CyberShake Science Plan
CyberShake Science Plan
This entry describes scientific elements and development plans for our CyberShake calculations as we prepare for a state-wide UCERF 3.0 CyberShake calculation.
CyberShake 1.0 was developed and run in 2009 completing 220+ sites using UCERF2.0 and CVM-S4 and Graves 1.0 Rupture Generator with up to 0.5Hz. Since the 2009 production run, we have continued to run CyberShake hazard curves for sites of interest including precarious rock sites, seismic station sites, and other sites of interest. We have continued to run CyberShake hazard curves, based on UCERF2.0 at 0.5Hz. These production runs have evaluated alternative aspects including the impact of updated rupture generator CyberShake 1.1, alternative CVM's CyberShake 1.3, and addition of 10Hz stochastic frequencies CyberShake 1.4.
CyberShake 2.0 is the current work activity. This calculation will use UCERF2.0, both CVM-S4 and CVM-H, and Graves 2.0 Rupture Generator up to 1.0Hz. This is what we are currently working on.
CyberShake 3.0 is a completion target for our PetaSHA3 Project. We are preparing to run this calculation on NSF and DOE computers. We plan to call the results the CyberShake 3.0 Hazard Model. Here are the current plans for the various science areas.
CyberShake 3.0 will use UCERF3.0 that is scheduled for release in June 2012. Our current system can convert UCERF2.0 to a 0.5Hz extended ERF. Unless more advanced rupture generators are developed, we will extend the current Graves 2.0 rupture generator to produce SRF-format rupture variations with frequencies up to 1.0Hz. In UCERF3.0, science changes that might affect our extended earthquake rupture forecast generator include Cascadia subduction events, and ruptures that involve fault-to-fault jumps.
Our current CyberShake 3.0 plan is to use a state-wide UCVM-CyberShake based on Po Chen’s inversion results from existing Southern and Northern California models, as presented at the 2011 CME meeting.
We plan to increase the current Graves 2.0 rupture generator output frequency from 0.5Hz to 1.0Hz. We need UCERF2.0 fault geometries at higher resolution to support 1.0Hz ruptures. OpenSHA group can provide those for both UCERF2.0 and UCERF3.0. Rupture generator will need to support new capabilities of UCERF3.0.
We plan to calculate 3 components SGT’s and calculate 3 component seismograms, minimizing to 2 components if necessary. Our deterministic seismograms simulations will use minVs of 500m/s with frequencies up to 1.0Hz.
We will add stochastic 10Hz seismograms using the Graves (URS) stochastic codes from the Broadband platform v11.2.
Our CyberShake 3.0 simulation goal is 5000 sites. This will give approximately 10km spacing across California plus precarious rock sites and seismic recording stations.
Data sets to be delivered include the following: the UCERF3.0 ERF, the UCERF3.0 Extended ERF, and the UCVM-CyberShake.
Resulting data sets to be delivered include the following: PSHA Hazard Map for California using UCERF3.0 and comparable to standard California PSHA maps.
For every rupture variation in UCERF3.0, plot a scenario ShakeMap based on 3d wave propagation.
For each site, we will calculate UCERF3.0-based PSHA hazard curves for intensity types PGA, PGV, Peak Spectral Accel from 10 seconds to 0.1 seconds period,1Hz and 10Hz seismograms for each rupture within 200km, and a database of peak ground motion amplitudes for all rupture in ERF affecting a site.