Southern California Earthquake Center CME Project AllHands Meeting

1
Southern California Earthquake CenterCME
Project All-Hands Meeting

• SCEC HQ _at_ USC
• July 17-18, 2006

2
Meeting Goals

• End-game plan for the NSF/ITR CME project
• Final report due Jan 1, 2007
• Input to SCEC3 science plan
• Promote further integration of CME into SCEC
  organization
• Revised plan for new PetaSHA project
• 2-yr funding period
• More specific CS objectives and budget
• For selling to NSF/OCI

3
Topics

• Earthquake System Science
• Need for Physics-Based Seismic Hazard Analysis
  (SHA)
• SCEC2-3 Transition
• SHA Computational Pathways
• SCEC Community Modeling Environment (CME)
• Need for petascale resources
• CME Computational Platforms
• OpenSHA, CyberShake, TeraShake, PetaShake
• SCEC Path from Terascale to Petascale
• PetaSHA Milestone Simulations

4
Earthquake System Science

• The science of complex natural systems (system
  science) seeks to represent nature by models
  that describe system-level (emergent) phenomena
  and predict their future behavior
• Challenges in predicting earthquake phenomena
  exemplify those encountered in the study of other
  natural systems
• SCEC has become a national leader in the
  development of system science
• The system is defined by the behavior it seeks to
  explain
• In seismic hazard analysis, the defining behavior
  is the shaking intensity at a geographic site
• SHA drives SCEC efforts in earthquake system
  science
• Model-based prediction of emergent behaviors
  plays an essential role in a continually iterated
  cycle of data gathering and analysis, hypothesis
  testing, and model improvement
• System-specific models are the basis for
  synthesize knowledge from different disciplines
  into a common understanding
• California is the natural laboratory developing
  the SCEC community models

5
Seismic Hazard Analysis

• Definition Specification of the maximum
  intensity of shaking expected at a site during a
  fixed time interval
• Example National seismic hazard maps

• Intensity measure peak ground acceleration (PGA)
• Interval 50 years
• Probability of exceedance 2

6
Phenomena poorly represented by standard
probabilistic seismic hazard analysis (PSHA)

• Source directivity
• Amplification of ground motions in sedimentary
  basins
• Rupture complexity and scattering by 3D geologic
  structure
• Magnitude saturation

Physics-based simulations ? computational
platforms
San Andreas fault
7

• Southern California Earthquake Center
• Involves 500 scientists at 55 institutions
  worldwide
• Focuses on earthquake system science using
  Southern California as a natural laboratory
• Translates basic research into practical products
  for earthquake risk reduction

SCEC Focus Groups
8
Science Plan in SCEC3 Proposal

• P1. Earthquake Source Physics
• Discover the physics of fault failure and
  dynamic rupture that will improve predictions of
  strong ground motions and the understanding of
  earthquake predictability.
• P2. Fault System Dynamics
• Develop representations of the postseismic and
  interseismic evolution of stress, strain, and
  rheology that can predict fault system behaviors
  within the Southern California Natural
  Laboratory.
• P3. Earthquake Forecasting and Predictability
• Improve earthquake forecasts by understanding
  the physical basis for earthquake predictability.
• P4. Ground Motion Prediction
• Predict the ground motions from realistic
  rupture models at frequencies up to 10 Hz for all
  sites in Southern California.

9
SCEC3 Science Priority Objectives

• Improve the unified structural representation and
  employ it to develop system-level models for
  earthquake forecasting and ground motion
  prediction
• Develop an extended earthquake rupture forecast
  to drive physics-based PSHA
• Define slip rate and earthquake history of
  southern San Andreas fault system for last 2000
  years
• Determine the origin and evolution of on- and
  off-fault damage as a function of depth
• Test hypotheses for dynamic fault weakening
• Assess predictability of rupture extent and
  direction on major faults
• Investigate implications of geodetic/geologic
  rate discrepancies for earthquake forecasting
• Develop a system-level deformation and
  stress-evolution model for earthquake forecasting
• Map seismicity and source parameters in relation
  to known faults
• Develop a geodetic network processing system that
  will detect anomalous strain transients
• Test of scientific prediction hypotheses against
  reference models to understand the physical basis
  of earthquake predictability
• Predict broadband ground motions for a
  comprehensive set of large scenario earthquakes
• Develop pseudo-dynamic source models consistent
  with dynamic rupture models
• Determine the upper limits of extreme ground
  motion
• Investigate the upper frequency limit of
  deterministic ground motion predictions
• Validate earthquake simulations
• Collaborate with earthquake engineers to develop
  rupture-to-rafters simulation capability for
  physics-based risk analysis
• Prepare post-earthquake response

10
(No Transcript)
11
SCEC Initiatives

• Networks as Research Tools (e.g., Earthquake
  Early Warning)
• Southern San Andreas Initiative
• Working Group on California Earthquake
  Probabilities (WGCEP)
• Next Generation Attentuation (NGA) Project
• End-to-End (Rupture-to-Rafters) Simulation
• Collaboratory for the Study of Earthquake
  Predictability (CSEP)
• National Partnerships through EathScope
• Multinational Partnership in Earthquake System
  Science (MPRESS)
• Extreme Ground Motions
• Petascale Cyberfacility for Physics-Based Seismic
  Hazard Analysis (PetaSHA)
• Advancement of Cyberinfrastructure Careers
  through Earthquake System Science (ACCESS)

12
SCEC Initiatives
Funded

• Networks as Research Tools (e.g., Earthquake
  Early Warning)
• Southern San Andreas Initiative
• Working Group on California Earthquake
  Probabilities (WGCEP)
• Next Generation Attentuation (NGA) Project
• End-to-End (Rupture-to-Rafters) Simulation
• Collaboratory for the Study of Earthquake
  Predictability (CSEP)
• National Partnerships through EathScope
• Multinational Partnership in Earthquake System
  Science (MPRESS)
• Extreme Ground Motions
• Petascale Cyberfacility for Physics-Based Seismic
  Hazard Analysis (PetaSHA)
• Advancement of Cyberinfrastructure Careers
  through Earthquake System Science (ACCESS)

13
SCEC Initiatives
Funded
Pending

• Networks as Research Tools (e.g., Earthquake
  Early Warning)
• Southern San Andreas Initiative
• Working Group on California Earthquake
  Probabilities (WGCEP)
• Next Generation Attentuation (NGA) Project
• End-to-End (Rupture-to-Rafters) Simulation
• Collaboratory for the Study of Earthquake
  Predictability (CSEP)
• National Partnerships through EathScope
• Multinational Partnership in Earthquake System
  Science (MPRESS)
• Extreme Ground Motions
• Petascale Cyberfacility for Physics-Based Seismic
  Hazard Analysis (PetaSHA)
• Advancement of Cyberinfrastructure Careers
  through Earthquake System Science (ACCESS)

14
Role of Simulation in Earthquake System Science

• System-level models are now important tools for
  basic earthquake science
• Facilitate for knowledge integration across
  different disciplines
• Quantify behaviors that emerge from complex
  interactions

• System-specific simulations are playing an
  increasingly important role in assessing
  earthquake hazard and risk
• Provide a quantitative framework for comparing
  hypotheses about earthquake behavior with
  observations
• Provide a physical basis for predictions in
  situations where little or no data exist

Empirical models
Intensity Measures
Earthquake Rupture Forecast
Attenuation Relationship
1
P(IMk)
P(IMk  Sn)
P(Sn)
15
SHA Computational Pathways
Standard seismic hazard analysis
1
Empirical models
Intensity Measures
Attenuation Relationship
Earthquake Rupture Forecast
Extended Earthquake Rupture Forecast
1
16
Physics-based PSHA requires prediction of
directivity and other rupture parameters
(extended ERF)
TeraShake Platform
Rupture direction NW?SE
Southernmost San Andreas M7.7 (Olsen et al. 2006)
Rupture direction NW?SE
Olsen et al. (2006)
17

• Southern California Earthquake Center
• Involves 500 scientists at 55 institutions
  worldwide
• Focuses on earthquake system science using
  Southern California as a natural laboratory
• Translates basic research into practical products
  for earthquake risk reduction

• SCEC Collaboratory
• Grid-enabled Community Modeling Environment (CME)
  developed under NSFs ITR Program
• Partnership with IT organizations in
  physics-based seismic hazard analysis

18
SCEC Community Modeling EnvironmentA
collaboratory for system-level earthquake science
Cyberinfrastructure layering of the SCEC
Collaboratory
19
CME Platforms
Implementations of computational pathways using
vertically integrated computational
configurations (hardware software wetware)
for physics-based seismic hazard analysis
P2 and P3 models P3 databases
TeraShake
PetaShake
P2 databases
capability computing
capability computing
CyberShake
Delivery to Users
Capacity data-intensive computing

• Attributes
• System-level scale range
• High-performance hardware
• IT/geoscience collaboration
• Validated software framework
• Workflow management tools
• Well-defined interface

OpenSHA
Data-intensive computing
20
Pathway 1 OpenSHA Platform
Time Span
OpenSHA A Computational Platform Seismic Hazard
Analysis
Earthquake- Rupture Forecast
IM
Rupn,i
Site
Type, Level
Sourcei
Intensity-Measure Relationship
Field, Jordan Cornell (2003)
21
TeraShake PlatformKinematic vs. Dynamic Rupture
22
CyberShake Platform

• Simulates ground motions for potential fault
  ruptures within 200 km of each site
• 12,700 sources in SoCal from USGS 2002 ERF
• Extends ERF to multiple hypocenters and slip
  models for each source
• 100,000 ground motion simulations for each site

23
CyberShake Platform
24
CyberShake Platform
25
SCEC3 Science Issues

• How reliable is the current generation of
  low-frequency ground-motion predictions
  (TeraShake, CyberShake)?
• Can large-event predictions be verified before
  the fact?
• What is the upper frequency limit for
  deterministic ground-motion prediction?
• How far can we extend this limit by improving 3D
  models of elastic structure?
• Is full-3D waveform tomography the most
  appropriate method for data assimilation into the
  CVM?
• Should new CVMs be based on the CBMs?
• What new data-gathering activities can elucidate
  key structural features?
• How should we deploy stochastic extensions for
  predicting ground motions at higher frequencies?
• What the physical limits of strong ground motions
  produced by large fault ruptures?
• How important are near-surface nonlinear effects?
• Are 1D models sufficient for ground-motion
  prediction?
• How do we convince earthquake engineers (and
  ourselves) that we can reliability predict strong
  motions?

26
PetaSHA Goals

• G1. Transform SHA into a physics-based science by
  deploying a cyberfacility that can execute SHA
  computational pathways and manage data volumes
  using the nations petascale computing resources
• G2. Use this cyberfacility to implement
  physics-based PSHA and validate the results with
  data from Southern California

27
Science Thrust Areas

• Investigate upper frequency limit of
  deterministic simulation
• Extend ground motion simulations to 3 Hz
• Investigate dynamic rupture complexity of large
  earthquakes
• Extend dynamic rupture simulations to outer/inner
  scale ratios of 104.5
• Compute physics-based PSHA maps for Southern
  California
• Validate them using seismic and paleoseismic data

28
PetaSHA Geoscience Objectives

• O1. Extend the upper frequency bound of
  ground-motion simulations (Pathway 2) from the
  current value of 0.5 Hz to 3 Hz. Investigate the
  upper frequency limit of deterministic
  ground-motion prediction by comparing the
  simulations with seismic data from Southern
  California earthquakes.
• O2. Extend the outer/inner scale ratio of dynamic
  rupture simulations (Pathway 3) from the current
  value of 103.5 to 104.5. Investigate the effects
  of realistic friction laws, geologic
  heterogeneity, and near-fault stress states on
  seismic radiation improve pseudo-dynamic rupture
  models and validate rupture models with seismic
  data.
• O3. Extend Pathway-2 simulations to broadband
  (0-10 Hz) using pseudo-dynamic rupture models
  improved via Pathway 3 and stochastic wave
  propagation methods. Provide broadband
  simulations to SCEC validation projects.

29
PetaSHA Geoscience Objectives

• O4. Incorporate additional geologic complications
  into the Pathway-2 and Pathway-3 simulations,
  including surface topography, non-planar faults,
  and nonlinear wave propagation effects, and
  assess their effects on simulation-derived hazard
  curves.
• O5. Demonstrate physics-based PSHA by calculating
  seismic hazard maps for Southern California using
  Pathway-2 simulations that adequately sample the
  NSHMP-2002 and WGCEP-2007 ERFs. Compare
  simulation-based hazard maps with those predicted
  by conventional PSHA. Validate the hazard curves
  using seismic data and paleoseismic constraints
  from SCEC studies of precarious rocks.
• O6. Provide digital libraries and computational
  capabilities that will facilitate the inversion
  of ground-motion data for 3D crustal structure
  and earthquake sources (Pathway 4), including the
  receiver Green tensors for rapid imaging of
  earthquake sources and the simulation capability
  needed for full-3D tomography.

30
Validation Using Seismic Data
Fontana 01/06/05
Yorba Linda 09/03/02
PGV Data (SH)
PGV Synthetic (SH)
31
Validation Using Precarious Rocks
UNR Database
32
PetaSHA Computer Science Objectives

• O7. Integrate PSHA workflows into the evolving
  national cyberinfrastructure, extending the
  scalability, usability, and robustness of the
  current CME. Achieve geoscience objectives by
  exploiting new high-end computing and storage
  resources.
• O8. Develop SHA codes that can efficiently
  utilize the tera-to-petascale computing resources
  that become available during the project.
• O9. Operate a distributed, high-capacity digital
  library that can manage petascale datasets from
  SHA simulations. Include facilities for
  replicating data, managing metadata and
  maintaining its integrity, and providing users
  with consolidated access across distributed
  storage resources.
• O10. Vertically integrate available
  cyberinfrastructure to create high-capability and
  high-capacity SHA computational platforms with
  workflow tools, grid-based middleware, advanced
  SHA application software, data analysis and
  visualization tools, digital libraries, and
  high-performance hardware (computing, networking,
  storage).
• O11. Develop science gateways utilizing
  service-oriented interfaces to enable seamless
  integration of SHA and PSHA data products and
  processes into the broader SCEC, EarthScope, and
  NEES Communities.

33
PetaSHA Workforce Development Objectives

• O13. Improve the CME successful collaborations
  between geoscientists and computer scientists
  directed toward socially relevant hazards
  research.
• O14. Cross-train diverse groups of undergraduate
  interns and early-career scientists in geoscience
  and computer science and motivate them to solve
  fundamental problems.

34
Petascale computing will be needed for SHA
simulations
Simulation Volumes V1 Northridge
domain V2 PSHA site volume V3 regional M7.7
domain V4 regional M8.1 domain
35
Meeting Goals

• End-game plan for the NSF/ITR CME project
• Final report due Jan 1, 2007
• Input to SCEC3 science plan
• Promote further integration of CME into SCEC
  organization
• Revised plan for new PetaSHA project
• 2-yr funding period
• More specific CS objectives and budget
• For selling to NSF/OCI

36
End