge Global information of course hides many of the challenges faced on the ground. Plus, increasing water productivity, while necessary, does not guarantee improved livelihoods. We thus need to attach some riders to this global challenge. Rider 1: Different Situations Different Needs

To meet human needs, river basins in the wold go through various stages of development. A basin is endowed with blue water that part of rain that contributes to river runoff, and green water the part that evaporates from land surfaces before reaching river systems. Over time to meet food, drinking, bathing, and other needs, humans tap into blue water by building dams, diversion structures, and conveyance infrastructure. We tap into green water by expanding agriculture and urban areas on the land surface. We can continue developing land and water up to a limit, set by the quality and quantity of land and water resources.

Figure 4. Stages of river basin development. Three stages can be identified. At first there is an initial development phase where people tap into resources to meet development needs. Much of Sub-Saharan Africa and parts of South and South East Asia are in this phase. We develop land and water resources to overcome basic problems of meeting food needs and providing access to water. The challenge in much of sub-Saharan Africa is to go through this development phase in a manner that meets the needs of the rural poor, and yet maintains fragile environments. A second phase is a utilization phase. In this period, significant amounts of infrastructure have been built, and much land is developed, and people are learning how to utilize these land and water resources. Typical challenges include saving water, increasing productivity of water, and solving allocation and distribution problems to assure equitable access. Many of the basins of South Asia, Latin America, and South East Asia are in this situation. A third phase occurs when utilizable land and water resources have been developed. Additional water cannot be sustainably tapped to meet additional needs, and competition increases. Allocation of water and increasing productivity of water are key concerns. Problems that appear when this phase is not well managed are: increased water pollution, groundwater depletion, and loss of access to water by the poor. This stage has been reached on the northern and southern edges of Africa, across Central Asia, Pakistan, Northwest India and the North China plain, and includes some of the worlds most important breadbaskets. The challenge is to reduce water use to sustainable levels while continuing to support rural livelihoods.

Rider 2 on the global challenge: Poverty alleviation

A simple focus on increasing overall productivity can increase food production to assure global national or regional food security but does not guarantee food security at the household level. That is a matter of access and distribution. Fortunately, we are learning how to target small farmers and alleviate poverty while increasing productivity. For example, more than 1.3 million treadle pumps have been sold in Bangladesh alone (Shah et al. 2000). The technology has reached a substantial number of rural poor in Bangladesh and irrigates about 600,000 ha of farmland and has raised the annual net household income by US$100 on the average. There is considerable activity in promising low cost drip irrigation technologies and water harvesting that offer hope in increasing access to water, water productivity, and income for the poor

Rider 3 on the global challenge: Managing for Multiple Uses Especially in areas of high water stress, a water action taken in one part of a basin, may have negative impact in another part of the basin. Capturing and using losses in irrigation may indeed leave less water to fisheries, to people, and to wetlands. Plus each drop of basin water can serve several purposes for fisheries, for drinking water, and for ecosystem services. For example, IWMI studies in Sri Lanka have indicated that lining canals in an effort to improve efficiency may cause problems for drinking water and health. A further example is the capture fisheries in Cambodias Mekong that are at risk from upstream developments along the Mekong17. A shift in institutional focus is required to manage water for its multiple uses. Understanding the tradeoffs, making sure that our actions improve overall basin water use, is an area that requires much more attention. Integrated Natural Resource Management methods offer great potential to help.

17

Capture fisheries production is important for Cambodias economy (valued at US$220-250 million at farmgate prices during the late 1990s) as well as for the protein intake and income generation of the countrys largely rural population.

Conclusion In conclusion, I have argued that how we manage water in agriculture holds the key to solving the water crisis. In essence, the global challenge for us is to grow more food with less water - decreasing water use in agriculture to meet environmental goals and other human needs, yet growing enough food, and improving livelihoods of the poor. This challenge requires substantial increases in productivity of water in agriculture. In 25 years, 60 percent increases of water productivity on irrigated lands, and 30% on rainfed lands will certainly go a long way to solving the crisis. It is not really up to us to decide how people in various countries and river basins use their water. People will have different ways they want to meet food and environmental security goals. But what we can do is offer feasible, sustainable choices and solutions that do not exist today. References
Alcamo, J., T. Henrichs and T. Rosch, 2000. World Water in 2025: Global Modeling and Scenario Analysis, in Rijsberman (ed), World Water Scenarios Analyses, Earthscan Publications, London. Cohen, Marc J., and Per Pinstrup-Andersen, 2001, The Case for International Agricultural Research in the 21st Century, note prepared for the Chairman of the CGIAR, 2001. FAO 2000. FAOSTAT Agriculture data base, www.fao.org, FAO: Rome FAO, 2000. Agriculture: Towards 2015-30 - Technical interim report. Rome, Italy: FAO Hong, L.; Li, Y. H.; Deng, L.; Chen, C. D.; Dawe, D.; Barker, R. 2001. Analysis of changes in water allocations and crop production in the Zhanghe Irrigation System and District - 1966 to 1998. In International Workshop on Water Saving Irrigation for Paddy Rice, 23-25 March 2001, Wuhan, China. Wuhan, China: China Rural Water and Hydropower. pp.4-18. IWMI, 2000. World Water Supply and Demand in 2025, in Rijsberman (ed), World Water Scenarios Analyses, Earthscan Publications, London. Molden, D., R. Sakthivadivel, and Z. Habib, 2001. Basin Use and Productivity of Water: Examples from South Asia. Research Report 49, International Water Management Institute, Colombo, Sri Lanka. Postel, Sandra, 1999. Pillar of sand: Can the irrigation miracle last? New York, NY, USA: W. W. Norton & Company. xvi, 313p. (Worldwatch book) Rockstrom, 1999. On-farm green water estimates as a tool for increased food production in water scarce regions. Physics and Chemistry of the Earth, Part B, Hydrology, Oceans and Atmosphere. 24(4): 375-384. Sakthivadivel, R., C. de Fraiture, D. J. Molden, C. Perry and W. Kloesen. 1999. Indicators of Land and Water Productivity in Irrigated Agriculture. Water Resources Development, Vol. 15, no. 1 and 2, 161-179. Shah, Tushaar, M. Alam, D. Kumar, R. K. Nagar, and M. Singh. 2000. Pedalling out of poverty: Social impact of a manual irrigation technology in South Asia. Research Report 45. Colombo, Sri Lanka: International Water Management Institute. Shiklomanov, L. A., 1999, World Water Resources: An Appraisal for the 21st Century. IHP Report. UNESCO, Paris.