Regional Boundary Conditions for Florida Bay

Regional Boundary Conditions for Florida Bay

Progress Report - August 1999

Principal Investigator(s):

Dr. Frank Aikman III and Dr. William P. O’Connor
NOAA/NOS/Coast Survey Development Laboratory
1315 East-West Highway, SSMC3
Silver Spring, MD 20910

Professor George L. Mellor
Atmospheric and Oceanic Sciences
Princeton University
P.O. Box CN710
Princeton, NJ 08544-0710

GOALS: Our long-range goal is to determine how open ocean forcing can be incorporated into a model of Florida Bay (FB) and to address the Bay's interaction and exchange with the adjacent shelf and Gulf of Mexico. This is a key component of the NOAA focus on the large-scale regional oceanographic, atmospheric, ecological, and fisheries context within which FB restoration will proceed.

OBJECTIVE: To develop, test, evaluate, and implement a Florida Shelf (FS) domain model that will provide accurate information on the boundary conditions for hydrodynamic models of Florida Bay.

RATIONALE: Any numerical model of the unique and complex FB hydrodynamics, such as the barotropic circulation model (RMA10) being applied by the U.S. Army Corps of Engineers (ACoE), requires adequate open ocean boundary condition information. In fact, one of the most difficult aspects of limited‑area bay modeling is the adequacy of the lateral boundary condition information available. For FB, the barotropic forcing from wind and tides is probably of primary importance, with the baroclinic forcing from, for example, meandering of the Loop Current or the influence of Loop Current Eddies most likely being of secondary importance.

FB regional observations have not had sufficient spatial or temporal coverage to provide adequate boundary condition information, and very little is actually known about the relative importance of the boundary forcing from the east and south (Florida Keys) versus the west (West Florida Shelf) of FB. However, recently obtained and on-going measurements of currents, salinity, temperature, and bottom pressure (Tom Lee [RSMAS]; Lisa Roig [ACoE]; Ned Smith [HBOI]) will prove useful for both defining what the conditions at the boundary are and for the evaluation of the boundary condition information derived from the National Ocean Service (NOS) model simulations.

METHODOLOGY: NOS is applying the Princeton Ocean Model (Blumberg and Mellor, 1987) for the FS domain (Figure 1) to simulate the barotropic boundary conditions for FB. Using tidal boundary conditions modified from Schwiderski (1980) and wind forcing from the 29-km Eta atmospheric forecast model (Black, 1994), the FS model has been calibrated for tides and is now being evaluated for wind-plus-tide-driven simulations. Output water level and barotropic current fields are provided to the ACoE to serve as boundary conditions for their barotropic model (RMA10) of FB. Analyzed (Eta 48-km) wind fields have been tested and evaluated using the FS model to see if this offers an improvement over the Eta forecast winds, and the assimilation of water level gauge data into the barotropic model will be tested for the FS domain.

STATUS and ACCOMPLISHMENTS:

1. An 18-month (1 September 1995 to 28 February 1997) barotropic FS model simulation of wind-plus-tide-forced water level and currents has been completed and we are in the process of evaluating the results using NOS water level gauge data, observations of bottom pressure and both moored and drifting buoy current meter data from Tom Lee (RSMAS) and Ned Smith (HBOI). At the request of the ACoE (Keu Kim, personal communication), the original 14-month FS simulation was extended four months to the end of February 1997, driven by analyzed Eta 48-km winds, and the output was interpolated to the RMA10 model grid and delivered to the ACoE. This will allow the ACoE to complete a 7-month (August 1996 - February 1997) simulation for salinity evaluation of their RMA10 model in FB.

(a) Estimates of the cross-FB sea level slope have been compared between observations and the model simulation. The results (Figure 2) indicate that the barotropic model water levels are in close agreement with the observations around FB (rms difference ~ 7-8 cm; correlation coefficient ~0.9), that the model cross-FB sea level slope is in qualitatively good agreement with the observations (Figure 3), and that the observed cross-FB sea level slope is primarily a barotropic feature.

(b) At a number of locations, where Ned Smith (HBOI) has processed moored current meter and bottom pressure records in the western part of FB, the model currents and water levels were subjected to harmonic (tidal) analysis and compared to the identical analysis applied to the observed data. This analysis was carried out by Ned Smith (personal communication). In general, and for all tidal constituents, the model tidal water levels and east-west velocity components are in good agreement with the observed tides while the amplitude of the model north-south tidal velocity components is about twice that of the observed. This may be explained, in part, by the poor resolution of the FL Keys in our FS model (5' resolution).

2. For the seven months of August 1996 - February 1997 the FS model has also been driven with Eta 48 km analyzed winds to compare the water level results over the same period of the model driven with Eta 29 km forecast winds. The results indicate that there is not a significant difference, although the correlation coefficients (observed vs. model water levels) are slightly higher and the rms differences are slightly lower with the model results driven by analyzed winds.

3. The hourly water level and barotropic current fields of the 18-month simulation have been interpolated to the RMA10 model grid and delivered to the ACoE (Keu Kim, WES, Vicksburg, MS). The ACoE is testing these as boundary conditions for their barotropic model simulations of FB.

4. Evaluation results and plans for further 2-D simulations and 3-D model development and simulations have been presented at the following meetings:

(a) 5th Estuarine and Coastal Modeling Conference, American Society of Civil Engineers, Alexandria, VA, October 22-24, 1997 (O’Connor et al., 1998a);

(b) 2nd Conference on Coastal Atmospheric and Oceanographic Prediction and Processes, American Meteorological Society, Phoenix, AZ, January 11-16, 1998, (O’Connor et al., 1998b);

(d) Marine Technology Society Ocean Community Conference, November 16-19, 1998, Baltimore, MD (Aikman and O’Connor, 1998).

FUTURE PLANS:

1. Analyzed wind fields will soon be available from the RUC-2 (Rapid Update Cycle) model, which will have the advantage of an hourly data assimilation cycle (versus 3 hours for the Eta model) and will be of 40 km resolution. We plan to test these wind fields with the FS model.

2. NOS is also examining the regional availability of real-time meteorological observations, especially on the West Florida Shelf (Mark Luther, USF, personal communication) and will consider developing our own analyzed wind fields for the region.

3. NOS will develop and test the assimilation of water level gauge data into the barotropic FS model using a number of different nudging techniques. Depending on the results, such techniques could be used for operational nowcasting using the real-time availability of NOS water level gauge data in the region.

4. A fully 3-dimensional version of the Princeton Ocean Model is being tested at Princeton University, which will encompass the extension of the Coastal Ocean Forecast System (Aikman et al., 1996) into the Gulf of Mexico (Figure 4). This model will be used to determine the importance of baroclinicity in the coastal circulation on the West Florida Shelf and to evaluate the baroclinic effects on FB currents and water levels. If required, 3-dimensional boundary condition information (including estimates of the 3‑D salinity, temperature and density) could be provided to the ACoE for FB, as well as for other regional studies associated with FB, the West Florida Shelf, and the Florida Keys National Marine Sanctuary.

REFERENCES:

Aikman, F. III and W.P. O’Connor. 1998. Model-based regional boundary conditions for Florida Bay. In: Proceedings of the Marine Technology Society Ocean Community Conference, November 16-19, 1998, Baltimore, MD.

Aikman, F. III, G. L. Mellor, T. Ezer, D. Sheinin, P. Chen, L. Breaker, K. Bosley, and D. B. Rao. 1996. Towards an operational nowcast/forecast system for the U.S. East Coast. In: Modern Approaches to Data Assimilation in Ocean Modeling. P. Malanotte-Rizzoli (Editor). Elsevier Oceanography Series, 61, 347-376.

Black, T. L. 1994. The new NMC mesoscale Eta model: Description and Forecast Examples. Weather and Forecasting, 9, 265-278.

Blumberg, A.F., and G.L. Mellor. 1987. A description of a three‑dimensional coastal ocean circulation model. Three‑Dimensional Coastal Ocean Models, 4, edited by N. Heaps. American Geophysical Union, 208p.

O’Connor, W.P., F. Aikman III, E.J. Wei, and P.H. Richardson. 1998a. Comparison of observed and forecasted sea levels along the West Florida coast. In: Estuarine and Coastal Modeling, Proceedings of the 5th International Conference, American Society of Civil Engineers, M.L. Spaulding and A.F. Blumberg (editors), October 22-24, 1997, Alexandria, VA, 601-612.

O’Connor, W.P., F. Aikman III, E.J. Wei, and P.H. Richardson. 1998b. Comparison of observed and forecasted sea levels for the Texas coast near Galveston Bay. Preprints of the 2nd Conference on Coastal Atmospheric and Oceanographic Prediction and Processes, American Meteorological Society, January 11-16, 1998, Phoenix, AZ, 23-29.

Schwiderski, E. W. 1980. On Charting Global Ocean Tides. Reviews of Geophysics and Space Physics, 18(1), 243‑268.