Posted on behalf of K. S. Jayaraman.
Indian meteorologists are under fire for having predicted a normal monsoon when the country is heading for a drought (see the Hindu).
The rains during the 4-month-long monsoon season (June to September) — accounting for more than 80% of India’s annual rainfall — are crucial for the agricultural economy. In April, the Indian Meteorological Department (IMD) predicted that the monsoon season would see normal or above-average rainfall. On 2 August, however, it confessed that more than half of India has received “deficient or scanty” rains, and that the monsoon rainfall for the entire country is likely to be 19.7% lower than normal.
“I admit this year we are going to go wrong!” Bhupendra Nath Goswami, director of Pune-based Indian Institute of Tropical Meteorology (IITM), whose seasonal prediction system was used by the IMD to make the forecast, tells Nature. “But blaming the meteorologists is not going to solve the problem,” he says. “You have to understand that regional climate like the monsoon has a potential limit on predictability due to the chaotic nature of the weather and climate system. Right now our models have skills significantly below this limit.”
Jayaraman Srinivasan, chairman of Divecha Centre of Climate Change at the Indian Institute of Science in Bangalore, agrees. “No forecasting centre in the world (American, European, Japanese or Indian) has been able to forecast monsoon rainfall accurately this year or in the past,” he says. “Hence the question to ask is, ‘what should scientists around the world do to improve the seasonal forecast of the monsoon rainfall?’” A similar question was posed in 2002, when the rainfall in July was 50% below normal (see ‘Climate model under fire as rains fail India’).
Srinivasan says that the accuracy of the forecast is hampered by uncertainties in conditions such as wind velocity, cloudiness and humidity. Gathering more data, using radars, for example, could reduce the errors, “but this will only improve the short-term forecast of less than one week”, he says. “The seasonal forecast will improve only if there is a dramatic improvement in our understanding of how clouds are born, grow and decay.”
New information about cloud processes provided by NASA’s TRMM (Tropical Rainfall Measuring Mission) and CLOUDSAT satellites has not yet “translated to tangible benefits for seasonal forecasting”, he says.
India’s Ministry of Earth Sciences recently launched a ‘Monsoon Mission’ to improve the prediction of Indian summer monsoon rainfall, says Goswami. The 5-year, US$75-million project aims to bring together researchers from “all relevant organizations and research institutes” and give them the best computational facilities to improve their meteorological models.
“Accurate prediction of monsoon rainfall may be as tough as finding the Higgs boson,” says Srinivasan, “and hence it will demand a large financial and human resource.”