TESTING STATISTICAL AND SEMI-AUTOMATED METHODS FOR ASSESSING AND PREDICTING KARST FEATURE DENSITY IN THE EDWARDS AQUIFER RECHARGE ZONE

Karst inventories provide information used to evaluate an area’s hydrogeologic connection to local and regional flow systems, as well as its environmental and ecological sensitivity. As urban development expands in karst regions such as the Edwards Plateau in central Texas, USA, related impacts to karstic groundwater systems are rapidly increasing. Identifying and classifying karst features during geological assessments, as mandated by the State of Texas to protect the Edwards Aquifer (EA) recharge zone, requires visual survey and interpretation methods may be biased, depending on the experience of the surveyor. Additionally, access to land in the recharge zone is limited prior to development being planned and approved. In light of this, it would be useful for planning, water conservation, and management purposes to have independent methods for estimating karst feature densities in un-surveyed regions of the recharge zone. The overarching question of this research is: can relationships be identified between predictor variables and karst feature density that allow estimation of karst vulnerability without physical surveys? Relationships between independent factors (e.g., surface geology, soil depth, and distance to the nearest water body) and karst feature density are well-documented in many regions, but the degree to which they affect karst feature density and distribution in the Edwards aquifer has not been well quantified. Utilizing the relatively undisturbed 17 km², Freeman Center (part of the EA recharge zone) of Texas State University in San Marcos, Texas, methods have been developed for a semi-automated karst detection workflow to identify sinkholes in the recharge zone of the Edwards Aquifer, and to quantify potential correlations between karst feature density and various geologic, soil, and geomorphic factors.

Interpretation of LIDAR data through a semi-automated sinkhole detection method, supported by an experimental and statistical karst feature survey design and analyzing resulting survey data, are expected to result in an assessment tool that can be used by other researchers to identify sensitive or vulnerable karst regions in the Edwards aquifer recharge zone, and to help prioritize target areas for visual surveys when time and resources are limited.