Integrated GPR attribute analysis for improved thermal structure characterisation of polythermal glaciers

This paper reanalyses and reinterprets the GPR data set collected in 2018 on the Von Postbreen polythermal glacier, in the Tempelfjorden region of the Svalbard Islands. Different Ground Penetrating Radar (GPR) attributes were exploited as tools for the detection of peculiar patterns and inapparent glaciological facies in typical reflection amplitude data. Spectral attributes such as dominant frequency, instantaneous phase and sweetness were specifically calculated. In addition, some texture attributes like chaos and entropy were integrated with the previous ones, so as to enable the detection of a peculiar anomalous zone (AZ). This zone is non-ubiquitous, concentrated in specific areas of the glacier, where the ice is thicker and close to the main bedrock topographic highs. By means of thermal models available in the international literature, this AZ was related to local temperature variations, which are the consequence of increased heat transfer mechanisms due to peculiar stress conditions. The proposed approach can be applied to other glaciological GPR surveys on polythermal glaciers in order to obtain improved and more constrained data interpretation and thermal facies imaging.