ShakeOut Earthquake Scenario Ground Motions Kenneth W. Hudnut U. S. Geological Survey Earthquake Research Affiliates Pasadena, California 9 May 2008 U.S. Department of the Interior U.S. Geological Survey

USGS Multi-Hazards Demonstration Project Lucy Jones, Chief Scientist Dale Cox, Project Manager Suzanne Perry, Staff Scientist Earthquake Scenario Coordinators: –Dan Ponti, Anne Wein, Rich Bernknopf, and Ken Hudnut (all at USGS), Mike Reichle and Jerry Treiman (CGS), Keith Porter and Dennis Mileti (Univ. of Colorado), Jim Goltz (OES), Hope Seligson (MMI Eng.), and Kim Shoaf (UCLA) ShakeOut Earthquake Contributors - Source Description, Surface Faulting and Ground Motions: –Brad Aagaard and Ned Field (USGS), Rob Graves* (URS), Lisa Star and Jonathan Stewart (UCLA), Thomas Jordan,* Gideon Juve,* Philip Maechling,* David Okaya,* Scott Callaghan* (USC), Jacobo Bielak,* Ricardo Taborda,* Leonardo Ramirez-Guzman,* Julio Lopez,* and David O'Hallaron* (CMU) and John Urbanic* (PSC), Geoff Ely* (SDSU/UCSD), Kim Olsen,* Luis Dalguer* and Steve Day* (SDSU), Yifeng Cui,* Jing Zhu,* Timothy Kaiser,* Amit Chourasia,* and Reagan Moore* (SDSC), Chen Ji (UCSB), Swami Krishnan, Matt Muto and Jeroen Tromp (Caltech) * participant in the SCEC/CME collaboration, funded by the National Science Foundation

Economic Activity Time Projected activity For one specific natural hazard event: ‘Disaster’ (a few yrs.) ‘Catastrophe’ (decades) event

USGS/CGS/SCEC & CEA - UCERF (4/21/08)

A San Andreas “Big One” is far bigger than Northridge 1994 What is meant by “Big One” exactly…? - fault length is 300 km (15 x) - rupture duration is 90 sec (10 x) - shaking duration is 180 sec (15 x) M 7.8 “great” >> M 6.7 amplitude (15 x) & energy (~30 x)

ShakeOut Scenario Earthquake Not a prediction - a plausible event First step: define the rupture length ? ?

Possible early/pre-historic earthquake rupture correlations: - Conservative correlations result in 6 events in 1200 years - Recurrence Interval ~150 years without latest ‘open’ interval - Currently, elapsed time of ~ 300 years appears longer than any previous recurrence interval A call for the most reliable data: Hard data versus soft data Examples: SoSAFE Project, B4 LiDAR project on San Andreas, multi-investigator trench studies Courtesy of Gordon Seitz, Pat Williams, Ray Weldon & Belle Philobosian Digging deep into the San Andreas fault, because the past may be the key to the future

ShakeOut - Static Rupture Description thrusting & folding ±310±316±322±628±7 Slip Rates: SE end Bombay Beach NW end Lake Hughes ±310±316±322±628±7 WGCEP Slip Rates: thrusting & folding rupture propagation direction slip (meters)

Kinematic Rupture Model (beta v.1) SSA 2007 presentation Aagaard, Hudnut & Jones web site w/ RefEq standard rupture model format file Slip (meters) SE (BB)NW CP SGP View is from a point to the NE and above the San Andreas fault

M w 7.8 ‘ShakeOut’ Scenario (GG’08 Nov. 13, 2008) San Andreas ‘Big One’ simulated earthquake Initiation near Bombay Beach (unilateral rupture to the NW) Disruption of critical lifeline infrastructure (freeway, internet, power and gas lines) along surface rupture Strong shaking throughout the region, including urban areas Surface Slip and Modified Mercalli Intensities for ShakeOut scenario earthquake. Credits: Rob Graves (SCEC-CME; URS / USC) Brad Aagaard (USGS); ShakeMap Integration thanks to Vince Quitoriano (USGS), Wan Lin (USGS), Keith Porter (Univ. of Colorado) and David Wald (USGS)

Kinematic rupture models Kinematic Rupture Description - Aagaard, Hudnut & Jones –Two SCEC workshops; CGS/USGS/SCEC - ‘A-faults’ & SoSAFE –1D slip distribution construction - background slip model –Mapped onto SCEC CFM and CFM-tin fault geometry –Source physics ‘first principals’ and empirical relations to construct details of the kinematic rupture description; presented at SSA 2007 –v 1.1 was superceded by v 1.2; v 1.2 included NW-to-SE and bi- lateral URS/USC - v 1.1, v 1.2 (SE-NW, NW-SE, bi-lateral) CMU/PSC - v 1.1 SDSU/SDSC - v 1.1 UCSB/Caltech - v 1.1, v 1.2 (SE-NW, NW-SE, bi-lateral)

ShakeOut v 1.1 vs. v 1.2 Surface slip - similar but ‘rougher’ in v 1.2 Ground motions - similar in v 1.2 and v 1.1; comparisons shown at sites of tall buildings Courtesy of Rob Graves; URS, SCEC/CME

Kinematic rupture models Kinematic Rupture Description - Aagaard, Hudnut & Jones –Two SCEC workshops; CGS/USGS/SCEC - ‘A-faults’ & SoSAFE –1D slip distribution construction - background slip model –Mapped onto SCEC CFM and CFM-tin fault geometry –Source physics ‘first principals’ and empirical relations to construct details of the kinematic rupture description; presented at SSA 2007 –v 1.1 was superceded by v 1.2; v 1.2 included NW-to-SE and bi- lateral URS/USC - v 1.1, v 1.2 (SE-NW, NW-SE, bi-lateral) CMU/PSC - v 1.1 SDSU/SDSC - v 1.1 UCSB/Caltech - v 1.1, v 1.2 (SE-NW, NW-SE, bi-lateral)

Kinematic rupture comparison

Rupture unzips SE to NW, taking 90 seconds; Shaking lasts >3 min.’s in LA & Ventura Southern California Earthquake Center

Kinematic –Forward model –A priori fault geometry (SCEC CFM), slip and rupture process –Ground motions computed are physics- based –4 groups took this approach and produced similar ground motions (URS/USC, CMU/PSC, SDSU/SDSC, UCSB/Caltech) Dynamic –Spontaneous rupture modeled –A priori fault geometry (vertical - simplified) and friction law on the fault specified –Match surface slip of ShakeOut kinematic rupture description –Work is ongoing by SCEC/CME group Dynamic vs. Kinematic

Dynamic Model (Kim Olsen, Luis Dalguer, Steve Day)

Onset and Duration of Shaking

Kenneth W. Hudnut, Ph.D. Geophysicist, USGS 525 S. Wilson Ave. Pasadena, CA (626)