Training of Sensors for Early Warning System of Rainfall-Induced Landslides

Abstract

Landslides have been a major issue in the Himalayan region where slopes are cut and reformed for construction practices for infrastructure development, deforestation, and many other human activities. In lieu of the mitigation measure for rainfall-induced landslides to improve the factor of safety against failure, several warning techniques have been suggested. However, they are quite expensive, resulting in an only limited application for infinite slopes. In lieu of the existing conditions, early warning systems (EWS) for detecting slope failure using the sensors have been found to be handy to control the fatality of the disaster. But, the various sensors have been used for these warning systems are not unique. Hence, they need to be trained for each type of soil and other favorable conditions. For the proposed study, Micro-Electro-Mechanical Systems (MEMS) based sensors have been used to predict the slope failures under rainfall conditions at controlled laboratory scale prototype and to perform a series of flume tests in order to develop the threshold for moisture levels and movement that can trigger the slope failure.