Testing ground-penetrating radar for resolving facies architecture changes - a radar stratigraphic and sedimentological analysis along a 30 km profile on the Karup Outwash Plain, Denmark

Knowledge of the spatial distribution of lithofacies, obtained through detailed modelling of the movement of water and solutes, may improve precise predictions of an area's vulnerability towards contaminants. This calls for information on lithofacies variability on several scales. The possibilities for upscaling lithofacies distribution from excavation scale to field scale or to an even larger scale with the use of ground-penetrating radar (GPR) is tested on one of the largest outwash plains in Denmark: the Karup Outwash plain in the western part of Jutland. The testing involves GPR surveys at 16 locations along a 30 km long profile following the ancient mean flow direction of the braided-river system, as well as combined detailed sedimentological investigations at selected locations. The GPR testing is divided into several phases of which this paper reports on the first two. In the fir st phase, GPR surveys and sedimentary investigations at gravel-pits and excavation sites are carried out at the same location with the aim of determining whether lithofacies can be related to GPR reflection sequences either directly or through facies architecture models. Studies of outcropping glaciofluvial deposits at a gravel-pit and a shallow excavation in the proximal and intermediate parts of the outwash plain, respectively, show that the sand-dominated outwash plain typically contains lithofacies types with bed thicknesses of up to 0.5 m and successions of beds of thicknesses up to 1 m. GPR surveys carried out at the same locations show that the observed lithofacies types typically have dimensions that, in a vertical direction, are the same as or less than the resolution obtained by a 200 MHz GPR system. In such sand-dominated outwash deposits, the GPR reflections may more often be caused by the bedding structures in successions of beds and of larger erosion surfaces separating different lithofacies, rather than by the lithofacies boundaries themselves. In the second phase, tests are conducted to determine whether facies architecture changes along the ancient mean flow direction in the outwash plain can be recognized using radar stratigraphic analysis. Radar stratigraphic analysis, so far limited to two GPR sections orientated parallel and perpendicular to the ancient mean flow direction at 15 locations, reveals a systematic change in the pattern of radar sequence boundaries and the occurrence of radar facies. The changes in radar sequences seem to reflect changes in lithology and sedimentary structures related to the overall fall in the depositional energy from the proximal to the distal part of the outwash plain. The results clearly indicate that radar stratigraphic analysis reveals realistic facies architecture models for the outwash plain, provided densely spaced GPR data are acquired and locally calibrated against excavations and coreholes. It is concluded that GPR can be used for upscaling lithofacies distributions in sedimentological settings like the present Karup Outwash Plain.