Developing a Systematic Approach to Tropical Cyclone Track Forecasting expert system-1: Information management, visualization, and proactivity considerations

To implement the Systematic Approach to Tropical Cyclone Track Forecasting concept (Carr and Elsberry 1994) under operational time constraints, a knowledge-based expert system is being developed that assists the forecaster in dealing with the highly complex information management, visualization, and decision-making challenges inherent to tropical cyclone track forecasting. This talk provides an overview of the features that have been built in the expert system software, which include: (i) a highly structured (but non-binding) procedural guidance with selectable levels of detail; (ii) a cluster-based objective assignment of the most probable error mechanisms to focus the dynamical model evaluation process; (iii) an error mechanism HELP feature that includes track graphics and field animations that illustrate the key indicators of the frequently occurring error mechanisms; (iv) a model forecast tracks display that includes cluster and individual model trend analyses; (v) a model forecast fields animation capability that instantly lets the forecaster switch from intra-model to inter-model comparisons; (vi) an interactive error mechanism assignment capability that lets the forecaster see the impact of selecting only certain dynamical model tracks for a consensus forecast; and (vii) an interactive Summary Sheet feature that records objective ensemble spread and cluster analysis data, objective and subjective error mechanism assignments, and forecaster comments to facilitate shift-to-shift information transfer.