Patterns of Earthquakes and the Effect of Solid Earth and Ocean Load Tides at Mount St. Helens Prior to the May 18, 1980, Eruption

Document Type


Publication Date


Digital Object Identifier (DOI)



Seismographs near Mount St. Helens Volcano recorded an earthquake swarm lasting nearly 2 months prior to the catastrophic May 18, 1980, eruption. The earthquakes are divided into four classes based on station CPW (Δ = 116 km) seismogram characteristics: (1) events with Sv:P amplitude ratio >3 and dominant frequency >3 Hz; (2) events with Sv:P ratio between 1 and 3 and dominant frequency >2 Hz; (3) events similar to characteristic 2 but with a strong (probably surface wave) phase just after the S phase; and (4) events with frequencies between 1 and 2 Hz lacking a clear S phase. The seismicity pattern for each of the four classes is unique, and we assume each group of events with similar wave-forms represents a common physical process, depth, source area, or fault orientation. Maximum likelihood b values range between 0.97±0.05 and 1.81±0.09 for the four groups. We calculate solid earth stress and strain tides at the average hypocentral depth of 4 km. We also calculate stress and strain tides induced by ocean loading; their amplitudes are typically 20–40% those of the solid earth tides at the location of Mount St. Helens. A weak but significant correlation exists between the latter two classes of events and the tides for a time interval of about 5 days preceding the first onset of volcanic tremor and about 5 days thereafter. The polarity of the correlation is opposite for the two classes of events. In each case, the phase of the correlation changes systematically with time, the changes coinciding with the onset of tremor on March 31 and with a pronounced decrease in earthquake energy release rate on April 3. There are no significant correlations between the tides and the number of events or energy release of these two classes of earthquakes during any other interval between March 20 and May 18, 1980. The first two classes of events show no evidence of significant tidal correlation at any time during the study period.

Citation / Publisher Attribution

Journal of Geophysical Research, v. 89, issue B5, p. 3075-3086