Page prepared by : Gene A. Ichinose
University of Nevada Reno Seismological Laboratory
and University of Nevada Reno Department of Geological Sciences
October 16, 1999
Updated 10/22/1999
On October 16, 1999, at 02:44 AM local time, an earthquake of magnitude 7.0 (Mw UNR) occurred in the Mojave Desert region of southern California near the location of the 1992 (Ms 7.3) Landers, California earthquake. The earthquake was felt strongly in Las Vegas, Nevada and Los Angeles, CA and as far away as northern Nevada and eastern Arizona.
Aerial photo of the southern segement of the Lavic Lake-Bullion fault where the apparent maximum slip was measured from the offset of a dirt road. The green lines mark the widest points of the "fault zone" where most of the motion is right-lateral. The fault trends north south at this location. Arrow points in the direction of north. Photo taken by Paul Otis-Diehl USMC out of Twentynine Palms Marine Base. More photos? For more information see the USGS links or the UNR press release. The following are the details of our preliminary analysis.
We model the long period seismograms recorded by regional seismic stations to obtain the moment magnitude and focal mechanism (geometry of faulting). The red triangles are 3 component very broadband IRIS, USNSN/USGS, and BDSN/UC Berkeley stations. The red lines are the source and receiver paths. The preliminary mainshock location (34.5955 °N -116.2681 °W) and depth (6 km) are fixed in the modeling (CIT/SCSN). The source is assumed to be a pure double couple. The following table lists the 1D velocity model from Ritsema and Lay, (1995) which is found to be a reasonable model for the western US.
| thickness(km) | P-wave velocity(kps) | S-wave velocity(kps) | density(g/cc) |
|---|---|---|---|
| 4.00 | 4.52 | 2.61 | 2.39 |
| 28.0 | 6.21 | 3.59 | 2.76 |
| 20.0 | 7.73 | 4.34 | 3.22 |
| 0.00 | 7.64 | 4.29 | 3.19 |
The red line are the synthetics and the gray line are the data. All stations are modeled within periods between 100 to 50 seconds. The fits are very good with stations distributed over half of the focal sphere. This solution is consistent with the Harvard CMT solution. We obtain a optimal depth of 6 km but depths from the iterating over 2 km depth increments suggest that the depth may not be resolvable from 6 to 14 km. I also maintain that the Mw is 7.0 from far regional stations. The Mw of 7.1 from the CMT method may be more accurate using teleseismic distances which may not be affected by the assumptions of the velocity model. The point source assumption may be violated even at far regional distaces which may also be a reason for the difference in magnitude. We use a trapezoid shape apparent source time function with duration between 8 and 12 seconds.
The misfit of the source parameters of strike dip and rake are plotted for 12000 iterations. The star marks the lowest misfit for a depth of 6 km and Mw 7.0. The depth resolution is 6 to 14 km.
The dislocation model for the mainshock consists of 8 segments each 4 km long from -116.207901 ° W 34.428684 ° N to -116.365547 ° W 34.739647 ° N. The right-lateral (-160 °) slip is distributed along a fault dipping 80 ° E roughly striking N 20 ° W. The fault segments were digitized from the mapped Lavic Lake and Bullion fault traces.
| Segment number | slip (m) |
|---|---|
| 1 | 1 |
| 2 | 2 |
| 3 | 2 |
| 4 | 3 |
| 5 | 3 |
| 6 | 5 |
| 7 | 3 |
| 8 | 2 |
Observed Global Positioning System (GPS) coseismic displacements were modeled using the above dislocation model [see table] in an elastic halfspace. The figure shows the observed GPS coseismic displacements (magenta arrows) and the predicted displacements (yellow arrows). There is good agreement at most stations within 95% confidence except for station OAES. More slip along the southern segment of the Bullion fault ( approx 8 meters maximum) may improve the fit which also is consistent with the finite fault slip inversion by Doug Dreger (UCB).
| site | latitude | longitude | East Pre | North Pre | East Obs | North Obs | |
|---|---|---|---|---|---|---|---|
| LDES | 34.2673 | -116.4328 | 61.9 | 150.8 | 56.9 | 179.6 | |
| PIN1 | 33.6100 | -116.4600 | 6.7 | 19.6 | 5.4 | 17.6 | |
| ROCH | 33.6110 | -116.6098 | 7.4 | 20.1 | 6.4 | 19.2 | |
| OAES | 34.1233 | -116.0633 | 26.7 | 9.4 | 24.1 | 28.6 | |
| BLYT | 33.6104 | -114.7149 | 7.3 | -3.5 | 6.0 | -2.7 | |
| SIO3 | 32.8600 | -117.2500 | 2.5 | 5.8 | 2.7 | 4.1 | |
| WIDC | 33.9347 | -116.3918 | 14.3 | 42.8 | 15.2 | 52.7 | |
| BBRY | 34.2642 | -116.8842 | 9.9 | 30.9 | 4.9 | 24.3 | |
| PMOB | 33.3572 | -116.8595 | 4.9 | 12.2 | 3.7 | 10.1 | |
| BSRY | 34.9186 | -117.0119 | -48.2 | 20.4 | -35.9 | 16.4 | |
| CTMS | 34.1241 | -116.3704 | 28.2 | 83.4 | 30.7 | 104.9 | |
| PSAP | 33.8192 | -116.4939 | 11.4 | 32.2 | 10.1 | 31.6 | |
| SCIA | 34.6074 | -117.3882 | -19.7 | 6.8 | -17.2 | 5.5 |
We compare the 1999 Hector Mine mainshock solution with fault geometries of other historical and modern large earthquakes in the Mojave Desert region (red quadrants are tensional and white quadrants are compressional). The circle symbols are ML > 4 and yellow squares are ML > 5 aftershocks. The static and/or dynamic stress interactions between the 1992 (Ms 7.3) Landers earthquake and todays (Mw 7.0) earthquake will be of most interest because of the possibility of spatial and temporal clustering. Earthquake clustering occurred between 1915 to 1954 where six magnitude > 6.5 earthquakes triggered within central Nevada. Earthquake geologists have trenched the faults in the Mojave region and may have found evidence for clustering from paleo-earthquakes [see references below]. This type of clustering would suggests the possibility for more large earthquakes in this area over the next few decades not predicted by probability theory.
Lindvall, Scott C. and Rockwell, Thomas K., 1993, Recurrent Holocene faulting along the Johnson Valley portion of the 1992 Landers earthquake surface rupture: Geological Society of America, Abstracts With Programs, v25, no. 5, p. 70.
Rockwell, T.K., Schwartz, D.P., Sieh, K.E., Rubin, C., Lindvall, S., Herzberg, M., Padgett, D., and Fumal, T., 1993, Initial paleoseismic studies following the Landers earthquake: Implications for fault segmentation and earthquake clustering: EOS AGU Fall Supplement, p. 67.
Herzberg, M. and Rockwell, T.K., 1993, Timing of past earthquakes on the northern Johnson Valley fault and their relationship to the 1992 rupture: EOS AGU Fall Supplement, p. 612.
We also examine the 2 largest aftershocks above magnitude 5 which occurred long after the coda of the mainshock. The 12:57 event (Mw 5.2) has a mechanism which is different in strike. The local magnitudes, which are not determined very well, are about 0.5 magnitude unit larger than the moment magnitudes in this study.
Change in Coulomb failure stress on planes of optimal failure caused by the 1992 Landers earthquake. The stresses are resolved onto planes of optimal failure at 6 km depth. These optimal planes are not shown but are consistent with the fault planes for the 1999 Hector earthquake. The Landers dislocation model was taken from an earlier geodetic study by (Hudnut et al., 1994). The arrows point to the northern and southern extent of the mapped 1999 Hector rupture. This suggests that the area of initial rupture may be in an area of positive Coulomb failure stress although other areas of the fault have negative values including the area of maximum slip.
Change in Coulomb failure stress on planes of optimal failure caused by the 1992 Landers and the 1999 Hector Mines earthquake.