Weathermen, compete

The Indian Meteorological Department8217;s preliminary monsoon forecast 8212; rains may be a shade below last year8217;s 8212; is based on a recently adopted model. The earlier one couldn8217;t, infamously, forecast the 2002 drought. Monsoon forecasting was started during the Raj, in 1886. One would have thought we would be pretty good at this by now. But we aren8217;t. IMD switched to a new eight-parameter model. This did fairly okay in 2003. But IMD8217;s 2004 forecast wasn8217;t awe-inspiring. It almost seems we need a model to predict whether IMD8217;s predictions will hold true. So why not privatise monsoon forecasting? The official view is that this will create confusion. That8217;s an extraordinary response. What do IMD8217;s bad predictions cause if not horrendous confusion? We need the best forecasters and if sarkari meteorologists can8217;t do the job, we should be looking at alternatives. To suggest one option, the US-based Scripps Institute of Oceanography has got a solid reputation. Surely, MPs won8217;t argue that an American institute will deliberately give wrong predictions. Its credibility depends on clients8217; satisfaction. IMD, being a government institution, has no concept of a client. At the very least, let others also predict rains. Surely, farmers8217; interests are primary here, not in some silly notion of self-sufficiency in weather science?

Indeed, if junking IMD8217;s monopoly is a radical change, it will be in keeping with another radical change 8212; India8217;s economic dependence on monsoons has sharply declined. Less than 19 per cent probably closer to 16 now of GDP comes from agriculture and less than 60 per cent probably closer to 55 now of India8217;s population is employed in agriculture. Unlike the fifties and sixties, economic growth is less closely associated with agricultural performance, robust growth in 2004-05 despite drought conditions being a case in point. However, there is still a strong link between agriculture and industry, on both the input and demand sides, with the latter possibly more important. But sensitivity of agriculture to monsoons also depends on commercialisation and diversification, since agriculture is not just foodgrain production. The problem is that only 40 per cent of India8217;s crop area is irrigated and reservoir water can8217;t compensate.

Also remember that in rain prediction spatial across 36 meteorological divisions and temporal July is crucial for kharif precipitation is more significant than aggregate levels. And if IMD goes wrong with the big picture, it8217;s hard to accept that it will be brilliant in disaggregated predictions.