The technique developed by Dvorak has thus far been regarded, by tropical meteorologists, as the best intensity identification scheme using satellite imagery. However, in recent years, several ideologies have arisen which discuss alternative means of determining typhoon rapid intensification or weakening. These theories include examining channel outflow patterns, potential vorticity superposition and anomalies, tropical upper tropospheric trough interactions, environmental influences (sea surface temperatures, effects of vertical shear, and air-sea interactions), and upper tropospheric flow transitions.

The goal of this work was to data mine atmospheric parameters responsible for typhoon rapid intensification/weakening and to validate the usefulness of using these parameters in the forecast process. Using the latest data mining software modules, this work uses components of the NOGAPS model along with selected atmospheric and climatological predictors in classification (i.e., conditional forecast decision tree) analyses. These analyses show interesting relationships between the various predictors and validate the notion that synergistic effects among parameters are responsible for rapid changes in typhoon strength. Although the conclusions are not traditional in the sense of statistical linear regression, relationships between the different perspectives are well established. This investigation also provides insight into current forecasting techniques and sheds light on new parameters which are beneficial to the forecast process.