On India Meteorological Department using statistical models instead of dynamical ones for forecasting Tell me which dynamical model has given correct forecasts for India? The forecasts for Indian monsoons given by the US, UK and Europe in April this year (using dynamical models) threw up different predictions. None of them could predict monsoons would be so low. But the India Meteorological Department (imd) did predict that monsoons will be below normal by taking inputs from all sources; for northwest India they specifically said monsoon would be 81 per cent below normal when most other agencies said it would be normal. In qualitative terms our forecast was accurate. The dynamical models do not give reasonably accurate forecasts at all times and for all regions. There is a study in a 2005 edition of the journal Science that says we will eventually have to depend on statistical models that use historical data.
On staff shortage in IMD We don't have enough people and are thinking of recruiting students right after graduation, train them for a year or two and provide some kind of employment guarantee after that. No recruitment has taken place in imd for some time. This ministry (of Earth Sciences) has been created to bring about a change in the functioning of such departments; the change will be clearly visible in a couple of years.
On weather forecasting at taluka level to help farmers For farmers to benefit, we will have to eventually provide forecasts at the village level. For that we need people who can handle high-resolution forecast models that give detailed predictions and sufficient number of computer systems. At present we are using forecast models that can predict weather for areas measuring 50 to 100 sq km (mesoscale models). That is why we give region-wise forecasts. Right now we are aiming for district-level forecasts in the next five to seven years.
On private forecasters taking over the business of weather forecast We have no problems with private players also giving forecasts. We also provide forecasts to specific sectors like civil aviation. We will be providing separate forecasts for five regions of Delhi during the 2010 Commonwealth Games.
On inability to provide taluka level forecasts while being able to give zone-wise prediction for Delhi We have the technology to provide taluka forecasts but we need more people. Region-wise forecast for Delhi is being done on an experimental basis. If it works, we will replicate it in other metros. imd has 240-odd scientists and I cannot increase the strength to 500 overnight. There are very few scientists in our country who can do modelling and assimilation of data. This deficiency has accrued over a period of time and will take time to rebuild.
On delay in installing weather stations; only 125 automatic weather stations have been installed so far We are in the process of installing 350 more weather stations. After procurement, the weather stations are calibrated in Pune before being taken to the site. I agree things are not moving as fast as they should. Of the 55 Doppler radars (used for assessing cloud type, wind speed and direction) sanctioned, 12 have been ordered and one has arrived. Radars require clearances; all 12 will be installed by next March.
On climate change research We have earmarked over Rs 70 crore for activities related to climate change. The money will be used for: (a) developing a climate change research centre to study the behaviour of weather (including monsoons) in different climate conditions and if weather predictability will change (b) introducing high speed computing for climate change predictions and (c) studying cloud-aerosol interaction and how it is influencing warming of the globe. The Indian Institute of Tropical Meteorology in Pune has already started the study.
On lack of communication among scientists/institutes working on climate change Researchers communicate through their published studies and during workshops. Forecasters and researchers have different roles to play and may disagree. For instance, if there is new research on rising sea-surface temperature increasing incidence of intense rainfall, the forecasters may say fitting this understanding into their mathematical forecast model is not giving accurate predictions. But I do not think there is any communication gap between the two groups of scientists.