Shishaldin volcano is among the most fascinating volcanoes in Alaska, at least from a seismic perspective. While other volcanoes typically exhibit a few earthquakes per day, between 2000 and 2004, Shishaldin commonly experienced hundreds to thousands of daily earthquakes, all of which took place with no other signs of volcanic unrest.
In collaboration with graduate student Tanja Petersen of the U. of Alaska Fairbanks, I have been investigating long-period earthquakes at Shishaldin. In particular, my interest has been piqued by earthquakes with an unusual waveform that Tanja has dubbed "coupled events".
 |
The figure to the left shows the different types of earthquakes commonly seen at Shishaldin. The uppermost figure is a "VT" or volcano-tectonic earthquake from a region ~10 km west of the summit. VT earthquakes are generally associated with brittle failure of rock...the same kind of thing you'd see on the San Andreas fault. The second event is an LP, or long-period earthquake. LP's are associated with fluid movement in the volcano and are often (mistakenly) assumed to indicate that an eruption is imminent. This is most decidedly not the case at Shishaldin. The bottom two quakes are coupled earthquakes. Note that each has two phases...a relatively higher frequency portion, followed some time later by a low-frequency phase that looks just like a "normal" LP. It is also important to note that although the high-frequency, or short-period (SP) phase is higher-frequency than the LP phase, it is still nowhere near as high as the VT. It does not appear to be a brittle failure event and is likely also associated with fluids in the volcano. | |
 |
The LP phase of the coupled earthquake is remarkably similar to a "normal" (that is to say, unassociated with a short-period onset), LP event. This suggests that the LP events share a common source, but in the case of a coupled earthquake, the LP phase is triggered by a short-period (SP) event. In the figure to the left, the yellow trace is a coupled event and the pink is a normal LP. In the filtered data (bottom panel) it is clear that there is no SP phase in the pink trace although both share identical LP signatures (top panel). | |
 |
We made use of the fact that the SP phase of Shishaldin coupled earthquakes are highly similar in time series to analyze long sequences of events. The figure to the right shows one day's worth of coupled-events, aligned by their SP phases. Note that while the SP phases are perfectly alighned (near 5 seconds into each file), the LP phases occur at a wide range of times. The top panel represents a stack of all of the SP signals. | |
