WEATHER/AGRICULTURE

In addition, IMD runs models specific to cyclones and hurricanes that generate forecasts with a lead time ranging from hours to days. Yet, why does IMD falter?

REASONS APLENTY

The main reason for IMD’s failure is the poor density of observatories. The government contends that the country needs 40,000 AWSs every 10 km in the plains and every 5 km in the hills. India has only a third of this, the report finds.

There is also a drastic difference in the scale of weather data collection infrastructure across the states. Kerala, for instance, has one

AWS every 87 sq km but Chhattisgarh has one every 2,703 sq km. Assam has one AWS per 472 sq km, Karnataka one per 167 sq km, Odisha one per 281 sq km, Bihar one per 313 sq km and Uttar Pradesh one every 1,005 sq km.

ARGs face a similar problem. Of the 8,000-odd ARGs in the country, 6,400 are in Karnataka alone.

This apart, IMD does not provide micro-level advisories. Its default scale for forecasts is the district level. This is insufficient with each district spanning hundreds to thousands of sq km, and significant weather variations found within short distances. No standard protocol for AWSs on data collection worsens the situation. “Only a part of the AWS network collects soil and agriculture-specific data,” says the

CSE report. “This lack of standardisation arises from the fact that the recent and ongoing expansion in the AWS network has proceeded with minimal coordination and oversight,” it says. A large number of AWSs and ARGs were set up in “project mode”, compromising on quality. The link between expansion and larger weather data

WATER PURIFICATION

WATER RECYCLING

WATER MANAGEMENT

WATER ECONOMICS