Modeling and Assessment of Weather-Induced Landslide Activity

In this paper, we use a data-driven modeling approach to examine the predominant contributing factors of precipitation-induced landslide activity, and the impacts those landslides have on anthropogenic systems. We forecast the occurrence of slides from a combination of variables derived from remote sensing data such as weather conditions, slope, soil density, vegetation, and drainage area. Supervised learning methodologies are used to identify long term and short term weather patterns that correlate with landslide occurrence. Landslide impacts are modeled using similar techniques applied to population density, land cover, land use, and other features of the anthropogenic environment surrounding each landslide event.