Unpredictable forecasts

The proverbial unpredictability of the weather no longer holds true when it comes to many global meteorological systems for which reliable forecast models are now in place. But it remains relevant for the Indian summer monsoon, which remains as mercurial as ever.

With long-range monsoon predictions going awry year after year, the credibility of these forecasts has been severely eroded. What is worse, different agencies create confusion by coming out with vastly divergent, even contradictory, forecasts""as has happened this year.

The south-west monsoon has already betrayed the predictions of its belated onset by hitting the Kerala coast on June 5, just four days behind schedule. The forecasts about the total quantum of rainfall during the four-month monsoon season (June to September) and its distribution over this period are, of course, yet to become known and the confusion over these is even greater.

While the India Meteorological Department (IMD) reckons the season's total rainfall to be 98 per cent of the normal, another government agency, the Bangalore-based Centre for Mathematical Modelling and Computer Simulation, has been wavering on this count.

So much so that last week it dramatically reversed its earlier forecast for June rainfall from "22 per cent surplus" to "34 per cent deficit"! With this kind of revision of the forecast, how can an agency expect its credibility to survive?

The record of the IMD is not much better, for it too has been erring more often than not. The recent instances have been in the years of 2002 and 2004, both of which experienced drought (19 per cent and 13 per cent rainfall deficiency, respectively) despite the IMD's projection of normal rainfall.

It is, therefore, no wonder that the corporate sector, agriculturists and even the stock markets have stopped taking these contradictory monsoon predictions too seriously. Of course, they do take the actual performance of the monsoon seriously enough as it unfolds over the four-month period. The Sensex itself now discounts the forecasts but takes note of real events.

What seems quite clear is that Indian meteorologists have been unable to develop a reliable monsoon prediction model though they have been trying to do so ever since the first monsoon forecast was made over 70 years ago, in 1932. This is despite the fact that after every major fiasco, the IMD invariably claims it has evolved a better prediction model.

The government, too, has been liberally investing in equipping it with better computers, satellites, radars, communication and other back-up facilities. It is of course true that the Indian monsoon is a complex global phenomenon involving the interplay of clouds, radiation, surface fluxes, air-sea interaction and other factors, but it should be possible to identify the deficiencies of our forecasting system and take corrective action.