The India Meteorological Department (IMD) is rolling out an artificial-intelligence-powered system to predict the arrival of the southwest monsoon at the sub-district level, an upgrade from the national and state-level forecasting that has kept farmers unsure about when to plant their crops.
Meanwhile, IMD will make the all-important announcement about the onset of the southwest monsoon on Thursday.
IMD will provide block-level monsoon onset forecasts at least 10 days in advance, covering more than 3,000 sub-districts across 16 states, including Gujarat, Maharashtra, Madhya Pradesh, and Uttar Pradesh. The system draws on AI-based models, extended-range prediction tools, and statistical techniques to generate probabilistic forecasts and will disseminate alerts through mobile applications, SMS, and local agricultural extension networks.
“The AI-enabled monsoon advance forecasting system will provide probabilistic forecasts of monsoon progression,” said Jitendra Singh, minister of Earth Sciences.
Singh said the system combines AI-based forecasting models, extended-range prediction systems, and statistical techniques to provide operationally useful forecasts for agricultural planning and preparedness.
IMD, Indian Institute of Tropical Meteorology (IITM), Pune, and National Centre for Medium Range Weather Forecasting (NCMRWF) have jointly developed the systems.
IMD has already forecast a below-normal monsoon this year, about 92% of the long-period average, raising concerns for the farm sector, which remains mostly rain-dependent. Only about 55% of India’s net sown area has access to irrigation, leaving nearly half its agricultural land entirely dependent on seasonal rainfall that accounts for more than 70% of the country’s annual precipitation.
“The block-level forecasting system is expected to offer more localised and precise information, especially for rain-fed agriculture regions where the timing of monsoon arrival plays a crucial role in crop productivity,” said Mrutyunjay Mohapatra, director general, IMD.
“Farmers often face uncertainty as far rainfall is concerned. If forecasts are available at the block level, they can take informed decisions regarding land preparation and sowing time and irrigation scheduling,” said Puneet Singh, a farmer from Ambala district in Haryana.
Timing is important
The government is also testing a rainfall forecasting system for Uttar Pradesh using , with plans for a national rollout as infrastructure expands. Together, the two initiatives mark a shift in how India approaches agricultural weather services.
Meanwhile, conditions are becoming favourable for the onset of the southwest monsoon, likely over parts of the South Bay of Bengal, the Andaman Sea and the Andaman & Nicobar Islands towards the end of this week. Monsoon reaches the Andaman & Nicobar Islands near the third week of May, after which it advances to the mainland and hits the Kerala coast around 1 June. Last year, the southwest monsoon arrived early in Kerala on 24 May, eight days ahead of its usual onset on 1 June.
Officials from the agriculture ministry said the initiative would help farmers determine the optimal sowing window, select crop varieties, and apply inputs such as fertilizers and schedule irrigation. The move assumes significance as weather variability and extreme increasingly impact agricultural operations across the country.
“Farmers often face uncertainty over the exact onset of monsoon rains in their areas, leading to delayed sowing or crop losses due to premature planting. Block-level forecasts are expected to bridge this gap by providing area-specific weather intelligence,” said a senior agriculture ministry official.
The enhanced forecasting system is likely to be integrated with the government’s digital agriculture initiatives and agrometeorological advisory services. may receive the information through mobile applications, SMS alerts and local extension networks.
Singh said the initiative in Uttar Pradesh would be particularly useful for agriculture, water resources, renewable energy, urban planning, disaster management, and infrastructure. He said farmers would now be able to take more informed decisions on sowing, irrigation, crop protection, and harvest planning, with far greater local precision.
Dr. Ravichandran, secretary, ministry of earth sciences, informed that the Uttar Pradesh pilot project demonstrates the capability of generating operational rainfall forecasts at 1-km resolution using dense observational networks and AI techniques. He added that similar services would gradually be expanded to other parts of the country as observational infrastructure continues to grow.