Private sector weather forecasting company Skymet stated that monsoon will be below normal this year, in its April projection. But the India Meteorological Department (IMD) said monsoon will be normal in the first of its two-phase estimate. IMD and Skymet have often differed in their monsoon calls. Sarthak Ray takes a look at probable reasons

The calls on Monsoon 2023

IMD expects a normal monsoon this year. On Tuesday, the government weather forecaster said that the southwest monsoon (June to September) will be 96%, with a model error of ± 5%, of the Long Period Average (1971-2020) of 87 cm.

Also Read: Food inflation falls sharper than headline rate in March

Skymet, on the other hand, says it will be 94% (±5% error), predicting a below-normal monsoon. In India, <90% of the LPA is ‘deficient’ rainfall, 90-95% ‘below normal’, 96-104% ‘normal’, 105-110% ‘above normal’, and >110% ‘excess’ rainfall.

The calls may seem very different, perhaps because of the terminology, but a Skymet blog post says: “The difference between the two amounts to … just 2 days of rainfall … There can hardly be any better consensus … between any two independent weather agencies.”

Why monsoon calls are critical

The monsoon irrigates around 60% of net sown area in the country. So, Indian agriculture’s fortunes are tied with its strength. Farmers, therefore, need accurate forecasts to take important decisions. In a bad monsoon year, for instance, a near-accurate forecast would help them pare down investment in commodities like seeds and fertilisers if they decide to scale back sowing.

There are knock on effects of a bad monsoon year for the rural economy (depressed farm wages, lower demand). Jefferies, in a recent research note, says demand outlook is something to watch out for this year given the prediction of a weak monsoon (Skymet).

In drought years, the government also has to import many crops and agri-products, with implications for inflation. A near-accurate monsoon prediction could help plan for this.

How have IMD & Skymet fared

The IMD has been forecasting monsoons for a very long time while Skymet started predicting monsoons only in 2012. In the last six years, IMD and Skymet have both, at times, lost the plot and also made fairly accurate predictions. In 2017, against the actual rainfall of 95% of the LPA, Skymet predicted 95% to IMD’s 98%. The very next year, both were wide off the mark—actual at 91% versus IMD’s 97% and Skymet’s 100%; 2019 also wasn’t a good year for both, with the actual at 110% versus IMD’s 96% and Skymet’s 93%. In 2022, IMD was closer to the mark (103% against the actual of 106%) versus Skymet’s 98%.

Also Read: India to have stable debt-to-GDP ratio: IMF

IMD has also had its fair share of misses (during 1999-2002, it overshot the actual sharply every year, as it did in 2009 and 2016). In 2002, 2004, 2009, and 2014, it failed to read the tea leaves on deficient monsoon, drastically so in 2002, 2004 and 2009.

How are monsoon forecasts made? Why do Skymet and IMD sometimes differ?

IMD generates its long-term predictions using the multi-model ensemble (MME) forecasting system. MME derives its forecast from coupled global climate models (CGCMs). India is a member of the World Meteorological Organisation and thus accesses a wide trove of weather related data. IMD’s forecasting is based on, among other things, monitoring of surface and upper air observations, real-time monitoring of monsoon with satellites and radars, analysis of different meteorological charts, and “guidance from various national and international weather forecasting models.” Skymet gets similar data from various sources.

The difference, according to experts, could lie in the weightage assigned to certain kinds of data rather than any sharp differences in the overall data pool. For instance, one sharp point of difference this year between the two is how they view the likely El Nino effect. IMD acknowledges its likelihood, but says not all El Nino years are bad monsoon years, relying on positive Indian Ocean Dipole as well as low snow cover in the northern hemisphere. However, a Skymet blog says: “Scare of El Nino this year is real and the 2nd half of the season is likely to be impacted strongly.” Meanwhile, Jefferies said: “… in prior El Nino years, IMD had to sharply revise its forecast downward from its April estimates.