Paper No. 6
Presentation Time: 9:00 AM-6:00 PM

MAPPING STREAM SUBSURFACE HETEROGENEITY USING GROUND PENETRATING RADAR IN GLACIALLY DOMINATED STREAMS


HAUGLAND, J. Oliver1, MALZONE, Jonathan2 and LOWRY, Christopher S.2, (1)Geology, University at Buffalo, 411 Cooke Hall, Buffalo, NY 14260, (2)Department of Geology, University at Buffalo, 411 Cooke Hall, Buffalo, NY 14260, john.haugland@marquette.edu

In this study a ground penetrating radar (GPR) system was used in streams to investigate changes in sediment type that affect the interactions between groundwater and surface water within the hyporheic zone. The GPR system was used in order to attempt to circumvent the limitations of mapping subsurface heterogeneity from surface exposures and core data in cobble bed streams. A Sensors and Software pulse EKKO Pro GPR unit was used with frequencies ranging from 100 MHz to 500 MHz to image stream bed sediments. GPR surveys were conducted in 3 streams within the Cattaraugus Creek watershed in western New York. These streams were chosen based on their variability in stream bed sediment and potential for hyporheic exchange. Data were collected both going longitudinally as well as laterally across the stream beds. Possible sediment interfaces could be seen in Cattaraugus Creek but with limited success. The finer sediment bed proved to yield more clarity in penetration. Virtually no structure could be seen in a 3rd order section of Elton Creek, likely due to the thick cobble bed of the stream. At a 1st order portion of Elton Creek resolution was also poor due to an abundance of clay and variable saturation of stream features. The 1st order site contained numerous dry gravel bars that diminished image quality. Overall image quality worked best in streams with saturated features and an absence of large cobbles. The problems that arose in viewing the subsurface became easily identifiable and classifiable. Streams with beds of thick cobble or large clay layers proved to be less optimal conditions for GPR. Another frequent problem that came up in these data are that of multiple or false reflections. This problem became easy to identify by analyzing the time at which both reflections occur. Streams with relatively fine grain and high contrast sediments without clay or cobble hold the best possibility for future GPR study however streams with heavy cobble beds or thick clay layers do not.