(a) Calculated coseismic Coulomb stress changes associated with the 1992 Mw=7.3 Landers (L), Mw=6.2 Big Bear (BB), and Mw=6.1 Joshua Tree (JT) earthquakes. Stresses are shown on the surface and through a cross-section that passes through the eventual rupture surface of the 1999 Mw=7.1 Hector Mine (HM) earthquake. Positive Coulomb stress changes indicate regions where neighboring faults (aligned parallel to the cross-section shown) have been pushed closer to failure. Note how a large reservoir of stress develops in the lower crust and upper mantle beneath the eventual Hector Mine hypocenter (star). Stresses in these regions cannot be sustained and begin to relax immediately after the Landers quake. (b) shows that after 7 years of relaxation (in this case the upper mantle is assumed to relax), much of the stress has transferred from the mantle to the upper crust leading to a build-up of Coulomb stress at the Hector Mine hypocenter. This transfer of stress is thought to have played an important role in explaining why the Hector Mine earthquake occurred only 7 years after the Landers quake despite both faults having repeat times in the 1000s of years.

Scientists have hypothesized for decades that one major earthquake can trigger another earthquake on a nearby fault through stress interaction. More recent studies have further suggested that this interaction may be delayed by the slow viscous creeping of rocks in the Earth’s lower crust and upper mantle. This is best illustrated by the 1999 magnitude 7.1 Hector Mine earthquake, which occurred only 30 km away from the 1992 magnitude 7.3 Landers quake, but seven and half years later. The delay between these events can be explained by viscous flow consistent with observations of continuous ground deformation following the Landers quake. In this study, we further calculated how the Landers, Hector Mine, and two other earthquakes in the Mojave Desert have changed stresses on the nearby southern San Andreas and adjacent fault systems. (a) shows calculated coseismic Coulomb stress changes caused by fault slip associated with the 1992 Joshua Tree (JT), Landers (L), and Big Bear (BB) earthquakes (green lines). The location of future 1999 Hector Mine (HM) earthquake (green dashed line) is also shown. (b) is the same calculation as (a) but stresses are shown for top surface and a cut plane (front) along the San Andreas fault. Brittle-ductile transition (b-d trans) and Moho depths are shown in cut plane. We calculated that these earthquakes and continuous viscous creeping at depth are causing a rapid increase of stresses on a section of the San Andreas Fault, called the San Bernardino Mountain segment, that is located only 80 km from Los Angeles. (c, d) show the projected stress state after about 3 decades of relaxation. Note how relaxation of stresses in the lower crust and mantle have lead to a dramatic increase in upper crustal stresses. The San Bernardino Mountain segment, where stress increases are particularly significant, is worthy of special attention because it is capable of producing major earthquakes with magnitude greater than 7. Since the last major earthquake on this segment was over 190 years ago, the fault may be late in its earthquake cycle, and thus the calculated ongoing stress increase on the fault is of added significance. In addition, we calculated that parts of the San Jacinto, Elsinore, and Calico faults are also experiencing accelerated stress buildup. In particular, the Calico fault, which lies just north of the Landers rupture (near Barstow), appears to have the calculated stress patterns and the observed post-Landers aftershock clustering quite similar to the Hector Mine region before the 1999 quake. This makes the Calico fault another candidate for a potential earthquake in the future, where seismic activity should be watched closely. Other faults: MS - Mojave segment, SBMS - San Bernardino Mountain segment, and CVS - Coachella Valley segment of San Andreas fault; SJF - San Jacinto fault, EF - Elsinore fault, CF - Calico fault, LF - Lenwood, and BWF - Blackwater fault.