Assessment of Sinkhole Vulnerability Indices Using a Probabilistic Approach for Cover Collapse and Solution Type Sinkholes in Florida

A vulnerability index can be defined as a quantifiable measure of conditions pertaining to some type of hazard. In areas with karst landscapes (i.e., geologic conditions leading to the dissolution of bedrock and subsequent sinkhole formation), engineers and researchers have developed several point-based indices to be used when assessing the vulnerability, a project site has to sinkhole formation based on the encountered subsurface conditions. In the state of Florida, engineers have historically used an index based on the subsurface layer thicknesses, called the raveling index (RI). In recent years, researchers have introduced more comprehensive point-based indices, the sinkhole resistance ratio (SRR) and the vulnerability geometric and raveling empirical indexing (VIG + VIR), incorporating encountered soil strengths and other empirically based data fitting techniques. This study investigates the effectiveness of these three indices against a database of sinkhole sites throughout Florida—most of which forming following either a solution type or cover-collapse mechanism. Using a simplistic probabilistic approach coupled with categorical analysis based on the characteristics and hydrogeologic conditions of each site, it was determined that all proposed indices are feasible in identifying relative sinkhole conditions in cover-collapse type sinkholes only, and may potentially not be effective for solution type. Furthermore, a "most effective" index was determined for cover collapse sinkholes, as well as recommended thresholds of each index to be used in future investigations.