RAIN may not be top on our minds when the sun is high in the summer sky. But in a building in Pune, a group of men and women sits obsessing over the monsoons. It’s that time of the year when the National Climate Centre at the India Meteorological Department (IMD) gets working round-the-clock to get the rain forecast right.
With the monsoon rains just a month away, the staff at the 60-year-old IMD building called Shimla Office—the name travelled with the office when it shifted from Shimla to Pune—are busy squeezing in extra hours to issue the monsoon forecast for June as well as to announce the onset date of the monsoons in India.
There is none of the paper clutter you would expect to see at a dingy weather office. There are no maps or strange equipment lying around. The Shimla Office is a swanky place where the weather data is available at the click of the mouse.
The office is abuzz with animated discussions. Some researchers are busy assessing data, while others check on the different indicators or parameters for the forecast as they gear up to ready the summary report. Accuracy being the key factor, they are making sure they get the predictions right.
“Accuracy and preciseness in both space and time” is the target Director General of Meteorology Dr Ajit Tyagi has set for the team.
That’s easier said than done though. It’s difficult to get long-range forecasts right all over the world. This year the IMD is using two different forecast models or parameters to meet the accuracy factor. While they have been using the statistical model for years to forecast the monsoon, this year they are using the dynamical model for better results.
“Our forecast last year for the country was 95 per cent but when the rain came, it recorded 105 per cent. Rainfall all over the country as a whole was in excess in June and September,’’ says Dr Madhavan Rajeevan, Director, National Climate Centre (NCC).
Associated with the IMD for over two decades now, he calls the long-range forecast a tough challenge for meteorologists. “Every minute detail has to be kept in account and every year, the aim is be as accurate as possible.”
So just how accurate have they been?
“Well, we have had our years when we have come really close to the actual figures. The year 2003 and 2005 were good. Our monthly predictions for July has improved over the years and so has our assessment for northwest India,” he says.
Dr Sivananda Pai, the other director of NCC, who too has put in more than 15 years in the department and joined the NCC for the long range forecast, tries to put it in simple words. “It is like you may know that summer is the season for chickenpox but you need to assess and anaylse when its incidence goes up and what are its indicators. Similarly, we have predictors which help us assess the rainfall,” he says.
But the process, he says, requires the staff to be very careful about the assessment of data and when they use the predictors like ocean surface temperatures or mean sea level pressure. “It’s team work and each person’s contribution is very important,” he adds.
The most important source for making any prediction, says Pai, is the data from their inhouse National Data Centre, all available in CDs now. The data at the Centre goes back to a hundred years.
“Relying on historical data we go ahead with our predictions using these models,” he adds. Apart from that the centre accesses maps from all over the world on the Internet. “Work gets divided with some staff members taking up two or three indicators or parameters and accessing the maps followed by discussion and then actual assessment,” adds Pai.
And things are hotting up in the office in the run-up to the monsoon but the workload will increase even further when the monsoon recedes from India in September. It’s then that the Department comes up with its analysis that is circulated to all the centres of the IMD as well as other organisations for comments and suggestions. With agricultural practices in the country strictly tied to the annual cycle of rainfall—the annual average rainfall of the country hovers around at 1,150 mm—the monsoon accounts for almost all the annual rain in 75 per cent of the geographical area and 78 per cent of the gross cropped area. It is this inter-annual variation of the monsoon that has been the first target of these predictions, says Pai.
The rain forecasters know they have to be as close to actual rainfall as possible since long range forecast has many social and economic impacts. “The total monsoon rainfall during the season has a statistically significant relationship with the crop yield, generation of power, irrigation schedule over the country,” says Pai. Statistical forecast models for the monsoon need to be scrutinised constantly, he adds.
‘‘The IMD also takes into account the experimental forecasts prepared by other national institutes like Indian Institute of Tropical Meterology, Indian Institute of Science Bangalore, Centre for Mathematical Modelling and Computer Simulation (CMMACS), Bangalore and the National Centre for Medium Range Weather Forecasting (NCMRWF),” adds Rajeevan.
Working with a staff of 12, he hopes to get more to join his team to firm up research.
So, who are the people who join the department? “Anyone with a physics, meteorology or mathematics background can join us. Then there are the institutes teaching meteorology in various parts of the country, and fresh talent needs to be absorbed in this area. There is so much research happening in the area, and fresh talent is always welcome. Coordinating with different research centres and updating ourselves is very important,” he adds.
The department is aware that people often don’t really believe in their predictions but with the IMD planning to put in place a modernisation programme, Tyagi is hoping their word on the weather will one day be the final word.