Physical properties of the firn layer without coring: Optimising field procedures for seismic refraction data acquisition on glaciers

Abstract:

An accurate understanding of the Antarctic firn layer is of growing importance for ice sheet modelling and remote sensing techniques, which define today’s mass balance estimations. Direct measurements of firn properties however are sparse, as the acquisition of firn cores requires a substantial amount of labor and logistical planning. Acquiring firn properties by utilising data from the passage of diving waves, recorded using seismic refraction field procedures, can represent a cheaper and quicker alternative. This is commonly done using an established two-step process: (i) determine seismic velocity and thickness using the 1D velocity-depth inversion proposed by Wiechert-Herglotz (WH), and subsequently (ii) determine firn density using an empirical relationship. The methods’ accuracy and performance under varying climatological conditions is however not well understood, which severely limits its informative value. Here, we investigate how well the method performs in different regions and for varying firn firn thicknesses and geophone layouts. To achieve this we make use of 19 datasets, contributed by British Antarctic Survey, Swansea University, Aberystwyth University and Alfred-Wegender Institute, that provide both direct firn density measurements and seismic surveys in close proximity. The data is complemented by 26 runs of a 2D finite difference P-wave model to analyse systematically the performance of WH for varying firn thicknesses and survey arrangements. We find that the WH method holds true for favourable measurement conditions, which depend mainly on there being an appropriate offset and the absence of summer melt. Both too short and too long offsets impact the method’s accuracy, highlighting that careful planning is necessary to achieve good results. Our results show that seismic refraction can be an efficient approach to acquire information about the firn layer. We provide recommendations such that future seismic surveys can optimise field data collection for the subsequent analysis of the firn layer.