Overview of the Navy's Coupled Mesoscale Modeling System

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Monday, 29 June 2015: 2:30 PM
Salon A-2 (Hilton Chicago)
James D. Doyle, NRL, Monterey, CA; and C. M. Amerault, S. Chen, S. Gabersek, T. Haack, E. A. Hendricks, R. M. Hodur, T. Holt, X. Hong, Q. Jiang, H. Jin, Y. Jin, W. Komaromi, J. Moskaitis, J. Nachamkin, P. A. Reinecke, J. Schmidt, D. Tyndall, K. Viner, S. Wang, and L. Xu

The Naval Research Laboratory (NRL) recently has added several new capabilities to the Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPSŪ) to provide more accurate analyses and predictions of the atmosphere and ocean at high-resolution. An overview of these new COAMPS capabilities will be highlighted in this presentation. COAMPS has been coupled in a two-way interactive sense to the NRL Coastal Ocean Model (NCOM) and the SWAN and WWIII wave models using the Earth System Modeling Framework (ESMF). A description of the coupled system will be presented along with examples from the coupled model. Other capabilities that will be highlighted include: i) a coupled tropical cyclone (TC) system and a TC ensemble capability, ii) a coupled ensemble system, iii) 4D-variational and ensemble Kalman filter data assimilation systems, iv) a nested adjoint and mesoscale observation impact system, and iv) new physical parameterization for the planetary boundary layer and moist processes. The tropical cyclone system, COAMPS-TC, has been demonstrated in real time over several different basins during the past 5 years, and transitioned to Navy operations in 2013 motivated by the fact that it was one of the top performing dynamical models for tropical cyclone intensity prediction. Lastly, a discussion of the future plans for the Navy's mesoscale modeling system will include a summary of the recent development of a new nonhydrostatic dynamical core model, NEPTUNE based on spectral element and discontinuous Galerkin techniques. Examples of the new dynamical core will be shown for idealized and real-data cases.