The Influence of Maximum Magnitude on Seismic-Hazard Estimates in the Central and Eastern United States

The purpose of this erratum is to correct several mistakes and to clarify the rationale for the analysis presented in Mueller (2010). These corrections and clarifications do not affect the quantitative results of the maximum-magnitude (mmax) sensitivity analysis that was presented in the article. In two places (first paragraph of the EPRI/SOG and USGS mmax Distributions for Two CEUS Sites section and last paragraph of the Summary and Discussion section) the author incorrectly states that the U. S. Nuclear Regulatory Commission (NRC) “required” updates to the mmax distributions in the application for a site. In fact, the NRC “requested” additional information about the updated mmax distribution used by the South Texas combined-license applicant as part of NRC’s ongoing review. The applicant recently (March 2010) replied to the request for additional information, and the NRC is currently reviewing the response. The author acknowledged support by the NRC (Mueller, 2010; acknowledgements section). More correctly, the work wassupportedbytheNRC’sOfficeofNuclearRegulatoryResearch,whichsupportscollaborative,long-rangeresearchthat is designed to inform future assessments of seismic hazards. This support does not imply endorsement of the results of the study by the NRC, especially in the context of seismic-hazard assessments at specific sites. The Electric Power Research Institute/Seismicity Owners Group (EPRI/SOG) mmax data from two nuclear power plantsiteswereusedtoinform thechoicesofalternativemodels for the sensitivity study. One of the sites was still under review by the NRC at the time of publication. The EPRI/ SOG mmax models are complex and difficult to summarize; broad distributions are used to reflect diverse approaches and large uncertainties. The reader should expect to be convinced that the range of alternative mmax models used in the analysis is reasonable, and scientific discussion is often improved through the use of site-specific examples. The two sites chosen represented two mmax “types” that helped to define the range of alternative models. The primary rationale for Mueller (2010) is to present the objective results of themmax sensitivity analysis. As noted in the article, there are other modeling differences that can obscure direct comparisons between U. S. Geological Survey (USGS)–based and EPRI/SOG-based hazard estimates. The article should not be read as expressing any preference for one mmax approach over another. This topic is beyond the scope of a sensitivity study (and the author resisted the suggestions of some reviewers to take the article more strongly in this direction). In particular, the last two paragraphs of Mueller(2010)illustrate thekindsofdifficultiesthatcanarise in defining—and especially in updating—seismic-hazard models. The final sentence is not meant to imply a preference for any particular mmax approach or model.