U.S. Geological Survey
Volcano Hazards Program
345 Middlefield Rd., Menlo Park, CA 94025
The year began with a series of small earthquake swarms located in the south moat of the caldera between the Convict Lake moraine and the southeastern margin of the resurgent dome. Swarm activity in this area gradually increased in intensity through February and March culminating in late March and early April with the most energetic earthquake swarm within the caldera since January 1983. This March 29- April 10 swarm included some 24 earthquakes of magnitude M=3 or greater. Three of these earthquakes exceeded magnitude 4. M=4.0 earthquakes occurred on March 30th at 7:22 AM and 3:15 PM (PST), and the largest earthquake of the sequence, a M=4.2 event, occurred at 5:50 PM on April 1st. Altogether this swarm included over 1,600 earthquakes large enough to be detected and located (M>0.5), and it had a cumulative seismic moment of roughly 5x1022 dyne-cm, or the equivalent of a single M=4.8 earthquake. The deformation monitoring instruments showed no indication of unusual ground deformation associated with this swarm.
Earthquake activity within the caldera gradually slowed following the March 29- April 10 swarm through the remainder of April and May. Activity increased again in June, however, with four bursts of swarm activity at 5-day intervals from June 9 through June 25. Swarms on June 9, 14, and 25 were located near the southwest margin of the resurgent dome (near the Highway 203-395 junction); the swarm on June 19/20 was located at the southeast margin of the resurgent dome (about 2 km north of the airport). The largest earthquakes in these swarms were a M=2.6 event on June 9, M=3.2 and M=3.5 events on June 14, and a M=3.3 event on June 19, respectively. The long-base tiltmeter, which is centered just one km southeast of the June 19/20 swarm, showed an 0.3-microradian tilt down to the northwest coincident with the June 19/20 swarm.
Small earthquake swarms on July 30th, August 7th , and 9th , all located near the southwest margin of the resurgent dome, were the last swarms to occur within the caldera for the remainder of 1996. Thus, in contrast to the first half of the year, the caldera was relatively quiet during the last half of 1996 producing only occasional small earthquakes (all less than M=3).
Occasional long-period (LP) volcanic earthquakes continue to occur at depths between 10 to 20 km beneath the Devils Postpile area just southwest of Mammoth Mountain. These events have become more frequent with time since their onset during the 1989 earthquake swarm beneath Mammoth Mountain. Minor volcano-tectonic earthquake activity in the shallow crust (depths less than 10 km) beneath Mammoth Mountain, however, has shown no significant changes in either rate or spatial distribution over the past seven years.
As defined by the two-color geodimeter measurements, the long-term uplift and extensional deformation of the resurgent dome has gradually slowed through the last half of 1996. This decrease in both the deformation rate of the resurgent dome and earthquake activity within the caldera over the last six months is reminiscent of the relative seismic quiescence and low deformation rates within the caldera that persisted from 1984 through mid-1989. The continuous deformation monitoring instruments showed no significant changes during the year (with the possible exception of the 0.3 microradian tilt accompanying the June 19/20 earthquake swarm note above). Note that the pronounced dip in the POPA dilatometer in mid-May is related to heavy runoff associated with a warm rain.
The dominant variations in carbon dioxide (CO2 ) soil-gas concentrations in the tree-kill areas around Mammoth Mountain reflect the blanketing effect of snow cover during the winter months. Results from the continuous CO2 monitors at Horseshoe Lake show a pronounced increase in CO2 concentrations in early February that persisted through the end of April. Concentrations gradually return to minimum values (roughly 10% CO2 at HS1, for example) by mid summer. The areas showing evidence of high CO2 soil-gas concentrations around the flank of Mammoth Mountain have changed relatively slowly since 1991. In the late summer of 1995, there were seven areas of CO2 induced tree-kill scattered around the southern, western, and northern flanks of the mountain covering a total area of about 150 acres. A series of small collapse pits extending from the southern-most tree-kill area at Horseshoe Lake merge with a crack in the bottom of Horseshoe Lake that was first recognized in late September. Whether this system of shallow fractures is somehow related to the anomalously high CO2 soil-gas concentrations in the adjacent Horseshoe Lake tree-kill area remains to be seen. In any case, this fracture system seems to explain the tendency of Horseshoe Lake to drain internally. As soon as the snow melts this spring, we will re-survey the quadralateral installed around Horseshoe Lake last October to determine if this fracture system is associated with local deformation.