Artificial intelligence has played a significant role in improving weather prediction for farmers in India, particularly with the recent successful forecast of a delayed monsoon onset in northeastern regions. This development allowed millions of smallholder farmers to adjust their planting schedules or select different crops, based on the information provided.
The AI-based weather model is a joint project between the University of California, Berkeley and the University of Chicago. It aims to provide more accurate forecasts for hundreds of millions of farmers across tropical regions where livelihoods depend on correctly timing crop planting with the arrival of monsoon rains. Climate change continues to affect these patterns, impacting nearly two-thirds of the global population living in monsoon-affected areas.
“This program harnesses the revolution in AI-based weather forecasting to predict the arrival of continuous rains, empowering farmers to plan agricultural activities with greater confidence and manage risks. We look forward to continuing to improve this effort in future years,” said Pramod Kumar Meherda, additional secretary at the Indian Ministry of Agriculture and Farmers’ Welfare.
The project involved collaboration among atmospheric scientists, AI experts, India’s Ministry of Agriculture and Farmers’ Welfare, and a nonprofit organization supporting smallholder farmers. The U.S. National Oceanic and Atmospheric Administration provided daily climate data that was crucial for making predictions.
UChicago’s Pedram Hassanzadeh worked with UC Berkeley’s William Boos to assess and utilize global AI weather models from Google and the European Centre for Medium-range Weather Forecasts (ECMWF). These models were trained on 40 years of global climate data. To tailor predictions for India, teams from both universities used 100 years of rainfall data from the India Meteorological Department.
Weekly forecasts were delivered via SMS to about 38 million farmers across 13 states in central and northeastern India—covering much of the core monsoon zone. The AI model offered up to four weeks’ advance notice about when monsoon rains would likely arrive in specific regions. According to William Boos, “Demonstrating that the long lead-time precipitation forecasts made by these AI models are of practical use in a tropical region where people live is a major step forward — no one really knew that before we did this work.”
The model predicted an unusual stall after initial rainfall hit southern India in early June—a pause lasting 20 days—which was not anticipated by other available forecasts but matched actual events.
Farmers such as Parasnath Tiwari from Madhya Pradesh benefited directly from these forecasts. Tiwari said: “Before this, I mostly relied on my own experience and local knowledge to know when the monsoon would arrive… The forecast about the arrival of the monsoon was accurate…. I have increased trust in the forecast, and I will rely on the information shared by scientists in the future.”
Boos noted that messages were tailored so that farmers could understand both what was being predicted and how certain those predictions were: “We actually gave farmers probabilistic forecasts, telling them how likely it was that monsoon rains would start in a particular week… By field-testing the SMS messages with farmers in advance, our team was able to tailor the language of the message so that they understood what was being predicted and the level of certainty of the prediction.”
Economic analysis led by Nobel Prize-winning economist Michael Kremer suggested that better predictions can benefit rural Indian farmers financially. Kremer stated: “Disseminating AI weather forecasts has an incredibly high return on investment, likely generating more than $100 for farmers for each dollar invested by the government… India is leading the way in using AI to improve people’s lives across many sectors, including agriculture.”
AI-based models like Google’s NeuralGCM and ECMWF’s AIFS were blended mathematically with historical rainfall data for improved accuracy. The result was a probabilistic model capable of providing useful 30-day lead time forecasts targeted at agriculture.
Government agencies distributed these forecasts through SMS platforms while organizations like Precision Development (PxD) helped design effective messaging strategies for smallholder farmers.
Early survey results indicated that up to 55% recalled receiving weather updates; nearly half who remembered specific monsoon onset forecasts used them to adjust their planting decisions. Many also shared these messages within their communities.
The initiative received partial funding from AIM for Scale—a program supported by entities including the Gates Foundation—that seeks scalable agricultural innovations for low- and middle-income countries. Efforts are now underway to expand similar programs globally and train meteorologists throughout developing nations on effective use of AI models.
