Link to NGI

Click for movie from
Super Computing 2007
(82.5MB)


VISUALIZATION TECHNIQUES FOR IMPROVING
PUBLIC UNDERSTANDING OF CATASTROPHIC EVENTS




PI: Robert Moorhead, Geosystems Research Institute, Mississippi State University


One of the greatest challenges to an appropriate public response to emergencies is accurate and easily understood information. The general populace can readily become so overloaded with information that individuals will either not realize the magnitude of the crisis and thus not prepare or respond adequately, or overreact and evacuate when such is not warranted. As modeling and forecasting improve, one important facet of public awareness that has not been sufficiently addressed is visualization of the data in such a way that the information is easily understood, and provides an accurate spatial depiction of the threat. This project will focus on developing new 2D and 3D visualization tools which produce visualization products that can be made publicly available, are easily interpreted by the non-technical public, and can be viewed on personal computers or used in television coverage.

The initial efforts will focus in two areas: storm surge and hurricane intensity/direction. This project will capitalize on the high performance computing and visualization capabilities at MSU, but will be closely linked to the severe weather modeling activities at MSU and FSU, and with partner activities at several NOAA units (including AOML, CSC, and NCDDC). It will focus on using HPC for both modeling and development of the visualization of model output, and will then create visualization products that can be produced as simple animations or static images.

The objectives of this project are to develop a hardware/software system which allows analyst with access to many and large data sources to see those many datasets in the viewing "environment" which allows them to extract the maximum amount of information and then knowledge from the datasets; extend existing visual analysis systems to ingest more data formats and types, perform data fusion in more automated ways, and display the data in more ways (e.g., 2D fields as contours, points, filled surfaces, glyphs); and study the optimal method to display various sets of multiple co-located datasets (topography, bathymetry, coastline, oceanography, and atmosphere) in the same view volume.