Enhancement of Ground Penetrating Radar Data through Signal Processing

Recently collected ground penetrating radar (GPR) data from a former lumber mill waste site near Boise, Idaho have proven to be a useful reconnaissance aid for delineating possible areas of fill from surrounding fluvial sediments. Several lines of data were collected at coincident stations using antennas with central frequencies of 50, 100, 200 MHz. Some ambiguities exist in the interpretation of the original 200 MHz data due to lack of a distinct fill signature in the data processed with standard seismic processing techniques. We present here an effort to enhance the original GPR data through air wave removal, bandwidth enhancement, and cross-correlation analyses. In general, these analyses appear to improve the usefulness of the data and in some environments may become standard techniques for future GPR data collection and processing.