Workshop on Fault Segmentation and Fault-To-Fault Jumps in Earthquake Rupture (March 15-17, 2006) Convened by Ned Field, Ray Weldon, Ruth Harris, David Schwartz, & David Oglesby on behalf of the WGCEP (

Agenda Weds & Thurs: Hear from experts and discuss Friday: Decide what WGCEP should do Workshop Goals To determine the spatial extent of all possible earthquakes on explicitly modeled faults in California (reevaluate current segmentation models and explore alternatives) time dependence at a future workshop

Working Group on California Earthquake Probabilities (WGCEP) Development of a Uniform California Earthquake Rupture Forecast (UCERF)

To provide the California Earthquake Authority (CEA) with a statewide, time-dependent ERF that uses “best available science” and is endorsed by the USGS, CGS, and SCEC, and is evaluated by Scientific Review Panel (SRP) and CEPEC Coordinated with the next National Seismic Hazard Mapping Program (NSHMP) time-independent model CEA will use this to set earthquake insurance rates (they want 5-year forecasts, maybe 1-year in future) WGCEP Goals:

Funding: Most from CGS, SCEC, & USGS $1,750,000 from CEA (22% of total)

NSF CEA USGS CGS SCEC MOC State of CA USGS Menlo Park USGS Golden Sources of WGCEP funding Geoscience organizations Management oversight committee WGCEP ExCom Subcom. A Subcom. B Subcom. C … … Working group leadership Task-oriented subcommittees Working Group on California Earthquake Probabilities WGCEP Organization & Funding Sources SCEC will provide CEA with a single-point interface to the project. SRP Scientific review panel

WGCEP Management Oversight Committee (MOC):  SCEC Thomas H. Jordan (CEA contact)  USGS, Menlo Park  Rufus Catchings  USGS, Golden Jill McCarthy  CGS Michael Reichle WGCEP Management: In charge of resource allocation and approving all project plans, budgets, and schedules Their signoff will constitute the SCEC/USGS/CGS endorsement

Responsible for convening experts, reviewing options, making decisions, and orchestrating implementation of the model and supporting databases Role of leadership is not to advocate models, but to accommodate whatever models are appropriate WGCEP Executive Committee:  Edward (Ned) Field; SCEC/USGS, Pasadena  Thomas Parsons, USGS, Menlo Park  Chris Wills, CGS  Ray Weldon, SCEC/UofO  Mark Petersen, USGS, Golden  Ross Stein, USGS, Menlo Park Key Scientists: Provide expert opinion and/or specific model elements - likely receiving funding & documenting their contributions. Contributors

Scientific Review Panel: Bill Ellsworth (chair) Art Frankel David Jackson Jim Dieterich Lloyd Cluff Allin Cornell Mike Blanpied David Schwartz CEPEC: Lucile JonesDuncan Agnew Tom JordanMike Reichle Jim BruneDavid Openheimer William LettisPaul Segall John Parrish This group will ultimately decide whether we’ve chosen a minimum set of alternative models that adequately spans the range of viable 5-year forecasts for California

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box (A) (B) (C) (D) UCERF Model Components Fault Model(s) Fault-slip rates & Aseismicity Estimates (at least) Long-term rate of all possible events (on and off modeled faults) Time-dependent probabilities Geometries (& alternatives)

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box (A) (B) (C) (D) Fault Model(s) Fault Section Database Paleo Sites Database GPS Database Historical Qk Catalog Instrumental Qk Catalog UCERF Model Components

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box (A) (B) (C) (D) Fault Model(s) Fault Section Database Paleo Sites Database GPS Database Historical Qk Catalog Instrumental Qk Catalog UCERF Model Components

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box (A) (B) (C) (D) Fault Model(s) Fault Section Database Paleo Sites Database GPS Database Historical Qk Catalog Instrumental Qk Catalog UCERF Model Components

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box (A) (B) (C) (D) Fault Model(s) Fault Section Database Paleo Sites Database GPS Database Historical Qk Catalog Instrumental Qk Catalog UCERF Model Components

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box (A) (B) (C) (D) Fault Model(s) Fault Section Database Paleo Sites Database GPS Database Historical Qk Catalog Instrumental Qk Catalog UCERF Model Components

Main Delivery Schedule February 8, 2006 (to CEA) UCERF 1 & Aug 31, 2006 (to CEA) Earthquake Rate Model 2 (preliminary for NSHMP) June 1, 2007 (to NSHMP) Final, reviewed Earthquake Rate Model 2 (for use in 2007 NSHMP revision) September 30, 2007 (to CEA) UCERF 2 (reviewed by SRP and CEPEC)

1)Statewide model 2)Use of SCEC CFM (including alternatives) 3)Use GPS data via kinematically consistent deformation model(s) 4)Relax strict segmentation (if appropriate) 5)Allow fault-to-fault jumps (if appropriate) 6)Apply consistent time-dependent models (esp. with (4) & (5)) 7)Include earthquake triggering effects 8)Deploy as extensible, adaptive (living) model. 9)Simulation enabled Issues/Possible Innovations

UCERF 1) Statewide model Issues/Possible Innovations WGCEPs:

2) Use of SCEC Community Fault Model (including alternatives) Issues/Possible Innovations

3) Use GPS data via kinematically consistent deformation model(s) Issues/Possible Innovations e.g., Peter Bird’s NeoKinema

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box Fault Models(s) We want: 1) Improved slip rates on major faults 2) Strain rates elsewhere 3) GPS data included GPS data from Geodesy group 4) Kinematically consistent 5) Accommodate all important faults 6) Can accommodate alternative fault models 7) Accounts for geologic and geodetic data uncertainties 8) Includes viscoelastic effects 9) Includes significant 3D effects 10) Statewide application

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box Fault Models(s) We want: 1) Improved slip rates on major faults 2) Deformation rates elsewhere 3) GPS data included 4) Kinematically consistent 5) Accommodate all important faults 6) Can accommodate alternative fault models 7) Accounts for geologic and geodetic data uncertainties 8) Includes viscoelastic effects 9) Includes significant 3D effects 10) Statewide application Are any existing models better than sticking to what we have? WGCEP recommending pursuit of: NeoKinema Harvard-MIT Block Model Parson’s FEM Shen & Zeng Perhaps others … No model has all these attributes

Black Box Deformation Model(s) Earthquake Prob Model(s) Earthquake Rate Model(s) Black Box Black Box Fault Models(s) Are any existing models better than sticking to what we have? WGCEP recommending pursuit of: NeoKinema Harvard-MIT Block Model Parson’s FEM Shen & Zeng Perhaps others … No model has all these attributes Delivery will be revised slip rates on modeled faults (and perhaps deformation rates elsewhere, and stressing rates on faults)

4) Relax strict segmentation Issues/Possible Innovations All previous WGCEPs have assumed strict segmentation, more recently allowing both single- and multi-segment ruptures (cascades). e.g., WGCEP-2002:

4) Relax strict segmentation Issues/Possible Innovations But... Viable interpretations of S. SAF paleoseismic data (Weldon et al.)

4) Relax strict segmentation Issues/Possible Innovations Does it matter (all models are discretized to some extent)? SJF; SB-SJV segment intersection SAF: Mojave-San Bernardino intersection 50% in 50 yrs

5) Allow fault-to-fault jumps Issues/Possible Innovations No previous WGCEPs have allowed such ruptures.

5) Allow fault-to-fault jumps Issues/Possible Innovations But...

5) Allow fault-to-fault jumps Issues/Possible Innovations Fault Interactions and Large Complex Earthquakes in the Los Angeles Area Anderson, Aagaard, & Hudnut (2003, Science 320, ) “… We find that … a large northern San Jacinto fault earthquake could trigger a cascading rupture of the Sierra Madre-Cucamonga system, potentially causing a moment magnitude 7.5 to 7.8 earthquake on the edge of the Los Angeles metropolitan region.

5) Allow fault-to-fault jumps Issues/Possible Innovations Can dynamic rupture modelers help define fault- to-fault jumping probabilities?

6) Figure out how to apply elastic-rebound-motivated renewal models “properly” Issues/Possible Innovations Problem: how to compute conditional time-dependent probabilities when you allow both single and multi- segment ruptures (let alone relaxing segmentation)? There seems to be a logical inconsistency with the way previous WGCEPs have modeled this …

From WGCEP-2002:

How then do we compute conditional probabilities where we have single and multi-segment ruptures? We have ideas …

7) Include earthquake triggering effects Issues/Possible Innovations CEA wants 1- to 5-year forecasts This is between the 30-year forecasts of previous WGCEPs (renewal models) and the 24-hour forecasts of STEP (the CEPEC-endorsed model based on aftershock statistics) We are attempting to design a framework that could accommodate a variety of alternative approaches (e.g., from RELM)

8) Deploy as extensible, adaptive (living) model. Issues/Possible Innovations i.e., modifications can be made as warranted by scientific developments, the collection of new data, or following the occurrence of significant earthquakes. The model can be “living” to the extent that update & evaluation process can occur in short order. CEA wants this.

9) Simulation enabled (i.e, can generate synthetic catalogs) Issues/Possible Innovations Needed to go beyond next-event forecasts if stress interactions and earthquake triggering are included Helpful (if not required) to understand model behavior Can be used to calibrate the model (e.g., % moment in aftershocks) If we can deal with simulated events, we’ll be ready for any real events

Implementation Plan Guiding principles: If it ain’t broke, don’t fix it Some of the hoped-for innovations won’t work out Everything will take longer than we think Build components in parallel (not in series) Get a basic version of each component implemented ASAP, and add improved versions when available We cannot miss the NSHMP and CEA delivery deadlines!

Black Box Deformation Model 1.0 Earthquake Prob Model 1.0 Earthquake Rate Model 1.0 Black Box Black Box Fault Model 1.0 UCERF 1.0 NSHMP-2002 Fault Model NSHMP-2002 Fault Slip Rates NSHMP-2002 Earthquake Rate Model Simple conditional probabilities based on date of last earthquakes By Feb. 8, 2006

Black Box Deformation Model 2.0 Earthquake Prob Model 2.0 Earthquake Rate Model 2.0 Black Box Black Box Fault Models 2.X UCERF 2.0 Revision of NSHMP-2002 Fault Model based on SCEC CFM (including alternatives) & any desired changes for N. California Revised slip rates for elements in Fault Model 2.0, perhaps constrained by GPS data. New model based on reevaluation of fault segmentation and cascades (e.g., based on Weldon et al.); may relax segmentation and allow fault-to-fault jumps. Application of more sophisticated time- dependent probability calculations We can at least deliver this for the NSHMP 2007 time- independent hazard maps (by June 1, 2007) & This to CEA By Sept. 30, 2007

Black Box Deformation Model 3.0 Earthquake Prob Model 3.0 Earthquake Rate Model 3.0 Black Box Black Box Fault Models 2.X UCERF 3.0 e.g., relax segmentation, allow fault- to-fault jumps, and use off-fault deformation rates to help constrain off-fault seismicity. e.g., enable real-time modification of probabilities base on stress or seismicity- rate changes. Use of a more sophisticated, California- wide deformation model such as NeoKinema.

Black Box Deformation Model 3.0 Earthquake Prob Model 3.0 Earthquake Rate Model 3.0 Black Box Black Box Fault Models 2.X UCERF 3.0 e.g., relax segmentation, allow fault- to-fault jumps, and use off-fault deformation rates to help constrain off-fault seismicity. e.g., enable real-time modification of probabilities base on stress or seismicity- rate changes. Use of a more sophisticated, California- wide deformation model such as NeoKinema. All of these are relatively ambitious and delivery cannot be guaranteed by 2007.

Agenda Weds & Thurs: Hear from experts and discuss Friday: Decide what WGCEP will do Workshop Goals To determine the spatial extent of all possible earthquakes on explicitly modeled faults in California (reevaluate current segmentation models and explore alternatives)

“Co-Seismic”? What’s the difference between “co-seismic” and “triggered quickly”? 1)Constructive interference (e.g., directivity) 2) My Earthquake Insurance: “One or more earthquake shocks that occur within a seventy-two hour period constitute a single earthquake” 3) California Earthquake Authority: “seismic event” = everything within a 15-day window

Current Practice (WGCEP-2002, NSHMP-2002, UCERF 1) ******************

NSHMP-2002 & UCERF 1 Southern California Type A Faults  = 0.12 Trunc +/-1.25 

NSHMP-2002 Bay Area Faults & WGCEP-2002  = 0.12 Trunc +/-2 

NSHMP-2002 B Faults & UCERF 1 2/3 Mo Char 1/3 Mo GR

An Alternative Approach(?) (extension of Andrews & Schwerer (2000) w/ elements of Field et al. (1999)) 1) Divide faults into small sections (~5 to ~10 km in length) 2) Define “all possible ruptures” as every possible combination of “contiguous” sections (“contiguous” means separated by less than ~10 km) Use matrix I rs to indicate whether r th rupture involves the s th segment 3) Get magnitude of each rupture from mag(area) relationship; this gives average slip (u r ) of each as well

4) Solve for the long term rate of each rupture (f r ) via linear inversion of: An Alternative Approach(?) f(m) = GR (for entire region) (e.g., trench constraint) f r ≥ 0 (positivity for all r) (moment balance) Anything else (e.g., rupture boundaries) … Solution space represents all possible models f r = f obs (e.g., Parkfield) A formal, tractable, extensible, mathematical framework

What can geology or dynamic modeling tell us about what ruptures are more or less probable?

How do we define and apply these constraints in our inversion? Have the simulators solved this problem?

An Alternative Approach(?) (extension of Andrews & Schwerer (2000) w/ elements of Field et al. (1999)) 1) Divide faults into small sections (~5 to ~10 km in length) 2) Define “all possible ruptures” as every possible combination of “contiguous” sections (“contiguous” means separated by less than ~10 km) Use matrix I rs to indicate whether r th rupture involves the s th segment 3) Get magnitude of each rupture from mag(area) relationship; this gives average slip (u r ) of each as well

4) Solve for the long term rate of each rupture (f r ) via linear inversion of: An Alternative Approach(?) f(m) = GR (for entire region) (e.g., trench constraint) f r ≥ 0 (positivity for all r) (moment balance) Anything else (e.g., rupture boundaries) … Solution space represents all possible models f r = f obs (e.g., Parkfield) A formal, tractable, extensible, mathematical framework

What can geology or dynamic modeling tell us about what ruptures are more or less probable?

How do we define and apply these constraints in our inversion? Have the simulators solved this problem?