QuakeSim: Enabling Model Interactions in Solid Earth Science Sensor Webs

QuakeSim is problem-solving environment for understanding earthquake processes through the integration QuakeSim: of multiscale models and data. The goal of QuakeSim is to substantially improve earthquake forecasts, which will ultimately lead to mitigation of damage from this natural hazard. Improved earthquake forecasting is dependent on measurement of surface deformation as well as analysis of geological and seismological data. Space-borne technologies, in the form of continuous GPS networks and InSAR satellites, are the key contributors to measuring surface deformation. We are expanding our QuakeSim Web Services environment to integrate, via Ontolody-based Federation, both real-time and archival sensor data with high-performance computing applications for data mining and assimilation. We are federating sensor data sources, with a focus on InSAR and GPS data, for an improved modeling environment for forecasting earthquakes. These disparate measurements form a complex sensor web in which data must be integrated into comprehensive multi-scale models. In order to account for the complexity of modeled fault systems, investigations must be carried out on high-performance computers. We are building upon our "Grid of Grids" approach, which included the development of extensive Geographical Information System-based "Data Grid" services. We are extending our earlier approach to integrate the Data Grid components with improved "Execution Grid" services that are suitable for interacting with high-end computing resources. These services are being deployed on the Columbia computer at NASA Ames and the Cosmos computer cluster at JPL.