New Delhi: Scientists at the Indian Institute of Technology (IIT) Mandi have developed a fully operational Landslide Early Warning System (LEWS) for the Indian Himalayan Region (IHR), aiming to improve disaster preparedness by providing daily forecasts of landslide risks during the monsoon season through a web-based platform, it was announced Wednesday.
The Indian Himalayan Region is among the most landslide-prone regions in the country, with changing climate patterns increasing the frequency of slope failures that result in heavy loss of life and property.
The research was led by Prof. Dericks Praise Shukla from the School of Civil and Environmental Engineering at IIT Mandi, along with research scholars Ankit Singh and Nitesh Dhiman.
The Landslide Early Warning System forecasts and monitors the probability of landslides by combining information on terrain susceptibility with real-time rainfall data. It issues location-specific warnings to help authorities and disaster management agencies take timely preventive measures.
Speaking about the initiative, Prof. Shukla said the system provides daily landslide forecasts through a web-based application from the beginning of the monsoon season, helping identify high-risk areas in advance and enabling authorities and communities to undertake timely evacuation and preparedness measures.
He said satellite-based early warning systems are among the most effective investments in disaster risk reduction as they convert scientific data into timely, actionable information. According to him, a region-wide forecasting platform has the potential to strengthen preparedness, improve emergency response and enhance coordination among disaster management agencies during periods of high landslide risk.
Unlike several existing landslide warning systems in India that are limited to smaller geographical areas, the IIT Mandi-developed LEWS covers the entire Indian Himalayan Region, making it one of the country’s most extensive operational landslide forecasting systems.
The researchers developed the system using a multi-stage methodology. They first identified nearly 26,000 historical landslides from the Geological Survey of India (GSI) database to prepare a landslide susceptibility map. Multiple landslide-triggering factors were then integrated using ensemble machine learning models.
The team also developed the Probability of Rainfall-Induced Landslides (P-RIL) model using data from NASA’s Global Landslide Catalogue and seven rainfall parameters obtained from IMERG satellite datasets. Since rainfall patterns change continuously, the model dynamically analyses rainfall recorded over the previous 15 days.
IANS
