ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Why and Where Are the Krishna Waters Disappearing?

The depletion of the waters of the Krishna river is changing the ecosystem of its basin. How can equity, sustainability and efficient use of the scarce water resources be ensured?


dry deserts in its central part to wet forest

Why and Where Are the

in the western Ghats but the climate is predominantly semi-arid to arid with

Krishna Waters Disappearing?

highly erratic rainfall concentrated during the monsoon between May and October. Potential evapotranspiration exceeds Jean-Philippe Venot precipitation (815 mm/year in average) in

The depletion of the waters of the Krishna river is changing the ecosystem of its basin. How can equity, sustainability and efficient use of the scarce water resources be ensured?

The author is grateful to the team of the International Water Management Institute, Hyderabad that made this research possible. Warm thanks are also due to Trent Biggs, Sylvain Massuel, Francois Molle, Madar Samad and Hugh Turral for thoughtful discussions in the earlier stages of this work.

Jean-Philippe Venot ( is a PhD scholar at the GECKO Laboratory, University of Paris X-Nanterre, France, and is presently based at the International Water Management Institute, Hyderabad.

Economic & Political Weekly

february 9, 2008

he Krishna river basin (2,58,948 sq km) is the fourth largest Indian river basin in annual discharge and ranks fifth in terms of basin area. The Krishna basin covers parts of three Indian states – Andhra Pradesh, Karnataka and Maharashtra. The Krishna river originates in the western Ghats of Maharashtra and Karnataka, drains the dry areas of the Deccan plateau, and forms a delta before discharging into the Bay of Bengal. The Krishna has three major tributaries, the Bhima river from the north, the upper Krishna from the centre-west, and the Tungabhadra river from the south (the figure, p 16). In 2007, the basin population has been evaluated at 79 million inhabitants [GoI 2001, assuming a 3 per cent annual growth], with 51 million in rural areas. The Krishna basin includes a full range of climates and ecosystems, from all but three months of the year making irrigation necessary for agricultural development [Biggs et al 2007].

Decreasing Discharge

The Krishna basin has seen an increasing mobilisation of its water resources and a dramatic development of irrigation since the 1850s [Wallach 1985; Venot et al, in review]. This phenomenon hastened after India gained independence and the total storage capacity (in large reservoirs) has multiplied eightfold since the 1950s to reach about 51.5 billion cubic meter (bcm) in 2002. The main irrigation and hydro-power projects include: Nagarjuna Sagar (built in 1967) and Sri-Sailam (1983) in Andhra Pradesh; Bhadra (1953), Malaprabha (1973) and Ghataprabha (1977) in Karnataka; Koyna (1964), Ujjani (1981) and Alamatti (1990) in Maharashtra. Though their extent is not well quantified,


small-scale irrigation projects commit significant volumes of water too. As a result, existing infrastructures regulate more than the 75 per cent-dependable annual

Figure: Krishna Basin in South India

Maharashtra Upper Bhima (Ujjani) Nira

Koyna Alamatti

Nagarjuna Sagar

Tungabhadra Vani Vilasa Sagar

Bhadra Karnataka

200 0 200 Kilometres

flow. This may not generate significant cuts in water supply in most years but it may lead to significant shortage downstream during dry years as upstream water supplies are not adjusted to reflect deficits in water availability. Along with this infrastructural development, irrigated areas have increased by about 50 per cent since the 1970s: they covered 3.4 million of hectares (mha) in 1996-2002. This increase is mainly due to the boom of groundwater abstraction thanks to private pumps, shallow tube wells and subsidised energy. Groundwater irrigated areas have more than doubled during the last 30 years: they supplied about 1.5 mha in 1996-2002.

As a result, the Krishna river discharge to the ocean gradually decreased, providing the first indication of river basin closure. A generally accepted definition of a closed river basin is one where all available water is depleted [Molden 1997], resulting in little or no discharge to the ocean during years with average precipitation. Consider the river discharge


measured at the head of the delta, after di-the Krishna basin. As basin closure intenversions to the Krishna delta project. Be-sifies interconnectedness between water fore 1960, river discharge into the ocean users [Molle et al 2007], localised interaveraged 57 bcm per year. Since 1965, it ventions tend to have unexpected consequences elsewhere in the basin. Further water resources development will be tantamount to sectoral and/or regional reappro priation of water and basin-wide strategies for water management and development are needed to mitigate so-

Andhra Pradesh

cial and environmental impacts of water


The table quantifies the changes of the

Krishna basin water balance since the

Prakassam Barrage

mid-1950s. It evaluates water depletion, defined as the use or removal of water from a water basin that renders it unavail-Main rivers able for further use [Molden 1997; refer

Major reservoirs N

to Venot et al (in review) for a detailed

Andhra Pradesh Sri Sailam Karnataka

presentation of the methodology used).


Krishna basin boundaries Table 1 confirms that the main trend is a strong increase of irrigation depletion that, overall, has more than doubled from

17.2 bcm a year in 1955-65 to 40.7 bcm a year in 1990-2000. Almost half this depletion (18.7 bcm/year) is sustained through groundwater exploitation that is taking place at an unsustainable rate: aquifers are over-exploited by 33 bcm/year. Though the nature and extent of ground

steadily decreased at an average of 0.8 water-surface water interactions are not bcm per year to reach 10.8 bcm in 2000 or well known, there are signs of declining less than 15 per cent of its historical baseflows due to groundwater over- value, and falling further, close to nil in abstraction, leading to water shortage 2004 (0.4 bcm). Only monsoon flows downstream [Hanumantha Rao 2006]. (July-October) reached the ocean and that Over the period 1955-2000, the total the peak outflow has been delayed by depletion increased by 23 per cent: it

about two months due to increasing upstream regulation.

In 2005-07, the Krishna basin witnessed high rainfall and the discharge to the ocean averaged 29 bcm/year: the Krishna river basin is under transition and defining management intervention for sustainable water use at the basin level is increasingly needed. This requires identifying and understanding the spatial and historical dynamics of water use in

Table: Water Accounting for Krishna Basin between 1955 and 2000

1955-1965 1970-1980 1980-1990 1990-2000

1 Mean annual rainfall 2,38,281 2,20,218 2,14,766 2,23,020

2 Surface water imports 49 0 0 0

3 Groundwater depletion 0 0 -1,356 -3,334

4 Net inflow [1+2-3] 2,38,330 2,20,218 2,16,122 2,26,354

5 Irrigation depletion 17,244 31,722 35,770 40,734

6 Rainfed agriculture depletion 62,630 62,208 68,331 67,781

7 Industry, domestic and livestock depletion 1,029 1,284 1,597 2,110

8 Natural vegetation depletion 62,525 63,125 61,739 65,046

9 Bare lands and reservoirs depletion 21,345 26,090 23,008 27,115

10 Total depleted [5 to 9] 1,64,774 1,84,429 1,90,445 2,02,785

11 Discharge to the ocean 73,539 32,393 22,005 19,625

12 Export to other basins 17 3,396 3,673 3,944

13 Total outflow [10 to 12] 2,38,330 2,20,218 2,16,122 2,26,354

14 Total depletion (percentage of net inflow) 69 84 88 90

15 Irrigation depletion (percentage of net inflow )

7 14 17 18

The unit of measure is the million cubic metre a year (mcm/yr). Average on 10-year periods has been calculated.

february 9, 2008

Economic & Political Weekly


amounted to 203 bcm during 1990-2000, i e, 90 per cent of the net inflow to the Krishna basin (the total depletion only represented 69 per cent of the net inflow of the Krishna basin in 1955-65). Depletion from rainfed agriculture (beneficial) and bare land (non-beneficial) slightly increased; low-beneficial depletion (natural vegetation) remained constant. In 1990-2000, natural vegetation and rainfed agriculture are the two main users of water due to their large area coverage: they consumed 29 and 30 per cent of the local rainfall. Irrigation depletion accounts for 18 per cent of the net inflow. Municipal and industrial uses remain negligible at thedbasin scale (less than 1 per cent of the net inflow in an average year) but are increasing. This might lead to conflict development during droughts as these needs are likely to be met first by transferring water out of agriculture. Eleven per cent of the net inflow is lost with no benefit whatsoever through depletion from bare soil and reservoir evaporation while exports to the west coast (production of hydropower) represent 2 per cent of the net inflow. Finally, the discharge to the ocean amounts to 8 per cent of the net inflow: the Krishna basin is a largely modified ecosystem.

Ensuring Sustainability

Despite early warnings of river basin closure, the three states that share the Krishna waters continue to promote their own agriculture and irrigation development. This development path can no longer be sustained without leading to a de facto redistribution of water among sectors and regions and further damage the environment. With increasing pressure on water resources, water users and managers generally develop informal adjustments, adapting to water scarcity and its socio-ecological ill-effects, in order to overcome institutional constraints and the lack of flexibility [Svendsen 2005]. Adjustments take place on the basis of current political, institutional and economic forces: the extent of stakeholder’s participation to the shaping of a particular waterscape is very variable as people who use or manage water have differing levels of access to natural

Economic & Political Weekly

february 9, 2008

resources, knowledge, political representation or courts.

To overcome the difficulties that uncoordinated adaptive mechanisms may create (rent seeking, competition among users, increasing inequalities) and to avoid conflict, there is a clear need to articulate options that preserve a balance between equity, sustainability and efficient use of scarce water resources for both human benefit and preservation of the environment [Molle et al 2007]. This could be done through the definition and the implementation of formal effective and adaptive water allocation mechanisms, within the framework of the present Krishna Water Disputes Tribunal in charge of apportioning the Krishna waters to the three states of Andhra Pradesh, Karnataka and Maharashtra. Following the example of Australian water allocation procedures, it is crucial that the current Krishna Water Disputes Tribunal ensures a fair degree of stakeholder participation across the board from the local users to the administrations of the different states involved to design allocation mechanisms that would (i) be defined at the basin level; (ii) be based on a comprehensive and transparent understanding of the hydrology; (iii) recognise the interactions between surface and groundwater; (iv) estimate long-term reliable supplies in any part of the basin in light of actual and projected use; and

(v) recognise customary rights, local strategies and local adjustments [Scott et al 2001]. Water development in the Krishna Basin has happened with little regard to the formal allocation mechanisms defined in the mid-1970s [GoI-KWDT 1973, 1976 for the description of these allocations]. This illustrates the need for a clear governance system with an identified river basin organisation responsible for the coordination and efficient implementation of such mechanisms. Moreover, river basin closure means that overall basin efficiency is close to maximum – the scope for effective water savings remains limited and all options of water management must be pursued together.

State governments as well as the World Bank are calling for the implementation of demand management options (participatory irrigation management, modernisation and rehabilitation of existing projects, water pricing strategies, technical on-farm improvements, conservation methods, etc) but supply augmentation projects (such as the Godavari-Krishna link in Polavaram) also enjoy a strong commitment from local governments. These supply augmentation projects will certainly re-open the basin but their social, ecological and economic consequences need to be carefully evaluated: mega-scale projects should not be used as justifications to disregard highly needed watersharing procedures. Finally, sustainable management of water resources cannot be achieved with policies limited to the water sector only: there is a clear need for strategies and policies that would ensure the rural population to make a successful transition beyond agriculture [Moench 2002].


Biggs, T W, A Gaur, C A Scott, P Thenkabail, P G Rao, G M Krishna, S Acharya, H Turral (2007): ‘Closing of the Krishna Basin: Irrigation, Streamflow Depletion and Macroscale Hydrology’, IWMI Research Report 111, International Water Management Institute, Colombo, Sri Lanka.

GoI (2001): ‘All India Census’, New Delhi.

GoI-KWDT (1973, 1976): ‘The Report and the Further Report of the Krishna Water Disputes Tribunal with the Decision’, New Delhi.

Hanumantha Rao, C H (2006): ‘Integrated Water Resources Development in Andhra Pradesh-Problems Faced in “Jalayagnam” and Solutions’, keynote address available online at, cited August 31, 2007.

Moench, M (2002): Groundwater and Food Security in India: Evaluating the Implications of Emerging Overdraft Concerns. Institute of Social and Environmental Transition, Denver, US.

Molden, D (1997): ‘Accounting for Water Use and Productivity’, SWIM Paper 1, International Water Management Institute, Colombo, Sri Lanka.

Molle, F, P Wester, P Hirsh (2007): ‘River Basin Development and Management’ in D Molden (ed), Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture, International Water Management Institute, London, Earthscan and Colombo, Sri Lanka.

Scott, C A, P Silva-Ochoa, V Florencio-Cruz, P Wester (2001): ‘Competition for Water in the Lerma-Champala Basin: Economic and Policy Implications of Water Transfers from Agricultural to Urban Uses’ in Hansen and Van Afferden, The Lerma-Champala Watershed, Kluwer Academic/ Plenum Publisher, New York.

Svendsen, M (2005): Irrigation and River Basin Management: Options for Governance and Institutions, CABI, Wallingford, UK; IWMI, Colombo, Sri Lanka.

Venot, J P, T W Biggs, F Molle, H Turral (in Review): ‘Reconfiguration of River Basins in South India: Trajectory of the Lower Krishna Basin’, Submitted to Water International.

Wallach, B (1985): ‘British Irrigation Works in India’s Krishna Basin’, Journal of Historical Geography, Vol 11, No 2, pp 155-73.

Dear Reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here

Back to Top