Section Description

Global Modeling (Code 7532)
Section Head: 7532@nrlmry.navy.mil

The Global Modeling Section develops and applies computer models for the dynamic prediction of atmospheric and oceanographic variables on a global scale and uses these models to improve our understanding of atmospheric behavior, with an emphasis on air-sea interaction, tropical meteorology, atmospheric predictability, and tropospheric/stratospheric processes. The primary model of interest is the NRL-developed global spectral model in the Navy Operational Global Atmospheric Prediction System (NOGAPS) which is currently operational at Fleet Numerical Meteorology and Oceanography Center. Almost all Navy environmental prediction and support systems require NOGAPS products; therefore, nearly every meteorological and oceanographic requirement benefits from NOGAPS and its continuing development. Other models of interest include the NRL Spectral Element Atmospheric Model, currently under development for possible future applications, and the Parallel Ocean Program (POP) model from Los Alamos National Laboratory, which is being coupled to NOGAPS.

The Global Modeling Section is working in these main areas:
Global Atmospheric Model Development. Advanced physical parameterizations (land surface, convection, radiation). Advanced numerical methods for global models (high order finite element methods, semi-Lagrangian formulation). Identification of systematic errors in the global prediction system. Development of a global coupled atmosphere/ocean data assimilation/modeling system. Extension of NOGAPS into the middle atmosphere. Transition of NOGAPS improvements to FNMOC. Software design for scalable, massively parallel computer systems.
Predictability Studies. Adaptive observing techniques using adjoint sensitivity and singular vectors. Forecast sensitivity to observations using the NOGAPS and NAVDAS adjoints . Mechanisms for rapid perturbation growth. Seasonal and interannual variations in predictability. Transient and asymptotic perturbation growth. Observing network design; application of adjoint in planning for and execution of THORPEX initial field phase. Ensemble forecasting techniques. Generation of initial perturbations. Methods for introducing model error perturbations. Use of ensembles in data assimilation.