Well begun

Well begun, it is often said, is half-done. While that may be true in many circumstances, it need not be the case with the Indian monsoon. The uncertainties of this great weather system, influenced by global factors, do not encourage a direct correlation between the timely onset of the monsoon and the total rainfall that occurs in the four-month rainy season (June to September). For that matter, there is little correlation between the timely onset of the rains and their distribution over time and space. As such, not too much needs to be read into the timely arrival of the monsoon in the Andaman-Nicobar region and the adjoining south Bay of Bengal Sea. It is mostly a matter of notional comfort that the kind of anxiety over the arrival of the monsoon that prevailed at this time last year is not there today. What is more genuinely reassuring this time is that the weather system is strong enough to last for four to five days and move forward. Normally, the first surge of the Bay of Bengal wing of the monsoon does not hold for long and is usually followed by a break in the rains. Equally comforting is the weatherman's projection that the cloudiness and rainfall over the south-central Arabian Sea and adjoining areas may gradually strengthen from May 23, paving the way for the monsoon to set in over the Kerala coast on May 30.

Though the India Meteorological Department (IMD) has forecast that total rainfall this year would be about 93 per cent of the long-period average of 89 cm for the entire country, that guarantees nothing. The forecast is based on a set of new statistical models which are yet to be validated. Indeed, the old 16-parameter statistical regression model, which had been giving fairly accurate forecasts ever since 1988, had to be given up as it began faltering in the 2000s. In fact, the IMD failed to foresee the drought of 2002. The new models, on the other hand, gave a correct prediction in 2003, the first year of their operation, but erred in 2004 by indicating 100 per cent normal rainfall while the actual turned out to be only 87 per cent. In 2005, though the projection was almost on the dot (prediction 98 per cent, actual 99 per cent), the season was marked by long dry spells and unusual cloud bursts, like the one that crippled Mumbai for several days. Indeed, the time seems to have come for the IMD to acquire the capability to warn about such catastrophic events well in advance. Most countries, especially those prone to frequent typhoons and cyclones, have the systems in place to caution the administration and public in advance about the impending weather onslaught. There seems to be little reason why the IMD cannot do so by either developing the needed technology or sourcing it from abroad. Besides, considering the inherent disabilities of empirical models, the IMD needs to switch over to the relatively reliable dynamic models that are in vogue the world over. Till that happens, monsoon forecasts will continue to be imperfect guides to reality.