India’s National Disaster Management Authority (NDMA) has highlighted landslides as a complex geological natural hazard. As a result, identifying hazardous zones under varied geometrical and geotechnical conditions has remained a complex problem that has received much attention in the classic era. This work aims to predict landslide hazard zones using spatial analysis by applying multi-criteria analysis methods. For the investigation, Ramban area of Jammu & Kashmir, India, has been considered. The locations of landslides were identified by analysis of Google earth images and field surveys. The major causative factors of landslides such as relative relief, slope, geological structure, lithology, soil thickness, hydrological condition, and land use and land cover were extracted from QGIS software tool, Google earth images, and field survey. The two major triggering factors such as rainfall and seismicity, were also included in the study. The cumulative effect of all these factors was considered for preparing the dataset that was scrutinized to produce a landslide hazard zonation map. It has become very much possible to develop intelligent models calibrated to experimental data to predict landslide zonation with the least potential errors with the advancement of machine learning-based computation technologies. In view of these observations, in this chapter: (i) the relevant spatial data are identified, processed, and analysed (ii) data is used to construct an intelligent machine learning model, namely the back-propagation neural network (BPNN), to predict the hazardous zonation. The study has been validated by comparing the landslide hazard zonation map with the actual occurrence of landslides. The outcomes of this research show that the designed AI-based model is quite promising and may be utilised successfully by practicing professionals to estimate landslide zonation with reasonable accuracy.
CITATION STYLE
Jaiswal, A., Verma, A. K., Singh, T. N., & Singh, J. (2023). Landslide Hazard Assessment Using Machine Learning and GIS. In Landslides: Detection, Predict. and Monit.: Technol. Dev. (pp. 389–399). Springer International Publishing. https://doi.org/10.1007/978-3-031-23859-8_19
Mendeley helps you to discover research relevant for your work.