IIT Mandi scientists develop landslide warning system capable of detecting minute shifts in ground

Shimla, Mar 5 (PTI) Researchers at the Indian Institute of Technology, Mandi, in Himachal Pradesh have developed a real-time landslide monitoring and early warning system.

Scientists said the technology was deployed in three landslide-prone areas in Mandi district and claimed that the model, based on machine learning and advanced analytics, achieved over 90 percent accuracy in landslide prediction.

The system, capable of detecting ground movements of less than one millimeter and uses sensors to transmit real-time data, has been developed in collaboration with the National Mission on Himalayan studies, IIT scientists said on Thursday.

During monsoon last year, Himachal Pradesh witnessed around 140 major landslides.

This new early warning system, which automatically activates cautionary mechanisms like hooters and blinkers and sends alerts to authorities enabling swift action and minimisation of losses and casualties, could prove to be very beneficial for the state, researchers said.

Prof Varun Dutt from IIT Mandi told PTI Videos that the system, which is installed on hills, is activated even if it detects a minute shift -- less than one millimetre -- in the ground, and sends alerts to 'warning poles' installed in vulnerable areas such as valleys or along road stretch around the hill.

It sets off a hooter and blinker, alerting people about possible landslide, besides sending alerts via SMS and an Android-based web application to the District Disaster Management Authority (DDMA) control room.

The system works in real time and provides resolution upto millimetres, as compared to satellite data which gives information with a resolution of around 20 meters, Dutt said.

He further said while satellite data helps track ground movements over a period of 14 days to two months, the new system can detect much finer changes.

The Government of India has launched the NASA-ISRO Synthetic Aperture Radar (NISAR) mission and its data will gradually become available, further improving the warnings, the researchers said.

Currently, satellite data is received once every two weeks. This data is being integrated with the system data to issue more accurate and timely warnings, they added. PTI COR BPL ARB ARB