13/05/09

Global warming may make monsoons harder to predict

The increased frequency and intensity of rainfall will make the tropical atmosphere even more unstable Copyright: Flickr/millzero.com

Send to a friend

The details you provide on this page will not be used to send unsolicited email, and will not be sold to a 3rd party. See privacy policy.

[NEW DELHI] Monsoons will be more difficult to predict in the future because of global warming, researchers have warned.

Scientists will need improved weather prediction models, conclude researchers from the Indian Institute of Tropical Meteorology (IITM), in a report published in Geophysical Research Letters last month (23 April). Reliable prediction of monsoon rains five to seven days in advance is crucial for farmers, for managing water resources and for disaster management, they say.

The researchers say that weather in tropical regions is inherently more difficult to predict than in regions further north. One reason is that daily changes in temperature and wind are small in the tropics and the signals are difficult to pick up with recording instruments. Another is the way wind systems are driven in the tropics, which makes the tropical atmosphere more unstable.

The increased frequency and intensity of rainfall due to global warming will make the tropical atmosphere even more unstable and prediction more difficult in the future, they say.

The IITM scientists analysed daily rainfall data over central India for 104 years, from 1901 to 2004. The frequency and intensity of rainfall has been increasing in tropical areas in general and over central India during the summer monsoon in particular (see Indian monsoon ‘intensified by climate change’).

They found that the predictability of monsoon rains over central India had decreased in the last couple of decades compared with the 1950s and 1960s. This coincides with an increased frequency of extreme weather events such as severe droughts and rains.

The changes brought about by global warming are also leading to errors building up faster in current prediction models. "Errors creep into prediction models both due to imperfect datasets and drawbacks in the model itself," IITM director Bhupendra Nath Goswami told SciDev.Net.

"Earlier, it would take three days for a small initial error to double. Now it takes only 1.5 days," he says. The larger the error build-up, the more inaccurate the prediction.

The findings have important policy implications, says Goswami. The first step is to recognise the problem that global warming will make monsoon prediction even more difficult, which has been demonstrated by the IITM team.

"Once you recognise this, one will have to double and triple our efforts at improving prediction. This means more funds for research on monsoon prediction models."

Link to abstract in Geophysical Research Letters