The weather mart

The weather mart

Archita Bhatta and Rohini Rangarajan

Met department is not the only one predicting weather in India. Forecasts are now sold by businesses to businesses

Photograph by Meeta Ahlawat   Also in this story   Cost of inaccuracy From what to sow to when to reap, farmers need weather forecast at every step. It can make or break them Nature’s forecasters Tribals read weather clues in birds, insects, wind and sky

At the peak of summer weathermen in Pune sat down to predict when the monsoon will hit India and how much rainfall it will bring. They fed data into forecast models, and out came the result rainfall will be 96 per cent of the normal, they announced on April 17. But in June the country received only 57 per cent of the normal rainfall for the month.

The government-run India Meteorological Department (imd) in its internal communication had said the June rainfall would be considerably less than normal, yet neither imd nor farmers across the country expected it to be so low. "The last time such a low rainfall was recorded in June was in 1926," said D S Pai, scientist at imd Pune who is in charge of forecasts.

On June 24, imd modified its forecast, saying the monsoon will be 93 per cent of the normal. Its scientists said their earlier forecast followed statistical probability that relies on historical data, not the prevailing weather conditions, to predict the trend. They did suspect that El Nino, a periodic weather anomaly, and heavy snowing in Eurasia would play spoilsport. But, according to Pai, two things happened in May that took them by surprise Aila cyclone hit West Bengal and western disturbance hit north India. Aila took away a large part of the moisture in the air and western disturbance brought drizzle cooling down the northern parts of the country. This weakened the monsoon, said the scientist.

Is weather forecast on the basis of historical data relevant any longer? Research shows trends of decreasing rainfall in June and July (see 'Monsoon earlyburly', Down to Earth, May 16-31, 2009). "We try to take the results from 10 dynamical forecast models, and often each model gives a different version of the weather," said Pai. "Unsure of the results we settle for the age old statistical model." Dynamical models take into account weather conditions like cloud movement and humidity.

Several imd scientists, who do not want to be named, say dynamical models are not reliable because they are not customized to Indian conditions. These models also need regular updates, which require research. imd scientists complained that though such research is happening in organizations like the National Centre for Medium Range Forecasting, iits, Indian Institute of Science and the Indian Institute of Tropical Meteorology, there is not enough interaction between the researchers and forecasters.

"There is almost a caste-like division among meteorological researchers in other organizations and forecasters in imd. So much so they do not publish in same journals," said an imd scientist. "Interaction is limited to seminars."

imd is not very happy about its seasonal predictions meant for the government to formulate its agricultural strategies. But it is confident about its medium- and short-range forecasts given five days and one day ahead respectively.

Enter the private forecaster
Knowing which way the wind blows--or does not--is not left to imd any more. Over the past six years a clutch of companies in India has entered the business of predicting and recording the weather every day, every hour, every 10 minutes. It depends on how frequently the client wants the data. These companies cater to the specialized needs of increasingly competitive sectors like power.

When sudden rain and thunderstorm brought down the temperature in Delhi a few notches on May 30 and air conditioners were switched off, distribution companies supplied considerably less power that evening, when demand usually peaks. Had North Delhi Power Limited (ndpl), a joint venture of the Tatas, not known in advance about the rain and thunderstorm, it would have lost between Rs 5 lakh and Rs 8 lakh.

Distribution companies buy power from generating companies a day in advance. So they have to estimate the demand for the next day. A sudden change in temperature can upset their calculations. Power distribution companies, therefore, make it their business to know what the weather is going to be like.

Delhi-based Skymet supplies customized weather data to ndpl and Reliance Infrastructure Limited, also a power distribution company. It started with a turnover of Rs 18 lakh in 2003 and today has a turnover of Rs 1.4 crore. Kanpur-based Weather Risk Management Services provides localized weather data to companies from its automatic weather stations installed in several places. The company helps compute weather-based insurance packages for farmers and calculates risk to crops from weather anomalies. Skymet even calls itself a "complete weather store". Weather forecasts are no longer confined to free-of-cost predictions by imd.

The profit or loss of many other businesses depends on how accurately these companies predict the weather. The companies are penalized for every weather forecast that goes wrong. In case of temperature, for example, the margin of error allowed is one degree.

"We are paying a private weather company for accurate forecast data because this data could prevent losses of up to Rs 20 lakh," Jayanta Chatterjee, assistant general manager of ndpl, said. "The price of one unit of power can vary from Rs 1.5 per unit to Rs 15 per unit depending on the demand. We calculate the power required for the next day based on that forecast and buy the power when the demand is still low."

Between May 28 and May 30, the maximum temperature in Delhi fell from 41C to 35C. The cost of a unit of power also fell from Rs 5.5 to Rs 1.08. "Weather conditions play a vital role in demand forecasting," a Reliance Infrastructure official explained.

Being forewarned about the weather also helps prepare for contingencies. When Power Grid Corporation, a public transmission utility, learnt of the high probability of fog last winter through Skymet, it hired helicopters to clean the transmission lines. This helped prevent tripping caused by a combination of fog and pollution. Fog in the presence of pollutants on surface of insulators provides an alternate low resistance path which causes flashovers and tripping of transmission lines.

Power Grid Corporation cleaned up the transmission lines intermittently between November 2008 and February 2009 when heavy fog had been predicted. The cost of cleaning came to Rs 8.2 crore, but it saved the business.

The survival of several other businesses is closely linked to advance warnings regarding the weather. Oil rigs, for example. Skymet provides Oil and Natural Gas Corporation (ongc) forecast for their off-shore rigs. "Along with surface weather parameters we track sea weather--wave, swell, gust, wind--for the oil company," said Jatin Singh, managing director of Skymet. The rigs have to be shut down if there is a cyclone. Otherwise, they may catch fire. Besides, ships ply to and from the rigs transporting oil; the company needs weather data to ensure the safety of these ships.

Some wind energy companies too have started buying weather data from private forecasting companies.

Check the cost

Current weather data Rs 12,000 - Rs 1,20,000 per year Forecast data Rs 14 lakh - Rs 18 lakh per year Forecast package for media Rs 18 lakh Daily forecasts for power distribution companies Rs 14 lakh a year (Penalty for each inaccurate forecast beyond a certain percentage is Rs 2-3. There are 400 such forecasts per day.) Wind pattern analysis for wind farms Rs 2-3 lakh, one time (Putting up a facility to analyse the wind at a place would cost Rs 6 lakh and one will have to wait for two years for the results.)

For setting up wind farms companies require data on wind speed, direction and pattern at the site. This helps them calculate how much energy a wind turbine will be able to generate.

"Earlier we were getting these data from imd and Structural Engineering Research Centre in Chennai. Now with more wind farms being set up, the 400 wind monitoring stations of imd are not sufficient," said A Raja Sukumar, vice-president, Indowind Energy, a Chennai-based company that produces wind energy. Private companies can analyse wind pattern quicker by combining imd information with satellite data.

For other businesses, like weather-based crop insurance, the companies set up automatic weather stations, thus, increasing weather coverage. These stations mainly consist of sensors mounted on a pole. In contrast to manual weather stations, automatic weather stations can measure temperature, humidity, wind speed, wind direction and rainfall every hour and send the data to computers via cable or satellite.

Automatic weather stations
Companies like National Collateral Management Services Limited (ncmsl) specialize in setting up automatic weather stations. The company was launched in 2004 with the support of several banks like hdfc, Bank of India and Canara Bank; iffco and the National Commodities and Derivatives Exchange to provide farmers with warehousing and certificate of crop quality so that they can get low-interest credit from banks. In 2004, weather-based crop insurance was introduced in India. The next year ncmsl diversified and started setting up automatic weather stations.

Srinivas Rao, vice-president, ncmsl, said the company has set up 400 weather stations across 16 states in India. It charges between Rs 1,000 and Rs 10,000 per month for the data. For data to be supplied every 10 minutes, the price is the highest Rs 10,000 per month.

"When a company wants data for a particular place we set up stations only if an imd station is not there. We complement the imd," said Rao, who imports these stations from the US. Each automatic weather station costs between Rs 1 lakh and Rs 1.5 lakh, depending on the type of data required. But ingen, a sister company of Weather Risk Management and Services, prefers to build its own stations, which comes to Rs 25,000 each.

There are 1,200 privately set up weather stations in India, and the number is increasing. Rao has orders to set up 50 more stations within July.

icici Bank's weather-based crop insurance scheme works around weather stations. "We buy weather data like temperature, rainfall, wind speed and wind direction from imd and private companies like ncmsl and ingen as these help us verify our claims," said Aditya Jain of icici -Rural that provides crop insurance in 15 states. Once the weather conditions cross or fall short of laid down parameters for a crop, the farmer can raise a claim.

icici Bank has grape cultivators in Maharashtra's Pune-Sangli-Nashik belt as clients. Grape farmers choose the crop phases they want to cover and also the location of weather station used for calculating the claim. They get weather updates from the forecasters the bank subscribes to. India's farmers are still not direct clients of private forecasters.

Increasing unpredictability of weather is another reason private forecasters are flourishing, Anuj Khambhat, director of Weather Risk Management and Services, believes. "More farmers are availing of insurance because of frequent crop loss due to frequent changes in weather," he said. Insurance firms, therefore, need more weather stations.

Pie in the sky
Around the time weather-based crop insurance gained ground, another boom was happening in the sky. News, it was being discovered, can be as marketable as soap operas. By 2005 all major satellite television broadcasters had at least one 24-hour news channel, and they were falling over one another to jazz things up. Weather was an important element of news in private channels, geared as they were to cater to changing consumer preferences in India. Fog-related information, for example, gained prominence during winter as the consumer was keen to know if it affected flight timings.

Weather forecasting companies saw an opportunity in the emerging news channel market. Attractively packaged weather-related data fetched good business. Viewers were not complaining as there were a lot more maps and graphics compared to what they were used to in the days when there was only Doordarshan. Today 70 per cent of the business generated by weather forecasting companies comes from media houses--both television and print.

"We take weather forecasts from private companies because it comes as a map with graphics, which is more suitable to us," said a Zee News executive. "There is less hassle in packaging the information at our end."

If packaging drove media companies from imd to private weather forecasting companies, power distribution companies switched mainly because private forecasters provided data more frequently. India certainly needs to improve forecasting infrastructure.

imd's modernization plan includes increasing the density of its automatic weather stations. In 2007-08, the Centre sanctioned a Rs 950-crore plan, under which 550 automatic weather stations, 55 Doppler radars for data on clouds, and 3,600 rain gauges will be installed across the country. Of these, 125 automatic weather stations have been set up.

But even these installations will not be enough. An imd official said regional changes are so palpable that every district needs at least four stations, which means a total of about 25,000 stations. "At present, we have installations at a distance of 100 km each. But we definitely need more," A B Majumdar, scientist at imd Pune, said.

Score points
	IMD	Private forecasters
Ground observation stations	525	1,200
Weather balloons	39	None
Doppler radars (for studying clouds and wind)	5	None
Microwave sounders (for temperature, humidity)	None, but planned	None
Satellite data	Get in real time	Get at intervals
Historical data	Of 100 years	No data
Research support	Can approach government institutes	No such system
Forecast models	Statistical and dynamical models	Dynamical models
Packaging	Poor	Good
Customized service	Do not provide	Available
Information update	Observations every hour. Forecasts thrice a day for the next day	Observations every 10 minutes. Forecasts every 15 minutes for next day

With more companies entering the business of weather forecasting there are fears of weather forecasting gradually turning into a paid service altogether--the US came close to privatizing its public weather forecaster in the early 1980s. The mention of increasing private business to imd officials tends to get their hackles up. For forecasting one needs historical data and current observations, Majumdar explained. "Nobody has either the infrastructure to record as many observations as imd or the historical data," he said (see Score points).

"Accuracy depends on infrastructure. If somebody is claiming that he is giving better predictions than imd, he is lying," said another official. At the same time, he added, while general forecasting is imd's responsibility, private agencies would do a better job of providing customized forecast data.

The growth of the private industry can, however, spur imd to upgrade its services. The World Meteorological Organisation says "no single government or agency has the necessary resources to address all the challenges on its own". It encourages public forecasters to work with international agencies, other organizations, academia, the media and the private sector to improve the range and quality of critical environmental information and services. At the same time capabilities of the national weather forecaster need to be upgraded since with changing weather pattern India's millions of farmers will need free forecasts more than ever before.

Cost of inaccuracy

Aparna Pallavi and Archita Bhatta

From what to sow to when to reap, farmers need weather forecast at every step. It can make or break them

The monsoon forecast in May said India would receive good rains this year, and the onset was predicted for the first week of June. Accordingly, farmer Lakshinarain Baghel bought 25,000 kg soybean seeds for his 25 hectares (ha) at Rs 30 a kg. He chose a long-duration variety that yields a good crop when water is sufficient. He prepared the field for sowing, but three weeks of June passed with no rain. Panic gripped the region in central Madhya Pradesh as yet another drought appeared imminent.

On June 25, Baghel sold his seed stock at Rs 22 a kg--the price of the long-duration variety had crashed with the drought-like situation. He purchased 25,000 kg of JS 9305 seed, a short duration drought-resistant variety, at Rs 35 a kg. The prices of drought-resistant varieties were soaring.What is more, he had to spend about Rs 30,000 for tilling his land with tractor all over again for sowing, and another Rs 31,250 on reapplying superphosphates. This fertilizer is used just before sowing for the best results, he explained. But if rains fail much of the fertilizer is lost. "I have already lost more than Rs 1 lakh even before sowing has begun," he said.

Baghel lost more than a lakh of rupees even before he began sowing
Photograph by Aparna Pallavi

On July 1, when this correspondent visited Baghel's village Khamkheda in Vidisha district, the sowing was half done, following three spells of light rain in as many days. But panic was still in the air because there had been no rain in the past two days. "Unless there is more rain soon, there will be more losses. Moisture has penetrated the soil only up to four fingers depth (seven-eight centimetres)," said village sarpanch Pradeep Singh Baghel. "Below that the soil is still dry, but we are sowing because it is already late. If it does not rain within a day or two the seedlings will die."

The delayed monsoon has spoiled farmers' calculations in Vidisha, an area known for highly mechanized and input-intensive commercial farming. Although the loss of sown seed has been minimal, farmers have spent an extra Rs 1,500 to Rs 3,500 per hectare, mainly on extra rounds of tilling and on superphosphates. Small farmers who do not own tractors have lost substantial amounts on renting tractors and on diesel and drivers' wages.

Farmers blame the agronomic setback on inaccurate weather information received through radio, TV and newspapers. "Forecasts of the meteorological department are not taken seriously because they do not work," said Govind Dewalia, who cultivates 40 ha in Ban village. "Even now for the past three days we have been hearing the monsoon has arrived in the Bhopal-Vidisha-Raisen region, but there is no rain," said Rajendra Sharma, former sarpanch of Deokhajuri village. Sharma rues that because of the May forecast the village is stuck with long-duration soybean seeds. Many people could not apply a second dose of phosphates because stocks in the market were exhausted. "Last year we were prepared for drought but this year there was no warning," he said.

During last year's drought, say farmers, agronomic disaster was averted in many parts of the district because in view of the low rains, farmers intercropped soybean with pigeon pea, also known as tur. But due to the early monsoon predictions this year, farmers did not bother to save tur for their kharif crop. According to Pradeep Singh Baghel, intercropping increases the cost of harvesting in areas where farming is mechanized. Farmers resort to it only when necessary. "The tur crop matures one month after soybean, so we have to go for expensive manual harvesting," he said.

The state meteorology department, however, insists it supplies exhaustive weather information through special farmers' bulletins twice a week through TV, newspapers and the All India Radio. Nine agro meteorological field units in the state provide district-level information on weather conditions, along with advice from agriculture experts, according to D P Dube, director of the Met centre in Bhopal.

H D Verma, in-charge of the Sehore field unit, under which Vidisha falls, said information is passed on by rural agricultural extension officers and through the Internet to contact farmers, who are supposed to spread the word. But the Sehore unit has contacted farmers only in Sehore district, he said. Verma could not give a definite percentage of farmers who received the information.

Both officials insist awareness among farmers is high regarding these bulletins since the field units and the Met centre in Bhopal receive phone calls from farmers seeking information on weather conditions. Dube gave the example of Swami Dharmachetan Virendra Kumar of Hirnai village in Vidisha tehsil, who calls him up regularly to get details of weather conditions.

Virendra Kumar, when contacted, said less than 0.1 per cent of the farmers in the state call up the field units for information. He complained that information disseminated through media is not farmer-friendly. "Most farmers are unable to access the weather bulletins because they are broadcast at inconvenient times when farmers are in the field," he said. "The Krishi Darshan programme on Doordarshan, for instance, is telecast at 6 pm, which is a very busy time for any farmer."

Kumar has been campaigning with agriculture ministers, officials of relevant departments and media persons, for allotting a regular column in newspapers for weather reports relevant to farmers. Farmers also complain about the jargon used in weather reports.

In effect the only weather information farmers have definite access to is the daily weather forecasts on TV and in newspapers. These are too generalized to be of help. Many farmers then fall back on their traditional knowledge of weather.

Nature's forecasters

Proof in hand Red velvet mites mean more rains are coming
Photograph by Aparna Pallavi

Aparna Pallavi

Tribals read weather clues in birds, insects, wind and sky

Tension was high in Manikwada village in Wardha district of Maharashtra. Weather forecasts by Akashvani and newspapers said contradictory things, promising early monsoons one day and delayed monsoons the next, heavy rainfall one day and scanty the next. Every day farmers waited for the rain, and postponed sowing for yet another agonizing day.

Finally, elderly farmer Jagannath Manmode decided to do what his ancestors had always done. On the evening of June 26, he observed the western sky carefully and announced to his family that the rain was coming. Next day he sowed cotton, tur and soybean in the dry earth. On June 28, the village as well as parts of the Vidarbha region received their first good rain of the season.

Others, who started sowing on June 29, were sheepish but refused to admit they had been outdone by the old man."We too had heard the perti wha bird, you know," said middle-aged sarpanch Keshav Dhole. "But we decided to wait for the rain to be on the safe side."

In the tribal-dominated eastern part of Vidarbha farmers still trust their ancient weather science for predicting rains, both long and short term. They read signs of nature, like the behaviour of birds, animals and insects, colours in the sky, formation of clouds and the colour of the falling rain.

The signs vary from village to village, even person to person. But in this tribal pocket sandwiched between the Nagpur, Amravati and Wardha districts, farmers identify certain signs that are trusted by almost everyone.

The most important sign is the perti wha bird. "It is a small brown bird with blue streaks," explained Dhole. "It must be migratory because it is seen in this region only just before the monsoons." The call of the bird, interpreted by villagers as perti wha, meaning 'let the sowing begin' in Marathi, is believed to be a sign that rains are on their way. Red velvet mite, a small fuzzy insect known as miragya kitak in the area, that shows up in the soil after the first rains, is a sign of more rains. In early July the insect was seen crawling in the farms there.

A year's rainfall can be predicted by looking at the way birds build nests. "If nests are towards the top of trees expect heavy rains," said Nagorao Pote of nearby Jamgaon village. "The lower the nest, the lesser the rains." This year, farmers said, the nests are high on the trees.

Rajesh Barsagade, a young researcher from Chandrapur, said the placement of nests, though considered an important sign, is interpreted differently by tribals in Chandrapur. "If the nest is in the centre of the tree, close to the trunk, it means heavy rains. But if birds build their nests on the tips of the branches, it means scanty rains."

The colour of the evening sky also offers clues, though again, interpretations differ. Red means more rain, white means a gap in rains, said Bapurao Dhole, an old farmer in Susundra village. For Manmode of Manikwada red, yellow and white all mean rains, but each colour combination indicates a different pattern of rain. "A sky mostly red, tinged with white and yellow means a jhad (gentle, steady rainfall that continues for days, even weeks)," he said. "But I have not seen that colour in the sky for over a decade, and we have had no jhads." A rainbow portends good rains.

Farmers say while they do pay attention to weather forecasts on the radio and in newspapers, nature's signs are more accurate than the weather bulletins. One reason is that nature's signs tend to be localized, while the Met predictions are generalized.

Nature's weather forecasters are becoming fewer. Miragya kitak, which were seen in clusters a few years ago, are now in far fewer numbers, and almost always singly. The numbers of the perti wha bird too have fallen drastically in the past decade. At this rate Manmode won't always be so lucky in surprising other farmers in his village.