AI-Based Weather Forecasting Program for Farmers

The Ministry of Agriculture and Farmers’ Welfare (MoAFW) launched an AI-based weather forecasting program that delivered monsoon predictions via SMS to 3.8 crore farmers across 13 states.

About AI-Based Weather Forecasting Program

• Launched by: Ministry of Agriculture & Farmers’ Welfare (MoAFW).
• Platform Used: m-Kisan SMS portal.
• Coverage: Forecasts sent to 3.8 crore farmers across 13 states.
• Timeline: Weekly updates during monsoon; forecasts up to 4 weeks in advance.
• Notable Achievement: Successfully predicted a 20-day pause in rainfall.
• AI Models Used:
• • Google’s Neural General Circulation Model (NeuralGCM)
  • European Centre for Medium-Range Weather Forecasts’ Artificial Intelligence Forecasting System (AIFS)

Significance for Farmers

• Rain-Dependent Agriculture: More than 50% of India’s net sown area is rainfed; Kharif farming highly monsoon-dependent.
• Advance Planning: Early & localized forecasts help in:
  • Crop choice & sowing time
  • Irrigation scheduling
  • Fertilizer & pesticide application
  • Reducing losses from erratic rains
• Climate Adaptation: Increasing weather variability due to climate change heightens the need for accurate predictions.

Challenges with AI-based Forecasting

• Data Quality & Availability: 
  • AI systems require long-term, clean, high-resolution datasets; 
  • In India, especially in remote and rural regions, weather and soil data often have gaps, inconsistencies, or poor quality → reducing forecast accuracy.
• Digital Divide: 
  • Many small and marginal farmers lack access to smartphones, internet, or may face language and literacy barriers.
  • Forecasts must be simplified into farmer-friendly advisories.
  • This reduces the last-mile impact of forecasts.
• Forecast Accuracy Limits:
  • Weather is inherently chaotic; long-range forecasts may still be prone to errors, leading to loss of trust among farmers if not communicated carefully.
• Infrastructure Gaps: 
  • Weak telecom networks in rural areas hinder delivery.
  • Limited local computing power means many weather centres rely on global agencies’ AI outputs.
• Black Box Problem:
  • Many AI models are complex and operate like a “black box”, making it hard to explain forecasts or justify errors to policymakers and farmers.

Other AI based Forecasting Initiatives

• IMD – AI Integration in Forecasting
  • India Meteorological Department (IMD) is experimenting with AI/ML to improve medium- and long-range forecasts.
  • Uses Ensemble Prediction System (EPS) enhanced with AI.
• Bharat Forecasting System (BFS)
  • Launched in May 2025 by the Ministry of Earth Sciences.
  • BFS is a high-resolution (6 km grid) numerical weather prediction system developed by IITM Pune to deliver more accurate, localized forecasts.

DOWNLOAD PDF