USING PROBABILITY DENSITY FUNCTION ANALYSIS TO QUANTIFY LANDSLIDE OCCURRENCE IN ALLEGHENY COUNTY, PENNSYLVANIA

The Pennsylvania Geological Survey’s goal is to develop a quantitative assessment of landslide occurrence in Allegheny County, Pennsylvania. We performed this assessment using an inventory of approximately 600 landslide occurrences compiled by the survey. In conjunction with landslide data, various raster datasets highlighted statistical relationships between occurrence and numerical coverages. Rasters included slope, elevation, topographic position, slope aspect, vegetation canopy height, soil water storage, regolith thickness, proximity to mapped formation contacts, and fold axes.

For each raster dataset, values were extracted from two sets of locations: the locations of known landslides and randomly generated control points. The ratios of landslide occurrence functions compared to the control location functions for each raster type were used to create a series of landslide correlation indices. Ratio values significantly greater than 1 indicated higher likelihood of landslide occurrence, while index values less than 1 indicated lower likelihood. We identified the most influential conditions contributing to landslide risk by examining the resultant ratios from different raster datasets. Unsurprisingly, steep land surface slopes were the primary predictor of landslide occurrence; however, elevation, topographic position, vegetation, soil characteristics, and geologic structure also played significant roles in landslide development. Overburden thickness and hillslope direction were found to have little effect on landslide occurrence.

A comprehensive landslide susceptibility map was created by aggregating these indices. This approach represents a crucial first step in developing a methodology to quantitatively assess landslide susceptibility. Our aim is to produce highly detailed landslide risk maps that can provide valuable insights to municipal planners and engineers for informed decision-making and enhancing our understanding of landslide susceptibility in Allegheny County, Pennsylvania. The outcomes of this research have the potential to significantly improve landslide risk assessment and management strategies in Southwestern Pennsylvania.