Geoinformatics: A Petascale Cyberfacility for Physics-Based Seismic Hazard Analysis (PetaSHA-2)
The SCEC Petascale Cyberfacility for Physics-Based Seismic Hazard Analysis (PetaSHA-2) (NSF EAR - 074493) Project, funded by the National Science Foundation (NSF), began work on April 1, 2008. This NSF-funded research Project performs basic seismic hazard research using high-performance computing technologies. The PetaSHA-2 continues and extends the work done on the PetaSHA-1 project, supports the research agenda of the core SCEC program, and complements the work done on the OCI-supported PetaShake Project. The PetaSHA-2 Project seeks to advance seismic hazard research through the use of Petascale computing facilities as they become available to the NSF research community.
Earthquakes are an emergent (system-level) behavior that arises from nonlinear, multiscale interactions within cmplex fault systems. Earthquake system science seeks a basic understanding of how matter and energy interact within the lithosphere to produce seismic phenomena. Its practical mission is to provide society with better predictions of earthquake hazards. Dynamic simulations of fault ruptures and seismic wave propagation are proving to be important tools, but they cannot yet achieve the capability and capacity needed to explore important domains of earthquake behavior. Our goal is to take earthquake system science to a new level using the petascale computational resources developed by NSF/OCI. In the 2-year PetaSHA-2 Project we will employ high-performance NSF computers to achieve four science objectives:
- Improve the resolution of dynamic rupture simulations by an order of magnitude to investigate realistic friction laws, near-fault stress states, and off-fault plasticity.
- Investigate the upper frequency limit of deterministic ground-motion prediction by simulating strong motions above 1 Hz using realistic 3D structural models for Southern California.
- Validate and improve the Southern California structural models using full 3D waveform tomography.
- Transform probabilistic seismic hazard analysis (PSHA) into a physics-based science.
This PetaSHA-2 Project, conducted by the cross-disciplinary, multi-institutional CME Collaboration is the follow-on to a 2-year EAR/IF project with the same name (PetaSHA-1), begun in October, 2006. We are developing a cyberfacility with a common simulation framework for executing SHA computational pathways, including two new production platforms, DynaShake and CyberShake, that employ advanced workflow management tools on TeraGrid and other machines to compute and store the large suites (>1K) of dynamic fault rupture (DFR) and anelastic wave propagation (AWP) simulations needed for physics-based PSHA mapping. These high-capacity, data-intensive computing platforms will deliver PSHA results to end-users through two flexible delivery platforms, OpenSHA, developed under ITR support, and BroadBand, proposed here. The EAR-supported platforms will be enhanced by the high-capability PetaShake platform we are proposing to build under NSF/OCI’s PetaApps program. The leveraging between EAR and OCI will be favorable to both programs.
The extensions of the PetaSHA cyberfacility will provide the common functionalities needed by all SHA computational platforms including verification and validation, computational scalability, and large-scale data management. Based on our recent success in using full 3D waveform tomography (F3DT) to improve the SCEC Community Velocity Model in the LA region, we plan to augment the PetaSHA facility with a F3DT platform that will allow researchers to improve 3D structural models by inverting large suites of observed broadband waveforms. We will move the PetaSHA platforms forward in complexity and scale through a graduated series of milestone calculations tied to a timeline with clear scientific objectives and quantitative measures of success. We will integrate the computational platforms with existing and future geoinformatics systems, and we will make the platforms and their databases available to the broader communities of SCEC, EarthScope, and NEES researchers through science gateways.