As the India Meteorological Department’s (IMD) continues to falter on its long range monsoon forecast over a wide geographical area, an increasing need is being felt for localized weather forecasts customised to the need of the consumers.
The IMD in its annual monsoon forecast this year first said that rains would be normal this year. However, when it didn’t rain adequately till end July it revised it downward to deficient rainfall, but since then the situation overturned completely and rains are now just 5 per cent less than normal. It was not able to correctly forecast the rainfall deficiencies in small pockets like in Saurashtra and Kutch.
Globally, the weather information market is worth around $6 billion, which also includes entities which disseminate forecasts available from various sources through multiple platforms. Industry players said that in US alone there are almost 400 plus private weather forecasting companies, apart from NASA, NOAA and US Navy, which too has a strong meteorological unit.
In India, the market is still small worth around Rs 10-20 crore comprising, but it has been growing at a fast clip, a reason why major global private forecasters like Foreca and Accuweather have started favorably looking at India as a market.
“In the US, government gives funds to private companies to develop their own forecasting models and also inputs to operationalise that model, to ensure that the end-user gets a perfect product,” said Angshujyoti Das, CEO of Expressweather, a domestic private weather forecasting company.
How the model works
As Jatin Singh of Skymet Weather Services, one of India’s oldest private weather companies explains private weather companies have their own weather stations spread across the length and breadth of the country. The weather stations have facilities to monitor temperature, wind speed, remote sensing etc.
The data is also sourced from publically available sources and then feed into a customized model to generate region-specific information be it on wind speed, rain, humidity etc. Skymet has around 300 such stations across the country.
In some cases, forecasters also purchase classified weather information like satellite images and radar information from other global agencies as weather is a trans-geographic phenomenon. “In the case of Express weather, we saw that numerical models presently available were mostly tailor-made for US. Hence, we joined hands with Jadavpur University in West Bengal to develop our own model of forecasting. The data that we get from various sources including our own 185-odd weather stations is fed into these models and results generate location specific-weather information,” Das said.
He said the more the data the better would be the end information. “We have developed separate algorithms (forecasting models) for Mumbai weather, fog in Delhi, seasonal rainfall over Madhya Pradesh and so on,” Jatin Singh said.