Extended Abstract (84K)P1.30
A Java Meteorological Data and Transport Model Visualization Tool (Formerly Paper Number 12.21)
Stephen Masters, ENSCO, Inc., Melbourne, FL; and M. Moore
Many of the transport-dispersion models, or other meteorological applications that process weather data, can be run as "black boxes." The user sends in weather observations or gridded forecast model data, and the application sends out the result of its calculations. These calculations may be the two- and three-dimensional air concentrations, locations of virtual particles, or the results of many other types of calculations. A critical part of the proper evaluation of sophisticated data processing tools is the comparison of the final output product with the meteorological fields driving the solution.
The Model Visualization Module (MVM) is a new application designed to overlay meteorological data from many different sources. These data may be pulled from local and remote formatted data files as well as Oracle databases on remote servers. The MVM is entirely written in Java and executes on PC, Sun Solaris, and SGI IRIX platforms. The MVM can current overlay the following types of data: 1) Surface and upper air observations, 2) 2.5-degree NCAR/NCEP Reanalysis, 3) 1-degree gridded data from NCEP and FNMOC, 4) gridded data from the RAMS mesoscale weather prediction model, 5) trajectories computed from four different atmospheric transport-dispersion models, 6) concentration isopleths from three transport dispersion models, and 7) isopleths from user-created data files.
In addition to the map overlays, the MVM can extract vertical profiles from upper air soundings and three-dimensional gridded datasets to be plotted on skew-T/log-P displays. Observed vertical data profiles may be quickly compared to their gridded counterparts. This tool also greatly aids the comparison between gridded data from different sources.
Each of the map overlays may be linked in time, allowing the user to scroll through multiple datasets with ease. Data may be easily viewed on multiple scales from global down to fine local scales. Using the MVM, the user can easily judge the correspondence between many combinations of meteorological data, such as concurrent displays from two global forecast models, profiles from a mesoscale forecast model and an actual sounding, and the observed surface winds and the airbone concentrations from a transport-dispersion model.
Poster Session 1, Poster Session
Wednesday, 16 January 2002, 1:30 PM-1:30 PM
Previous paper