Chino Hills --- A highly computationally efficient 2 Hz validation exercise

The 2008 Chino Hills earthquake was the largest earthquake in the Los Angeles metropolitan region since the 1994 Northridge earthquake. With a magnitude Mw 5.4, the July 29, 2008 Chino Hills earthquake was recorded by most networks in the area. Its occurrence constitutes an excellent opportunity to study the response of the greater Los Angeles basin and to test the most common assumptions for crustal structure and material properties under ideal conditions of anelastic modeling due to a kinematic point source excitation. We present here a preliminary validation study for a set of simulations of the Chino Hills earthquake using Hercules---the parallel octree-based finite element simulator developed by the Quake Group at Carnegie Mellon University. In the past, we have reported on the simulation capabilities of Hercules for more complex---yet hypothetical---earthquake scenarios such as TeraShake and ShakeOut. For the latter, we have also conducted a comprehensive verification of results in collaboration with other SCEC simulation groups, using different methodologies. With this new simulation we attempt to come full circle in the verification and validation paradigm as understood by the modeling and simulation community. The results presented here correspond to a set of four different simulations, the most challenging one with a maximum frequency of 2 Hz and a minimum shear wave velocity of 200 m/s. These particular values of these two critical parameters help us explore the influence of higher frequencies and lower velocity profiles on ground motion. The extension to these parameters is becoming possible in our simulations thanks to the latest computational improvements we are implementing into Hercules. While our focus is on the physical interpretation of the results of our simulations and their comparison with observations, we also report on the computing resources employed. Our preliminary results suggest that extending the maximum frequency beyond the de facto assumed barrier of 1 Hz for deterministic simulations is an effort worth pursuing.