The India Meteorological Department (IMD) has forecast a ‘normal’ monsoon for this year. In the agency’s parlance, normal implies that the country will get 96% to 104% of the 88 cm that it gets from June-September. This quantity, called the Long Period Average (LPA), is a mean of monsoon rainfall from 1961-2010. The IMD, for over 20 years now, follows a two-stage monsoon forecast system. After the prognosis in April, it gives an updated estimate in late May or early June. This includes an estimate of how much rain is likely in: northwest India, northeast India, central India and southern peninsula. Numbers are also given for July and August, which see two-thirds of the monsoon rains and are the most important months for sowing. This year, there will be forecasts for June and September too, to be given in May and August, respectively. Historically, predicting rain for June and September is challenging as it corresponds to the monsoon’s entry and exit. There will also be forecasts for what is called the monsoon core zone, which represents most of the rainfed agriculture region in the country. All of these updates are an extension of the IMD’s increasing reliance on dynamical monsoon models. Unlike the traditional statistical models, which are based on a fixed set of meteorological variables that have historically been correlated with variations in monsoon rainfall, the dynamical models generate forecasts based on evolving weather patterns. The IMD has been testing such models for many years but it is only in the last few years that it is finding use for practical weather forecasts.
A ‘normal’ monsoon forecast this year is primarily predicated on ‘neutral’ surface temperatures in the Central Equatorial Pacific. In 2019 and 2020, the IMD forecast normal rains but India ended up with 110% and 109% of the LPA. This year, a warming El Niño is unlikely, and another ocean parameter closer home, the Indian Ocean Dipole, too is expected to be unfavourable for excess rains and so the IMD seems more confident that its calculations are not going to be as wrong. However, the models already show a good chance of ‘above normal’ rain in central and southern India. While forecasts are a critical aspect of India’s disaster preparedness, there should also be more focus on incorporating these forecasts down to municipal and block-level planning. The monsoon forecasts were primarily evolved to assist with agriculture and it is only now getting more urban-focused because of the natural disasters that accompany even ‘normal’ monsoons. Several business and service sector industries need weather products and in terms of science and infrastructure, few have the resources the IMD has. The IMD must continue to aid on all these fronts.