The ‘ICCS Nowcasting System’ offers real-time, hyper-local rainfall forecasts aimed at improving disaster preparedness and agricultural planning
The Institute for Climate Change Studies (ICCS), an autonomous body under the Kerala government based in Kottayam, has developed an artificial intelligence-powered rainfall forecasting platform named ‘Eye in the Sky’ ICCS Nowcasting System. The new system leverages machine learning to provide real-time, hyper-local weather updates, predicting rainfall up to two hours in advance.
The innovation is seen as a significant step in applying AI for climate resilience and disaster management, particularly in monsoon-prone regions like Kerala. The system was unveiled earlier this month and is currently undergoing beta testing in selected districts.
AI Brings Micro-Forecasting to Kerala
Unlike traditional weather forecasting models, the ICCS Nowcasting System uses a neural network-based AI model trained on years of meteorological data sourced from Doppler radars, satellite imagery, and ground-based weather stations. It can deliver rainfall predictions for regions as specific as 12-square-kilometre zones—providing crucial lead time for civic agencies and citizens alike.
The tool works by analyzing atmospheric patterns in real time and identifying the probability and intensity of precipitation. With Kerala’s unique topography and frequent shifts in microclimate, such detailed granularity is expected to fill a critical gap in weather intelligence.
ICCS officials said the model is capable of learning from each weather cycle to improve its predictions over time, making it more reliable with continued use. The AI component specifically helps in recognizing non-linear weather patterns and anomalies often missed by conventional statistical forecasting methods.
Mobile Alerts and Dashboard Integration Underway
The Eye in the Sky platform is also being developed into a mobile application and a web-based dashboard to make rainfall alerts accessible to the public and authorities. The forecast interface will display updates every 10 minutes, with visual markers indicating varying levels of rainfall intensity.
ICCS is working closely with the Kerala State Disaster Management Authority (KSDMA) to implement the system in flood-prone districts. Alerts will follow a colour-coded format—green for normal, yellow for moderate rainfall, orange for heavy showers, and red for severe rainfall—to aid in decision-making.
Pilot versions of the alert system are being tested in Thiruvananthapuram, Kottayam, and Ernakulam. Full deployment across the state is expected in the coming months, pending feedback from the pilot phase.
Wide-Ranging Applications for Public and Agriculture Sectors
The ICCS Nowcasting System is expected to play a vital role in disaster preparedness, especially during the monsoon and cyclone seasons. Real-time alerts can assist emergency services in deploying resources more effectively, while helping local governments to proactively close schools, reroute traffic, or initiate evacuations in vulnerable areas.
Farmers, too, stand to benefit from this tool. Hyper-local forecasts will enable them to time irrigation cycles, pesticide spraying, and harvesting activities more precisely, minimizing crop losses due to unpredictable rainfall.
An ICCS meteorologist noted that Kerala’s 2018 and 2019 floods highlighted the urgent need for more agile and predictive weather systems. “This tool is not meant to replace the India Meteorological Department but to complement it with short-term, localized alerts that the public and field officials can act on immediately,” the official said.
Commitment to Open Data and Ethical AI
True to ICCS’s ethos of public service and scientific collaboration, the core model and forecast outputs of the Nowcasting System are planned to be made open source in the near future. The organization believes in fostering innovation by allowing external researchers, universities, and developers to adapt the model for their own regions or use cases.
The platform is built to be privacy-conscious. Since it uses only publicly available atmospheric and geospatial data and does not collect any personal information, it aligns with India’s digital governance frameworks.
Strategic Alignment with National Climate Goals
The launch of the Eye in the Sky tool aligns with India’s broader push to use emerging technologies like AI for sustainable development. It also contributes to the National Disaster Management Plan’s emphasis on early warning systems as critical infrastructure.
This development follows similar moves by the India Meteorological Department, which is increasingly adopting AI for cyclone tracking and urban weather modeling. However, ICCS’s model stands out for its focus on ultra-local precision and practical deployment across community-based alert systems.
What’s Next
The ICCS Nowcasting System will continue to undergo refinements based on pilot feedback. Once fully rolled out, it could serve as a benchmark for other Indian states dealing with climate volatility. The organization is also exploring the integration of the tool with Kerala’s existing flood alert networks and public warning systems.
In the long run, ‘Eye in the Sky’ has the potential to be scaled across other parts of the country and possibly exported as a climate tech solution for other monsoon-affected regions globally.