Wake Up to Realities of River Basin Closure

Abstract

As societies develop, river basin water resources are increasingly controlled, diverted and consumed for agricultural, domestic and industrial purposes, hence reducing the ability to meet the growing demands from various sectors and interests. Basins are closed when additional water commitments for domestic, industrial, agricultural or environmental uses cannot be met during all or part of a year. Basin closure is already prevalent in the world today, with 1.4 billion people living in areas that have to deal with the situation. Societies may adapt to this in various ways, with reallocation of water, demand management or interbasin transfers as the primary means of dealing with the problem. However, ‘quick-fix’ measures such as further groundwater or surface water exploitation or ill-planned water appropriation that unfairly reallocates water from one user are common. Symptoms of poorly managed closed basins include groundwater overdraft, limited or no environmental flows, pollution and inequitable allocation of water. Thus, a pertinent question is whether there will be a hard or soft landing in closed basins—will the resource base fail to meet basic requirements causing undue hardship, or can societies adapt to achieving a soft landing. Surprisingly, limited attention has been given today to this urgent water situation.

Background

In August 2006 at the World Water Week in Stockholm, a seminar was jointly organized between the Stockholm International Water Institute (SIWI) and the Consultative Group on International Agricultural Research (CGIAR) to discuss the development of closed basins, the symptoms of closure, the social and ecological impacts of closure, and the adaptation processes required to achieve a soft landing. Six different river basin cases were presented to capture different processes of closure and draw out lessons to help society better adapt and achieve soft landings.

This paper summarizes some crucial points from these discussions, their relationships, and some conclusions that can be drawn from an issue of fundamental importance to the future of humanity.

Size of the Closure Phenomenon

River Basin Closure

Societies divert water from rivers and aquifers to meet their urban, industrial or food needs. But this process can happen only up to a limit, beyond which river basin water cannot perform important functions. To understand this, consider Figure 1. Committed outflows from a sub-basin include flows required to meet downstream allocations to meet societal needs, to dilute pollution, to meet environmental flow needs including sustenance of estuarine and coastal ecosystems, flushing sediments and controlling saline intrusion. A basin is said to be closing when these commitments cannot be met for part of the year, and closed when commitments cannot be met over the entire year (Molle et al., 2007) as shown in Figure 1. This definition evolved from the original definition given by Seckler (1996), stating that a basin is closed when no usable water leaves the basin, to the present definition that includes a more explicit inclusion of water rights and environmental flows.

Figure 1 The process of closure over time. In open basins more water can be allocated and diverted, while in a closed basin, flows are over-allocated and diversions of water have impacts on the levels committed for environmental flows and downstream users. Source: From Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture (http://www.earthscan.co.uk)

If environmental flows are judiciously defined, a well-managed closed basin can sustainably support agriculture, urban and other ecosystem services. However, if a basin goes past the point of closure, the basin is in a danger zone, and a question of sustainability arises, because it is not possible for river basin water to adequately support its many functions (Figure 2), and ecosystem services will be lost unless societies figure out a way to curtail water use or increase supply. Increasing supply through interbasin transfers is a common response to reopen closed basins, and desalinization to increase supplies is a much-discussed option. Demand management approaches which curtail diversions and depletion of water may decrease the amount of water diverted and depleted by rivers.

Figure 2 Development of basin closure. The stair-step line represents the addition of hydraulic structures over time, developed to make water accessible for human (domestic, industrial and agricultural) use, and the curve line represents depletion of water. Depletion can exceed available water, and even renewable supplies, when water is removed from storage, exceeding the rate of replenishment as is often the case with groundwater. A basin is closed when depletion reaches the utilizable amount of water. In this hypothetical case, the basin closed, and then depletion dropped.

A typical pattern emerges where as society grows and prospers, more water is diverted and depleted, so reducing river flows (Molden et al., 2005). In open basins, more structures can be added to tap into utilizable flows to meet additional demand. But there comes a point when there is no more utilizable flow left and the basin becomes closed. In spite of this, in many instances river basin exploitation continues and the depletion surpasses the utilizable flow, a potentially unsustainable situation. Depletion at some point will drop and land back to sustainable levels, but the question of this paper is whether this will be a hard or soft landing.

When river basins move past the point of closure, a major question is, if and how societies adapt. Will there be a hard landing with increased pollution, loss of ecosystem services, increased competition and inequitable sharing of benefits? Or will the landing be soft, with equitable, sustainable and efficient use of water?

The Global Extent of Basin Closure

There is tremendous concern that over the last 50 years many river basins supporting important economies—many of the world's breadbaskets—have reached the closed limit. Flows of the Colorado River into its delta dropped drastically from the 1930s pertaining to upstream water development, and eventually fell to near zero levels in the 1960s. Inflows into the Aral Sea dropped to one-tenth of their pre-1960 levels with intensive development of upstream irrigation in the 1960s and 1970s, leading to shrinking of the size of the Aral Sea. The number of no-flow days at the mouth of the Yellow River increased from 19 in the 1970s to more than 200 in 1997, a consequence of upstream development, especially for agriculture in one of China's important breadbasket areas. The list of rapidly closing and closed basins goes on to include river basins in this issue—the Jordan River, the Krishna River, Lerma-Chapala and many more, including the Murray-Darling River in Australia and the Indus River in India and Pakistan. A good indication of closed river basins is provided by the map of environmental water stress (see the map presented by Smakhtin, 2008, this issue) where areas with a high water stress index are likely to have met the point of closure.

Sharing a Limited Supply amongst More People: Water Crowding

The difference between a hard or soft landing is the ability to effectively manage water in situations of high water stress. The Comprehensive Assessment of Water Management in Agriculture (Molden et al., 2007a) defined the term ‘physical water scarcity’ as areas that lack enough water to meet demands, indicated by situations when the use to availability ratio exceeds 70%, a proxy for closed basins. The type of challenge will vary with the level of competition indicated by water crowding (the number of people that collectively depend on a finite amount of available water).

Figure 3 demonstrates two dimensions of water scarcity (Falkenmark, 1999; Falkenmark et al., 2007):

Figure 3 The difference between demand-related water stress and population-related water shortage. The diagonal lines show water demand in m³/per year. The boxes show the number of people living in different situations, with over 1.4 billion living in over-appropriated basins with high water stress or severe water shortages (based on IWMI analysis).

•	Demand-driven water stress: where there is a high usage compared to the availability of water.
•	Population-driven water shortage: where there are many people dependent on the availability of water.

High water stress occurs when there is high usage compared to the amount of water available. But the most challenging situation is one of severe water shortage, which develops when there is high water stress and more than 1000 people have to share each flow unit (1 million m³ of water per year). In other words, the higher the level of water crowding, the more difficult it will be to manage basin closure. The water management and governance required for a soft landing will need to be more sophisticated.

For example, at a population level of 2000 people per flow unit of 1 million m³/per year (p/flow unit), high water stress (40% use-to-availability ratio) already develops at 200 m³/per year in water demand. Previously this level has been referred to as the ‘water barrier’ (Falkenmark, 1989). If 30% of the water has to be left for environmental flows, obtaining 200 m³/per year will not be possible once a country has passed the water crowding of 3500 p/flow unit.

Venot et al. (2008, this issue) report that population growth in the Lower Jordan basin resulted in a sharp decrease in the per capita water availability from 3600 m³/per year in 1946 to only 163 in 2000, and it is expected to decrease further to 90 in 2025. What has contributed to this is a twelve-fold increase in the population since 1943 due to both natural reasons and successive waves of migration. This corresponds to an increase in water crowding from quite a moderate level of 300 people per flow unit to 6000 in 2000, and is projected to increase to 11 000 in 2025. 6000 p/flow unit is already far beyond the water barrier just mentioned. Together with this very high level of water shortage there has been a strong development of urban areas and improving living standards. Irrigated agriculture is now the main user of the country's scarce water resources and it has a very high consumptive use of water that leads to depletion of river flows and aquifers.

A recent assessment by de Fraiture et al. (2007) provides an estimation of the scale of the populations concerned. By 2000, 1.4 billion people were living in closed river basins (as defined by a rough proxy of more than 70% use to availability). Out of these, 1.1 billion lived in basins where there was already a severe water shortage. Another 180 million lived in closed basins where a severe water shortage was approaching (more than 600 p/flow unit). Using another method, Smakhtin et al. (2004) also estimated the number of people living in areas of environmental water stress in 2004, giving cross-verification on the size of the phenomenon.

Driving Forces: Avoidable and Unavoidable

Population growth, poor management, climatic change and vagaries of weather, the growing needs of cities and the large share of water used in agriculture are some of the reasons commonly cited as being the roots of water scarcity. These drivers lead to water and land-use change, which in turn leads to increased consumptive use by evaporation and transpiration, which in turn leads to a decline in river flows. In addition, there are societal demands and political pressures—people want more water resources for drinking, food production, recreation and aesthetics, and water development is at the top of the political agenda.

Responding to Increasing Water Demands

A common response to meet increasing water demands is to build more hydraulic works, and ultimately push river basins past the point of closure. A major reason for this is that it is politically easier to develop more water in the short term than take action in foresight for future problems. Molle (2008, this issue) points at a series of main drivers contributing to the so-called over-building of basins by which more water is made accessible for use facilitating increased water demands:

•	The construction of hydraulic works is facilitated by the convergence of interests of very influential actors including the state, private sector and development agencies. For the state, large-scale projects serve as political icons of importance for political support, private construction companies see them as business opportunities, and development banks have preferred large projects for relatively fast and large-scale changes.
•	Prestigious engineering works have always been key elements of state building (for example, the aqueducts built during the Roman Empire).
•	Fuzziness of water rights and unclear water accounting is linked to a tendency of over-allocating water to satisfy more water users, because of an unclear picture of supplies, their variability, and who really has rights to those supplies.
•	Malleability of cost-benefit analysis provides opportunities to make projects attractive even though long-term costs can be prohibitive.
•	Socio-economic concerns of poorer regions are given as a reason to develop more water for economic development.
•	High subsidies for projects, such as public irrigation schemes and the lack of sanctions for failed or sub-optimal projects, lessen risks and responsibilities for implementing agencies.
•	The push factor of agrarian pressure and shock events also contribute, such as El Niño-related climatic perturbations with prospects of famine or social unrest.
•	Lopsided governance and weak participation

One example is the Bhavani basin, tributary to the Cauvery River in Southeast India (Lannerstad, 2008, this issue). The recognized formal allocations within the lower basin already exceed the average renewable water resources. The increased diversions from the mid-basin Kodiveri weir over time pushed the demand past the point where available water could meet all the allocations made, and as Lannerstad states, the basin was closed ‘allocation-wise’. This included the planned allocations in the original plan for the Lower Bhavani Project command area.

Lannerstad also shows how the closure in one basin can spill over to another basin through interbasin water transfers. In this case the Noyyal basin closed, and to supply more drinking water a link was made to transfer water out of the water-stressed Bhavani basin. This case also shows how the strength of the urban economy in one basin can drive new allocations, depriving irrigation of water in the neighbouring basin. In this case, the neighbouring basins tend to see the Bhavani River system as a potential new water source.

Increasing Consumptive Water Use

Consumptive water use by evaporation and transpiration, especially by agriculture, leads to increased river depletion in three parallel ways:

•	Irrigation, where blue water from rivers and groundwater is evapotranspired in the process of plant production and is no longer available for use in the basin.
•	Crop-per-drop policies that tend to reduce blue return flows by involving efforts to make more of the blue water withdrawn available for the plants.
•	Land-use changes that capture an increasing percentage of rain and convert it to evapotranspiration, thus reducing runoff generation. Increases in afforestation and rainfed crop production have this effect in dry areas (Batchelor et al., 2003).

The Yellow River (Yang & Jia, 2008, this issue) offers a dramatic example where irrigated agriculture expanded considerably for increased food production, leading to significantly increased consumptive use in the upper and mid sections of the basin in the 1980s and 1990s. While this has been a plus for short-term economic development, it has led to critical concerns about long-term economic development and the environment.

In the Lower Jordan basin (Venot et al., 2008, this issue) agricultural abstractions today account for almost 80% of the total withdrawals. Irrigated agriculture competes with growing municipal and industrial sectors, and is increasingly the recipient of treated city effluents. In addition, irrigated agriculture is one of the main activities responsible for the current dramatic over-depletion of the basin's groundwater aquifers.

The Lerma-Chapala River basin in Mexico, as presented by Wester et al. (2008, this issue), is an important agricultural and industrial area and the water source for 15% of the country's population, including out-of-basin transfers to two major cities (Mexico City and Guadalajara). Since the 1980s the water resource has been over-committed and there is no reservation for environmental flows. The groundwater in the region is over-exploited and water tables are falling by 1–5 m per year. Overall there is river depletion of 104%, and no reservation has been made for environmental flows. In this case the closure of the upstream basin of the river, which ends in Lake Chapala, has led to drying of this 0.12 million hectare lake, one of the largest shallow lakes in the world, rich in biodiversity, a centre for tourism, and an economically important fishery.

Bioenergy Clouds Future Outlook

Biomass production for energy to partially replace fossil fuels is seen by many as an important response to climate change. But according to Berndes (2008, this issue), the potential scale of production of biomass for bioenergy (such as crops and trees), as indicated in a study by the International Institute for Applied Systems Analysis (IIASA) and the World Energy Council (WEC), might involve a new appropriation of ET that can be as large as the present global crop production. This will put further pressure on rivers. Similar to crop-based agriculture, increased consumptive use for bioenergy production in rainfed and irrigated areas will decrease river flows, either by reducing runoff to rivers, or through increased irrigation. For example, irrigating sugarcane and maize for bioenergy could substantially increase irrigation withdrawals in some basins.

Avoidable versus Unavoidable Dimensions of Basin Closure

Many of these driving forces towards river basin closure are difficult to avoid. Water development for poverty reduction has a strong justification. Water for domestic purposes for health and hygiene; water for industrial development expected to generate income and employment in poor countries; and water for producing food in low-income regions unable to import their food is a top priority for developing countries.

Of course, to a certain degree climate change is unavoidable, and will be influencing future streamflow. This may accelerate the closing process in the semi-arid tropics where the climate is expected to become drier with decreasing rainfall and decreasing runoff generation.

However, many driving forces are at least to a degree, avoidable. The many forces behind basin over-building, encouraging increased water demands, which are highlighted by Molle (2008, this issue), are possible to moderate. The water implications of biofuel production, including competing demand for water and land resources, must be highlighted to develop measures to mitigate negative consequences of bioenergy production.

Symptoms and Consequences

In the discussion of the consequences of river basin closure processes, the differences should be made clear between increased withdrawals and increased consumptive use, depleting the water resource. On the one hand, increased water demands will involve increased withdrawals and wastewater generation, and on the other hand, increased consumptive water use will deplete river flows and reduce the dilution capacity of the water resource. Increased water pollution due to increasing water demands, especially where the water resource is shrinking due to depletive water use, is an important consequence of basin closure.

Over-allocation and Increasing Demands

The situation in the Lower Jordan basin, described by Venot et al. (2008, this issue) illustrates the gradual change of this river basin from a somewhat natural state, to one where there is almost total control over the water resource. In 1950, only 10 000 hectares were irrigated, groundwater was untapped and abundant freshwater flowed to the Dead Sea. Today, 46 000 hectares are irrigated, nearly all surface water is tapped, and groundwater is severely mined. The Lower Jordan River basin is now a closed basin where no uncommitted outflows remain and the river valley, the highlands, agriculture and the cities are all interconnected and interdependent.

In the present situation in this basin there is a spatial interconnectedness of uses and flows between river valleys and highlands, and between water for agriculture and water for municipal supply. In the future, urban water needs will have to be covered by reallocations from agriculture or by bringing water from another source through interbasin transfers—an expensive proposition. There is no slack in the system, so that the brunt of any hydrologic fluctuations and supply uncertainty will have to be borne by agriculture and other important ecosystems.

Across National Boundaries

A common situation illustrated by the case studies is where an upstream country implements a large-scale water resources development, creating a situation for the downstream country that may be extremely challenging, such as the case in the Jordan Valley. When other countries are involved the situation becomes more complex because each country responds differently. Increasing water scarcity places more strain on individual countries attempting to solve their own goals, which simultaneously affects regional development goals.

The Lower Jordan River basin itself is an extreme case of downstream vulnerability (Venot et al., 2008, this issue). In this case a large-scale out-of-basin transfer upstream into Israel's National Water Carrier initiated the problem in the 1960s by considerably reducing river flows downstream. Moreover, water diversions in the Upper Yarmouk in Syria led to significantly less water availability in Jordan. In the 1950s, approximately 600 Mm³/yr entered Lower Jordan, and nearly 1300 Mm³ flowed into the Dead Sea (Courcier et al., 2005). Today, the contribution from Upper to Lower Jordan is about one-tenth of the previous amounts, and the inflow into the Dead Sea is about 275 Mm³.

Consumptive Use and Out-of-basin Transfers

The Musi basin in the Lower Krishna basin in Southeast India is a basin where decreasing inflow to two mid-basin reservoirs is causing concern, since well-established downstream areas, including the city of Hyderabad, depend on these reservoirs (see Box 1). Groundwater withdrawals support market-oriented agricultural production to meet a continuously rising urban demand for vegetables and fruits. As groundwater levels decrease and more groundwater is being withdrawn, the inflows to the two mid-basin reservoirs are being reduced and during some years the reservoirs do not fill up. Inflows to the two reservoirs are decreasing fast even though there has not been a large deficiency in the rainfall.

In the Yellow River (Yang & Jia, 2008, this issue), the key water control projects, hydropower plants and storages are in the upstream part. The remainder of the river lacks flow control capacity. With the economic development and population growth, the water withdrawal and consumptive water use have been increasing, mainly driven by irrigation, responsible for 80% of the depletion. As a result, water shortages have intensified over the years, which has even lead to problems in terms of sustainability of the economic development. The seasonal drying up in the downstream part began in the 1970s, and the river was left dry for more than half a year in the late 1990s.

Box 1. Musi River Groundwater may be an important source of water for both cities and agriculture. The link between large-scale groundwater withdrawals and its consequences in terms of decreased inflow to downstream reservoirs was illustrated by a study in the Upper Musi basin by Venkateswara Rao et al. (2006). This is a sub-basin of the Krishna River, in Andhra Pradesh, India, and an important source of drinking water for the city of Hyderabad (van Rooijen et al., 2005) and at the same time a major source of irrigation water for many villages downstream. As groundwater use has been increasing, inflows into two reservoirs have been observed to decrease, because of consumptive/depletive water use and the interconnectedness between surface water and groundwater. In addition, increased water conservation works to harvest more rain has also led to a decrease in flows supplying the river and the reservoirs. A major driver of increased water use is the growing demand for fruit and vegetables in the city, which require a source of blue water. Farmers respond by tapping into groundwater, or harvesting rain, to provide irrigation for their crops—important for their livelihoods. The results of 20 years of intensive groundwater monitoring were presented at the seminar, focusing on actual groundwater recharge in the upstream basin as indicated by rising water tables during the monsoon season. The results indicated good conditions for rapid groundwater recharge, suggesting that the groundwater use was in fact locally sustainable, except for some locations where groundwater over-exploitation suggested non-sustainable situations. Consequently, increased consumptive use of groundwater upstream, even if locally sustainable, has contributed to basin closure by ‘robbing’ water that would otherwise have contributed to reservoir inflow. In spite of the basin being closed, groundwater sources are relatively easy for farmers to use. It has been extremely difficult to control the use of these additional sources of water, and the connection between additional groundwater use and declining reservoir supplies is not well recognized and difficult to manage. This study underlines the need to consider all sources of water within a basin including groundwater and rainwater, and not just the river flows and to bring these into the overall context of water resource management for proper planning. Source: Venkateswara Rao et al. (2006)

This river depletion in the Yellow River basin has several serious consequences, including the seasonal dry ups in the lower section and also the reduction of the capacity of the river to carry and flush out its heavy sediment load to the sea. The risk of floods increases in the lower reach where sedimentation continues to raise the bottom of the river to well above elevations of the surrounding floodplains. The depletion of the Yellow River has led to the loss of virtually all water outflows to the coast, causing severe ecological consequences in the river delta and for coastal fisheries.

In the case of the Lerma-Chapala basin, Wester et al. (2008, this issue) shows the implications of upstream river depletion for Lake Chapala. Excessive upstream water use nearly resulted in the drying up of that lake, which has lost 90% of its volume in two decades.

The Dead Sea at the downstream end of the Jordan basin is another example of a large-scale drop in lake levels. Before 1950, 1250 Mm³/yr entered the Dead Sea, which is now reduced to less than 300 Mm³/yr, and the lake level has dropped over 15 m during the same time period (Courcier et al., 2005). The Jordanians now foresee a future where irrigated agriculture is uncertain, where the olive trees are seen as too thirsty and the depletion of aquifers has meant they have become saltier, which will jeopardize their use for domestic water supply.

In the contrasting case of the Bhavani basin, presented by Lannerstad (2008, this issue), development began downstream, mainly with irrigation development, and moved upstream towards more forested regions. Two major drivers of this downstream to upstream development are:

•	Over-commitment in the lower basin by expanding an irrigated command area, creating disputes between established farmers with specified water rights and the latecomer farmers who do not have well-established, formal water rights. During drought years the farmers have turned to groundwater, solving short-term water stress problems, but this has depleted the aquifer.
•	Continuing development in the upstream basin in spite of closure, involving both small-scale development for food production and rural water supply, and out-of-basin transfers to the cities Coimbatore and Tiruppur, resulting in further stress for the Bhavani basins.

Policy Implications and Response

How can a soft landing be achieved? The case studies showed several suggestions, and what is clear is that there are few examples of societies successfully coping with closure. It is also apparent that there is no single solution, rather several possibilities that could be adapted to each situation.

Responses to Closure

Collaboration, negotiation and water allocation

A major implication of basin closure is that changes in basin management require a reallocation of water (Molle et al., 2007). Where an additional allocation is given, for example, to cities or the environment, a reduction in another use is required. Where water saving practices are put in place, the saved water needs to be reallocated. Reallocation to balance needs and re-establish flows out of the Yellow River has been a central function of the Yellow River Conservancy Commission (Yang & Jia, 2008, this issue), considering equity, efficiency and economic development. In the Jordan River, as in many river basins, there has been an increase in allocation from agricultural to urban users. In over-allocated basins, some users will have to give up part of their supply. Hard choices have to be made.

Without adequate institutional mechanisms, reallocations are readily made from less to more powerful users (Molle et al., 2007). For equitable and productive water use, an institutional framework is required that takes into consideration the objectives of various users. Participatory approaches to reduce allocations would require considerable negotiation, and such negotiation processes do not imply at all that win-win situations can always be achieved. As shown by the Lerma-Chapala experience (Wester et al., 2008, this issue), even where collaborative governance and negotiation processes are put in place, these are easily derailed by power politics. Turton & Ashton (2008, this issue) point out that many closed and closing basins, especially in Africa, are transnational, implying that negotiation processes must be in place between countries. The results of negotiations will rest upon the ability and power of countries to negotiate. Where there are winners and losers, this requires a means to share benefits and compensate losers. When considering Southern African basins, Turton & Ashton conclude that incentives to seek consensual management are high where countries have water constraints to future development.

Water rights are a key element in reallocation processes. However, these are rarely in place to serve as a basis for negotiation (Merrey et al., 2007). If they are in place, there could be mechanisms for trading or transferring rights with compensation from one user to the next. For example, in the case of the Yellow River, measures are being taken whereby industries and cities can invest in water saving projects for the right to use water (Yang & Jia, 2008, this issue). If they are not in place, there is the difficult task of deciding who gets what rights to water.

Secure access has been found to be a key means of using water to alleviate poverty (CA, 2007). In the first instance this means finding sources of water for the poor. However, in closed basins (Molle et al., 2007) this means making sure that access to water is retained as water is reallocated from the rich to the poor. South Africa has instituted a water reserve for the poor to help ensure this takes place.

Environmental flows

Water allocation procedures should include a notion of environmental flows (Smakhtin, 2008, this issue), but allocations for environmental flows were not made in any of the case studies in this issue. In closed basins, even if an amount of environmental flow could be identified, and livelihood and ecological consequences could be predicted and monitored more easily, there is a difficult negotiation process to release water to secure the required amount of environmental flow. Establishing environmental flows was possible in Australia where the institutional set up allowed for negotiations and the setting of rules, and the government essentially bought water rights from farmers to release water for environmental flows. However, it is more difficult in other locations with weaker institutional frameworks and where funding would not be high on the agenda for environmental flows. The case of the Lerma-Chapala illustrates the serious renegotiation that must take place to establish the environmental flows necessary to maintain lake levels.

Water accounting

As basin water resources become tighter, there is a need for better accounting of water flows, and a better understanding of the complex flow network that develops (Molden & Bos, 2005). It is very easy to make mistakes thinking that water is saved or new water has been created—that results in one user giving up water to another user without compensation (Molle et al., 2004). Adequate water accounting is essential to backup negotiations.

In the case of basin closure it is the water depleted through evaporation, water pollution or water flows to sinks that becomes important as it sets upper limits on the use of water (Molden, 1997). Reuse and recycling are important considerations. Changes in land use, through changes in evapotranspiration, impact water flows. While in open basins it was enough to measure withdrawals, in closed basins it is essential to understand depletion of water, and ideally to allocate water based on depletion. As pointed out in the Lerma-Chapala example by Wester et al. (2008, this issue) “Water Accounting is not rocket science—it's harder!”

Demand management and water savings

An obvious approach is for users to curtail demand on water resources across all sectors—industrial, household and agricultural use—in order to reduce withdrawals and depletion of water. Demand management has been the focus of water resource management for many years and is seen as a solution to increased water scarcity. This was featured in the case studies in this volume. However, unless demand management is placed in the overall context of water resource management, it is unlikely to lead to success in achieving a soft landing. Demand management measures need to become part of the reallocation of water supplies, as discussed below.

Demand management approaches include economic incentives: charging or taxing for water use, or providing positive incentives to curtail water use. According to Venot et al. (2008, this issue), in Jordan the government has actively pursued this policy to tax for groundwater sold for industrial or aesthetic purposes, or tax on the volumes pumped over a given threshold for agricultural wells. Market-driven approaches include a means of trading and pricing water, where users could pay for additional supplies of water. However, these have not yet lived up to expectations because there are prerequisites of adequate regulation, questions of capturing benefits (who wins, who loses), and water rights issues must be dealt with (Merrey et al., 2007). Administrative approaches included policies aimed at reallocation of supplies of water. Water authorities reallocate water from one user to the next, and could determine a compensation mechanism for those who lose water.

Because irrigation is almost always the largest user of water in terms of both evapotranspiration and withdrawals, and is perceived to waste a lot of water, it is almost always the target of water saving practices. Water saving practices were emphasized in almost all of the case studies. There are a range of practices that do reduce the amount of water delivered to irrigation—drip and sprinkle irrigation and more precise field irrigation practices. Because these practices have a cost, additional incentives are required to put them in place—taxes or subsidies, or delivering less water and pressurizing farmers to employ the practice to use limited supplies of water more effectively. However, the amount of real savings that can be achieved is often much less than as anticipated in closed basins (Molden et al., 2007b; Venot et al., 2008, this issue). A major reason is that, as water resources become scarcer farmers do respond, either by increasing reuse of water or employing more efficient practices. At a basin, or sub-basin scale, surprisingly it is found that in most closed basins effective efficiency is already quite high and there is little further scope for water savings (Molden et al., 2007b). An alternative approach to focusing on field level practices is to take a basin perspective to find opportunities for real water savings, and then find an appropriate mix of technical, administrative and demand management responses.

Improving agricultural water productivity—growing more food with less water—becomes imperative as part of the demand management response (Molden et al., 2007b). This is particularly true in view of the demands for more water by agriculture. In addition to water savings, this approach also looks at the possibility to grow more food with the same or less amount of water. However, farmers who save water or increase productivity have an incentive to use the ‘saved water’ for themselves, which further exacerbates the problem of basin closure. Either water savings or increasing water productivity must have a water allocation process in place to effectively use the water in the basin where it is most desired. When water is reallocated out of agriculture, it makes sense to compensate farmers, or provide technologies to enhance water productivity.

Approaches outside agriculture

The closure of international basins presents special problems, as illustrated in the Southern African case. Turton & Ashton (2008, this issue) argue that countries with greater reserves of social, economic and technical assets are more likely to achieve a soft landing than those that do not.

As agriculture is typically the largest user of water, a response will be to have less agriculture and release water to other users. This has been the response in more developed countries. In these cases the countries can afford to buy water rights and there are many other opportunities other than farming, and food can be purchased from a variety of sources. The choice is more difficult in developing economies where agriculture is an important part of the economy, and there are few opportunities outside agriculture. Economic development thus seems to be one route to opening closed basins. However, somewhere in the world some basins will have to use more water for agriculture to meet growing global food and energy demands (de Fraiture et al., 2007), and to replace the food production that was reduced to solve basin closure problems in other basins.

Supply-side Approaches

More water being transferred to closed basins is a way to ease tensions, and often the preferred way. There are already many interbasin transfers in place and in operation. Megaprojects such as the South-North diversion in China (Yang & Jia, 2008, this issue), or India's National River Linking Project, are ways to get water from wet regions of countries to dry areas, adding more water to closed river basins. However, these projects also come with significant social and environmental considerations.

Conclusions: Hard or Soft Landing?

River basin closure has developed into a sizeable challenge of extreme importance, yet the phenomenon is a major blind spot in natural resource management. It has silently crept up in many river basins of the world, and has now reached such an extent that 1.4 billion people live in river basins that are closed or facing closure soon. Moreover, these areas are typically significant food breadbaskets, and must be addressed to solve some of the water and food issues of the future. The future prospect is that pressure on closing basins will only increase with additional food and bioenergy needs.

Symptoms of basin closure include the over-appropriation of water for people, resulting in inadequate or no environmental flows—an unsustainable situation in the long-run, and heavy competition for resources leading to reallocation of supplies—often unplanned, and with unforeseen consequences and increased pollution. However, the first response of societies appears to be actions that buy more time, or to cover-up some of the symptoms. Surprisingly, increased appropriation of water, especially groundwater abstraction, continues in many basins to alleviate short-term problems. If the situation continues, river basins could make a hard landing with possible consequences such as the collapse of agriculture, increased health risks and the potential for conflicts and migration. The challenge is to arrive at a soft landing where changes can be made in an equitable and sustainable manner, and continue to support human and ecological needs.

There is little evidence to show that current institutional arrangements for water resource management have been adequately able to deal with issues of basin closure. Managing basin closure will require systems analysis, seeing the basin as a complex socio-cultural-political-natural resource system, understanding how a change in water and land use in one part of the basin impacts others in the basin, involving diverse groups of users in informed decision-making processes.

The SIWI/CGIAR seminar fell short in answering the question whether there will be a hard or soft landing in closed basins, but an effort was made to highlight this ‘blind spot’ so that significant efforts can be made to solve one of the complex problems of our times.

Acknowledgements

The papers for this issue were originally presented at a seminar jointly organized by the Stockholm International Water Institute (SIWI) and the CGIAR through the International Water Management Institute, the Comprehensive Assessment of Water Management in Agriculture and the Challenge Programme on Water and Food.

This paper contributes to the Comprehensive Assessment of Water Management in Agriculture, with financial support from a range of donors including core support from the governments of the Netherlands, Switzerland and the World Bank in Support of Systemwide Programs. Additional support for the preparation of this paper comes from the Swedish Water House in support of the Assessment.