JP1.4
Applying Knowledge Discovery from Databases (KDD) to Combined Satellite and High Resolution Numerical Model Data
Paul M. Tag, NRL, Monterey, CA; and R. L. Bankert, M. Hadjimichael, A. P. Kuciauskas, W. T. Thompson, and K. L. Richardson
The DaFWA (Data Fusion for Weather Assessment) project addresses the U.S. Navy need for accurate nowcasting of sensible weather parameters by combining satellite data from various platforms with numerical model data from the Navy's high-resolution regional model, COAMPS (Coupled Ocean/Atmosphere Mesoscale Prediction System). The triply-nested (81, 27, and 9 km resolution) model is run for three areas: U.S. west coast, Adriatic Sea, and South Korea. From these multiple data sources, various parameters are extracted/computed at METAR observation sites within the 9 km domains of the three regions. KDD methods are applied to that database which includes actual METAR observations supplying the ground truth for the parameter of interest. Data mining techniques help determine relationships within the database that will aid the user in nowcasting this sensible weather parameter, specifically in a data-void region. This presentation will include a discussion on the design and creation of the database, the data mining techniques as applied in the KDD methodology, and the derived relationships that result for various parameters (e.g, cloud ceiling).
Joint Poster Session 1, (Joint with 10th Conference on Satellite Meteorology and Oceanography and Second Conference on Artificial Intelligence)
Tuesday, 11 January 2000, 4:30 PM-5:45 PM
Previous paper Next paper