INVESTIGATING COARSE-GRAINED GLACIO-FLUVIAL OUTWASH WITH GROUND PENETRATING RADAR (GPR), SUSQUEHANNA VALLEY, HARPURSVILLE, NY

The relationship between Ground Penetrating Radar (GPR) and sedimentary facies was examined in coarse-grained fluvio-glacial outwash exposed in two quarries in the Susquehanna River Valley between Harpursville and Windsor, NY. Knowledge on the ability and limitations of GPR to monitor sedimentary structures is a prerequisite for successful GPR investigations and is relevant to e.g. mining, archeology and modeling of groundwater flow. Common Offset Profiles were taken with 110, 225, and 450 MHz antennae along the top of quarry faces, and processed with Seismic Unix software modified for application at the scale of GPR-reflections. Two-way travel time was converted to depth through the use of Common Midpoint Profiles (CMP's) and Normal Move-out (NMO) Corrections, and spectral analyses were used to visualize the effect of processing steps on the data. A systematic sensitivity analysis of both collection and processing parameters indicates that the quality of the subsurface reflections is most sensitive to step size and antennae orientation, whereas a combination of bandpass filtering, muting of direct waves, and Automatic Gain Control yielded the best visualization, therefore facilitating interpretation of the profiles. However, the low residual water content of the deposit and the abundance of large clasts, or point scatterers, limits visualization of GPR reflections to such a degree that only large scale sedimentary features and large scale sedimentary stratification could be identified with certainty. The investigation indicates that CMP's and NMO corrections are needed to determine the possible variation in radar wave velocity, hence depth, and, the success of GPR investigations within gravel to cobble sized outwash sediments strongly depends on water content of the deposit, and the amount and size of point scatterers. The success of GPR to image large and medium scales of stratification allows for the method to be used in projects aimed at determination of scales of stratification in 3D. More knowledge on the effectiveness of collection and processing methodology is needed for research in areas where limitations exist.