High-resolution predictions of coastal storm surge, inundation, and waves generally require substantial computational resources, particularly in a forecast, real-time mode. Regional forecast systems using the ADCIRC tide, storm surge, and wave model have been operating in several risk-prone areas of the eastern US and Gulf coasts for the past several years using the Adcirc Surge Guidance System (ASGS). ADCIRC uses triangular finite element to represent the spatial domain, with the grids generally high-resolution regional discretizations of the topography, bathymetry, critical hydraulic structures, and land cover. Each ASGS instance uses locally provisioned compute resources and publishes its output in a standardized way. However, the distributed nature of ASGS instances requires an efficient way to discover, gather, organize and visualize the distributed results from multiple forecast system instances and other models.
UNC’s Renaissance Computing Institute (RENCI) has developed a system to federate ASGS instances and provide uniform access to forecasts on unstructured model grids. This simplifies knowing which systems are running and when solutions become available. This data federation is enabled by using netCDF4 and UGRID (a netCDF Climate and Forecast metadata convention extension for unstructured grids, http://bit.ly/ugrid_cf) standards for file formats and metadata content; THREDDS and OPeNDAP for data cataloging, data discovery, and data access; and the NOAA-funded MATLAB toolbox NCTOOLBOX (http://code.google.com/p/nctoolbox/), to access the ASGS predictions. A simple cataloging mechanism maintains a catalog file on the THREDDS server, and will be extended to harvest simulation information from a collection of THREDDS servers.
To access and visualize ASGS results, we have developed AdcircViz, a MATLAB-based visualization application that provides robust access to ASGS forecasts in the data grid. This application is not limited to ADCIRC model output. Any coastal ocean model that publishes its output to the data grid is accessible by applications that are aware of the UGRID standard. In this poster, we overview the methods used, the ASGS system outputs, and use of UGRID to enable rapid and robust application development.