MULTI-OFFSET ACQUISITION AND PROCESSING OF GPR DATA

Most GPR surveys are acquired with a constant transmitter-receiver offset and often investigators apply little or no processing to generate a final image. This mode of operation can provide useful information, but does not take full advantage of the information carried by the GPR signal. It is possible to significantly increase the amount of available subsurface information by acquiring data in continuous multi-offset (CMO) mode. With this method, I acquire several traces with varying source-receiver separations at each point along the survey, as opposed to a single trace at each point in a constant-offset survey. Improved subsurface characterization arises from three key features: 1) through processes such as stacking and velocity filtering, coherent and random noise is significantly attenuated resulting in subsurface images that are easier to interpret, 2) CMO data enable measurement of vertically and laterally continuous velocity distributions, and 3) CMO data enable observation of reflected wave behaviour (ie variations in amplitude and spectrum) at a common reflection point for various travel paths through the subsurface - quantification of these variations can be a valuable tool in material property estimation. Although there are a few examples in the literature, investigators rarely acquire CMO GPR data. This is, in large part, due to the fact that CMO acquisition with a single channel system is labor intensive and time consuming. At present, no multi-channel GPR systems designed for CMO acquisition are commercially available. Over the past 8 years I have designed, conducted, and processed numerous 2D and 3D CMO GPR surveys using a single channel GPR system. The majority of the work was for hydrogeologic applications with particular emphasis on contaminated site characterization. I have developed field procedures that enable a three man crew to acquire CMO GPR data at a rate comparable to a similar scale multi-channel seismic reflection survey. In data processing, I focus primarily on pre-stack imaging and analysis including reflection tomography, pre-stack migration, and amplitude vs offset analysis. Through several field and laboratory examples, I demonstrate that CMO acquisition and processing significantly improves image quality and measurement of subsurface variability.