Seismology and Probabilistic Hazard for Waste Repository Siting

Seismology and Probabilistic Hazard for Waste Repository Siting

Click to View the Next Photo in this Album

The IRIS Consortium

Other Photos in this Album
Remote Earthquake Triggering by the Denali Earthquake


Probabilistic seismic hazard curve for Yucca Mountain, Nevada, showing peak ground acceleration (PGA) plotted against the annual probability of exceedance (P.E.). Most buildings are designed to withstand shaking with a P.E. of around 10% in 50 years and life-safety designs (avoiding catastrophic building collapse) are for a P.E. of around 2% in 50 years. The design criterion for the proposed long-term nuclear waste repository at Yucca Mountain is 1 x 10-8/yr (Plot from J.S. Stepp and I.G. Wong, 2003. Probabilistic seismic hazard analysis for Yucca Mountain, Presentation to the Nuclear Waste Technical Review Board, February 24, 2003.) Photos: Balanced rocks can be used to constrain limits on peak ground motions observed over geological time scales (bottom). The Kashiwazaki-Kariwa Nuclear Power Plant in Japan (top and middle) was damaged during the July 16, 2007, Mw 6.6 Chuetsu earthquake. A transformer at the site caught fire and leaked fluids and gases. The ground motion in this earthquake exceeded the design criteria and the reactor, Japan’s largest, is presently closed. (Top photo by Japanese Coast Guard via Bloomberg News. Middle photo by Tokyo Electric Power Company from World Nuclear Association Picture Library. Bottom photo courtesy of M. Purvance.)

27 National Seismic Hazard Maps ( provide a probabilistic assessment of strong ground motions for any location across the United States and drive the design criteria for new construction. The catastrophic nature of structural failure at critical facilities, such as nuclear power plants and long-term waste repositories, requires construction to even higher standards of ground-shaking tolerance, as would be generated by large events with very small probabilities of occurrence. Historic records only document recent earthquake activity, even when supplemented with mapping and trenching of fault zones, and may not include the largest possible events in any given region. It is thus necessary to extrapolate hazard curves to estimate ground accelerations and velocities to levels that have never been historically recorded. Advances in our understanding of the physics of earthquake rupture, combined with massive computational capabilities, allow exploration of the range of physically plausible ground motions. The challenges include generating reasonable slip time histories and accounting for the very small-scale, near-surface structure. Model validation and verification requires multiple modeling approaches and additional model constraints, respectively. Validation of numerical codes uses some of the largest computing facilities currently available. The search for constraints on prehistoric ground motion is also underway. For example, precariously balanced rocks, precipitous cliffs, and fragile geological formations can be used to set bounds on ground motions experienced over geologic time scales.

Date Taken: February 18, 2009

This photo has been tagged with

Seismological_Grand_Challenges, Long_Range_Science_Plan,

Request a Hi-Res Version of this Image / Feedback on this Image