The 140-year-old Indian Meteorological Department (IMD) forecast a deficient monsoon this year, only to be proved wrong by the 16 per cent surplus rains in June and the above-normal reservoir levels countrywide as of July 23. Not many were surprised however, as the IMD’s success rate for monsoon forecast is barely 50 per cent — in recent times, it failed to predict the droughts of 2002, 2004 and 2009. Interestingly, directly contradicting the official weathermen, a barely 12-year-old private forecaster, Skymet Weather Services, had predicted a normal monsoon this year.
Similarly, thanks to an app from Kolkata-based Express Weather, potato growers in West Bengal not only know when it will rain next but also the humidity levels, temperature and other climate parameters for the week ahead. This helps them plan their schedule for spraying pesticides, irrigation, sowing or harvesting. With its weather updates, the app makes agriculture preventive, rather than corrective, and non-dependent on IMD’s forecast.
In Chennai, transporter PR Venkatesan is busy instructing his truck drivers to cover their vehicles with waterproof sheets as Weather Risk has alerted him about the likelihood of heavy rainfall en route to Delhi. “With the detailed information, I can equip my drivers suitably and avoid wastage of goods as well as delay in transportation,” he says.
In a country with vastly differing climatic regions — from tropical in the south to temperate in the north and heavy snowfall in elevated areas — accurate weather forecasting is emerging as a key requirement for businesses and agriculture alike. Several enterprises have emerged in recent times to fill this need.
“The demand for forecasting services is increasing. From aviation and shipping to farmers and governments, everybody needs accurate weather predictions,” says Skymet founder and CEO Jatin Singh.
He set up the Noida-based company after his former boss at a Hindi news channel complained about the lack of reliable weather data in India. As Singh’s father happened to be a supplier of computers and other equipment to IMD, he was well aware of the dynamics of weather prediction. In 2003, when Singh launched his weather-forecasting service he found ready subscribers among the rapidly mushrooming news channels back then.
For Express Weather founder and CEO Angshujyoti Das, it was the complete lack of weather data on the Sunderbans that led him to the business of forecasting. “We were implementing a sustainable ecotourism project in Sunderbans and Unesco asked for some wind data. On approaching the IMD, we were told they didn’t have the data,” he says.
Though the Sunderbans project was eventually shelved, Das ventured deeper into weather data sourcing and research, and combined it with crop science data to support decision-making by farmers.
Formula for the day
Express Weather boasts three areas of expertise — it generates seven-day weather forecasts using open-source numerical models developed in the US and modified for Indian conditions; it has developed an app called Farmneed to combine crop science and micro weather data, and this is being used by over 25,000 farmers; and it installs and manages automatic weather stations, which collect data related to temperature, humidity, aridity and other environmental factors. The company has 300 such stations today.
Weather Risk, founded by Sonu Agrawal, has a network of nearly 1,000 weather stations across nine states including Rajasthan, Uttar Pradesh, Bengal, Bihar, Madhya Pradesh and Maharashtra. His clientele includes farmers, transporters, banks and other financial institutions, insurance companies as well as FMCG players. The company’s primary focus, however, is crop insurance. Based on an agro-meteorological model, the company captures drought risks in the form of a formula. For instance, if the rainfall in a village in July is less than, say, 100mm then the crop could fail and result in huge losses. “We would need to put up a weather station in that village; and to price the insurance product, we would need forecast based on historical weather data as well as an understanding of what the situation could be in July,” says Agrawal. Besides farmers in 15 states, power companies utilise his services in a big way.
Powered by data
Power distributors use weather forecasts to estimate electricity demand over the next ten days. “With our research we know clearly when it is going to be hot and humid, leading to higher power consumption for air-conditioners; when the temperature falls below 28 degrees Celsius, the demand goes down. Power companies plan their supply accordingly,” says Agrawal.
Additionally, the power companies can manage costs better. They usually procure power from the grid a day ahead, as spot buying is costlier. However, purchase of extra power leads to penalties, which the companies can avoid with the help of demand forecasts from Weather Risk.
The key to accurate forecast lies not only in sourcing data accurately but also processing it well. Dr Kanti Prasad, weather scientist and former director-general of IMD, explains that modern-day weather forecast is based on mathematical models, which are run on supercomputers. “The data for running these models is generated by observatories across the globe. Additionally there are conventional weather stations, while remote-sensing data is available from satellites, aircraft, radar, land and ocean-based stations.”
This data is free of cost and flows into the global telecommunications system maintained by the World Meteorological Organisation and accessed by forecasting centres all over the world. The number churning then takes place and the computers generate forecasts for the short range, medium range and long range. Weather forecasting companies customise the results for users and price them accordingly.
It is here that private weather forecasting services have an edge over IMD. They not only have meteorologists, scientists, programmers and developers on board, they are also focused on reaching out to user groups. Not surprisingly, all these companies are profitable.
It’s raining money
Skymet’s Singh says his company was profitable from day one. Ditto for Express Weather and Weather Risk. That also explains the interest shown by private equity firms. Skymet raised ₹4.5 crore in its Series A round of funding from Omnivore in August 2011. Last year, it raised ₹27.6 crore in its Series B round led by Asia Pacific Pte Ltd. Das, too, is “in discussions with four to five large players — both strategic and financial investors”.
The private players also boast higher success rates for their prediction. “In the long range, we have not got a single monsoon wrong yet,” says Singh, claiming a 75 per cent success rate for his company’s monsoon prediction. Das says at Express Weather the accuracy is plus-minus 2 degrees Celsius for temperature, 5-7 per cent for relative humidity, and 80-83 per cent for rainfall.
But unlike Singh, he is unwilling to dwell on IMD’s inaccuracy. “I don’t want to see the IMD as a competitor as it is an old organisation with huge infrastructure and financial muscle. Any private player cannot think of creating such infrastructure,” Das says.
Also, unlike IMD, the private forecasters do not wish to limit themselves to India. Express Weather is already operating in foreign countries, including Zambia. “We are creating infrastructure to bring in data services and integrate farms to markets,” Das says.
Agrawal’s Weather Risk plans to double the number of its stations to 2,000 within a year. Outside India, it is working in Africa, Bangladesh and Cambodia. Skymet’s Singh, on the other hand, prefers to stay focused on India for the moment. “I very well understand this subcontinent, the monsoon and winter pattern, and so on. Weather is highly geography-specific and it takes at least two years of R&D to enter a new country,” he says.
But the forecaster that he is, his reading is that in the long term Skymet will be a global weather brand capable of long-range forecasting for the rest of the world.