India Unveils AI-Powered Monsoon & Rainfall Forecasts for Hyper-Local, Impact-Based Weather Services

• AI-enabled Systems Introduced by IMD to Provide Hyper-local Weather Forecasts. 
• Advanced Forecast Systems to Provide Localised Weather Information Up to 10 Days in Advance. 
• Government Introduces AI-enabled Monsoon Forecasting Platform for 16 States and Over 3,000 Sub-districts
• Union Minister Dr Jitendra Singh Launches AI-based Monsoon Advance Forecast System and 1-km Resolution Rainfall Forecast for Uttar Pradesh
• Dr Jitendra Singh Says IMD Has Become an Essential Part of India’s Everyday Governance and Public Decision-making

India has launched two landmark AI-enabled weather forecasting systems—an AI-driven monsoon advance forecast and a high-resolution rainfall model for Uttar Pradesh—marking a decisive shift towards hyper-local, impact-based climate services designed to aid farmers, disaster managers, and policymakers.

The systems have been developed jointly by the India Meteorological Department (IMD), Indian Institute of Tropical Meteorology (IITM), Pune, and National Centre for Medium Range Weather Forecasting (NCMRWF).

The two systems are –

1. AI-enabled Monsoon Advance Forecast
   • Provides probabilistic forecasts every Wednesday up to 4 weeks in advance.
   • Covers 16 states and 3,000+ sub-districts.
   • Designed to support farmers’ sowing, irrigation, crop planning decisions.
2. High Spatial Resolution Rainfall Forecast (Pilot in Uttar Pradesh)
   • Generates 1-km resolution rainfall forecasts up to 10 days ahead.
   • Uses AI-driven downscaling techniques integrating data from Doppler radars, AWS, ARGs, and satellites.
   • Expected to expand to other states as infrastructure grows.

Benefits Across Sectors

• Agriculture: Farmers gain precise, localized forecasts for sowing, irrigation, crop protection, and harvest planning.
• Disaster Management: Improved early warnings for floods, cyclones, and extreme rainfall events.
• Urban Planning: Helps cities prepare for drainage, infrastructure resilience, and water resource management.
• Renewable Energy: Supports solar and wind energy forecasting, stabilizing grid operations.

Technological Advancements

• Expansion of Doppler Radars: From 16–17 a decade ago to ~50 today, with another 50 planned under Mission Mausam.
• Forecast Accuracy Gains: Severe weather forecast accuracy improved by 40% in the past decade; cyclone track predictions improved by 30–35% in the last five years.
• Digital Dissemination: Forecasts shared via mobile apps, SMS, WhatsApp, Kisan portals, TV, and radio for last-mile connectivity.

Comparison of New Systems

System	Coverage	Forecast Horizon	Resolution	Primary Use
AI Monsoon Advance Forecast	16 states, 3,000+ sub-districts	Up to 4 weeks	District/block level	Agriculture planning, disaster preparedness
Rainfall Forecast (UP Pilot)	Uttar Pradesh (pilot)	Up to 10 days	1 km	Hyper-local rainfall prediction, urban planning

Strategic Importance

• Strengthens climate resilience and citizen-centric governance.
• Aligns with PM Modi’s modernization drive under Mission Mausam.
• Positions IMD as a decision-support system for governance, agriculture, and infrastructure.

The monsoon advance forecasting system would now provide granular forecasts on monsoon progression at district-level scales, while the Uttar Pradesh pilot project demonstrates the capability of generating operational rainfall forecasts at 1-km resolution using dense observational networks and AI techniques.

Secretary, Ministry of Earth Sciences, Dr. M. Ravichandran said that similar services would gradually be expanded to other parts of the country as observational infrastructure continues to grow.

Dr. Jitendra Singh said the newly launched forecasting products represent another important step towards building a climate-resilient, digitally empowered and citizen-centric weather service system for the country, where scientific advancements directly contribute to societal and economic benefits.

Scientists Surprised As India Saw Hottest February Since 1877, April-May Will Be Severe

This year, the month of February was the hottest in India since 1877 and the average maximum temperature was recorded at 29.54 C. The Meteorological Department provided this information, linking it to Global Warming and climate change. Scientists are also surprised by this change. India Meteorological Department (IMD) said that temperatures in most parts of the country are expected to be higher than the normal, while the southern peninsula and parts of Maharashtra are likely to survive harsh weather conditions.

SC Bhan, Scientist & Head of IMD's Hydromet and Agromate Advisory Services, said in a virtual press conference that Lu's chances are low in March, but most parts of the country m experience extreme weather conditions in April and May.

Probability forecast of season March to May 2023 (Image - mausam.imd.gov.in) 

In response to a question linking the developments to Global Worming, Bhan told reporters that the monthly average maximum temperature was the highest in February this year since 1877.

Asked whether high temperatures are a sign of climate change, Bhan said that the whole world is in a period of global warming. We are living in a hot world!

This year, the average maximum temperature in Delhi was recorded at 27.7 ° C in February and this is the highest temperature in February this year, for the third time in the last 63 years.

"During the upcoming hot weather season [March to May (MAM)] , above normal maximum temperatures are likely over most parts of northeast India, east and central India and some parts of northwest India. Normal to below normal maximum temperatures are most likely over remaining parts of the country, " said a recent press release by IMD. 

The figure illustrates the most likely categories as well as their probabilities. The dotted less rainfall and the whitea rea show n in the map climatologically receives very shaded areas within the land areas represent climatological probabilities. ( * Tercile categories have equal climatological probabilities, of 33.33% each ).