ICEquakes: Identification, Classification and Evaluation of cryogenic events in seismological overservatory data
Final Report Abstract
Global warming causes increased mobility of glaciers worldwide resulting in a variety of cryogenic seismic signals (e.g. ice berg calving, basal sliding and fracturing of ice). While the source mechanisms of individual cryogenic seismic signals are still discussed their potential to monitor active glacier dynamics has been recognized widely during the last decades. However, the manual detection process is only feasabile for short observation periods and in order to monitor glacier dynamics it is necessary to analyse long-term seismic records. Following this the aim of this study was to exploit the application of an automatic event detection procedure to continous seismic data recorded at the Neumayer seismological observatory in Antarctica. We used a stochastical classifer based on Hidden Markov Models for a routinely discrimination of cryogenic and tectonic seismic events and identified and classified icequakes and changes of their occurrence in space and time for a one year data record. First we identified two types of cryogenic seismic signals manually which we then detected throughout the year automatically. Manual spot-checks confirmed a very good performance of the automatic system. We found several interesting patterns. Event occurrences correlate with the fortnightly and semidiurnal variations in sea level and can be explained by bending effects of the floating ice tounge. However, not all observed pattern can be explained by tide induced bending. We observe large variations in event occurence from one day to another. Also we found a slight seasonal variation. Hence the observed pattern suggests an additional triggering factor of observed events which might be related to other factors such as wind or temperature and should be investigated in future research.
