AN IMPROVED DEM PROCESSING MODEL TO MORE ACCURATELY LOCATE SINKHOLES IN SOUTHWESTERN OHIO

The Ohio Geological Survey has developed a new ArcGIS model to delineate karst terrain. The new model improves upon previous digital elevation model (DEM) processing methods by using an automated series of geoprocessing tools to extract the enclosed depressions from the DEM. This processing reveals that Ohio’s southwestern karst region has significantly more sinkholes than the northern region; these results were unexpected because karst data previously compiled by the Ohio Geological Survey indicated that the northern karst region had the most sinkholes. The new model also shows that sinkholes in the southwestern region appear to be smaller in extent and more clustered compared to other regions. With completion of the majority of central and northern Ohio karst regions, and the first four 7.5-minute quadrangles of southwestern Ohio, DEM processing and field mapping has begun on two more quadrangles of the southwestern region. The new model applied to these two quadrangles allows shallow depressions up to 1 foot in depth to be processed. Previous model versions showed only depressions 1 foot in depth or deeper. Such refinement allows better detection of very small features—and more complete delineation of sinkhole extents—although it also adds many shallow depressions that are not karst related. Furthermore, additional steps have been added to the model to automatically remove very shallow isolated depressions. Field verification helps to identify these and other false positives, including old house foundations, broken field tiles, uprooted trees, steep-sided stream banks, and breached ponds. Field verification also reveals new karst features, such as springs, caves, small depressions, and recently formed features, which do not appear on the model results or the DEM.