In any riparian dispute, the contending parties each try to 'win', through a decision that is favourable to them. This is understandable - we all want victory on our part of the river (upstream or downstream), and politicians want to show that their side is winning. In Maharashtra, we see this happening within the state itself. But this kind of stance distracts the attention of both sides from the real long-term solutions that they can pursue TOGETHER, in ways that BOTH WILL BENEFIT. That is the key to the solution.

In my view there are four things that can be done to address the problem.

(a) Irrigation takes up 65% of the Cauvery water, whereas globally the share of agriculture in water use in a river basin is in the 20-30% range. This is a huge difference (for comparison, Bengaluru uses 4% of the Cauvery water, and all the other towns put together get another 3-4%). If we make agriculture even slightly more efficient in water use, we will never need to worry about drinking water shortages for a long time.

Karnataka should seek, from the Supreme Court and from the Cauvery River Authority, an independently monitored process by which agriculture in the basin of the Cauvery - including in our state - is made less dependent on water. The choice of crops, the techniques of cropping, etc. should get more and more water-smart over time.

Both sugarcane and rice can be grown with a lot less water than Indian farmers use, but the adoption of techniques like SRI (System of Rice Intensification) has been extremely patchy; there has been some adoption of SRI, but not nearly enough. The extension services that the government operates are most defunct, and may as well be closed down. That money can instead be given to organisations that are committed to soil health, improved livelihoods for farming, innovations, and such things.

(b) The Centre should fund a massive overhaul of water infrastructure in the metropolitan Bengaluru region, which currently gets about 4 per cent of the river's water. This scheme should have three aims - one, to plug all leaks in BWSSB's creaking infrastructure, which today results in the loss of nearly 20-30% of the piped water; second, putting at least 40% of all Kaveri-sourced water to multiple use through dual piping in homes and offices; and third, the establishment of more nimble infrastructure to source water for the city from the 300-odd lakes in and around it.

A scheme like this will double the piped water supply in Bengaluru. It will additionally have the effect of infusing some financial and technological life to the moribund BWSSB, which today operates like an unscientific infrastructure company rather than an intelligent manager of water resources.

(c) We have to rethink the role and location of reservoirs, and of hydro-electric power generation in both states from such storage. The long-delayed proposal to establish more reservoirs along the river - two in Karnataka and two in Tamilnadu - should be taken up in earnest. Additional storage points on the river will allow a more geographically concentrated distribution of water for sub-basin needs, and also provide greater storage to meet water needs during lean periods.

In parallel, there should also be a sincere effort to restore the flow in the many tributaries of the river. The long-overdue de-silting of all reservoirs should also be taken up - THIS CAN QUICKLY RAISE THE STORAGE in the existing reservoirs.

Power sources can be substituted, but water is needed as it is. It only makes sense, therefore, to use water in the reservoirs to generate electricity if there is no supply weakness for the water. When the water itself is scarce, there is not much point in trying to balance the two competing uses. This problem exists in the management of nearly all the rivers of India, not just Cauvery.

With Koodankulam now operational, some of this should reduce. It would be wiser to raise Tamilnadu's allocation from the nuclear plant by this amount, and have better control of the reservoir waters for irrigation purposes only.

Karnataka has a much higher dependence on hydel, but here too a policy of prioritising water supply over power supply can be established for such facilities. Our state too could get a higher allocation from Koodankulam, under such an approach.

(d) The current approach to weather forecasting is inadequate. In India, we have got used to season-long forecasts, and some ambiguous effort to guide farmers' planting decisions using these. That program needs to be consigned to the dustbin, and replaced with something more dynamic and more accountable for results. Week-by-week, sub-regional assessments have to be developed. The consequences of neglecting this are evident, but as with so many other things in the country, the diligent and much-needed alternative has not been considered seriously enough.

It is time we stopped trying to 'decide' the answers to large questions, assuming that good outcomes will flow from that. In many complex problems, in fact, the opposite is true - we must enable the emergence of the right macro design by pursuing smaller, specific outcomes that add up. That will not help answer the immediate problem facing the two states, but it will surely help ensure that the problem doesn't visit us again and again.