Recent Progress and Challenges in Tropical Cyclone Intensity Prediction Using COAMPS-TC

The rapid intensification (RI) of tropical cyclones (TCs) is notoriously difficult to predict, even for current state-of-the-science TC prediction systems. The Coupled Ocean/Atmosphere Mesoscale Prediction System for Tropical Cyclones (COAMPS-TC) has been developed for forecasting tropical cyclone track, structure, and intensity and has been run at high resolution in Navy operations at the Fleet Numerical Meteorology and Oceanography Center since 2013. In this presentation, we will provide an update on the latest advancements to the COAMPS-TC system in 2018 and 2019 including: i) several advancements to the physical parameterizations, and ii) improvements to the vortex initialization. An evaluation of a large sample of the real-time forecasts since 2013 in the Atlantic, E. Pacific and W. Pacific basins reveals markedly improvements in COAMPS-TC intensity predictions, although many challenges remain, particularly with regard to rapid intensification. Regional dynamical prediction models such as COAMPS-TC have great difficulty accurately predicting the onset and subsequent rate of TC intensification for post-genesis, weak TCs. Average intensity errors are highest, and the preponderance of very large intensity errors is greatest, for model forecasts of initially weak TCs. Examples of these RI events will be discussed. Improved understanding of the physical processes contributing to the timing of intensification onset and subsequent intensification rate is ultimately needed to help improve dynamical model predictions of RI.

New capabilities under development for COAMPS-TC will be highlighted including an ensemble Kalman Filter (EnKF) data assimilation system capable of all-sky radiance assimilation, which will be particularly important for representing the initial state of TCs prior ro and undergoing RI. In addition, results and operational transition of a high-resolution (4 km) 11 member COAMPS-TC ensemble over the W. Atlantic, E. Pacific, and W. Pacific basins will be presented. The ensemble results continue to show considerable promise for probabilistic intensity and track prediction.