The Hindu : Opinion / Editorials : STILL A LONG SHOT

Online edition of India's National Newspaper
Wednesday, Oct 06, 2004

Opinion

Opinion - Editorials

STILL A LONG SHOT

HAD THE INDIA Meteorological Department been a bookmaker, it would have lost a packet. In April it predicted that rainfall in this year's southwest monsoon from June 1 to September 30 would be normal, in fact, 100 per cent of the long-term average. The chances that the rainfall would be less than 90 per cent of the average were set at just 4 per cent. In June, it raised the stakes further, postulating the odds of such a reduction in rainfall at a mere 3 per cent. As it turns out, the monsoon has performed 13 per cent below normal. For the second time in three years the Department has got it strikingly wrong, not having foreseen the drought of 2002 when the deficiency was 19 per cent. This time the Department can cite some extenuating aberrations in the monsoon schedule: the monsoon actually broke on May 18 over Kerala, 12 days ahead of schedule, and whatever rain came in May has not been taken into the southwest monsoon folder. Further, the monsoon has not yet closed its books, with rain falling fairly heavily in many parts of the country in the first week of October. Some dribbles of comfort, maybe, for farmers in some States but none for the Department that will need to get back to questioning the validity of the 10 Parameter Regression Model it uses to predict monsoon performance. Or was it that its interpretation of the data went awry? Data on the sea surface temperatures in the Tropical Pacific Ocean known as the Nino 3.4 region from January to June (one of the 10 parameters on which the forecast is built) had turned out to be "unfavourable" but the Department chose to belittle their effect.

It is not just in the overall number that the forecast was way off target; it went seriously wrong in projecting the distribution of rainfall over the country. Northwest India was projected to get 103 per cent of its long-term average rainfall; it got barely 80 per cent. Northeast India was the only region with the right number. When forecasts go so wayward, it raises the question whether they provide any value to India's agriculture that depends so vitally on the performance of the monsoon. It was the Great Indian Drought of 1877 that prompted the Department to issue its first seasonal monsoon forecast in 1884. Yet 120 years later and despite extensive multinational research, monsoon predictive capabilities are still too green, uncertain, and bereft of much practical value at least to individual farmers.

Long term predictions for the country as a whole might offer some inputs for policy making at Krishi Bhavan in New Delhi and for punters on Dalal Street. But they have never meant anything to farmers who have relied on gut instinct, tradition, and soothsayers to tell them how the seasonal rains would turn out. It must be conceded that there has been more reliability and practicality about the Department's short-term forecasts in recent times. The National Centre for Medium Range Weather Forecasting, which provides countrywide regional forecasts three to five days ahead, has claimed better results. "Farmers who follow our advisories earn about twice as much profit as those who do not," is one such claim. Reliable wind and cloud forecasts three days ahead can no doubt help farmers reduce costs of irrigation, fertilizer, and pesticide application. To be fully effective for precision farming, these forecasts must be valid at the local village level. Given the complexity of the monsoon and our limited understanding of it, none will bet on such a service being available in the near future.

Printer friendly page
Send this article to Friends by E-Mail

Opinion

News Update

Stories in this Section

STILL A LONG SHOT

MR. INDONESIA

Sorry, you're not part of the plan

E.U.'s real problem with Turkey