Abstract
Using geospatial technologies to assess geological hazard risk has been proved feasible and effective. In this paper, a geospatial information quantity model is proposed to assess landslide risk, which includes nine triggering factors: slope, aspect, cumulative catchment area, formation lithology, seismic intensity, distances to water, precipitation, vegetation, and land use/land cover type, in which the last three triggers are dynamic ones and need to be extracted from up-to-date remote sensing images. These triggering factors are then taken as geospatial information quantities and used to construct an information quantity-based model to assess and predict the landslides in Fuling District, Chongqing City, China, resulting in a risk distribution map. Finally, ROC curve is used to validate the model. With the AUC of success-rate ROC of 0.839 and the AUC of prediction-rate ROC of 0.807, the model is proved reliable to interpret and predict the landslide occurrences in the study area.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call