The Geographical Journal, 2010, doi: 10.1111/j.1475-4959.2010.00384.x Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India geoj_384 1..11 JEAN-PHILIPPE VENOT*, LUNA BHARATI†, MARK GIORDANO‡ AND FRANÇOIS MOLLE§ *International Water Management Institute, PMB CT 112, Cantonments, Accra, Ghana E-mail: [email protected] †International Water Management Institute, Kathmandu, Nepal ‡International Water Management Institute, Colombo, Sri Lanka §Institut de Recherche pour le Développement, Montpellier, France This paper was accepted for publication in August 2010 As demand and competition for water resources increase, the river basin has become the primary unit for water management and planning. While appealing in principle, practical implementation of river basin management and allocation has often been problematic. This paper examines the case of the Krishna basin in South India. It highlights that conflicts over basin water are embedded in a broad reality of planning and development where multiple scales of decisionmaking and non-water issues are at play. While this defines the river basin as a disputed ‘space of dependence’, the river basin has yet to acquire a social reality. It is not yet a ‘space of engagement’ in and for which multiple actors take actions. This explains the endurance of an interstate dispute over the sharing of the Krishna waters and sets limits to what can be achieved through further basin water allocation and adjudication mechanisms – tribunals – that are too narrowly defined. There is a need to extend the domain of negotiation from that of a single river basin to multiple scales and to non-water sectors. Institutional arrangements for basin management need to internalise the political spaces of the Indian polity: the states and the panchayats. This re-scaling process is more likely to shape the river basin as a space of engagement in which partial agreements can be iteratively renegotiated, and constitute a promising alternative to the current interstate stalemate. KEY WORDS: Krishna – India, river basin management, allocation, politics, scale Introduction A s demand and competition for water resources increase, a strong international consensus has been created regarding the need for Integrated Water Resources Management (IWRM) (GWP 2000). In this context, the river basin is invariably singled out as the natural unit for organising water management (Warner et al. 2008). There is, however, little agreement on how IWRM initiatives should be designed, implemented, and evaluated (Watson et al. 2007). Conventional approaches see river basins as rational units where technical ingenuity strives to ensure that supply remains in line with societal demand (Molle 2007). Designing appropriate basin-wise allocation mechanisms takes centre stage (Allan 2006) and holds the promise of addressing the impacts of water The Geographical Journal, 2010 development, land use and climate change on livelihoods and limiting the scope for conflict over scarce and disputed water resources. From a resource availability perspective, the river basin approach is appealing. Sharing and integrating data may show where there is balance or imbalance in resource allocation, but solutions rest in a much larger world than the river basin itself. A clear discrepancy between the rhetoric and the reality of basin water management worldwide has led many authors to question the river basin approach to sustainable water management on several grounds. First, delineating river basins is not easy, as groundwater basins and surface watersheds rarely match, making integrated management a challenge (Warner et al. 2008). Second, river basins are not neutral spatial units. Contrary to the common hydrologic vision that presents © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) 2 Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India them as natural and non-political, river basins are disputed arenas that overlap political boundaries and are appropriated – or not – differently by different users pursuing multiple, and possibly conflicting, objectives (Molle 2009; Warner et al. 2008). Third, natural resources management takes place across multiple scales and institutional levels. Water use and demand are, for example, driven by dynamics that go beyond the hydrological and topographic boundaries of water availability. The discourse around the naturalness of the river basin tends to erase this complexity, possibly leading to unsuccessful interventions and to the persistence of environmental degradation. Understanding the process of river basin development and the scope for future management strategies requires going beyond a water-centred and basin-centred vision of water management (Molle 2007). Following Watson et al. (2007) who call for a critical analysis of IWRM, this paper engages with the political dimension of river basins and the multiple drivers of water development and management by using the notion of space. It focuses on the rhetoric, practices and limits of current interstate allocation mechanisms to achieve equitable and sustainable basin management in the federal setting of India. The case of the Krishna basin in South India illustrates that interstate water disputes over the use of water are a persistent phenomenon despite calls for IWRM and existing basin water allocation procedures. Negotiations over basin water allocation have reached a stalemate, and there is little sign of improvement in the near future (Richards and Singh 2002). This paper aims at understanding why and at identifying possible alternatives. The paper provides a socio-technical analysis of water availability, taking into account the social forces that shape water use and allocation. Methods used are multiple and characteristic of interdisciplinary research. They include a literature review, secondary data analysis for land use change and water accounting, interviews with key informants, and analysis of policy document. The second section gives the analytical lens of the study. The notions of space and scale are used to shed light on the root causes of basin water management challenges in India. The third section describes the trajectory that is the history, the politics and the multiple drivers of water development in the Krishna basin. The fourth section highlights the shortcomings of current mechanisms for interstate negotiation and basin water allocation in the Indian context. The fifth section explores possible alternatives to current basin water negotiations and investigates the potential for establishing new negotiation platforms in order to achieve sustainable water management. A short conclusion reflects on the Krishna case study. Recognising the river basin as a political arena, among many others, in which water-related decisions are taken and going beyond The Geographical Journal, 2010 a water-centred vision is needed to attain sustainable water management. The river basin: space of engagement or dependence? Space and scale are concepts that have long been used in the field of geography: Space is a [social] product [. . . it] serves as a tool of thought and of action [and . . . ] in addition to being a means of production it is also a means of control, and hence of domination, of power. Lefebvre (1974, 26) In the literature, there has been a lively debate about the politics of scale (Smith 1992), surrounding the idea that scales are socially constituted in historically specific ways (Perkman 2005). Beyond their apparent naturalness as ‘relatively differentiated and selfenclosed geographical units’ (Brenner 2001, 599), scales are social and political constructs mobilised by individuals, groups, or the state in order to frame problems and solutions in particular ways and to favour specific courses of action (Delaney and Leitner 1997; Lebel et al. 2005; Marsden 2000). This reveals three interconnected aspects of scales (Perkman 2005): 1 their nature as social formations, implying the need for political mobilisation for the construction of a new scale; 2 their function as sites of governance, dependant on the existence of institutional arrangements for channelling interests and coordinating decision making; and 3 their construction as an unit for intervention via narratives and discourses. Space then becomes an open and ongoing production (Lefebvre 1974; Massey 1993) subject to re-scaling processes; that is, the emergence of new territorial scales (Brenner 2001). This ‘continuous reshuffling and re-organization of spatial scales is an integral part of social strategies and struggles for control and empowerment’ (Swingedouw 1997, 141 as quoted in Brenner 2001, 608). The river basin has emerged as the natural unit for the planning and the management of water resources (see Molle 2009 and Warner et al. 2008 for critical reviews). It is a ‘new scale’ but it is also one way among others to structure space. Its uses reveal particular visions of space and of its functions. Ghiotti (2006), for example, highlights that the river basin can be seen and used as: 1 an operational entity to structure social and economic territories, design policies, appropriate and control Space; © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India 2 a platform to articulate and coordinate processes that are designed at – and affect – other scales; and 3 a space whose control is disputed by other constructions of space such as the nation, the state, the community. These three aspects of the river basin can be brought in alternatively or concomitantly by different actors whose agendas unfold at multiple levels. They reveal that river basins are political constructs. Using the terminology of Cox and Mair (1991), the first two dimensions of the river basin as identified by Ghiotti (2006) indicate a space of ‘engagement’ (structuring and operational entity) and a space of ‘dependence’ (coordinating platform), respectively. Both dimensions are needed for basin management mechanisms to be effective. This also requires the existence of institutions constituting the river basin as an object of governance and a social support to provide resources and legitimacy to those institutions. The next section focuses on the Krishna basin and examines whether these requirements are met in the Indian context. Multi-level trajectory and space of dependence: the example of the Krishna basin, South India The Krishna river basin is the fifth largest river system in India. It drains an area of 258 514 km2 in three South Indian states, namely Andhra Pradesh, Karnataka and Maharashtra (Figure 1). In 2008, the basin population was 74 million (Venot 2009). The Krishna River originates in the Western Ghats, drains the dry areas of the Deccan plateau, and forms a delta before discharging eastwards into the Bay of Bengal. Since the 1850s, the Krishna basin has witnessed a dramatic development of irrigation and water use with the construction of large-scale projects, minor irrigation schemes and a booming groundwater economy. In the early years of the twenty-first century more than 130 medium and major irrigation projects tame the Krishna River and its tributaries. Together, they can store about 54 billion cubic metres (Bcm), or 95% of the pre-1965 runoff when few dams were in existence. A further 175 000 minor irrigation structures have a significant impact on the overall river basin balance (Venot 2009). Historical basin-wise water accounting (Figure 1, see Molden 1997 for details on the methodology) highlights that total water depletion, defined as the use or removal of water from a river basin that renders it unavailable for further use downstream (this includes direct rainfall evaporation) rose by 19% over the period 1955–2000. It averaged 181 Bcm/year in 1990–2000. The increase was primarily due to an expansion of irrigated agriculture and to widespread supplemental irrigation of formerly rain-fed crops. As a result, the discharge to the ocean declined from 44.3 Bcm/year in 1955 to 17.1 Bcm/year in 2000. In © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) 3 2001–4, during an extended period of low rainfall, the outflow to the sea dropped further to 0.8 Bcm/year (Venot 2009). Shiva (1991) highlights that water demand is at its highest in Andhra Pradesh (47% of the total demand in the Krishna basin), followed by Karnataka (33%) and Maharashtra (20%), while Maharashtra and Karnataka, in this order, are the main contributors to surface water flows due to higher rainfall and lower water uses. Under these conditions, the planning and development of water projects in the Krishna basin have traditionally led to major interstate disagreements that contribute to intense inter- and intra-state political struggles. This tension is expected to further sharpen as demands for water increase. In the early stages of basin development, in 1969, the Government of India (GoI) set up the Krishna Water Disputes Tribunal (KWDT) as an attempt to solve enduring interstate water disputes. In 1976, the KWDT allocated the 75%-dependable flow (flow exceeded in 3 out of 4 years) between the three states. The ruling was provisional and expired in 2000. Interstate negotiations to revise the agreement have been ongoing since 2004 but seem to have reached a stalemate. The Krishna river basin is a hydrological and topographical reality but the planning and management of basin water allocation and uses are embedded in a broader political context. Processes unfold at multiple scales and numerous non-water issues are at play. The next sections describe these multi-level and political processes at the federal, state and local levels and how they shape the river basin as a space of dependence rather than a space of engagement. This sets limits to what can be achieved through basin water allocation and adjudication mechanisms – tribunals – that are too narrowly defined. Federal level The Indian water resources governance structure and policy process remains one of centralisation and hierarchy, dominated by water bureaucracies reluctant to devolve their power (Mollinga 2005). Therefore, institutional and policy initiatives such as decentralisation and participatory management in the mid 1990s and the National Water Policy of 2002 have failed to shape river basins as spaces of engagement. These new scales are yet to be socially appropriated by users as operational platforms. However, river basins clearly constitute spaces of dependence. The above water accounting shows that spatial and historical dynamics of basin water result from agricultural and rural development policies that have been promoted since India gained independence. A first group of policies has concentrated investments on groups and areas that offered the highest potential for development and political gains, leading to medium- and large-scale irrigation projects in downstream regions. Other policies promoting minor irrigation, watershed and The Geographical Journal, 2010 4 Maharashtra Chhattisgarh Pune Warangal Solapur Vishakhapatnam Bijapur Kris a n Kolhapur Goa Hyderabad Krishna Raichur Dharwad h Vijayawada Cumulative water depletion vis-à vis total rainfall in the Krishna Basin, South India Andhra Pradesh Bellary Bay of Bengal 1955-1965 Nellore © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) Mangalore 50 0 100 150 200 250 1990-2000 Krishna River Basin Chennai Bangalore Karnataka Billion cubic meters Mysore Coimbatore Kerala Comparative water depletion in the Krishna basin: Evapotranspiration from different land covers and human consumptive uses Tamil Nadu Cochin Irrigated agriculture 10 20 30 0 10 20 30 0 10 Trivandrum Source : J.P. Venot, International Water Management Institute (Hyderabad office). M&I uses 1955-1965 % 0 200 km Natural vegetation Rainfed agriculture Discharge to the Ocean (%) 1990-2000 Layout and design: Cyrille Suss and Claire Levasseur Figure 1 The Krishna basin: setting and historical water accounting 20 0 4 0 10 20 30 Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India The Geographical Journal, 2010 Orissa Mumbai Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India drought-proofing programs, or facilitating access to credit, targeted upper secondary catchments that had less benefited from the Green Revolution package, leading to increased groundwater use. State level State policies, priorities and politics drive water resources development and define the river basin as a space of dependence rather than a space of engagement too. In all three states of the Krishna basin, the protective approach to irrigation development that consists of spreading available water thinly over a large area and to a large number of farmers remains central despite its practical failures (Mollinga 2003). Indeed, it is a way to secure the political support of rural constituencies that constitute the true spaces of engagement in the democratic setting of India and provides convenient legitimacy to the state and its irrigation bureaucracy for infrastructural development on the basis of equity and poverty-alleviation principles. The latter also motivate credit policies and electricity subsidies that fuel an uncoordinated development of local groundwater. In the Krishna basin, groundwater depletion tripled from 1955 to 2000 and accounts for 40–50% of all irrigation depletion, with dramatic impacts on downstream water availability and reliability (Venot 2009). Other state agendas drive water development regardless of basin boundaries and water availability. Producing hydro-electricity by transferring water to the west coast is a priority for Maharashtra; limiting fiscal deficit is important for all three states and has been a major driver for the promotion of participatory irrigation management (Gulati et al. 2005); in Andhra Pradesh, preventing regional tensions and state implosion (notably under the pressure of independence claims from the Telangana region) has long been and still is a major driver of infrastructural development (Venot 2009), as is meeting the water needs of its capital city, Hyderabad, through water transfers (Celio et al. 2009). Given those multiple rationales that underpin water availability, use and distribution, water resources management strategies that define the river basin as the sole scale of intervention – such as allocation procedures in India – are likely to fail. 5 causing shortages further down the irrigation system. However, managers may have to change their practices during low flow years, introducing rotational supply or deficit irrigation to meet crucial demand of standing crops. This induces farmers’ adjustments such as shifting cropping pattern and calendar, resorting to groundwater use, suspending cultivation, livestock sale, out-migration, and tampering with the irrigation system (Venot et al. 2010). On the one hand, uncoordinated strategies can subvert management and allocation efforts at the basin/project level, generate environmental problems, and induce a social selection as not all farmers have the ability to adjust to scarcity. On the other hand, understanding and accounting for local innovativeness can help in the implementation of basin allocation mechanisms. Challenges to river basin management in India There is no river basin authority and there has been no basin-wide planning Iyer (1994,191) As early as 1956, with the Rivers Board Act, the concept of integrated basin development and management was seen as a possible response to water problems in India (Iyer 2003). The idea is reiterated in the National Water Policy of 2002 (GoI 2002). As most Indian rivers cross state boundaries, interstate negotiations are key elements of sustainable basin management. The Constitution of India provides the overall framework for the apportionment of water among states. As a last resort – after direct negotiations among states have failed – interstate water disputes are meant to be adjudicated by tribunals set up by the GoI at the request of at least one of the riparian states. But most tribunals have reached a stalemate and have failed to shape the river basin as a space within which stakeholders can engage. In the words of a former senior public servant of the Indian water sector, interstate conflicts over Indian river waters are becoming intractable (Iyer 2003). This section shows how this intractability is linked to technical and institutional shortcomings of tribunals but also to their failure to recognise that politics are an integral part of river basin management. Local level Cumulated local level practices impact water allocation and use at higher scales, hence defining the river basin, once again, as a space of dependence. In average years, farmers tend to intensify production. This, in turn, requires a more generous and flexible water supply and has social and environmental externalities because of a decline in downstream water availability. Managers may meet those demands when there is little competition for water. This leads to increasing project-level water use while © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) Hydrological complexity: technical limits Other than the 2007 Cauvery award, most tribunal awards were enacted in the 1970s and the early 1980s. They were based on an incomplete understanding of river basin hydrology and they need to be revamped. In the case of the Krishna, several technical shortcomings explain why the tribunal could not manage and accommodate change, therefore leading to over-commitment, conflicts and environmental degradation. The Geographical Journal, 2010 6 Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India First, the KWDT explicitly recognised the notion of ‘surplus water’ as all water in excess of the 75% dependable flow and stated that the downstream riparian, Andhra Pradesh, was free to use it with the caveat that it shall not acquire any formal right to it. But as environmental concerns take centre stage – and the need for instream water flows is increasingly recognized – policies, rules and infrastructure based on this notion become counter-productive. They indeed create new demands for water already committed to the environment and are likely to compound conflicts. Second, the KWDT allocated fixed volumes of surface water on the basis of the 75%-dependable flow. It did not offer an adequate platform to manage low-flow years, an issue of utmost importance for the downstream riparian. A proportional apportionment scheme was contemplated but never implemented due to the early opposition of Andhra Pradesh that felt its irrigation projects – already well developed at that time – would be constrained in the case of scarcity (Sajjan 2005). Given upstream water development, Andhra Pradesh might be willing to reconsider such an arrangement. Third, groundwater use was neglected by the KWDT when it stated that the three states ‘will be free to make use of underground water within their respective territories in the Krishna river basin [and] use of underground water shall not be reckoned as use of the water of the river Krishna’ (GoI-KWDT 1976). Recent trends in groundwater abstraction in secondary upstream basins lead to streamflow depletion, lower downstream availability and reliability, and constitute new challenges that can’t be addressed by current allocation procedures. Lastly, the KWDT mentioned that ‘beneficial use shall include uses . . . for domestic, municipal, irrigation, industrial, production of power, navigation, aquaculture, wildlife protection and recreation purposes’ (GoI-KWDT 1976). However, it did not mention the relative shares allocated for these consumptive and non-consumptive uses. As new water claims arise, allocation mechanisms must move from allocating surface ‘blue’ water between irrigation projects to more inclusive mechanisms that account for multi-level and multi-sectoral dynamics. This requires a better understanding of the hydrology, its variability, and the interactions between surface, ground and ‘green’ water (soil moisture and rainfed evaporation) on the basis of transparent and effective data collection. negotiations. Delays in the establishment and the proceedings of a tribunal, and in the giving and the notification of its award, are seen as further flaws of the current approach (Iyer 2003). More significantly, tribunals accommodate different water rights regimes. First, they endorse the riparian rights of the states. Second, they explicitly recognise prior appropriation rights by protecting existing use. Third, they sanction the rights of the states to further develop water resources by considering planned future uses (Venot 2009). Multiple rights regimes shape the river basin as a space of dependence. In the case of the Krishna basin –where the award was provisional – the three states engaged in massive developments of their hydraulic infrastructure to lay claim to water resources and ensure they would be holding a prevailing position during the renegotiation. This led to an overcommitment of basin water where water demand exceeds the KWDT allocations. Another recognised limitation is the institutional deadlock the actors find themselves in if one of the parties does not accept the tribunal award. Although declared as binding and final, the award is often questioned by states who subsequently file review petitions, seek the intervention of the Supreme Court or of the federal government (Iyer 2003). This has been observed after the recent adjudication of the Cauvery (Pani 2009) where Karnataka and Tamil Nadu contested the award and repeatedly called for further intervention by the Prime Minister of India, without any foreseeable solution. The most common suggestions to address those institutional shortcomings include the creation of a national and independent water commission and a permanent federated structure incorporating river basin authorities and water user associations, as well as the provision of effective judiciary mechanisms to make the award of the tribunal truly binding (for instance, Richards and Singh 2002). Others highlight that the system is not badly designed but has been wrecked by ‘politics’ and that what is needed is a respect of the constitutional provisions (Iyer 2003). But such proposals fail to acknowledge that conflicts between state governments are not the only form of conflict. Conflicts are multiple in nature and scale; they can occur between the centre and the states, between regions of a state, between agriculture and urban users, and between users and the state. To address those issues, other resolution mechanisms – at other scales than the river basin – are required. Identity and politics: a disputed space Institutional shortcomings: dependence rather than engagement Richards and Singh (2002) highlight that the process of adjudication (tribunals) is ambiguous and opaque, and inevitably leads to adversarial positions and maximum claims that impede any middle ground The Geographical Journal, 2010 Interstate water disputes are entangled in the broader Indian polity. Complex financial, political, and developmental relations between states and the federal government underpin interstate water conflicts. They are, nevertheless, not addressed by the adjudication process. D’Souza (2002) indeed highlights that legal © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India adjudication through tribunals tends to insulate the allocation process from the socio-historical context of basin development. It depoliticises the issues by framing them in terms of equitable apportionment that technology and science could solve while, she argues, interstate water disputes are an expression of neocolonial development conflict pressed upon states by the federal government (D’Souza 2002). Less forcefully, it is widely admitted that tribunals are institutions characteristic of a strong ‘paternalistic’ federal government and ‘childly’ non-assertive states (Iyer 2003; Anand 2004; Maitra 2007). The KWDT, for example, internalised the development project of a young nation, championed at the federal level by the Planning Commission of India that emerged as the main agency for dam planning and development (D’Souza 2006). With a strong involvement of the federal government through the discourse of development and the financing of irrigation projects, it is no surprise that the KWDT was ineffective in curbing water abstraction. It can then be said that the river basin approach remains an artifact shaped by the federal government (D’Souza 2002). The river basin becomes a disputed space through which the federal government tries to assert its control over the territory of India through water negotiations. But the political configuration of India has changed. Economic processes remain piloted by the Planning Commission of India and tend to a greater centralisation of decisionmaking. Nonetheless, coalition and regional politics are assuming an increasing importance and the push towards democratisation and decentralisation means that states are assuming an ever prominent role in Indian politics (Kohli 2001). This induces a disconnect between the dispute resolution mechanisms (the tribunals) and their operational environment (Maitra 2007). States are keeping the upper hand on their water resources. In addition to the fact that the ‘development mission’ of the central and state governments sets limits for basin authorities to regulate water use, the latter are perceived by state governments as possibly eroding their authority to the benefit of the central government. Hence, calls for basin agencies have remained dead letters since the River Board Act of 1956 (Iyer 1994). This political tension is clearly illustrated by the endless debate over the status of water and interstate rivers in the Indian constitution. Water-sharing principles have always been highly controversial and national policy documents have mostly steered clear of the idea (Iyer 1994 2003). Identity and politics define the river basin as a disputed space that states refuse to engage with. This may be due to the ‘quasifederal’ nature of India centred on the notions of regional distribution, central planning, and political integration (Maitra 2007). This calls for exploring alternatives to the current basin-centred vision of dispute resolution. © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) 7 Opportunities for governing basin water in India Multiple agendas drive water-related decisionmaking at multiple scales, hence shaping the river basin as a space of dependence. Institutional arrangements such as tribunals narrowly focus on sharing surface blue water among states and fail to acknowledge the multilevel drivers of basin development and to give any role to actors other than water bureaucracies. They therefore fail to establish the river basin as an arena the society can truly engage in. As different actors give different weight to different problems, extending the domain of negotiation from that of a single river basin to multiple scales and to non-water sectors by including, for example, agriculture, energy, trade, and environmental issues, could foster cooperation. These re-scaling processes, we argue, are more likely to craft the river basin as a space of engagement for integrating and sharing information, and accommodating the conflicting positions of the three riparian states of the Krishna basin. This implies adopting a dynamic vision of scales and space in which the focus is about understanding and accommodating nested arrangements. The next sections distil some elements for an alternative approach to interstate water negotiations in India as a means to design socially and environmentally acceptable water management strategies for the future of the Krishna basin. Common to the options discussed below is a call for direct dialogue among the states and between the local users and the states – through their political representatives at the Panchayat level and the states and the states within the overall framework set up by the Constitution of India, that is, the tribunals. Beyond blue water: environment and multiple sources of water There is increasing evidence of the adverse impacts of water and land degradation on people’s livelihoods and downstream ecosystems. Environmental concerns have started to gain strength and the notion of environmental flows that embodies a compromise between water development for productive use and ecosystems’ protection is beginning to generate significant interest among Indian policymakers. But the volumes of water required are significant (Smakhtin and Anputhas 2006) and environmental flows determine how much water supply can be further developed. They are also often used by downstream states to justify higher water demands while they must be monitored in the basin as a whole. Mainstreaming environmental issues in the Indian water sector agenda remains a challenge, notably in fully committed river basins such as the Krishna. Decisionmakers need to be convinced that letting free flowing water in the river channels can yield benefits to the society as a whole. Discussions over environmental flows could be linked to the debate on agriculture and payments The Geographical Journal, 2010 8 Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India for environmental services. This would also mean opening up the space of water negotiations to environmental groups of the civil society, and to the ministry and the states departments of environment and forest. Agriculture and food policies designed at the federal level have strong impacts on basin water development and the environment. They are often at loggerheads with objectives of water demand management and environmental preservation (Venot 2009). Procurement and minimum support price policies, for example, act as incentives to farmers to cultivate water-consuming crops such as sugarcane, paddy and wheat, regardless of water availability. In upstream secondary basins, this has led to significant land use change and environmental degradation such as declining baseflows and groundwater overabstraction. But, rainfed agriculture still remains the main productive water user at the basin level (Figure 1) and is the major livelihood of the poorest communities. The environmental and social significance of rainfed agriculture need to be recognised and discussed together with blue water allocation. This would provide an opportunity to link and rationalise water interventions and agricultural and rural development planning by involving users and decisionsmakers from all relevant agencies, from the local to the federal level. For instance, interstate cooperation could be as follows: downstream riparians could participate in the financing of watershed programs that enhance green water use in upstream states in exchange for reliable and timely blue water supplies. Delays in runoff would also have additional benefits and perform environmental services such as flood mitigation. Enhancing green water use in upstream watersheds might result in decreased runoff downstream and potential conflicts. A proper understanding of basin flows in relation to consumptive water use of green and blue surface and groundwater (see Molden 1997 describing the water accounting framework) and the related spatial and social trade-offs should be the basis of such negotiations. Reforming the procurement system through which India redistributes grains from surplus to deficit regions by enhancing direct trade linkages between surplus (Andhra Pradesh, downstream) and deficit (Karnataka and Maharashtra, upstream) states could also help in negotiating an agreement between the three riparian states, with the federal government playing a regulatory role. In India, grain procurement dates back to the 1960s. By the early 2000s, the Food Corporation of India was purchasing about 20% of the total net production of cereals in India (50% of all cereals in the market) and Andhra Pradesh notably contributed to 30% of the total procurement of rice (Landy 2009). With direct procurement between states being possible since 2002, Andhra Pradesh surplus could meet the rice deficits of both Karnataka The Geographical Journal, 2010 and Maharashtra. Upstream states could commit themselves to securing surplus paddy production downstream through adequate blue water supplies; the downstream state would commit itself to selling paddy, that is, virtual water, at a preferential rate to upstream states. This move towards a more decentralised procurement has already been advocated at the state level and could contribute to solving some of the challenges of the current system (transport, stocks; Landy 2009). While politically sensitive, it has been shown that virtual water trade has the capacity to limit the impacts of local water scarcity (Allan 2007). Beyond water: sharing the benefits of the energy–irrigation nexus and other uses Power-generation projects are also a source of conflicts between states as the most efficient option for hydropower generation is transferring water out of the Krishna basin towards the west coast, in Maharashtra. The first KWDT accounted for this sensitive issue when designing its award in the 1970s (GoI-KWDT 1976). However, increasing hydropower production remains high on the Maharashtra agenda. It is strongly opposed by the two downstream states as transfers would affect downstream water availability. All three states have increasing power needs. Coupling water and energy issues in multi-sectoral negotiations among relevant agencies could allow reaching an agreement whereby water would be transferred out of the basin. This would limit the risks of floods and part of the energy produced would be supplied to downstream states through an extension of the power network partly funded by the upstream riparian state. Finally, power supply policies are effective tools for groundwater demand management (Shah 2009) and could constitute an innovative instrument for interstate basin water allocation if they were to be considered by tribunals. Municipal and Industrial (M&I) uses have tripled during the past 50 years. They only represent 1% of all depleted water in the basin (Figure 1) but already drive re-allocation from rural to urban areas (see for instance Celio et al. 2009 on the case of Hyderabad). Because of their economic and political importance, M&I users get the upper hand and meeting their demands becomes the first priority of governments (Molle and Berkoff 2006). De-facto reallocations and their impacts on rural areas are likely to become more common in the near future. Recognising this reality in the fast-developing states of the Krishna basin is required. When planning re-allocation to M&I users, more attention needs to be given to the consequences of such transfers. This can be achieved through more direct negotiation among the concerned communities. Finally, linking water negotiations to trade discussions needs to be explored as trade agreements have been shown to drive changes in water use patterns (for © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India instance, Neir and Campana 2007 on the case of NAFTA and US–Mexico water disputes). Promoting new platforms of negotiation The above considerations highlight the need for multilevel cooperative negotiations among the three riparian states of the Krishna basin, whereby the federal government holds a regulatory role. At present, tribunals give little incentive for direct cooperation at the state level (Anand 2004). However, such direct negotiations are made necessary because of the changing politics of India, with increasingly stronger states. The recent past provides examples of such negotiations. In the late 1990s to early 2000s, the state of Karnataka decided unilaterally to increase the height of the Alamatti dam located on its territory. This would have had significant impacts on water availability further downstream, in Andhra Pradesh (Venot 2009). This move sharpened states’ disputes over the Krishna waters. But pressure by decisionmakers of Andhra Pradesh, epitomised by the lodging of an appeal in the Supreme Court of India, led to a bilateral agreement between Karnataka and Andhra Pradesh on the height and the operation rules of the Alamatti dam. This happened in the absence of a basin-wise regulatory framework as the first KWDT tribunal award expired in 2000 and the new tribunal had not reached a decision. Hyderabad provides another example whereby Andhra Pradesh signed two separate agreements with the states of Maharashtra and Karnataka to secure water transfers from the neighbouring Godavari river basin to its capital city (Celio et al. 2009). In both instances, state-level – rather than basin-level –agreements were reached. The Tungabhadra irrigation project gives another example of a state-designed agreement between Karnataka and Andhra-Pradesh (Mollinga 2003). Tribunals should offer a formal arena where agreements among states regarding specific projects can be negotiated. Such direct inter-state agreements – recognised and sometimes initiated by the federal system – are common in the United States (interstate compacts) and Australia (water-sharing plans). Linking negotiations over the Krishna to other spaces and notably other river basins, such as the neighbouring Godavari, could also offer new perspectives on interstate allocation. India is currently contemplating the construction of a national water grid called the National River Linking Project that could provide the needed framework, but this would also require opening the scale of negotiations to actors located beyond the topographic and hydrologic boundaries of the Krishna basin and a careful consideration of equity and environmental concerns. Generally speaking, there is a need to enlarge the platform of negotiation beyond the irrigation bureaucracies of the three states and the experts nominated by the federal government. Within the state apparatus, © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) 9 this means extending the negotiation ‘horizontally’ so that members of other departments such as the energy, agriculture, rural development and environment departments participate in the discussions. The other possibility is to entrust broader portfolios to a limited numbers of negotiators with different backgrounds, so that sectoral interests do not dominate the negotiation process. Involving third parties from other river basins or other states who would have less at stake in the negotiation could also help in de-emotionalising the dispute and reaching an agreement (Anand 2004). In Australia, it has been observed that the possibility of a third party state intervention shapes the river basin as a space of engagement and pushes negotiating states to reach an agreement. Beyond the state apparatus, the negotiation process needs to be extended ‘vertically’ so that water users can directly voice their demands. This could be done through their local representatives: the Panchayats that are the true space of political engagement in the democratic setting of India and whose importance is deemed to increase in the current context of decentralisation, as mentioned in the National Water Policy of 2002. Agreements at the local level are prerequisites for socially and environmentally acceptable water allocation mechanisms at larger scales. This has already been tested in the case of the South Indian Cauvery river basin, though farmers’ input to the basin negotiation process has been indirect and consisted of better awareness (Pani 2009). Solutions emerge at the local level too. Farmers’ adjustments are major buffers during low-flow years and offer an opportunity for adaptable allocation frameworks. But mere coping adjustments need to be converted into pro-active strategies through services supporting the agricultural sector, such as better access to credit, improvement of market chains, extension services, better access to information, supporting diversification, adequate planning of conjunctive water use (Venot et al. 2010). Finally, economic diversification can help in overcoming water scarcity as other sectors consume less water than agriculture. Policies should focus on turning the benefits derived from water use into other forms of capital and provide means to make this new capital productive and people less dependent on the resource base. Diversification offers the opportunity and the resources that are needed to achieve water management solutions by providing rural dwellers opportunities outside the agricultural sector. Conclusion IWRM with a focus on the river basin as the primary unit for planning, development and management is heralded as a solution to water problems worldwide. While appealing in principle, practical implementation has often been problematic. This paper uses a case study of the Krishna basin in South India. It highlights that interstate water conflicts remain largely The Geographical Journal, 2010 10 Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India unresolved in India despite such long-held calls for IWRM and the existence of basin water allocation mechanisms. Framing the river basin as a social and political space sheds light on the root causes of continued interstate water disputes in India and allows opportunities for sustainable water management to be identified. The terminology of Cox and Mair (1991) is useful to assess the constraints and opportunities for basin water planning in India. The river basin is a hydrological and topographical reality. It defines water flows and, together with climate, constrains the amount of water that is available without resorting to transfers from other basins, hence connecting users via the drainage network. This casts the river basin as a ‘space of dependence’, regardless of the political structures that manage water resources. But, the river basin is also entangled in a broader polity and constitutes a disputed space through which different actors try to assert their control via water planning. Finally, while it has been proposed as the fundamental unit of water management for over a century, the river basin largely remains a ‘new space’ without social reality. It is not yet a ‘space of engagement’, a necessary condition for basin-centred mechanisms to promote sustainable water management. There are contradictions between the physical organisation of water and the socio-political reality that craft the river basin as a disputed space of dependence, and the policy rhetoric presenting the river basin as the unique space of engagement for water planning and management. Current institutional arrangements such as tribunals fail to address this multiplicity of scales, therefore falling short of shaping the river basin as a true space of engagement. They also frame water disputes in technical terms that science is expected to resolve and remain narrowly focused on sharing surface blue water. They leave out – or assume that federal and state administrative and political authority can solve – the hydrological complexity of the river basin and the multi-level financial, political and developmental relations that underpin water disputes. They show high degrees of inertia and fail to recognise that power and struggle are at the core of water allocation. In the case of the Krishna, this led to water over-commitment, conflicts, and environmental degradation as the KWDT internalised the Indian development project. The failure of interstate allocation and basin planning is not only due to institutional weaknesses and the absence of a firm hand of governance – as is often and rightly pointed out – but also to a lack of space for meaningful engagement of multiple actors. This sets limits to what can be achieved through basin water planning alone. This paper does not call for yet another layer of new institutional arrangements but rather, and within the limits set by the Constitution of India, for tribunals to account for the true spaces of political engagement that characterise The Geographical Journal, 2010 the Indian polity: the states and the panchayats. By recognising the role and the importance of these spaces and by facilitating direct interactions among them rather than focusing on surface basin water, tribunals would shape the river basin as a space that multiple actors de-facto engage with. The need for meaningful engagement and interactions among actors and the problems of scale are recognised in the IWRM literature (Watson et al. 2007). We argue that a reorganisation of space, that is, broadening the scope of the negotiations to go beyond water and incorporate agriculture, energy, trade and environmental elements, and beyond boundaries by considering the multiple scales of decisionmaking is a promising alternative to induce the necessary engagement for IWRM. This also means that natural resources management should not only aim at managing the resource base but also at promoting alternative sustainable livelihoods, less vulnerable to water stress. Rather than a silver bullet and an apolitical chimera, river basin management should be seen as a practical and political activity that is, like any others, subsidiary to national, regional and local concerns. Institutional arrangements for basin management should provide an umbrella under which actors engage politically at other scales. Politics are at play, yes, but this illustrates the working of the Indian democracy. Acknowledgements This research was supported by a grant from the Australian Centre for International Agricultural Research (ACIAR). The authors thank Trent Biggs, Frederic Landy, Sylvain Massuel, Hugh Turral and two anonymous reviewers for their helpful comments and suggestions on earlier versions of this paper. References Allan J A 2006 IWRM: the new sanctioned discourse in Mollinga P P, Dixit A and Athukorala K eds Integrated water resources management: global theory, emerging practice and local needs Sage, New Delhi 38–63 Allan J A 2007 Beyond the watershed: avoiding the dangers of hydro-centricity and informing water policy in Shuval H and Dweik H eds Water resources in the Middle East Springer, Berlin 33–9 Anand P B 2004 Water and identity: an analysis of the Cauvery River water dispute BCID research paper 3 University of Bradford (www.brad.ac.uk/acad/bcid/research/papers/Paper3.pdf) Accessed 15 October 2008 Brenner N 2001 The limits to scale? Methodological reflections on scalar structuration Progress in Human Geography 25 591– 614 Celio M, Scott C and Giordano M 2009 Urban-agriculture water appropriation: the Hyderabad, India case The Geographical Journal 176 39–57 © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) Beyond water, beyond boundaries: spaces of water management in the Krishna river basin, South India Cox K R and Mair A 1991 From localised social structures to localities as agents Environment and Planning A23 197–213 Delaney D and Leitner H 1997 The political construction of scale Political Geography 16 93–7 D’Souza R 2002 At the confluence of law and geography: contextualising inter-state water disputes in India Geoforum 33 255–69 D’Souza R 2006 Interstate disputes over Krishna waters – law, science and imperialism Orient Longman, Hyderabad Ghiotti S 2006 Les territoires de l’eau et la décentralisation – La gouvernance de bassin versant ou les limites d’une évidence Développement durable et territoires Dossier 6 GoI (Government of India) 2002 National water policy GoI, New Delhi GoI-KWDT (Government of India–Krishna Water Disputes Tribunal) 1976 The further report of the Krishna Water Disputes Tribunal with the decision GoI, New Delhi Gulati A, Meinzen-Dick R and Raju K V 2005 Institutional reforms in Indian irrigation Sage, New Delhi GWP (Global Water Partnership) 2000 Integrated water resources management TAC Background Papers No. 4 Water Partnership, Stockholm Iyer R R 1994 Indian federalism and water resources Water Resources Development 10 191–202 Iyer R R 2003 Water: perspectives, issues, concerns Sage, New Delhi Kohli A 2001 The success of India’s democracy Cambridge University Press, Cambridge Landy F 2009 Feeding India – the spatial parameters of food grain policies Manohar-CSH, New Delhi Lebel L, Garden P and Imamura M 2005 The politics of scale, position, and place in the governance of water resources in the Mekong region Ecology and Society 10 18 Lefebvre H 1974 La production de l’espace [The production of space] Anthropos, Paris Maitra S 2007 Inter-state river water disputes in India: institutions and mechanisms Japanese Journal of Political Science 8 209–31 Marsden S 2000 The social construction of scale Progress in Human Geography 24 219–42 Massey D 1993 Politics and space/time in Keith M and Pile S eds Place and the politics of identity Routledge, London and New York Molden D 1997 Accounting for water use and productivity SWIM Paper No. 1 International Water Management Institute, Colombo, Sri Lanka Molle F 2007 Scales and power in river basin management: the Chao Phraya river in Thailand The Geographical Journal 173 358–73 Molle F 2009 River-basin planning and management: the social life of a concept Geoforum 40 484–94 Molle F and Berkoff J 2006 Cities versus agriculture: revisiting intersectoral water transfers potential gains and conflicts Research Report No. 10 Comprehensive Assessment Secretariat, Colombo, Sri Lanka © 2010 The Authors. The Geographical Journal © 2010 Royal Geographical Society (with The Institute of British Geographers) 11 Mollinga P P 2003 On the waterfront: water distribution, technology and agrarian change in a South Indian canal irrigation system Orient Longman, New Delhi Mollinga P P 2005 The water resources policy process in India: centralization, polarisation and new demands on governance ZEF Research Report 7 Center for Development Research, Bonn Neir A M and Campana M E 2007 The peaceful resolution of U.S.–Mexican transboundary water disputes The Economics of Peace and Security Journal 2 41–8 Pani N 2009 Institutions that cannot manage change: a Gandhian perspective on the Cauvery dispute in South India Water Alternatives 2 315–27 Perkman M 2005 The construction of new scales: a framework and case study of the EUREGIO Cross-Border region CSGR Working Paper No. 165/05 May 2005 (www2.warwick.ac.uk/ fac/soc/csgr/research/workingpapers/2005/wp16505.pdf) Accessed 15 November 2008 Richards A and Singh N 2002 Interstate water disputes in India: institutions and policies International of Water Resources Development 18 611–25 Sajjan S 2005 Availability of water in Krishna basin and scope for scheme-B irrigation projects. Fifth International R&D conference 15–18 February, Bangalore Shah T 2009 Taming the anarchy: groundwater governance in South Asia Resources for the future, Washington DC and International Water management Institute, Colombo, Sri Lanka Shiva V 1991 Ecology and the politics of survival: conflicts over natural resources in India United Nations University Press, Tokyo Smakhtin V and Anputhas M 2006 An assessment of environmental flow requirements of Indian River basins Research Report No. 107 IWMI, Colombo, Sri Lanka Smith N 1992 Geography difference and the politics of scale in Doherty J, Graham E and Malek M eds Postmodernism and the social sciences St Martin’s Press, New York 57–79 Swingedouw E 1997 Neither global nor local: ‘glocalization’ and the politics of scale in Cox K ed Spaces of globalization Guilford Press, New York 137–66 Venot J P 2009 Dynamics and new challenges in the Indian water sector: trajectory of the Krishna Basin South India in Molle F and Wester P eds River basins trajectories: societies, environments and development IWMI, Colombo, Sri Lanka and CABI Publishing, Wallingford, UK Venot J P, Ratna Reddy V and Umapathy D 2010 Coping with drought in irrigated South India: farmers’ adjustments in Nagarjuna Sagar Agricultural Water Management 97 1434– 42 Warner J, Wester P and Bolding A A 2008 Going with the flow: river basins as the natural units for water management? Water Policy 10 (Suppl. 2) 121–38 Watson N, Walker G and Medd W 2007 Critical perspectives on integrated water management – editorial The Geographical Journal 173 297–9 The Geographical Journal, 2010