Researchers at the Indian Institute of Technology (IIT) Mandi have developed an advanced Landslide Early Warning System (LEWS) to improve disaster preparedness in the Indian Himalayan Region. The platform generates daily landslide forecasts through an online portal and sends location-specific WhatsApp alerts, allowing authorities and residents to respond more effectively during the monsoon season.

Growing Need for Early Warning

Climate change has increased the frequency of extreme rainfall events across many parts of the world, leading to a higher occurrence of landslides. The Indian Himalayan Region is especially susceptible because prolonged and intense monsoon rainfall often destabilizes mountain slopes, causing damage to infrastructure, transportation networks, and human settlements.

To address these challenges, IIT Mandi has developed a fully operational forecasting platform capable of monitoring landslide risk across the Himalayan region every day.

Research Team

The project was led by Professor Dericks Praise Shukla from the School of Civil and Environmental Engineering at IIT Mandi, along with research scholars Ankit Singh and Nitesh Dhiman.

How the Landslide Early Warning System Works

The warning system combines long-term information about landslide-prone terrain with continuously updated rainfall observations. By analyzing both geographical conditions and recent precipitation patterns, the platform estimates the probability of landslides and identifies areas facing elevated risk.

The researchers believe satellite-based forecasting systems are valuable tools for disaster risk reduction because they convert scientific observations into practical information that supports faster and better-informed decision-making.

Coverage Across the Indian Himalayan Region

Unlike several existing landslide warning systems that focus on specific locations, the IIT Mandi platform has been designed to cover the entire Indian Himalayan Region, making it one of India's most extensive landslide forecasting initiatives.

To build the system, the research team analyzed approximately 26,000 historical landslide records from the Geological Survey of India (GSI). These records were used to prepare a detailed landslide susceptibility map by evaluating multiple terrain and environmental factors through ensemble machine learning models.

The rainfall forecasting component, known as the Probability of Rainfall-Induced Landslides (P-RIL) model, was developed using NASA's Global Landslide Catalogue together with seven rainfall parameters obtained from IMERG satellite datasets. The model continuously incorporates rainfall data from the previous fifteen days to generate updated risk assessments.

The final daily forecast combines the static susceptibility map with the dynamic rainfall model through a probabilistic approach. Risk levels are displayed using percentile-based categories to make interpretation easier for users.

Web Portal and Alert Features

The research team has also developed a Google Earth Engine (GEE)-based web portal to provide easy access to forecasts and related information.

• View landslide forecasts for the current day and the previous three days.
• Download daily forecast bulletins in PDF format.
• Receive location-specific landslide alerts through WhatsApp.

Expected Benefits

The operational Landslide Early Warning System is expected to significantly strengthen disaster preparedness across the Indian Himalayan Region. By providing timely and location-specific forecasts, the platform can support quicker emergency response, improve coordination among agencies, and help reduce casualties, infrastructure damage, and economic losses associated with rainfall-induced landslides.