Icequakes and ambient noise sources detected by a geophone array at the Kaskawulsh glacier

Monday, 14 December 2015
Poster Hall (Moscone South)
Naofumi Aso1, Victor C Tsai1, Christian Schoof2, Arran Whiteford2 and Gwenn E Flowers3, (1)California Institute of Technology, Pasadena, CA, United States, (2)University of British Columbia, Vancouver, BC, Canada, (3)Simon Fraser University, Burnaby, BC, Canada
Both sliding and meltwater drainage processes of glaciers are expected to generate seismic signals. The confluence of the North and Central arms of the Kaskawulsh glacier in the Yukon Territory is an especially attractive place to study such phenomena not only because of the confluence but also because a nearby ice-dammed lake fills and drains rapidly every summer.

We analyzed geophone data from nine stations at the Kaskawulsh glacier during the summer of 2014 to detect and locate icequakes and ambient noise sources. We first detected icequakes automatically by picking arrivals. Then we located events using differential arrival times between stations obtained precisely by cross-correlations, and also applied a double-difference relocation technique. During the 1-month observation period, we found 183 events that clustered near the medial moraine. More icequakes are observed from midnight to noon, potentially due to lower noise levels. These events are distributed on a dipping plane sub-parallel to the glacier flow direction. The depths below the surface range from 200m on the shallower side to 500m on the deeper side. This structure may correspond to the basal slope of the medial moraine and implies that these icequake signals come from either shear basal sliding or an englacial splay fault.

We also determined ambient noise source locations for each 1-hour record sequence using the same process as for the icequakes. We located 31 sequences, among which more sequences were observed in the afternoon, possibly related to melting of the glacier. Most of the ambient noise sequences were located in two vertical clusters, with each cluster potentially corresponding to a crevasse or a moulin. We interpret this ambient noise as being produced by meltwater drainage. In both analyses, we find that inter-station differential arrival times obtained by cross-correlations provide effective information to locate sliding or meltwater drainage processes.