An Experimental Basin-Scale HWRF Analysis and Prediction System for Model Development and Satellite Data Assimilation Research

Reliable probabilistic forecasts for tropical cyclones remain limited by the suboptimal quantification of initial condition uncertainty and the misrepresentation of physical processes in numerical weather prediction models. While these error sources are often difficult to distinguish from one another, sequential ensemble-based data assimilation methods provide a good starting point for rigorously testing components of current forecast systems and validating upgrades introduced to models, observing systems, or data assimilation systems using observation-space diagnostics.

This study summarizes recent progress at the NOAA Hurricane Research Division of AOML, developing a testbed for HWRF model development based on the above paradigm. Unlike the operational HWRF prediction system, the experimental system performs data assimilation continuously throughout hurricane seasons, assimilating satellite and conventional observations every six hours with the GSI ensemble Kalman filter. In addition, the system uses an extensive “basin-scale” regional domain, which is large enough to capture interactions between tropical cyclones and the synoptic scale environment represented across multiple hurricane basins. The resulting configuration provides the necessary framework for evaluating deficiencies in the current HWRF model and investigating new strategies for assimilating satellite observations more effectively in operational prediction systems. In this presentation, we will highlight results obtained by running the experimental HWRF system during the 2017 hurricane season. To the best of our knowledge, these experiments provide the most extensive testing of the HWRF model using cycling data assimilation.