ABOUT 500 active users utilise more than 10,000 items of data each year in India. And, about 72 per cent of the data is from IRS-1A and IRS-1B satellites, with the government subsidising the space segment. Cost-benefit analyses have been made in some applications. For example, using satellite data for flood mapping of the river Godavari in Andhra Pradesh during the 1986 monsoons proved to be 15 times cheaper and 20 times faster than using conventional methods.

For agricultural and crop yield estimations, which require a large quantity of repetitive data covering a wide geographical area and mode available in a short time, satellites are ideal.

This view is opposed by M S Bhatia, an economic adviser at the ministry of agriculture. "For a crop like wheat," he said, "perhaps remote sensing may prove adequate and because it is grown largely as a monocrop in contiguous and irrigated areas, it is easier to estimate the hectarage. Also, as wheat is a winter or rabi crop, cloud cover is less of a problem. On the other hand, when an area has various crops with multiple water sources, satellite estimation becomes more inaccurate."

The ministry of agriculture is evolving a wheat production information system in Punjab and Haryana that will enable pre-harvest estimates of production. This would enable timely forecasts and decisions on storage, imports and public distribution. For instance, in a reduction of 1.4 million metric tonnes of wheat imports would save India's foreign exchange to the tune of about $15 million at 1976 prices.

The total cost of the information programme is estimated at Rs 12.35 lakh, about half of which is the cost of satellite data.