The southwest monsoon and Narendra Modi have not always been the best of friends. The Modi government’s first innings had begun in 2014 on a drought note, which was followed by another bad monsoon in 2015. There had been only four such instances of back-to-back failures since the last century: In 1904-05, 1965-66, and 1985-87 (three successive years).
While the monsoon season is from June to September, it also leaves its imprint on the ensuing months. Within the season, too, there are long dry spells followed by periods of excessive rainfall; so, you could well have a drought and a flood in the same region. And climate change, of course, has made the monsoon more volatile.
Over a longer period, the monsoon has had a pattern, with certain epochs – a decade or even two – having a higher frequency of droughts. Thus, during this century, there have been five droughts (2002, 2004, 2009, 2014 and 2015) and five below-normal years (2000, 2001, 2012, 2017 and 2018). The average monsoon rainfall in the current decade (2011-2018) has been only 835 mm, as compared to the normal long-period average (LPA) of 887 mm. This would indicate that we are now in a low rainfall epoch.
The above fact was borne out even in 2016, which was a La Niña year. Yet, despite predictions of a bumper/surplus season, the 2016 monsoon ended up with cumulative country-wide rainfall only at 97% of the LPA. Although at the lowest end of the range for a “normal” monsoon, it led to drought conditions in parts of Punjab, Haryana, Gujarat and North-East India. 2017 provided no respite either. It was a below-normal monsoon year, with Madhya Pradesh and Chhattisgarh joining the drought-hit areas of Punjab, Haryana and Uttar Pradesh. Last year also, the monsoon was 9.4% deficient and only narrowly escaping the definition of drought. While Punjab and Haryana managed to bounce back, the whole of eastern India (Bihar, Jharkhand and West Bengal), Gujarat and large parts of the South slipped into scarcity.
Against this background, the severely deficient rains in June seemed to suggest that 2019 would join the ranks of below-normal monsoon years. But as it turned out, this monsoon has defied all expectations and patterns.
On April 3, Skymet had forecast this year’s monsoon to be below normal at 93% of LPA. Our prediction was made amidst an ongoing El Niño event. In recent years, it has been seen that even a “weak” El Niño can adversely impact the monsoon, as much as a strong event did in the drought of 2004. The 2019 monsoon, in contrast, has behaved somewhat on the lines of 1997: That year witnessed a “strong” El Niño and yet it had a normal monsoon, with rainfall at 102% of LPA.
Skymet has been using the Climate Forecasting System (CFS V2) global model. While our model has successfully predicted monsoon rainfall in the past, it is, however, very sensitive to El Niño. The El Niño this time was expected to last through the season, albeit on a weakening note. But it started collapsing after June. At the same time, the Indian Ocean Dipole (IOD) factor seems to have played a dominant role this season. IOD, by itself, is usually not considered a strong driver of the monsoon, particularly during El Niño years. Most models, including Skymet’s, underestimated its role. The IOD, right from the beginning, kept growing stronger (positive). In combination with the Madden-Julian Oscillation, as and when it visited the Indian Ocean, this enhanced the overall rainfall.
Skymet normally reviews the monsoon’s performance halfway through the season. But we decided against it this time, notwithstanding indications of better rains in the second half of the season. What held us back was the volatility of numerical models and the historical record. Our own model volatility grew so much that projections for the second half showed large variations. At best, we could have revised the forecast to 99% of LPA, although with low confidence. In hindsight, had we done that, our short-range forecast for August-September as well would have gone wrong, as it happened with the India Meteorological Department’s (IMD).
Also, historically since 1871, whenever June had a rainfall deficit of 30% or more, the season invariably ended up as below normal, if not recording a drought. This time, we had a rainfall deficiency of 33% in June. Given the past record, a recovery from such a huge deficiency is nearly impossible. Monsoon dynamics, moreover, do not suggest enduring good rainfall performance for three consecutive months. In 2019, however, we have precisely had that: July, August and September have got progressively better and the cumulative rainfall for the monsoon season from June 1 has ended with a surplus of 10%.
The Modi government has finally seen its first good monsoon. 2019 is possible the best monsoon year since 1994. What is more, the September rains have broken the record of the last 102 years, drenching most parts, especially central India. It has taken every forecaster by surprise. We at Skymet had assessed September to be the best month of this monsoon, with rains at 102% of LPA. But the actual number has been 153%, the highest since 1917.
Equally heartening is the fact that the spatial distribution of rainfall has been good this year. Perennially water-stared Gujarat, Maharashtra and Madhya Pradesh have, in particular, received excellent rains. With water levels in India’s 113 major reservoirs also at nearly 87% of their live storage capacities – as against the last 10-years’ average for this time – the Modi government can take comfort from a great start in its second innings. The excess rainfall, especially in September, would help in the accumulation of residual moisture in the soil that will prove beneficial for the coming rabi winter-spring crop. The replenishment of the water table and aquifers may even see us through till the next monsoon.
The IMD deserves congratulations for predicting a near-normal monsoon at 96% of LPA, even if this may have been the lowest possible normal. Even on August 1, the IMD had revised and retained its forecast at this level. Its August and September projections, at 99% and 102% of LPA, also went off the mark by a huge margin. At the end of it all, none of us, the IMD included, were able to capture what has been the best monsoon at least since 1994.
The writer is founder and managing director of Skymet Weather Services Pvt. Ltd