Florida Bay Physical Science Team
Draft Summary Report
March 22, 1999 Meeting - NOAA/AOML
The Florida Bay Physical Science Team (PST) first met September 4th and 5th of 1997 to compare initial results, identify deficiencies and needs of the program, and make recommendations. A second meeting was organized for the spring of 1998 but had to be canceled due to schedule conflicts of the participants. The purposes of the March 22, 1999, meeting reported here were to review progress to date, consider how research efforts are addressing the central question(s) of the Strategic Plan for the Interagency Florida Bay Science Program, and discuss future plans. In addition the group was tasked to consider recent results of the RMA10 hydrodynamic model and FATHOM water budget model.
The central question of the Program most directly related to PST efforts is Central Question #1: How and at what rates do storms, changing freshwater flows, sea level rise and local evaporation/precipitation influence circulation and salinity patterns within Florida Bay and the outflow from the Bay to adjacent waters.
SEAKEYS update - Sandy Vargo:
The northwest Florida Bay weather station was installed by the Coast Guard last summer south of East Cape and data came online October 1998. The data include standard atmospheric variables (wind speed and direction, air temperature., humidity, and precipitation) plus enhanced oceanographic measurements of sea level, conductivity and water temperature. The Long Key station has enhanced ocean measurements of water temperature, salinity, sea level, PAR, transmissivity and fluoresence. The Sombrero station is scheduled for the same instrument configuration as Long Key but, at present, the only ocean sensors working are temperature, salinity, and sea level as a result of problems related to the National Data Buoy Center’s support for basic control units. Ocean sensors may also be added to Sands Key station in cooperation with Charles Yentsch (NASA). Table 1 provides a description of instrumentation at each SEAKEYS station.
It was pointed out that the combined sea level measurements at Long Key and Sombrero provide an estimate of the cross-Keys sea level slope, which is believed to drive net flows through the Keys passages and with calibration could serve as a monitoring strategy for interactions of Florida Bay and reef tract waters.
Seasonal Variation of Water Mass Properties - Elizabeth Johns:
Using available small boats, salinity surveys were begun in Florida Bay and adjacent areas in 1995 then expanded in September 1997 with bi-monthly interdisciplinary surveys to include the larger Atlantic and Gulf regions involved in the transport and dispersal of Everglades and Florida Bay waters. Results from these surveys provide boundary conditions for use in NOAA/NOS’s Florida Shelf Domain model and the ACoE’s small-scale Florida Bay model. During this time a typical seasonal pattern is difficult to determine for in 1995 there was record fresh water discharge and in1997/1998 there occurred one of the strongest El Nino's on record and La Nina is dominating 1999. The data clearly show the complexities of the spatial and temporal variations of the low-salinity plume resulting from fresh water discharge management, tide, wind and coastal current forcing.
Everglades fresh water discharge from the Shark River complex into the coastal waters west of Cape Sable forms a low-salinity plume that is advected toward the southeast by wind-driven and Gulf-driven coastal currents. These tend to trap the plume against the coast of Cape Sable and the shallow boundary of western Florida Bay which provides a secondary northwestern low-salinity end member to Florida Bay.
The largest uncertainty affecting our understanding the plume dynamics is lack of quantitative fresh water discharge data. The U.S. Geological Survey (USGS) is measuring discharge in several of the larger rivers and has agreed to provide these data in the near future.
Expansion of water mass surveys to include the Florida Keys and the southwest Florida Shelf west to the Dry Tortugas and north to Naples has revealed strong vertical stratification of temperature and salinity that develops in the mid-shelf region in the summer. This appears to be a seasonal process due to increased heating and decreased wind mixing but is aided by significant low-salinity input in the upper layer. At present, the source of the low-salinity input is unknown, but due to its large volume, mid-shelf location, and northward extension along the shelf, it appears to derive from rivers in the northeast Gulf of Mexico, possibly even the Mississippi River. Comparison of observations with USF/ECOHAB may clarify this question.
Current Variability - Tom Lee:
The intensive measurement period began September 1997 and consists of an array of moored current, salinity, temperature and bottom pressure sensors maintained at the southwest Florida shelf, the Everglades low-salinity coastal band, western Florida Bay, and the Keys reef tract. Advection/dispersion pathways of Shark River discharge are estimated from surface trajectories of satellite tracked drifters. Tidal and wind forced volume transport exchanges between Florida Bay and the Keys reef tract are estimated from shipboard ADCP transects across the Keys passages.
Current variability on the southwest Florida shelf seems to be primarily tide and wind driven. The east-west flows are dominated by the semi-diurnal tides. The north-south flows are near equally tide and wind forced. Also observed is a significant slowly varying southward residual flow that is more persistent nearer the bottom and further offshore. The cause of this flow is unknown at present, but it does not appear to be related to wind forcing and may be connected to large-scale coupling of the Gulf and Atlantic. The net effect of this residual flow may contribute to the observed southeastward mean flow transporting Florida Bay and west Florida shelf waters toward the reef tract. Surface drifter trajectories also show strong correlation to local wind forcing, as well as a generally southerly net flow connecting the Shark River discharge with the reef tract.
Subtidal flows through the Keys passages can be either toward the reef tract or toward Florida Bay depending on wind forcing. The magnitude of these flows can reach 1400 m3/s from the combined outflows through Channels 2, 5 and Long Key. Seven Mile Bridge Channel can be of similar magnitude. Flows through the Keys passages are driven by the sea level slope across the Keys. Measurement of sea level variations on either side of Long Key clearly show that the subtidal (low-passed filtered) sea level slope variations are highly correlated to the alongshore wind stress over the reef tract. Results indicate that wind forced coastal current dynamics in the reef tract control subtidal sea level variability southeast of the Keys while local wind induced setup/setdown is the primary cause of low-frequency sea level variations in western Florida Bay. The net effect is to cause Florida Bay waters to stand higher than the coastal waters of the reef tract during summertime southerly winds, thus causing an outflow of Florida Bay water toward the reef. Strong winds from the east or northeast that occur in the fall and winter can cause the Keys coastal waters to stand higher than Florida Bay and, thus, reverse the flow through the Keys passages toward Florida Bay.
Water mass exchange (Western Boundary) - Ned Smith:
Long-term current meter measurements in Long Key Passage have been converted to volume transport time series using current meter and Acoustic Doppler Current Profiler transects across the channel for calibration. Average flow over a 8 month period was toward the reef tract at 233 m3/s from current meters and 260 m3/s from ADCP calibrations, showing very good agreement between these data and those of Lee et al. High coherence was found between subtidal volume transport fluctuations and winds in the direction of 100 and 280 degrees. A next step is to compare the volume transport time series to sea level difference and wind time series.
Long-term current measurements of flows across the western boundary of Florida Bay at 81E05' W longitude from stations near the north, central, and southern parts of the boundary show inflow to the Bay in the north and central stations and persistent outflow in the southern stations. Similar results are shown in recent ACoE measurements along this boundary using ADCPs. The estimated tide induced net flow across this boundary is 1400 m3/s to the Bay, which will cause significant tidal pumping of sea level inside the Bay which is higher than previously reported. Several of the Everglades National Park (ENP) tide gauges within the Bay have now been reference leveled and these data will permit comparison of observed tidal pumping with estimates of tide induced net flows (Dewitt Smith). A significant concern is whether the RMA10 model reproduces these pictures.
Hurricane Georges' wind fields and precipitation over Florida Bay - Craig Mattocks:
The atmospheric data for Hurricane Georges is being compiled for the period of influence over the Keys and Florida Bay. Wind fields are being gridded for use with Bay Models. The South Florida Water Management District (SFWMD) precipitation grid over south Florida has now been merged with NEXRAD derived precipitation fields over Florida Bay to produce a grid over the Bay and Everglades with a two mile resolution that will be useful for models. This was done for a single atmospheric event but will be expanded over a longer period for model use. Local weather services are now putting wind and precipitation maps through LAPS. The Miami Weather Service should be approached as to their activities or plans for LAPS. However, these data are insufficient to provide wind and precipitation fields for model runs.
Evaporation computations and predictions - Craig Mattocks:
The Advanced Regional Prediction System (ARPS) mesoscale meteorological model has been initialized with surface property data on a 2 mile grid to produce 2 mile grid evaporation/precipitation fields over South Florida and Florida Bay. The model clearly shows the influence of agricultural activities in shifting hot spots (compared to runs without man's influence) and divergence in wind and rainfall patterns. These gridded fields will be supplied to the ACoE for use in their Bay modeling effort. Evaporation/precipitation fields show considerable north-south gradients between the Everglades and Florida Bay and within the Bay that are not incorporated in ACoE's hydrodynamic model.
Smaller-scale, process-oriented studies of advection/dispersion (inner basins) - Ned Smith:
Long-term current meter deployments in three channels into Rabbit Key Basin all indicate a steady mean outflow with low-frequency variability that appears wind induced. The steady mean outflow suggests that the inflow occurs over the surrounding shallow banks on incoming tides. These waters are trapped within the basin on falling tides and forced out the deeper channels. More measurements of this type are necessary, however, in other representative basins. In addition, the Intracoastal Waterway appears to be an important conduit for water exchange in the eastern Bay, but not much is known about this process at present.
In general it was agreed that application of a 2-dimensional hydrodynamic model to the Bay was appropriate. 3-D water mass structures and circulations may occur at times on small spatial scales near banks or fresh water outflows or near the western extreme of the domain. The effects are not considered to be long-lasting, nor to generate significant baroclinicity.
FATHOM Mass Balance Model - Jack Cosby:
FATHOM is a sophisticated mass balance model that divides Florida Bay into 44 basins separated by line segments paralleling the shallow banks. The model estimates salinity in each of the basins or representative groups of basins on a monthly basis to look at seasonal, annual, and interannual changes. The approach also assumes each basin is well-mixed.
FATHOM is driven by freshwater inflow and sea level setup and assumes salinity in the Gulf and Atlantic is constant at 35 psu. It applies tide and wind induced sea level at the open boundaries to the Key West annual cycle of sea level and computes flow across segments (banks) from Mannings equation. However, it ignores flow through channels. Measured rainfall estimates are used and considered uniform across the Bay. Seasonal and annual evaporation rates are derived from a combination of data sources and assumed to be constant across the Bay. While salinities were estimated for the period 1973 - 1995 and explained approximately 50% of the observed variance, modeled salinities were too high during wet season. Uncertainties in the model may be the result of poorly known freshwater discharge, unresolved evaporation and precipitation patterns, and mixing and freshwater input along the western boundary.
There are also plans to use the model to estimate residence times and initial estimates indicate residence times for the northeast portion of the Bay of about 7-8 months. FATHOM is run on a PC and can be used to evaluate different water delivery strategies for the Bay.
Since freshwater runoff that occurs in the northeast portion of the Bay is a small fraction (about 10%) of the precipitation that falls over the Bay as a whole, in FATHOM it affects only the northeast corner of the Bay.
Discussion of models - Bill Nuttle:
While estimates of annual evaporation rates in the south Florida region range from 75 to 207 cm/year, evaporation in Florida Bay is still poorly resolved and spatial and temporal scales are uncertain. Precipitation is also poorly resolved but better than evaporation. Accurate estimates of Bay salinities resulting from future water delivery actions are needed to advise managers. Both the ACoE hydrodynamic model and FATHOM require better resolution of evaporation and precipitation to make reliable forecasts of Bay salinity. Approaches to obtaining better evaporation/precipitation estimates are planned. These include:
1. Craig Mattocks’ 2 km ARPS model using relative humidity, temperature and wind measured at CMAN and Everglades National Park sites for evaporation and NEXRAD supplemented by Everglades National Park gauges while available (see #4 below) for precipitation.
2. Eric Swain of USGS running regressions with soil evaporation - correlation method.
3. Peter Swart’s isotope method was discussed but was felt to be of marginal utility.
4. DeWitt Smith’s enhanced measurement network i.e., Dewitt is planning to install an additional evaporation/precipitation station in the central Bay east of Whipray Basin. Also Ned Smith is installing a relative humidity sensor on the Long Key SEAKEYS station. These stations together with the Everglades National Park’s Joe Bay station and the new SEAKEYS Northwest Florida Bay station south of East Cape forms more of a triangle through the Bay allowing better north-south and east-west coverage. Each station will measure precipitation, radiation, humidity, temperature and winds for estimating evaporation. Consideration should also be given to more sophisticated methods of estimating evaporation, possibly using an eddy-correlation method.
Feedback to Modeling Groups - Bill Nuttle:
The two modeling efforts underway: the ACoE/WES hydrodynamic model and FATHOM have agreed to estimate salinity in Florida Bay as a result of different management scenarios: 1) changing fresh water inflows; and 2) removing the railroad causeway at Lignumvitae.
The ACoE has issued a formal report showing results of the RMA10 model evaluation and verification and discussing possible impacts from changing fresh water deliveries to Florida Bay. A similar report will soon be out for the FATHOM model.
Only a few members of the PST have reviewed the ACoE model report dated Oct. 1998. However it was pointed out that although the hydrodynamic model show very promising results in comparisons of modeled and observed sea level, currents, volume transports and salinity at several stations at the boundaries and interior of the model, there were also many stations where the agreement was poor. This indicates that the model calibration and verification has not been completed and there needs to be a continued effort of model adjustments (roughness coefficients for example) to improve the comparisons with observations. The model calibration effort must be conducted as a cooperative effort between the modelers and observationalists to avoid problems.
It was pointed out that the magnitudes of the fresh water inflows to Taylor Slough and Shark River Slough that were used in the ACoE model verification runs were a factor of 10 or more larger than observed. This caused salinities in the northeast portion of the Bay to be much fresher than ever observed.
It was agreed that the PST would review both the COE/WES Florida Bay model report and the FATHOM report and send comments to Bill Nuttle for synthesis and transmittal to COE and FATHOM investigators.
PHYSICAL SCIENCE TEAM MEETING AGENDA
March 22, 1999
Attendees: Peter Ortner, Dawn Welcher, Tom Lee, Libby Johns, Joe Boyer, Bill Nuttle, Jack Cosby, Frank Aikman, Ned Smith, Ed Patino, Dewitt Smith, Peter Swart, Sandy Vargo, John Wang, Hugh Willoughby, Craig Mattocks, Eric Swain, Dave Gruber, and Joan Browder.
0930 Introduction (Ortner and Lee)
0945 SEAKEYS update (Vargo)
1000 Current variability (Lee)
1020 Water mass exchange - Western Boundary (Smith)
1040 Seasonal variation of water mass properties - SW Florida Shelf and the Keys (Johns)
1100 Smaller-scale, process-oriented studies of advection/dispersion (Smith [inner basins] and Lee [Shark River discharge])
1130 Hurricane Georges' wind fields and precipitation over Florida Bay (Mattocks)
1200 Working Lunch
1230 ACoE's modeling status
1300 FATHOM Mass Balance Model (Cosby)
1315 Evaporation computations and predictions (Mattocks)
1330 Discussion of models (Nuttle)
*Water budget differences (FATHOM and ACoE)
1500 Final comments