CREATING A SINKHOLE GIS DATABASE: CAN ALSM DETECT SINKHOLES IN URBANIZED FLORIDA?
After filtering the data to remove vegetation and buildings, we were left with more than 100 million measurements of bare-earth x,y,z coordinates. We divided these point data into manageable sizes, imported them into ESRI ArcGIS 8, and constructed a grid with 2.133-m (7-ft) spacing using the inverse-distance weighting (IDW) technique. The IDW technique easily captures subtle discrete changes in a uniform background. We found that IDW was the best technique to use in Pinellas County, due to the flat topography punctuated by sinkhole development. Contouring the grid with a 0.304-m (1-ft) contour interval produced numerous closed-contour depressions (CCDs). These apparent sinkholes, stood out in sharp contrast to the surrounding flat terrain. In order to assess accuracy, we compared the contour map with color aerial photos within a GIS. If we confirmed that a CCD was most likely a sinkhole, we converted it to a shape file and added it to a Pinellas County sinkhole GIS database.
Urban development was a definite impediment to identifying remnant sinkholes. Buildings that were filtered out created large voids in the point data. Many swimming pools appeared as CCDs. Highly reflective surfaces produced multi-path laser returns. Anthropogenic modifications to the land surface disrupted the uniformity of the background signal. We conclude that, although ALSM is a valuable tool for locating sinkholes, in general, it is certainly best utilized in undeveloped areas.