Predictability of Tropical Cyclones: Theoretical and Practical Aspects

Abstract : The long term goals of this program remain largely unchanged, namely to define, and attempt to approach in practice, the limits to predictability of tropical cyclones, beginning with track forecasting and later extending the work to intensity and structure. The scientific objectives of the effort are to apply the techniques of non-linear systems analysis to estimating the so-called intrinsic limits to predictability for (i) tropical cyclone mean forecast position errors in the first instance, and (ii) to tropical cyclone intensity and intensity change in the second part of the program. The intrinsic limits are to be compared with the results being obtained in practice and the size of the disparity represents the gains in predictive skill that are still achievable. It is of fundamental importance to have some idea of how large the gap is between that being obtained and the ultimately achievable in order to justify the continued allocation of resources to the various problems. Additionally, it is worthwhile recognizing that the critical problem of landfalling tropical cyclones is now capable of being addressed with the new data sources and improved models now available.