OS41B-07
Live from the Seafloor: Seismic Signals associated with the 2015 Eruption of Axial Seamount

Thursday, 17 December 2015: 09:30
3009 (Moscone West)
William S D Wilcock, University of Washington Seattle Campus, Seattle, WA, United States, Maya Tolstoy, LDEO-Columbia Univ, Palisades, NY, United States, Charles Garcia, University of Washington Seattle Campus, Oceanography, Seattle, WA, United States, Yen Joe Tan, Columbia University of New York, Palisades, NY, United States and Felix Waldhauser, Columbia University, Lamont-Doherty Earth Observatory, Palisades, NY, United States
Abstract:
Axial seamount is the most active volcanic feature on the Juan de Fuca Ridge having previously erupted in 1998 and 2011. In 2014, the Ocean Observatories Initiative installed a multidisciplinary, cabled observatory on Axial seamount that includes a compact network of 7 seismometers spanning the southern half of the summit caldera with another seismometer located at the base of the volcano. Real-time data from the seismic network has been archived at the IRIS Data Management Center from mid-November 2014. We have implemented an automated algorithm to detect earthquakes, pick P and S wave phases, determine hypocenters and estimate magnitudes. Over 5 months to mid-April earthquake rates increased from an average of ~100 per day to >500 per day. Most of the earthquakes were small (MW ≤ 1.5) and occurred in swarms beneath the east and west walls of the caldera at depths clustering around 1.5 km. On April 23, 2015 at 1700-2100 GMT there was a sizeable swarm of earthquakes near the east wall of the caldera with most activity north of the network. The main seismic crisis, which coincided with ~2 m of deflation at the center of the caldera (Nooner et al., this meeting), commenced at 0500 GMT on April 24. For the first 2 hours epicenters were confined near the east wall north of the network. The southern limit of seismicity then migrated ~3 km south over one hour to beneath the network where very high levels of seismicity (~500 per hour) persisted until 1500 GMT. During the seismic crisis, the caldera seismometers were swamped by nearby earthquakes, including ~100 per hour with MW ≥ 2; additional analysis is required to determine if seismicity also migrated outside the caldera. In the following weeks, earthquake rates beneath the caldera decreased to <10 per day. An exciting discovery during this period was the detection of a large number of impulsive water borne signals, each comprising a train of water column multiples with arrival times consistent with sound sources on the seafloor 15-20 km to the north of the network. These events commenced at the end of the seismic crisis, peaked at a rate of >2000 per day on April 26 and continued at decreasing levels until mid-May. Recent seafloor mapping and observations have confirmed that these impulsive events mark the site of an eruption on the north rift of the volcano (Kelley et al., this meeting).