Imaging Magma Under St. Helens

Seismic studies in the Mount St Helens area have largely used naturally occurring seismicity recorded by Pacific Northwest Seismic Network (PNSN) and Cascades Volcano Observatory (CVO) permanent stations. The bulk of the network around Mount St Helens was installed in the months following the May 18, 1980, Mount St Helens eruption, and has recorded ~25,000 M > 0 earthquakes within 50 km of Mount St Helens (~30% of all M > 0 earthquakes located by the PNSN). This data from this network has been used in many different types of seismic studies, including eruption dynamics and forecasting (e.g., Malone et al. 1981; Malone et al. 1983; Endo et al. 1990; Sabra et al. 2006; Harrington and Brodsky 2007; Moran et al. 2008; Waite et al. 2008; Matoza and Chouet ), modeling of the underlying Mount St Helens magmatic system (e.g., Scandone and Malone 1985; Lees 1992; Moran 1994; Wiemer and McNutt 1997; Musumeci et al. 2002; Waite and Moran 2009; see Fig. 3), and models of the relationship between Mount St Helens and regional tectonics (e.g., Weaver et al. 1987; Stanley et al. 1996; Giampiccolo et al. 1999; Moran et al. 1999).

To date seismic-imaging studies around Mount St Helens have produced only P-wave velocity models for the mid-to-upper crust, albeit with locally excellent resolution (~0.5 km: Lees 1992; Moran et al. 1999; Waite and Moran 2009). The PNSN/CVO network dominantly consists of short-period vertical- component sensors in the Mount St Helens area, precluding S-wave tomography as well as teleseismic-source based studies. In addition, no active-source experiments have been carried out near Mount St Helens due to complex logistics, with the exception of a 1994 east-west refraction-reflection line on the far northern edge of our study area (Parsons et al. 1998). The passive and active-source experiments in this proposal will therefore provide fundamentally new knowledge of the Mount St Helens area through models of S-wave structure throughout the crust, interface geometry and orientation, anisotropy, and high-resolution models of whole crustal structure from the active-source experiment.

Figure 3: Upper crustal seismic-velocity cross section (W-E) through Mount St Helens (Waite and Moran, 2009), highlighting the shallow low velocity anomaly directly beneath the volcano. Scale in km, no vertical exaggeration. Note the image extends only to 10 km depth.

 

 

Lithosphere-scale seismic imaging

Broadband earthquake arrays illuminate the whole crust and upper mantle, anchored by the 70 km USArray grid. For example, several travel-time tomography studies illuminate the fragmenting Juan de Fuca / Farallon slab as it descends and interacts with Yellowstone (e.g., Xue and Allen 2007; Schmandt and Humphreys 2011). Both body waves and surface waves show low velocities beneath and behind the Cascades arc at 50-100 km depths (Burdick et al. 2008; Yang et al. 2008), indicating a warm possibly melt-rich mantle source for Cascades volcanics. At a finer scale, scattered wave images (Bostock et al. 2002; Nicholson et al. 2005; Abers et al. 2009) provide evidence for a serpentine mantle wedge, which, if extrapolated along arc, would have an eastern terminus near Mount St Helens. They also image the Juan de Fuca crust subducting beneath the Coast Ranges, with indication of continuation at 25-30 degree dip at least as far east as the Arc. How these structures relate to the much steeper slab imaged below 100 km depth (Obrebski et al. 2010; Schmandt and Humphreys 2011) is unclear. There remains a substantial “resolution gap” between the 70 km USArray grid spacing, and the local studies described above.

Figures.  Receiver-function stacks of data from a USArray experiment in the Mount St. Helens area.