Sinkhole Hazard Mapping Using Frequency Ratio and Logistic Regression Models for Central Florida

摘要

Sinkholes are among the most frequent hydrogeological hazards in the area of Central Florida. It is important to identify the sinkhole-prone areas in order to minimize the loss and damage occurred due to such events. The main objective of this study is to produce a sinkhole hazard map, focusing on Central Florida, based on quantitative methods within a GIS environment. Multiple sinkhole hazard models including frequency ratio (FR) and logistic regression (LR) were constructed to assess the multivariate statistical relationship between the sinkhole occurrence and sinkhole-affecting factors. Seven hydrogeological factors that potentially contribute to the sinkhole development in the region were identified and analyzed. The sinkhole hazard index (SHI) and the probability of the sinkhole occurrence were calculated using the developed hazard models, and then used to produce the sinkhole hazard map for the Central Florida region. Both sinkhole hazard maps show a strong correlation with the reported sinkholes.