Description
The Elsinore fault zone, extending over 250 km from the Los Angeles basin in southern California southeastward into Mexico, has been largely quiescent of major earthquakes during the historical period. Other than the 1910 M6 Temescal Canyon earthquake along the Glen Ivy segment and the 1892 M7.2 Laguna Salada earthquake south of the border, historical seismicity is characterized by small to moderate earthquakes, generally not exceeding M5. Paleoseismic trenching results, microseismicity studies, and the geometry and continuity of individual fault strands within the overall fault zone have resulted in segmentation models to both explain the historical observations and to attempt to forecast potential future ruptures. Lacking from these analyses were any paleoseismic data from the entire 120-km-long central portion of the fault zone. Herein, we present new paleoseismic results from three trench sites, spaced 12-20 km apart, along the central Elsinore fault zone. Specifically, we excavated trenches near Pala, at Lake Henshaw, and near Julian to date the timing of the most recent earthquakes and establish average recurrence intervals. One fundamental purpose of this study is to test current segmentation models by resolving whether or not the Wildomar and Julian segments have distinct (different) rupture histories. The trenches near Pala, along the southern reach of the Wildomar segment, cut Holocene sediments and degraded bedrock, exposing the fault and evidence for the last earthquake at one end of our trench. A detrital charcoal 14C date above the faulted units yielded an age of 170 ± 45, and a sample from below was dated at 255 ± 50. These dates indicate that the uppermost faulted units postdate 1655 A.D., which along with the construction of the Pala Mission, constrains the timing since the last earthquake to between 190 and 344 years B.P.. Although no penultimate event was recognized, an average return time of 170- 400 years was calculated using a 5 ± 2 mm/yr slip rate and assuming a 51 km rupture length along the Temecula segment of the Elsinore fault. This recurrence interval is close to the 550-600 year interval estimated from other trenching at Pala. Several trenches were emplaced near the shore of Lake Henshaw. Two of the trenches show a sequence of surface soil and two buried soils deformed and displaced across the fault zone. The ages of the inferred events are 1.5-2 ka for the most recent earthquake and 5±1 ka for the penultimate event. These data suggest a long recurrence interval for the Julian segment, perhaps 3-3.5 ka. A trench near Julian, excavated across a sidehill depression, revealed evidence for only one late Holocene event between 0.7 and 1.8 ka. In that the Julian and Lake Henshaw trenches are only 12 km apart, and because the timing of the most recent event at both sites are indistinguishable, they are interpreted as the same event. The above observations support a segment boundary in the Palomar Mountain area, as suggested by all segmentation models. Analysis of the timing of past earthquakes along each proposed segment of the Elsinore fault best supports the model that separates the Glen Ivy fault as an independent segment.
