Bio-physical ocean modeling in the Gulf of Mexico

The Naval Research Laboratory (NRL) Oceanography Division has implemented a 1/25? horizontal-resolution numerical ocean model for the Gulf of Mexico (GOM). The model domain encompasses the entire GOM extending from 18? to 31? North and from 77? to 98? West. The physical formulation is based on the Naval Coastal Ocean Model (NCOM) configured with a 40 level ?-z vertical structure: 19 terrain-following ? (sigma) levels at the top of the water column, and 21 z (depth) levels at the bottom. The terrain-following levels, reaching from the surface to about 137 meters, allow higher vertical resolutions for resolving mixed layer/shallow waters, while the z-levels are used in the stratified ocean. The ocean bathymetry is constructed from the NRL 2-minute database with the coastline set at a depth of 2 meters. The physical model is one-way coupled to a 13-component ecosystem (biogeochemical) model [Chai et. al., 2002, 2003, 2007] that includes nutrients, two classes of plankton, as well as O and CO2. The current configuration is tailored, but not limited to, real-time (RT) prediction; providing nowcasts (current state of the ocean) and up to 120-hour forecasts for the region. In this configuration, the model receives (initial) boundary information from the operational 1/8? Global NCOM, and it is forced by 3-hourly 1/2? momentum and heat fluxes from the Naval Operational Global Prediction System (NOGAPS). The NCOMGOM model assimilates daily surface/subsurface temperature and salinity generated by the Modular Ocean Data Assimilation System (MODAS), which regresses satellite derived Sea Surface Temperature (SST) and Sea Surface Height (SSH) data to obtain T&S synthetic profiles. The model was initialized on January 1, 2009 from the operational 1/8? Global NCOM physical state and from the World Ocean Atlas 2005 and Carbon Dioxide Information Analysis Center (CDIAC) biogeochemical fields. Results from the real-time NCOMGOM nowcasting/forecasting ocean modeling system are comp- ared and evaluated against in-situ and remotely sensed observations, which include bio-optical products processed by the NRL Ocean Color Section. Google Earth/Ocean is used as a platform for viewing the model results interactively and dynamically in real-time. Initial assessment of the model prediction skill is presented along with future plans for improvements and enhancements. The suitability of the system as a tool for decision management is discussed, outlining processes localized to particular areas such as hypoxia, dead zones, wetland loss and degradation, harmful algal blooms, as well as tropical storms and related issues that affect all coastal regions of the Gulf of Mexico.