Passive seismic recording of cryoseisms in Adventdalen, Svalbard

Abstract

Abstract. A series of transient seismic events were discovered in passive seismic recordings from 2-D geophone arrays deployed at a frost polygon site in Adventdalen, Svalbard. These events contain a high proportion of surface wave energy and produce high-quality dispersion images using an apparent offset re-sorting and inter-trace delay minimisation technique to locate the seismic source, followed by cross-correlation beamforming dispersion imaging. The dispersion images are highly analogous to surface wave studies of pavements and display a complex multimodal dispersion pattern. Supported by theoretical modelling based on a highly simplified arrangement of horizontal layers, we infer that a ∼3.5–4.5 m thick, stiff, high-velocity layer overlies a ∼30 m thick layer that is significantly softer and slower at our study site. Based on previous studies we link the upper layer with syngenetic ground ice formed in aeolian sediments, while the underlying layer is linked to epigenetic permafrost in marine-deltaic sediments containing unfrozen saline pore water. Comparing events from spring and autumn indicates that temporal variation can be resolved via passive seismic monitoring. The transient seismic events that we record occur during periods of rapidly changing air temperature. This correlation, along with the spatial clustering along the elevated river terrace in a known frost polygon, ice-wedge area and the high proportion of surface wave energy, constitutes the primary evidence for us to interpret these events as frost quakes, a class of cryoseism. In this study we have proved the concept of passive seismic monitoring of permafrost in Adventdalen, Svalbard.

Long term analysis of cryoseismic events and associated ground thermal stress in Adventdalen, Svalbard

“…The aim of this study was to test this hypothesis by analyzing the spatial and temporal occurrence of these events. This study is highly complementary to the previous work of Romeyn et al (2021). While the spatial and temporal wavefield sampling of the SPITS array is much coarser than the temporary array they deployed, a much longer and nearly continuous record is available.…”

“…Based on previous work further down-valley in Adventdalen, e.g. Matsuoka et al (2018) and Romeyn et al (2021), we hypothesized that the short duration events at SPITS might be dominated by frost quakes initiated by thermal contraction cracking in the vicinity of the array. The aim of this study was to test this hypothesis by analyzing the spatial and temporal occurrence of these events.…”

Case study of combined marine‐ and land‐based passive seismic surveying in front of Nordenskiöldbreen outlet glacier, Adolfbukta, Svalbard

“…Broadband fk ‐analysis can give more accurate estimates of the azimuth and apparent velocity of the waves, but again, overlapping phases and multipathing in the water layer make the waves from calving events incoherent and unsuitable for such analysis. As the OBNs also record a lot of high‐frequency noise, which seems uncorrelated across the receivers, attempts at source localization using matched field processing (similar to Romeyn et al., 2021) fail. For distant events, like earthquakes, the wavefield appears much more coherent over the OBN array and fk ‐analysis works well.…”