Faculty and Collaborators: Norm Abrahamson (Pacific Gas and Electric), John Anderson (Nevada-Reno), Ralph Archuleta (UCSB), Jacobo Bielak (Carnegie Mellon), James Brune (Nevada-Reno), Jim Davis (CDMG), Steve Day (SDSU), Bill Foxall (LLNL), Joan Gomberg (USGS), Robert Graves (Woodward-Clyde), Tom Hanks (USGS), Tom Heaton (Caltech), Don Helmberger (Caltech), Tom Henyey (USC), Gene Humphreys (OSU), Dave Jackson (UCLA), Bruce Luyendyk (UCSB), Mehrdad Mahdyiar (Vortex Rock), Bernard Minster (UCSD), Mark Petersen (CDMG), Mike Reichle (CDMG), Chandon Saikia (Woodward-Clyde), Paul Somerville (Woodward-Clyde), Ross Stein (USGS), Jeff Stevens (Maxwell Labs), Lynn Sykes (Columbia), Mladen Vucetic (UCLA), Steve Ward (UCSC), Steve Wesnousky (UNR)

Postdoctoral Fellows: Jishu Deng (Caltech), Ned Field (USC), Peggy Johnson (USC), Yan Kagan (UCLA), Peng-Cheng Liu (UCSB), Stefan Nielsen (UCSB), Kim Olsen (UCSB), Zhen-Kang Shen (UCLA), Jamie Steidl (UCSB), Mark Stirling (Nevada-Reno), Feng Su (Nevada-Reno), Alex Tumarkin (UCSB), Yue-hua Zeng (Nevada-Reno), Lupei Zhu (USC)

Students: Carmen Alex (UCSB), Rasool Anooshehpoor (Nevada-Reno), Fabian Bonilla (UCSB), Wen-Xuan Du (Columbia), Leo Eisner (Caltech), Javier Favel (Caltech), Karen Gerst (UCLA), Leland Green (UCSB), Jeanne Hardebeck (Caltech), Eleanor Jewel (UCSB), Daniel Lavallee (UCSB), Lowell Kessel (UCSB), Grant Lindley (UCSB), Jackie Moccand (USC), David Oglesby (UCSB), Javier Santillan (UCSB), Lisa Sarma (Columbia), Brian Savage (Caltech), David Valentine (UCSB), Danielle Vanderhorst (UCSB), Paul Vincent (Colorado)

There have been three major emphases in 1998. First, several investigators including Dolan, Field, Hanks, Jackson, Kagan, Petersen, Stein, Stirling, Ward, and Wesnousky have studied the magnitude distribution of southern California earthquakes and the role of these distributions in testing seismic hazard models. Second, Day, Deng, Hardebeck, Humphreys, Kagan, Harris, Stein, Sykes, and Ward have investigated (1) models for estimating stress on faults resulting from tectonic motions, past earthquakes, and viscoelastic effects, and (2) the influence of these stresses in triggering earthquakes. Third, Abrahamson, Mahdyiar and Petersen have studied the influence of various assumptions regarding seismic sources and strong ground motion propagation on earthquake hazard estimates.

Several research efforts were carried out to help resolve the apparent discrepancy between the historical earthquake record and the magnitude-frequency relationships predicted by the Phase II and other earthquake hazard models. Stirling and Wesnousky considered whether the magnitude distribution implied by frequently used relationships between magnitude and fault length were consistent with the historical earthquake catalog. They concluded that the discrepancy could be explained at least partly by random errors in the geologic data underlying the forecasts, and the randomness of the earthquake process itself. The random simulations of the observed magnitude distribution and a suite of Monte-Carlo models randomized from the Phase 2 report overlapped at the 95% confidence level of each. Stein and Hanks argued that the earthquake catalog used in the Phase 2 report was not complete, and that rates inferred from the more complete recent data were higher and more nearly in agreement with the Phase 2 model. Field, Jackson, and Dolan (submitted, 1998) show that by allowing for randomness in the relationship between fault length and magnitude, and by using an improved model for "cascade" earthquakes, theoretical models based on geology can be constructed that agree well with the historic earthquake catalog. All of these studies agree in reporting that geologically based models can and should agree with the historic earthquake catalog. These studies leave open the question of the largest magnitude that might occur in southern California: present data neither confirm nor reject the assertion that earthquakes over magnitude 8 might occur here.

SCEC has long emphasized stress accumulation as a promising criterion to recognize faults or regions most likely to experience earthquakes. A major SCEC contribution in this field has been to produce the important ingredients of stress evolution models for southern California: earthquake catalogs, fault maps, and deformation data.

Thanks to Ruth Harris, Joan Gomberg, and numerous authors, 13 papers resulting from a 1997 SCEC conference on this subject will soon be published in a special edition of JGR. This year SCEC sponsored another such conference, hosted by Ross Stein, who wrote a more detailed summary in the SCEC Quarterly Newsletter. Examples of sequences of quakes, each encouraged by Coulomb stress increments from previous events in the sequence, continue to accumulate. In addition, global and regional studies of complete catalogs show that main shock and aftershock occurrence are both favored in regions where shear stress has been increased by previous events. Attention has focused on the relative effectiveness of shear stress and normal stress. The studies of complete catalogs, and targeted studies of quakes on mature strike slip faults, generally show little influence of normal stress. On the other hand, some studies of other faults imply that a reduction of normal stress may act to increase seismicity. One problem that has complicated stress evolution calculations is the "nodal plane ambiguity:" two orthogonal planes, each giving identical displacements for point sources, respond differently to Coulomb stress. Hardebeck and Hauksson have made progress in providing statistical tests that are tolerant of this ambiguity; Kagan has used tensor invariants to circumvent the ambiguity; Venkataraman, Mori and Kanamori have shown that broad band seismic data can resolve the fault plane ambiguity for earthquakes as small as magnitude 5.

SCEC's primary interest in stress evolution results from its possible predictive power. Several investigators have combined stress increment calculations with Dieterich's "rate-state" friction models to calculate theoretically the influence of Coulomb stress increments on individual fault segments. A result of the recent conference is ongoing discussions on ways that the predictive power of stress evolution calculations can be quantified and tested using data from future earthquakes.

Mahdyiar studied the effects of various source models, attenuation relations, and site conditions on theoretical seismic hazard calculations. The Phase III working group, based on the review of the regional earthquake ground motion and site conditions data, has developed a regional site condition map and the related site response amplification factors for seismic hazard analysis. Mahdyiar investigated the effects of this new information on the probabilistic seismic hazard analysis (PSHA) of southern California.

The Phase III report defines three site categories, Quaternary (Q), Tertiary (T), and Mesozoic (M), for characterizing site conditions. Each site category is identified by a shear wave velocity for its top 30-m materials. The report provides site response amplification factors for these three site categories as a function of frequency and peak ground acceleration. Specific amplification factors, for long period motions, are also introduced to account for the basin-effects on earthquake ground motions.

He used three attenuation relationships (Geomatrix, Joyner/Boore/Fumal (JBF), and Abrahamson/Silva) to test for their influence on PSHA. SCEC's site amplification factors are reported as correction factors for the Geomatrix rock-site attenuation equation to account for different site conditions. SCEC defined local site amplification factors with respect to the Geomatrix attenuation model, as used in the Phase 2 Report. The sensitivity of probable acceleration levels to reasonable choices in the analysis was explored by plotting spectral acceleration levels with 10% probability in 50 years, calculated in different ways, along three cross sections passing through sedimentary valleys, mountains, and faults.

Sites along three cross sections crossing major faults in southern California are selected for the PSHA. For ground motions calculations, the Q- and T-sites are characterized as soil-site conditions and the M-sites are characterized as rock-site condition. For the JBF attenuation equation, the site amplification factors are calculated based on the average 30-m shear wave velocities of different site categories. We use spectral acceleration values with 10% probability of exceedance in 50 years to compare and evaluate the results of the PSHA based on different attenuation equations and site condition information.

The results of the analysis are presented by a number of hazard plots over different cross sections. This provides direct comparison between the results of the PSHA based on different attenuation equations and site condition information.

Model Y2, based on the Geomatrix attenuation model with a correction for basin depth and three site correction categories, is the most complete model we tried. It applies only at 0.3 Hz frequency. For Mesozoic sites, the JBF attenuation equation predicts noticeably lower spectral acceleration values than all other equations at 0.1 and 0.3 s periods. However, at 1.0 s period all the attenuation equations predict very similar values. For tertiary sites at 0.3 Hz, the SA values from Y2 and Geomatrix PSHA become rather close. For quaternary sites at 0.3 Hz, basin effects and site corrections essentially cancel each other, and the Y2 model predicts nearly the same response as the Geomatrix model. However, the basin effect in the Y2 equation is depth dependent, so that at some locations the Y2 equation calculates up to 50% higher spectral acceleration values than the Geomatrix soil-site equation. This is an important consideration for the design of large scale structures such as water tanks.

Abrahamson/Silva attenuation equation calculates higher ground motion values at sites over the hanging walls of the thrust/reverse faults. The results of this study indicate that the hanging-wall effects at sites close to reverse/thrust faults are significant. The hanging-wall effects are in the same order of magnitude and in cases higher than the site condition effects on the PSHA.

[Thrust 2] [Thrust 3] [Thrust 4] [Thrust 5]

Back to Executive Summary