by Paul Somerville

The development of new knowledge about earthquakes and their effects is an important role of SCEC, but not its only role. Because earthquakes have major impacts on society, SCEC must also transfer knowledge about earthquakes and their effects for use in earthquake risk mitigation. This article describes a subset of SCEC's implementation interface program: the joint activities between SCEC and PEER (the Pacific Earthquake Engineering Research Center), headquartered at the Richmond Field Station of U.C. Berkeley.

(Editors note: Paul Somerville (URS Corp.) is working with the SCEC Communication, Education and Outreach program to facilitate SCEC's interactions with technical users of SCEC products, especially research engineers.)

PEER PBEE Methodology Testbed Program

The PEER Testbeds are real facilities to which PEER performance based earthquake engineering assessment (PBEE) and design methodologies are being applied. The primary focus of the testbed program is to assess the applicability of the methodologies and to foster their refinement. The testbeds will serve supplementary purposes such as focusing and integrating the research, promoting multidisciplinary research interactions, emphasizing systems level research, and involving interested earthquake professionals and decision makers. The testbed program includes the following four individual structure testbeds: a Van Nuys hotel, damaged by both the 1971 San Fernando and 1994 Northridge earthquakes; a new laboratory building at UC Berkeley; the Humbolt Bay Bridge, an older bridge which has been retrofitted by Caltrans; and a new bridge on Interstate 888 in Oakland. Paul Somerville provided ground motion time histories for these four testbed structures under a contract from PEER, and in a SCEC project is interacting with the teams of researchers for all four testbeds to make sure that the time histories are properly interfaced with the geotechnical and structural engineering analyses.

3D Basin Ground Motion Modeling Project

An ongoing program of calibration of 3D basin modeling codes is being funded by SCEC and the PEER Lifelines Program. To date, the program has focused on validation of five different computer codes. Under the leadership of Steve Day at SDSU, this program has been spectacularly successful, to the point where all significant differences in the results of five different codes have been wrung out by a series of carefully designed tests. The next steps, which are under way, are to test the codes against recorded data of the 1994 Northridge and 2001 Hollywood earthquakes, and to use the codes in the calculation of the ground motions of important earthquake scenarios.

Framework for Object-Oriented Seismic Hazard Analysis

Allin Cornell (Stanford), who is involved in the PEER Core Program, and Norm Abrahamson (Pacific Gas & Electric), who is involved in the PEER Lifelines Program, are providing review and guidance of the Framework for Object-Oriented Seismic Hazard Analysis that is being developed by Ned Field (USGS Pasadena). A review meeting, attended by these three individuals together with Tom Jordan (USC and SCEC Director) and Paul Somerville, was held at USGS on June 17 to discuss issues related to the development of the Framework. The Framework will provide an important vehicle for the experimental use of RELM hazard models and the SCEC Community Models. Earthquake source and ground motion models that are develop for the user community will need to undergo rigorous review before being made available for use.

Probabilistic Vector-Valued Ground Motion Hazard

Allin Cornell (funded by PEER) and Paul Somerville (funded by SCEC) are participating in the development and use of a vector-valued representation of seismic hazard for use in the prediction of building response. The vectors will initially consist of spectral acceleration at two or more periods, with a plan to eventually use the peak velocity and period of the near-fault rupture directivity pulse. The use of vector-valued hazard is expected to improve the accuracy and the efficiency of prediction of building response. The project will be performed within the context of the PEER PBE Methodology Testbed Program, specifically, the Van Nuys hotel testbed. In this program, ground motions are being specified using the conventional scalar-valued hazard approach. The goal of the collaborative project is to concurrently test the application of vector-valued hazard to this project, and compare its efficacy with that of the standard scalar-valued hazard approach.

Other SCEC and PEER Lifelines Activities

The PEER Lifelines program includes many activities in which SCEC has an interest. Recently completed activities include the validation of 1D strong motion simulation procedures against recorded data. Current activities include the testing of rupture dynamics models and ground motion models using foamquake data, the development of near real time ground motion maps using rapid source inversion and forward ground motion calculations, the validation of probabilistic seismic hazard calculation codes (in which Ned Field is participating), and the validation of 3D basin modeling codes (described above).