Don't bet on Met for accurate monsoon forecasts

July 1, 2010, governancenow.com
Deficient rainfall delays sowing in north India


The Indian Meteorological Department may let the farmers down once again. As against its prediction of a normal rainfall this year (102 percent to be precise), June has witnessed scanty rainfall in the grain bowl of India—Punjab, Haryana, Uttar Pradesh and parts of Rajasthan and Madhya Pradesh.

According to the met department’s own record, these states have not only received scanty rainfall, the rain bearing clouds have disappeared altogether. Unless these clouds reappear in the next few days it would jeopardize sowing of paddy and sugar crop. What happens next is anybody’s guess.

Particularly disturbing is the low rainfall in the week between June 17 and 24--21 percent. This takes the average shortfall for June to 13 percent, which is not much in absolute terms but is crucial because it adversely affects sowing of crops and there are tell-tale signs that the monsoon is slowing down. The met department hopes that the clouds would reappear from July 5 onwards. But anyone who has been following the met office knows that its prediction can never be relied on.

Take the case of last year. The met department predicted a “near normal” rainfall at 94 percent. That was revised to 91 percent when the rain clouds played hide and seek. The downward revision continued but couldn’t match the actual deficiency and the year ended as one of the worst drought-hit in recent years.

Given the fact that our agriculture is largely dependent on monsoon, it is crucial that our met department provides accurate and timely information—a task it has singularly failed to perform for decades. Worse, the predictions are never specific in terms of the days or the weeks of a rainy month and a block or taluk of a district where farmers wait for the signal from the sky to begin their farming activities.

As a minister in charge of the met department during the previous UPA regime, Kapil Sibal had tried to completely revamp the weather prediction system. More than five years down the line, little progress has been made.

All that the met department has got so far is a “super computer” that can deal with additional data required to predict rainfall. This was installed only this January, but the Doppler Weather Radars, which are to provide a crucial chunk of weather-related data, are yet to be in place. Less than half-a-dozen radars have been installed so far--of 55 that are required. According to the officials, additional six radars would be installed by end of this year. The plan is to complete the task by the end of 2012, which seems a bit far-fetched now.

Sibal had also tried to appoint scientists from abroad to give a fillip to the way the weather-related data are collated and analyzed to predict rains but this was not allowed because of protests from the bureaucratic set up. Assuming that all the technologies are to be in place in the next years, it is difficult to imagine our homegrown scientists used to antiquated methods—largely dependent on weather balloons that are released twice a day at 36 places in the country to collect weather updates and some information that are collected through weather satellites--would actually be able to put the superior technology to good use.

There are other issues too, like the mathematical model used for the purpose of predicting rain. This model keeps changing from year to year and not a single one has come close to making accurate predictions.

So, the farmers are condemned to pray the weather gods. And so is the government, which is heavily banking on a good monsoon, and consequently a good harvest, to bail it out of the crisis caused by food inflation.