The Western Ghats account for nearly 60% of India’s reported landslides, with the majority triggered by intense and prolonged rainfall | Photo credit: K. MURALI KUMAR
Researchers at the National Institute of Technology Karnataka (NITK), Surathkal, have developed an integrated landslide early warning system designed specifically for the Western Ghats, one of India’s most landslide-prone regions.
The system, called Slope Vulnerability and LandSlide Assessment (SVALSA), combines rainfall analysis, real-time monitoring of soil behavior and surface movement, and machine learning to provide reliable landslide warnings while reducing false alarms.
Why existing warnings often fail
The Western Ghats account for nearly 60% of India’s reported landslides, with the majority triggered by intense and prolonged rains. Recent disasters, including the July 2024 Wayanad landslide, have highlighted the limitations of existing warning systems and the need for more accurate site-specific warnings.
Currently, landslide warnings in India are largely based on rainfall thresholds, where alerts are issued when rainfall exceeds a certain threshold of intensity or duration. These systems, while useful, often fail to explain what is happening inside the slope itself. As a result, they can generate frequent false alarms or miss impending failures when soil conditions deteriorate even with moderate rainfall.
Except for rain only warnings
The SVALSA framework addresses this gap by going beyond just rain warnings. It integrates hydrologic data, soil strength behavior, and visible surface deformation into a single decision system that reflects how slopes actually fail in the Western Ghats. More than 90% of landslides in the region occur in residual soils composed of weathered rock, where changes in moisture content and soil suction play a critical role in slope stability.
The SVALSA device is currently the subject of a patent application. The research was developed by Varun Menon under the supervision of Sreevals Kolathayar with financial support from the Ministry of Science and Technology (DST), IMPRINT (Impacting Research Innovation and Technology) and the National Technical Textiles Mission (NTTM) under the Ministry of Textiles.
Three-stage warning system
The system works through a three-phase warning mechanism implemented as a Python-based algorithm on a compact processor unit.
In the first phase, rainfall data from government records and past landslides are analyzed using a machine learning method called K-Nearest Neighbor (KNN). The model compares current rainfall with past landslide-triggering events and filters out low-risk situations, reducing unnecessary alerts. Tests have shown that the method is highly accurate.
If precipitation conditions appear risky, the second stage assesses soil stability using a modified version of the simplified Bishop method that accounts for soil moisture and suction based on unsaturated soil mechanics. Laboratory tests have confirmed that slope stability decreases as the soil absorbs more water.
The final stage monitors surface motion using particle image velocimetry (PIV) image analysis. A sudden increase in ground motion has been found to be a reliable warning sign of an impending landslide, often before visible failure occurs.
According to the researchers, using all three indicators together makes the warning system far more reliable.
The SVALSA framework also includes a low-power, deployable monitoring device that integrates rain sensors, soil moisture probes, imaging units, and a compact processor capable of generating real-time alerts and remotely communicating with authorities.
Where it can be used
Researchers say the framework is particularly suitable for hill roads, highways, railway cuttings, settlements located on steep slopes and critical infrastructure corridors across the Western Ghats. Its adoption could improve disaster preparedness, promote timely evacuations, and significantly reduce loss of life and property.
Published – 28 Dec 2025 19:56 IST
