The Uniform California Earthquake Rupture Forecast, Version 3 (UCERF3) by the Working Group on California Earthquake Probabilities (WGCEP) presented here provides the rate of earthquakes projected over a long time span (independent of the date of the most recent earthquakes) on major faults and “background seismicity” that may occur on minor faults or unknown faults that are not modeled individually. This model represents a significant update in the input data and major methodological changes from the previous model (UCERF2, published in 2008). The new model uses an approach that takes real world observations and finds the range of possible solutions, known in technical terms as an inverse problem. In UCERF3, the available data that can be used to project future earthquake rates: fault locations and slip rates, deformation rates from geodetic measurements, rates of historic earthquakes, and rates of pre-historic earthquakes from paleoseismic investigations, are used in a system-level “grand inversion” to find the rates of earthquakes that are consistent with the input data. This inversion allowed the WGCEP to avoid assumptions, including fault segmentation and magnitude-frequency distributions that were applied to all faults in previous models. The “grand inversion” is a large and underdetermined inverse problem, so samples a range of possible models rather than solving for a single correct model. The inversion also accounts for epistemic uncertainties, using 1,440 alternative logic tree branches. Consideration of epistemic uncertainties in input values and sampling of the wide range of possible output required supercomputers to generate the resulting model.
Notable achievements of the UCERF3 model include an evaluation of the frequency that multiple sections of faults, or multiple faults, will rupture together in larger earthquakes. Most large multi-fault ruptures are extremely unlikely, but including them in the model will allow for evaluation of the consequences of large, rare events. Input to the model includes three deformation models based on kinematically consistent inversions of geodetic and geologic data, which for the first time provide slip-rate constraints on faults previously excluded because of lack of geologic data. The model does not include input models following the hypothesis that individual faults have a Gutenberg-Richter magnitude-frequency relationship, because such models could not be reconciled with the other input data, demonstrating serious challenges for that hypothesis. The UCERF3 model results are constrained to be as close to the UCERF2 model without violating any of the data. Nevertheless, UCERF3 removes the apparent UCERF2 overprediction of M6.5-7 earthquake rates and also includes types of multifault ruptures seen in nature. The grand inversion constitutes a system-level framework for testing hypotheses and balancing the influence of different experts and types of data. The most influential uncertainties include alternative deformation models (fault slip rates), a new smoothed seismicity algorithm, alternative values for the total rate of M≥5 events, and different scaling relationships, virtually all of which are new. The UCERF3 model represents the most complete evaluation of California earthquake probabilities to date, and will form the basis for revisions to upcoming versions of the National Seismic Hazard Maps.
Special Report 228, The Uniform California Earthquake Rupture Forecast version 3.
Click here to download Special Report 228 (PDF Document, 32 MB) also published as USGS Open File Report 2013-1165. The report includes the appendices listed below. Click here to link to the appendices.
A. Updates to the California Reference Fault Parameter Database: UCERF3 Fault Models 3.1 and 3.2 by Timothy E. Dawson
B. Geologic Slip Rate Data and Geologic Deformation Model by Timothy E. Dawson and Ray J. Weldon, II
C. Deformation Models for UCERF3 by Thomas Parsons, Kaj M. Johnson, Peter Bird, Jayne M. Bormann, Timothy E. Dawson, Edward H. Field, William C. Hammond, Thomas A. Herring, Rob McCaffrey, Zhen-Kang Shen, Wayne R. Thatcher, Ray J. Weldon, II, and Yuehua Zeng
D. Compilation of Creep Rate Data for California Faults and Calculation of Moment Reduction Due to Creep by Ray J. Weldon, II, David A. Schmidt, Lauren J. Austin, Elise M. Weldon, and Timothy E. Dawson
E. Evaluation of Magnitude-Scaling Relationships and Depth of Rupture by B.E. Shaw
F. Distribution of Slip in Ruptures by Glenn P. Biasi, Ray J. Weldon, II, and Timothy E. Dawson
G. Paleoseismic Sites Recurrence Database by Ray J. Weldon, II, Timothy E. Dawson, Glenn P. Biasi, C. Madden, and Ashley R. Streig
H. Maximum Likelihood Recurrence Intervals for California Paleoseismic Sites by Glenn P. Biasi
I. Probability of Detection of Ground Rupture at Paleoseismic Sites by Ray J. Weldon, II, and Glenn P. Biasi
J. Fault-to-Fault Rupture Probabilities by Glenn P. Biasi, Thomas Parsons, Ray J. Weldon, II, and Timothy E. Dawson
K. The UCERF3 Earthquake Catalog by K.R. Felzer
L. Estimate of the seismicity rate and magnitude-frequency distribution in California from 1850–2011 by K.R. Felzer
M. Adaptive Smoothed Seismicity Model by K.R. Felzer
N. Grand Inversion Implementation and Exploration of Logic Tree Branches by Morgan T. Page, Edward H. Field, and Kevin R. Milner
O. Gridded Seismicity Sources by Peter M. Powers, and Edward H. Field
P. Models of Earthquake Recurrence and Down-Dip Edge of Rupture for the Cascadia Subduction Zone by Arthur D. Frankel and Mark D. Petersen
Q. The Empirical Model by K.R. Felzer
R. Compilation of Slip in the Last Event Data and Analysis of Last Event, Repeated Slip, and Average Displacement for Recent and Prehistoric Ruptures by Christopher Madden, David E. Haddad, J. Barrett Salisbury, Olaf Zielke, J. Ramón Arrowsmith, Ray J. Weldon, II, and Javier Colunga
S. Constraining ETAS Parameters from the UCERF3 Catalog and Validating the ETAS Model for M≥6.5 Earthquakes by Jeanne L. Hardebeck
T. Defining the Inversion Rupture Set via Plausibility Filters by Kevin R. Milner, Morgan T. Page, Edward H. Field, Thomas Parsons, Glenn P. Biasi, and Bruce E. Shaw