4

Seismic Source Characterization

Chapter 4 of the Senior Seismic Hazard Analysis Committee's (SSHAC) report, entitled “Methodology for Characterizing Seismic Sources,” describes the key elements of a seismic source characterization (SSC): the seismic source requirements for a probabilistic seismic hazard analysis (PSHA), the uncertainties in seismic source characterization, and guidance on expert elicitation for seismic source description. The chapter presents a good description of the state of practice for SSC in a PSHA, as shaped chiefly by guidance on methodology from the seismic hazard programs of the Lawrence Livermore National Laboratory and the Electric Power Research Institute (EPRI), as well as from other PSHA exercises modeled on those programs, for many other critical facilities. In the panel's judgment, practitioners of PSHA should be aware of and free to use other valid approaches to SSC.

SCIENTIFIC VALIDITY AND CLARITY OF PRESENTATION

A primary concern of the panel is the overall scientific validity of the procedures recommended by SSHAC. The basic methodology for SSC described in the SSHAC report has been validated by extensive peer review of prior projects in which such a methodology was used. The SSHAC report correctly states that a seismic source is a construct developed for seismic hazard analysis as a means of approximating the locations of earthquake occurrences. Insofar as SSC involves a simplified representation of real-world complexity, the validity of the simplifications is always an issue. Such validity is generally tested as part of sensitivity analyses, which are an essential part of a PSHA, as correctly advocated in SSHAC's report. With regard to modeling real-world complexity, the



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 41
Review of Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts 4 Seismic Source Characterization Chapter 4 of the Senior Seismic Hazard Analysis Committee's (SSHAC) report, entitled “Methodology for Characterizing Seismic Sources,” describes the key elements of a seismic source characterization (SSC): the seismic source requirements for a probabilistic seismic hazard analysis (PSHA), the uncertainties in seismic source characterization, and guidance on expert elicitation for seismic source description. The chapter presents a good description of the state of practice for SSC in a PSHA, as shaped chiefly by guidance on methodology from the seismic hazard programs of the Lawrence Livermore National Laboratory and the Electric Power Research Institute (EPRI), as well as from other PSHA exercises modeled on those programs, for many other critical facilities. In the panel's judgment, practitioners of PSHA should be aware of and free to use other valid approaches to SSC. SCIENTIFIC VALIDITY AND CLARITY OF PRESENTATION A primary concern of the panel is the overall scientific validity of the procedures recommended by SSHAC. The basic methodology for SSC described in the SSHAC report has been validated by extensive peer review of prior projects in which such a methodology was used. The SSHAC report correctly states that a seismic source is a construct developed for seismic hazard analysis as a means of approximating the locations of earthquake occurrences. Insofar as SSC involves a simplified representation of real-world complexity, the validity of the simplifications is always an issue. Such validity is generally tested as part of sensitivity analyses, which are an essential part of a PSHA, as correctly advocated in SSHAC's report. With regard to modeling real-world complexity, the

OCR for page 41
Review of Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts classification of seismic source types (Section 4.2) is nonunique, and the categories described in the report are admitted to be arbitrary. Nevertheless, they provide a useful framework for discussion and guidance on methodology. The practitioner experienced in PSHA will have no trouble understanding SSHAC's Chapter 4. However, the nonpractioner scientist may be confused by the subtleties between differing concepts of a “seismic source ” presented in chapters 4 and 5. Chapter 4 describes a seismic source as a geologic structure or as a domain within which the spatial and temporal occurrences of earthquakes are approximately uniformly distributed.Chapter 5, on ground motion, describes seismic source basically as a dynamic excitation in the earth that causes ground motion at the surface. Readers of the SSHAC report should be aware that two different terms, upper-bound and maximum magnitude, and two symbols, mu and Mmax, are used Section 2.1 and in Chapter 4 to denote the largest-magnitude earthquake that a particular seismic source is capable of producing. This magnitude is the upper bound of the frequency of occurrence magnitude curve used in the analysis. A value for this parameter must be specified in order to carry out the integration over all relevant magnitudes when calculating seismic hazard. The problems encountered and conventional procedures used in the selection of Mmax (mu) and the specification of the substantial epistemic uncertainty often associated with it are discussed in Sections 4.2.2 and 4.3.2 of the SSHAC report. If one accepts the basic formalism of uncertainty analysis presented in Section 2.2 of the SSHAC report, the approaches for characterizing uncertainties in SSC (Section 4.3) will seem logically consistent and well established in practice. Similarly, the guidance described in Section 4.4 for the expert elicitation process follows one's acceptance of the decision science methodology laid out in Chapter 3. A notable gap in Chapter 4 of the SSHAC report is the absence of discussion on and guidance for earthquake catalogs. In Section 4.4 the technical facilitator/integrator (TFI) or the technical integrator (TI) is given responsibility for providing a comprehensive and uniform data base to the experts for use in the PSHA. The only guidance given, under the subheading “Area Sources” in Section 4.2.3, is the recommendation that “seismicity catalogs should be reviewed for uniformity in designation of magnitudes and for completeness as a function of magnitude, location, and

OCR for page 41
Review of Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts time. The association of older historical events with particular seismic sources should be assessed bearing in mind the location uncertainties. ” Earthquake catalogs can play a major, even dominating, role in determining the outcome of a PSHA, particularly in the central and eastern United States, where information on active faults and other geologic structures is generally lacking. There are many problems hidden in earthquake catalogs that need be sought out and identified. There may be improper or mistaken entries, particularly for historic earthquakes. In many cases, locations and sizes were assigned to historic earthquakes based on inadequate or incomplete information. Unfortunately, modern earthquake catalogs often do not indicate which events have been critically reexamined and which have been carried forward without question from original catalog compilations. Uniformity of the data with time is also variable even in times of instrumental monitoring. Changes in network configurations and sensitivity and changes in the procedures for computing event magnitudes reported in earthquake catalogs (often not documented in an easily available form) should be sought out and carefully considered in a PSHA. Tests are available for identifying time-varying systematic shifts in reported magnitudes. Declustering or decomposing earthquake catalogs into main and secondary events (foreshocks, aftershocks, swarm events) is a nontrivial procedure that also requires careful attention. Recognizing that earthquake recurrence relationships based on seismicity depend critically on factors such as those described above, EPRI undertook major efforts to address these and other earthquake data base issues, which are still of great importance in PSHA—both in principle and in continuing practice. Those who utilize the SSHAC procedures should be aware of these requirements for preparation of their earthquake catalog for PSHA. To the panel 's knowledge, a comprehensive study of the effects of systematic changes in earthquake catalogs on the results of a PSHA has not been done. Most of Chapter 4 of the SSHAC report is well organized and well written, and the presentation should be easy for general readers to follow. The text refers to Appendixes H and I, each of which provides some ancillary pertinent material. Appendix H describes the results of a workshop on expert elicitation of seismic source (zone) information, while Appendix I describes effects of a nonuniform spatial distribution of seismicity in a seismic source (zone). Both of these appendixes are informative.

OCR for page 41
Review of Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts The table in Section 4.2.1 is important for guidance, but it is confusing. The lines beginning with “Faults” and “No faults” should be understood to be “if” statements, recognizing “fault” to mean a “Type 1 seismic source” (i.e., “If no Type 1 fault source within 50 km of a site, then . . .”). Because the SSHAC report is intended for general PSHA guidance, the following question arises: Is the EQPARAM code (which is introduced as an important element of the methodology in Section 4.3.5) readily available or is it proprietary to EPRI? If the latter, it should have been described as such. This question illustrates the concerns of the panel about software availability expressed in the previous discussion of documentation. CONTRIBUTIONS TO THE DEVELOPMENT OF PSHA Because SSC is such a major component of a PSHA, the comprehensive methodology for expert elicitation presented in Section 4.4 of the SSHAC report is an important contribution. On first reading, the material in Chapter 4 may appear to be just a restatement of Chapter 3. However, SSHAC is correct in noting in Section 4.4 that the elicitation procedures and methods for SSC differ from those for ground motion characterization. Further, “lessons learned” from past SSC exercises are incorporated into major PSHA projects (Appendix H). Another important contribution of Chapter 4 and its accompanying appendixes is the practical guidance provided for carrying out sensitivity analyses to determine “what drives the seismic hazard” and “what contributes significantly to uncertainties in hazard.” Basic discussion relevant to SSC is presented in Section 4.3.6, but important details are given in Appendix G and Section 7.8. A third major contribution of Chapter 4 is the exposition in Section 4.3.5 (bolstered by Appendix I) of the effects of spatial variations in seismicity within a seismic source vis-à-vis the assumption of homogeneous seismicity. The analysis techniques date from the EPRI program (EPRI, 1989, as cited in the SSHAC report), but the detailed discussion and examples presented there forcefully demonstrate how the usual assumption of homogeneous seismicity for seismic sources can, under certain predictable cases, significantly affect both the mean seismic hazard and its statistical uncertainty.

OCR for page 41
Review of Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts THE OUTLOOK FOR EVOLUTION OF SSC While affirming the scientific validity and practical effectiveness of the SSC methodology set forth in the SSHAC report, the panel recognizes that the scientific community will naturally strain against the confines of SSHAC's prescriptions for SSC. The panel applauds SSHAC's perspective that “[its] formulation should not be viewed as an attempt to ‘standardize' PSHA in the sense of freezing the science and technology that underlies a competent PSHA, thereby stifling innovation” (Section 1.2 of the SSHAC report). A few brief examples suffice to illustrate current trends in the scientific community that may influence the evolution of SSC. Diverse trends lead to advocacy for both greater simplification and greater complexity. Frankel (1995) proposes a method for PSHA that uses spatially smoothed representations of historic seismicity instead of seismic source zones to directly calculate probabilistic seismic hazard. Insofar as he demonstrates the capability to produce values of mean seismic hazard similar to those from the more complicated EPRI methodology, his simple methodology offers understandable attraction. The applicability obviously pertains to cases where seismicity “drives the hazard”—either for specific regions or for definable exposure periods. In terms of modeling earthquake occurrence with greater complexity, one example is the multidisciplinary approach (e.g., Ward, 1994), in which data from space geodesy and synthetic seismicity are added to the traditional information from geology, paleoseismology, and observational seismology. Main (1995) examines the implications if earthquake populations are really an example of a self-organized critical phenomenon. If this is correct, the a priori assumption of the Gutenberg-Richter frequency-magnitude distribution is no longer valid in some cases, and Main provides evidence for questioning the use of only the Poisson distribution in seismic hazard analyses, based on the accumulating evidence of local or long-range interactions of earthquakes. It should be pointed out that PSHA is not limited to the use of the Gutenberg-Richter relationship. Alternate estimates of the frequency-magnitude distribution are, and have been, used in probabilistic analyses. Main (1995) also discusses an independent approach to the vexing problem of estimating the maximum-magnitude earthquake that is “ credible” for a seismic source zone, based on his suggested distribution of moment release and the long-term slip rate on the causative fault

OCR for page 41
Review of Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts system. Geophysicists are becoming increasingly aware of the nonstationarity of earthquake occurrence, particularly in light of observations of fault interactions leading to “triggered” or “encouraged” earthquakes. As earth scientists improve their ability to assess time-varying earthquake potential on active faults, SSC will evolve correspondingly. Indeed, “time-variable seismic hazard” is already a topic of special sessions at geophysical society meetings.