Caplan-Auerbach, J., Thelen, W.A. and Moran, S.C., 2009, An unusual cluster of low-frequency earthquakes at Mount Baker, Washington, as detected by a local broadband network: AGU Fall meeting, Abstract #V23D-2111

An unusual cluster of low-frequency earthquakes at Mount Baker, Washington, as detected by a local broadband network

A recent cluster of shallow low-frequency earthquakes on Mount Baker volcano marks one of the most seismically active periods in the volcano’s instrumented history (since 1972). Although Mount Baker, the northernmost of the U. S. Cascade volcanoes, has a history of recorded unrest (including an episode of geothermal unrest in 1975-6), it has never exhibited high levels of seismicity. Most of Baker’s seismicity has been associated with glacial earthquakes and deep long-period events. However, between June and September 2009 at least 39 low-frequency events were recorded at Mount Baker, 21 of which were located by the Pacific Northwest Seismic Network (PNSN). Locations are shallow and are scattered over a 5 x 5 km area around the southwest flank of the edifice. However, waveform similarity between many events suggests that most are located fairly close together and that the scatter apparent in PNSN locations is largely because of picking errors and a sparse network. To better constrain earthquake locations and source mechanism, a network of five broadband seismometers was deployed on Mount Baker between July and October 2009. This network greatly reduced the magnitude threshold for locatable events, with approximately three times as many earthquakes located by the local network than with the existing regional network. The additional stations also provided better depth constraints. The local network detected a larger number of events than identical temporary networks deployed in 2007 and 2008, suggesting that the increase in seismicity is real. Earthquakes located with the addition of data from the local network still locate at shallow depths beneath the southwest flank, but location uncertainty is significantly improved. We are using waveform similarity to evaluate relative event locations and investigate possible source mechanisms for the earthquakes, and are developing a more accurate velocity model that includes station elevations. This will better determine the absolute depths of the earthquakes and will also address the question of whether these events have a glacial or volcanic source.