[Group A | Group C | Group D | Group E | Group F]

The primary research areas are 1) 3D Deterministic Prediction of Ground Motion, 2) 3D Stochastic Prediction of Ground Motion and 3) Quantification of Site Effects

The ultimate objective is to compute ground motion in a realistic 3D structure from scenario earthquakes in southern California. In all tasks, the use of data, including that from Izmit, Turkey, Chi Chi, Taiwan, and Hector Mines, California, is encouraged in validating the different approaches.

Task B1: 3D Deterministic ground motion. Verify and validate methods of computing 3D waveforms, amplitude and phase for scenario earthquakes in the frequency band 0.0 < f <0.5 Hz using the latest version of the SCEC southern California velocity model. This will include an exercise for those who are doing 3D calculations to participate in: (i) Verifying 3D propagation methods against standard analytical methods for halfspace and layered media. (ii) Comparing 3D codes using the SCEC southern California velocity model. (iii) Studies directed at model validation including point source and extended source calculations. What is the best quantitative measure for determining the optimal fit between two time series? (iv) Sensitivity studies focused on source and/or path effects. (v) Extensions and improvements to models and methodologies. (vi) Computations of ground motion for scenario earthquakes-on the a San Andreas fault and on faults within or bounding the Los Angeles Basin. Proposers should be prepared to participate in (and help define) a coordinated program of computations, some of which may require use of standardized source and/or seismic velocity models, with the objective of developing one or more SCEC consensus products.

Kim Olsen's ground motion simulations

Task B2: Stochastic broadband (0 < f < 25 Hz) ground motion. Verify and validate methods for computing stochastic ground motion with application to scenario earthquakes. Which attributes (e.g., peak acceleration, peak velocity, response spectra, Fourier amplitude or integral values like CAV or Arias intensity) of ground motion time histories are best determined by stochastic models? In validation studies how should the misfit between computed and observed attributes be defined? How are empirical Green's functions to be used for strong motion prediction? That is, how is nonlinear site response accounted for when EGF's are used? Those submitting proposals should be prepared to participate in a working group to define parameters for scenario earthquakes and compute at least 10 realizations of broadband ground motion from at least three sources.

Task B3: Site effects -- predictability and reproducibility of site response. Work may include assembly of existing data, collection of new data where needed, statistical analysis of site response data, and data analysis and modeling directed at improved understanding of site response. Particular emphasis is on documenting and characterizing nonlinear site response. SCEC may support research to identify nonlinear effects in recorded ground motion, relate these effects to the dynamic properties of soil, and validate computational models for nonlinearity.

Task B4: Near-source effects on strong motion. Promising approaches to the prediction of near-source effects, including innovative empirical, semi-empirical, numerical or theoretical approaches, are of interest, and may be considered for support.

Task B5: Legacy document. Proposals are encouraged from investigators who will synthesize and summarize the primary accomplishments of activities in Group B. The major areas are 3D deterministic time histories, broadband stochastic time histories, site effects and scenario earthquakes.