S51D-2722
Deep long-period earthquakes (DLPs) beneath Mount St. Helens

Friday, 18 December 2015
Poster Hall (Moscone South)
Jiangang Han1, John Emilio Vidale2, David A Schmidt2, Kenneth C Creager1 and Heidi Houston1, (1)University of Washington, Department of Earth and Space Sciences, Seattle, WA, United States, (2)University of Washington, Seattle, WA, United States
Abstract:
The volcanic deep long-period earthquakes (DLPs) have been observed for a long time but remain poorly understood. Hypotheses associated with magmatic process have been proposed for the mechanisms of these DLPs, including dehydration embrittlement, flow of magma and/or magmatic fluid and cooling of magma. DLPs are commonly characterized by weak signal on the waveforms, deficiency in high-frequency energy, long-duration coda and their rare occurrence. They are located at 10-35 km depth, which are the mid- to lower-crust and/or uppermost mantle.

The imaging Magma Under St Helens (iMUSH) experiment began in late June 2014, and since then the broadband seismometers have recorded six DLPs, two of which are also captured by dense array of Nodal stations. We use the iMUSH data and seismic data from nearby network stations to study the DLPs beneath St. Helens. Catalog DLPs are taken as templates to search for repeating events that might be too small to be detected otherwise. So far, we have searched for cross-station correlation detections for four template DLPs for the period 2007 to 2015. Three of the four seems to be isolated one-offs, while the fourth has at least 56 repetitions, three times more than were already in the catalog, and hints of many more. Many of the DLPs have several bursts within tens of seconds or several minutes.

Overall the DLPs show an episodic activity with a period of roughly sixteen months. Several, but not all, episodes are temporally correlated with the subduction zone tremor activity west of St. Helens (Figure 1), which we are still investigating. We are locating these detections, and preliminary results suggest concentrated loci within a distance of one or two kilometers.

We will conduct correlations between all detections, search farther back in time, and search with other templates as well, to better characterize their timeline and fine-scale geometry and analyze the waveforms to understand their physical mechanisms and the complicated pattern of arrivals observed to traverse the midcrust directly under the volcano.