|How our changing climate is, and will be, affecting the Indian monsoon
||[Sep. 8th, 2009|04:42 pm]
Rainspotting in Bangalore
|The monsoon season is now drawing to a close, and the newspapers are rife with articles on how India will wrestle with the drought that was this year declared in 25% of her districts. Yet all summer the meteorological offices were defiant of any suggestion that the monsoon is becoming increasingly erratic, maintaining that the season has always displayed such unpredictable behaviour and that everything was hunky dory.
I met with the Director of the Indian Meteorological Department (IMD) in Thiruvananthapuram, Kerala, in the middle of the season and he also echoed this stance. An imposing man who seemed slightly nervous, M.D. Ramachandran’s ready agreement to meet me came as a pleasant surprise, particularly as I showed up unannounced. My friend, an established Indian journalist who took me there, said that he was rarely granted such an impromptu audience.
“Every year, people come to us saying ‘This summer it is very hot! There is no rain!’ There will also be others who say ‘This year is the heaviest rainfall I have ever seen!” he said, leaning back in his chair and chuckling. “But weather is variation only. Research may be showing that there will be changes from this global warming, but our day-to-day data is not showing any change.”
To be fair, the weather station in Thiruvananthapuram does not conduct research, consigned instead to the short-term prediction of Keralan weather based on a system of plotting areas of high and low pressure and anticipating weather events across their lateral gradients. To do this, they receive meteorological data from about fourteen observatories and seventy raingauge stations across the state and coastal islands, and any inaccuracies require an explanatory report to the Director General of Meteorology in Delhi. Ramachandran pointed out that prediction of weather patterns, an influential and serious task at the best of times, is more problematic in equatorial regions such as Kerala, as weather changes much more rapidly than in areas of greater latitude. Still, he said, their June predictions for Kerala had an accuracy of 89-95%.
That said, he was almost too hasty to put down my questions. No, the average annual rainfall has not changed but no, there has been no seasonal variation either. No, there is no evidence that the spatial distribution of rainfall is changing. No, the melting of the arctic ice caps is not affecting our area, and no, ocean currents do not directly affect the monsoon. I got the impression he felt so consistently badgered about changes in the weather these days that he had become defensive.
But the truth is, there have been observed changes in the monsoon patterns in India. In Kerala, for example, there has been a decreasing trend in the amount of rainfall received by the south of the state over the last 100 years (see graph, top left), particularly over the slopes of the Western Ghats. The onset date of the monsoon has not moved from its 1st June average, however, and there has been no such long-term trend observed for northern Kerala.
Looking at the rainfall for all of India, it seems that there is no trend in the amount of water that falls during the monsoon, although there is, of course, interannual variability.
However, it would be foolish to assume this means there has been no change at all. Studies by Professor B.N. Goswami, who heads the Indian Institute of Tropical Meteorology – the place where they do do research – and other scientists have shown that there has been an increase in the occurrence of very intense rainfall in the period 1951-2003 (see graph, right). Both ‘very heavy’ events (more than 150mm of rainfall) and ‘heavy’ events (100-150mm of rainfall) show an increasing trend, while ‘low and moderate’ rainfall events (5-100mm) are on the decline. Combined, these figures produce an apparently unchanged amount of overall rainfall.
A similar change has been observed with cyclones. India gets about six tropical cyclones a year, whirling in from the Arabian Sea and Bay of Bengal in the months either side of the monsoon season. Professor P.V. Joseph of Cochin University, who kindly provided me with this data, has shown that cyclone occurrence shown no trend of change in the long-term, although their frequency does seem to follow a 36-year oscillation. However, within this, the amount of those cyclones that are classified as ‘severe’ has been increasing over the last 100 years. Cyclone Aila, which slammed into West Bengal in May of this year, inundating the Sundarbans region with 20ft of water and claiming more than 200 lives in total, can certainly be classified as severe. There has also been revealed a long-term decrease in monsoon depressions (see graph, below left), with a 36-year oscillation period within that. Monsoon depressions are associated with large areas of heavy rainfall – 20 to 30cm per day - and whereas roughly twelve monsoon depressions would occur during the 1900s; now there are only about four. Many of the people we spoke to for Rainspotting attested to rain now only falling in ‘pockets’, whereas before it would rain uniformly over large areas. Depressions form over the Bay of Bengal and head west-north-west, bringing heavy rains to Rajastan, in which 26 of the 33 districts have been declared as drought-hit this year.
Joseph and colleagues have also observed a steady weakening in both the lower tropospheric monsoon winds (the low level westerly jetstream, running from the surface to about six kilometres height) and the wind flow at heights of 12 to 16 kilometres (the tropical easterly jetstream), over the past five to six decades.
So what of the future? Climate models have never been able to predict the monsoon climate and its variability with accuracy, and calculations by Goswami indicate the monsoon weather has now become twice as difficult to predict as a result of climatic changes. The climate modeling community is small in India, and huge improvements would be required both in observations and prediction models, as well as computational power.
The Walker Institute for Climate System Research at the University of Reading draws together a number of renowned climate system research groups and centres from all over the world. Dr Andy Turner’s work focuses in particular on the changing climate in the Indo-Pacific area, and its effect on the South Asian monsoon. He is of the opinion that the extremes of weather – floods and drought – experienced in this year’s monsoon could well be a sign of things to come.
“Modelling results show that the wet areas over India are likely to get wetter overall, though the rain is likely to come in shorter, heavier bursts with longer dry periods in between,” he says. “The result may be both increased flooding and, paradoxically, increased drought. So we could see more of the sort of conditions we’ve seen this year in India.”
There is a strong consensus that the variability of the monsoon will manifest itself on both a year-to-year and day-to-day basis. The yearly variation of the monsoon will result in greater occurrence of flood and drought years, while the daily undulations mean that rain days may occur less often, but it’ll really flog it down when they do.
“Heavy events will likely get worse in the future, as extreme rainfall becomes both heavier and more frequent,” predicts Turner. “Rainfall events are essentially heavier because the Indian Ocean is warming up and this means there is more moisture available for convention over India.
“In addition, there are variations during the summer known as active-break cycles: short periods of heavy rain (active) or dry (break) conditions, and much more work is needed to see what will happen to these events with climate change. It’s important to note that ‘drought’ does not necessarily mean no rainfall at all, but Indian agriculture and economy are so well tuned to normal conditions that these sorts of events can cause immense problems.”
Top left: Over the last 100 years, monsoon rainfall in south Kerala has shown a long-term decrease, the best fit line of which has a gradient of -0.3588
Middle right: Although the average rainfall for all India has shown no overall change from 1951-2001, within that there has been a decrease in low and moderate rainfall events, and an increase in heavy and very heavy rainfall events. A further increase in the intensity of extreme rainfall events is predicted by climate models.
Bottom left: The top graph in this images displays the decreasing trend in frequency of monsoon depressions over the last 100 years (plus). The graph at the bottom left of the picture shows that, during the first half of the period, depressions decreased at the rate of 0.29 per decade. The graph at the bottom right of the picture shows that during the second half (1951 onwards), it decreased at the accelerated rate of 0.93 per decade. A 36-year oscillation is seen superposed.
With thanks to Prof. P.V. Joseph, Dept of Atmospheric Sciences, Cochin University of Science and Technology, Kerala, for data provision.