Using Radar and Seismic Methods for the Determination of Ice Column Properties and Basal Conditions at Jakobshavn Isbræ and the NEEM Drill Site

Abstract

The development of preferred crystal orientation fabrics (COF) within the ice column can have a strong influence on the flow behavior of an ice sheet or glacier. Typically, COF information comes from ice cores. Observations of anisotropic seismic wave propagation and backscatter variation as a function of antenna orientation in GPR measurements have been proposed as methods to detect COF. For this investigation I evaluate the effectiveness of the GPR and seismic methods to detect COF by conducting a seismic and GPR experiment at the North Greenland Eemian Ice Drilling facility (NEEM) ice core location, where COF data is available. The seismic experiment was conducted 6.5 km North West of the NEEM facility and consisted of three multi-offset seismic gathers. The results of the anisotropy analysis conducted at NEEM yielded mean c-axes distributed over a conical region of I angle of 30 to 32 degrees. No internal ice reflectors were imaged. Direct COF measurements collected in the ice core are in agreement with the results from the seismic anisotropy analysis. The GPR experiment covered an area of 100 km2 and consisted of parallel, perpendicular, oblique and circular (radius: 35 m) acquisition patterns. Results show evidence for COF for the entire 100 km2 area. Furthermore, for the first time it was possible to image three different COF (random, disk and single maxima) and their respective transition zones. The interpretation of the GPR experiment showed a strong correlation with the ice core measurements. Glacier basal drag is also an important, and difficult to predict, property that influences glacier flow. For this investigation I re-processed a 10 km-long high-resolution reflection seismic line at Jakobshavn Isbræ, Greenland, using an iterative velocity determination approach for optimizing sub-glacier imaging. The resultant line imaged a sub-glacier sediment layer ranging in thickness between 35 and 200 meters. I interpret three distinct seismic facies based on the geometry of the reflectors as a basal till layer, accreted sediments and re-worked till. The basal till and accreted sediments vary in thickness between 4 and 93 meters and are thought to be water-saturated actively-deforming sub-glacier sediments. A polarity reversal observed at one location along the ice-sediment interface suggests the presence of water saturated sediments or water ponding 2-4 m thick spanning approximately 240 m across. Using information from the seismic line (bed geometry, ice thickness, till thickness) as well as information available for the area of study (ice surface elevation and ice flow velocity) we evaluate the effect of sub-glacier sediment viscosity on the basal drag using a linearly viscous model and the assumption of a deforming bed. Basal drag values estimated for the study area fall within the range of physically acceptable values. However, the analysis revealed that the assumption of a deforming bed might not be compatible for the area of study given the presence of water at the ice/bed interface.