New Delhi: India plans to offer monsoon prediction and climate modelling services to neighbours such as Nepal, Sri Lanka and Afghanistan in a bid to access valuable data from these countries and put its recently procured supercomputing systems to use.
“Though we have good scientists and are upgrading our infrastructure, we need data from other countries,” said an official familiar with the matter, who didn’t want to be identified.
“To that end we can develop forecasts for them and maybe over time, get some return on our supercomputing investments,” the official added.
The systems were procured at a cost of $28 million (Rs127.4 crore) from International Business Machines Corp. They were installed at various climate research institutes, including the Delhi headquarters of the India Meteorological Department (IMD). Key officials from the ministry of earth sciences (MoES) are expected to meet delegates South Asian Association of Regional Cooperation (Saarc) nations in Pune next month for a conference for preliminary discussions. Saarc groups Afghanistan, Bangladesh, Bhutan, India, the Maldives, Nepal, Sri Lanka and Pakistan.
The weather data that India is looking to access will help improve the accuracy of its global circulation models (GCMs), two MoES officials said. GCMs are next-generation weather forecasting systems on which scientists simulate the atmosphere at any given point of time and run the resultant data on powerful computing machines that try extrapolating weather behaviour over the next few months.
Although GCMs allow accurate forecasts over a week, they often fail in predicting monsoon trends even over a month’s span. In 2008, GCM models predicted south India will witness only three-quarters of normal rainfall. The region posted a 25% excess.
“There are still weaknesses in this approach (GCM). That’s why we still depend on the statistical data. But given that there’s a worldwide push to GCM forecasting, we must improve this approach. For that we need a continuous flow of high quality data,” said D.S. Pai, director (forecasting) at India Meteorological Department, Pune.
IMD still relies heavily on statistical models in preparing its monsoon forecasts, though since 2004, it has gradually started incorporating results from locally developed GCMs too.
Other officials also see India’s initiative as a strategic step. A key objective of the conference in Pune is to help develop a Saarc monsoon model and the World Meteorological Organization has tasked India with taking the lead in the region.
“The WMO has asked us to take the initiative. Our key objective will be to improve our observation systems, gather better data and use that to improve our own monsoon forecasts,” said Shailesh Naik, secretary, MoES.
Better models will also help better forecast the impact of climate change and its effects over South Asia. In the absence of quality data generated locally, policy makers and scientists have often relied on predictions made by Western institutions on the impact of climate change.
“That could be useful given that we are in the process of strengthening our in-house modelling capabilities and trying to depend less on Western predictions of climate change,” said M. Rajeevan, a climate specialist with the Indian Space Research Organisation.