MoES: AI/ML-based early warning and improved weather forecasting under Mission Mausam

What happened: MoES/IMD uses AI/ML to strengthen forecasting and early warning

The Ministry of Earth Sciences (MoES) announced that India is leveraging Artificial Intelligence (AI), Machine Learning (ML), and Big Data to improve the accuracy and timeliness of weather forecasting and early warning for extreme weather events. MoES and its research and operational agencies—especially the National Centre for Medium Range Weather Forecasting (NCMRWF) and the India Meteorological Department (IMD)—are integrating AI/ML-based forecast guidance with operational data assimilation, ensemble prediction, coupled Earth System modelling, High Performance Computing (HPC), and conventional Numerical Weather Prediction (NWP) models. The AI/ML-generated forecast guidance is stated to be used by IMD to enhance forecasting skills across different spatial and temporal scales for multiple types of weather extremes.

MoES also stated that AI/ML-derived data products have been integrated into an indigenously developed GIS-based Multi-Hazard Early Warning Decision Support System, strengthening how early warnings are produced and used for multiple hazards.

Background and earlier position: conventional forecasting plus operational integration

India’s weather forecasting ecosystem uses numerical weather prediction models and assimilation of observational data to generate guidance. MoES stated that AI/ML is now being used to complement conventional Numerical Weather Prediction (NWP) rather than replace it—by accelerating forecast generation, improving accuracy, reducing systematic biases, and producing probabilistic guidance for extreme events.

The government further stated that AI/ML systems are currently mostly in experimental and evaluation stages, so a separate quantitative assessment of independent contribution to reducing loss of lives, property, and livelihoods is not yet available.