Indian Meteorological Department · Monsoon

IMD’s 1st Long Range forecast for SW monsoon 2020

IMD’s 1st Long Range Forecast for SW Monsoon 2020 on Apr 15, 2020[i] says:

– Rainfall will be normal (100% of LPA: Long Period Average of 88 cm). IMD says Neutral El Nino Southern Oscillation prevails in Pacific Ocean and Neutral Indian Ocean Dipole conditions prevail in Indian Ocean, likely to remain throughout the monsoon as per most models, some indicate weak La Nina conditions in Pacific Ocean in second half of monsoon.

– FORECAST BASED ON MONSOON MISSION COUPLED FORECASTING MODEL There is high probability (70%) of monsoon rainfall being above average to excess (over 104% of normal)

– FORECAST BASED ON OPERATIONAL STATISTICAL ENSEMBLE FORECASTING SYSTEM: 9% probability of Deficient monsoon (Over 10% below normal); 20% probability of below normal (90-96% of normal) rainfall; 41% probability of normal monsoon (96-104% of normal rains); 21% probability of above normal (104-110% of LPA rains) and 9% probability of excess (over 110% of normal) rains.

LPA Changed One of the major changes IMD has effected in its April 2020 forecast is that it has reduced the Long Period Average (LPA) rainfall from 89 cm that was till last year (based on average rainfall for 1950-2000) to 88 cm this year (based on average rainfall for the period 1960-2010). It could have easily updated it right till 2019 rainfall, but it is not clear why it chose to do it this year and that too till 2010.

Onset and withdrawal dates changed The other major change in IMD’s Long Range forecast this year is that it has changed what is normal dates of onset and withdrawal of monsoon for different cities and states[ii], factoring in the likely impact of climate change, which has altered the way the south-west monsoon progresses over India. IMD’s new onset dates are based on an analysis of monsoon data from 1961 to 2019, and withdrawal dates are based on data from 1971 to 2019 by scientists in IMD, Pune. The onset date over Kerala continues to remain June 1 and withdrawal from the south peninsula October 15.

“The data we were using for monsoon onset and withdrawal was very old and based on only a few stations. It’s better to use latest data as much as possible. We don’t know if the changes in monsoon dates are a result of climate change. It is a possibility. It could also be a normal change in monsoon cycle or what we call natural variability. These things will have to be studied further,” said M Rajeevan, secretary, ministry of earth sciences. The new monsoon arrival date in Northwest India is earlier and withdrawal dates are later than old dates, which means monsoon will stay longer in Northwest India.[iii]

However, the IMD would continue to account for monsoon rainfall as the rainfall that the country received from June 1 to September 30, even though, as an official said, the monsoon appeared to be lingering in many places until October 15. “We have been discussing this matter, of the monsoon’s delayed withdrawal and whether we should have a different accounting system for monsoon rainfall keeping in mind the new withdrawal dates. No decision has been taken,” said Madhavan Rajeevan, Secretary, Ministry of Earth Sciences, which oversees the IMD.[iv]

– Monsoon will withdraw from Kerala on Oct 8.

– Monsoon will arrive in Ahmedabad on June 21 instead of June 14. In Bhopal on June 22 in place of June 15.

The new forecast dates show delay in Chennai (3 days, new date June 4), Pune (1 day, new date June 10), Mumbai& Kolkata (1 day, new date is June 11), Lucknow (3 days, new date June 23) and Delhi (4 days, new date: June 27). However in case of Chandigarh, the new arrival date is June 26, five days earlier than the July 1 that prevailed so far.[v]

The full table from IMD, as published by the Times of India, is given here.

IMD SW Monsoon NEW onset and withdrawal dates, Apr 15, 2020 p1
IMD SW Monsoon NEW onset and withdrawal dates, Apr 15, 2020 p2
IMD SW Monsoon NEW onset and withdrawal dates, Apr 15, 2020 p3
IMD SW Monsoon NEW onset and withdrawal dates, Apr 15, 2020 p4

The Application Laboratory of Japanese national forecaster Jamstec has extended the wetter-than-normal rainfall outlook for the South-West monsoon in India to the North-East monsoon as well[vi]. Most parts of South-East Asia, the Philippines and Sri Lanka will also experience a wetter-than-normal condition during the summer monsoon. The same forecast is now being extended into the autumn for India (right until December), most parts of South-East Asia, the Philippines and Indonesia.

Skeptical Editorials The editorial in Business Standard (soft copy not available, see image below) of Apr 17, 2020 highlights that with new arrival and withdrawal dates, the length of the monsoon has increased by 15, days to Oct 15, 2020. The Edit is critical of credibility of IMD’s long range forecast of monsoon, saying how it has been proved wrong in the past including the stage 2 forecast. Moreover, the distribution of rain, the most important aspect for farmers, is forecast when monsoon is well underway already: “That is too late to be of much avail for the farmers and policy planners.” The Edit thus concludes: “A good deal, therefore, stil needs to be done to sharpen the IMD’s monsoon rainfall foretelling models and improve the utility of its predictions for the various stakeholders.”

The Business Standard in fact reported on Apr 16 2020[vii], showing the difference between IMD’s first monsoon forecast and actual rains in last seven years since 2012: 2012 (IMD’s 1st LR forecast in April: 99%, Actual rainfall: 92%); 2013 (98, 106); 2014 (96, 88); 2015 (93, 86); 2016 (106, 97); 2017 (96, 95); 2018 (97, 91); 2019 (96, 110). So the difference between forecast and rainfall has been: -7, 8, -8, -7, -9, -1, -6, 14. So the highest difference between forecast and actual rainfall was in just the last year! There has been only one instance in these eight years when IMD forecast 100% or above rainfall, in 2016, when the actual rainfall was 9% below forecast! In fact, in six of the eight years, the actual rainfall has been below the forecast rains. So Business Standard’s skepticism about IMD forecasts is not entirely misplaced.

BS Edit however, observes: “The optimism on this count is supported by the remarkable current water profile of the country.” The Edit that goes on to count the water stock in 123 reservoirs as THE indicator for these remarkable. Clearly, the EDIT is way off the mark here. The reservoirs indicate a tiny part of the reality, and the state of water resources in India does not provide much reason for such optimism.

The Hindu in its editorial on April 18, 2020[viii] does not find IMD’s April forecast “much to go by especially if the agency declares it ‘normal’. It in fact encourages IMD to go for a “truly momentous break from the past” by moving “to a new monsoon-accounting calendar, instead of the century-long tradition of June-Sept”. But editorial hinted that such momentous break is not possible considering that IMD is “a part of a hierarchical government set up” rather than an independent scientific agency that it ought to behave like. Indeed, that break from the past would be a momentous one, one that is highly required, long overdue.

IMD also needs to provide state wise of at least Meteorological sub division wise rainfall forecast and also remove some of the anomalies in the way IMD has divided India into 36 sub divisions.











Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.