Part I. Broadband modeling of aftershocks from the Joshua Tree, Landers, and Big Bear sequences, southern California. Part II. Characteristics of the June 28, 1992, Big Bear mainshock from TERRAscope data: evidence for a multiple-event source

Citation

Jones, Laura Ellen (1995) Part I. Broadband modeling of aftershocks from the Joshua Tree, Landers, and Big Bear sequences, southern California. Part II. Characteristics of the June 28, 1992, Big Bear mainshock from TERRAscope data: evidence for a multiple-event source. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/r040-cd43. https://resolver.caltech.edu/CaltechTHESIS:11302011-093359502

Abstract

The Joshua Tree, Landers and Big Bear earthquake sequences recorded on the broadband TERRAscope array in Southern California provide an unusually large data set of high-quality three-component broadband waveforms from small to moderately sized earthquakes. This data set offers the opportunity of detailed large-scale studies of these smaller but nonetheless potentially damaging earthquakes. What follows is a detailed study of over sixty M > 3.8 aftershocks in three regions: north and south of the Pinto Mountain fault in the Mojave desert (associated with the Joshua Tree and Landers sequences), and within the San Bernardino mountains block (associated with the Big Bear sequence). Source parameters, including depths, moments, and durations, for sixty M_w > 3.8 earthquakes from the Joshua Tree, Landers, and Big Bear sequences are presented here. These events occurred between April of 1992 and November of 1994; the list of events comprises nearly every aftershock above M3.8 for which we could obtain coherent TERRAscope data and accurate timing and location information.

Choice of velocity model affects the accuracy of the source estimations and the error associated with estimations of moment, though it appears that for Landers events recorded at stations within or near the Mojave region, a simple one-dimensional velocity model is adequate. To minimize model-associated error, however, a velocity model for the Mojave region is developed and presented. This model is used in the computation of the synthetic Green's functions used in estimates of source parameters for many of the earthquakes presented here.

The existence of such a large data set from events in the same region (the Mojave desert) also allows systematic investigation of station effects for the five TERRAscope stations used in this study; Goldstone (GSC), Isabella (ISA), Pasadena (PAS) , Pinyon Flats (PFO), and Seven Oaks Dam (SVD). For each event, moments and durations are computed for each station, and these examined for systematic variations of moment with azimuth and with source-receiver distance.

Moments and durations are computed for each aftershock we study, and stress-drops inferred from these appear to vary with location, with respect to previous seismic activity, and proximity to previous (i.e., Landers) rupture. A strong correlation of increased stress-drop with depth is noted for the Big Bear region; the same is not observed for Mojave aftershocks.

The June 28, 1992, Big Bear earthquake is commonly considered to be an aftershock of the earlier M_w = 7.3 Landers mainshock, and as such has been perhaps overlooked. However, it is a significant and enigmatic event in its own right. Its rupture history was obscured by controversy over epicentral location, lack of observed surface rupture, and the complexity of source suggested by the mainshock waveforms themselves. From overall pat terns of seismicity and long-period focal studies, rupture is generally assumed to have propagated northeast. However, mainshock locations from both strong-motion and TERRAscope data are consistent and do not lie on this assumed fault plane. Further, directivity analysis suggests significant energy propagating northwest along the presumed antithetic fault-plane .A combination of directivity analysis, point-source empirical Green's function analysis, and line-source directivity analysis together indicate that a two-fault event is necessary to produce the waveforms observed during the Big Bear mainshock. These results suggest that the Big Bear earthquake comprised at least two substantial subevents, with the initial subevent rupturing towards the northwest on the presumed antithetic fault plane. Several seconds later, rupture initiated on the northeast striking plane.

Thesis Files