Principal Investigator: Dr. Mark D. Powell
Collaborating scientist(s):
Dr. Mark DeMaria
Dr. Craig Mattocks
Samuel H. Houston
Eric Swartz
Paul Hebert
Objective: The primary objective of this research project is to reconstruct episodic/catastrophic meteorological events and local weather regimes which critically affect the South Florida ecosystem.
Rationale: Hurricanes, tropical storms, and frontal passages are believed to exert considerable influence on the health of Florida Bay. Wind fields associated with these storms generate surface stress and an associated response in the bay circulation patterns and sediment transport. Because the "multiple pond-bank" nature of Florida Bay inhibits sediment transport forced by relatively weak wind conditions, hurricanes are believed to play a critical role in flushing the bay. Severe winds associated with tropica l cyclones may also contribute to ecosystem health as a result of the post-storm decay of organic material damaged by the wind and storm surge. In order to assess the response of Florida Bay to episodic wind events, circulation and ecological modelers would benefit from wind field analyses based on reconstruction of past events.
Method: (1) Episodic Wind Field Reconstruction -- Gridded wind, precipitation, and temperature fields will be generated from real case analyses, numerical model simulations, and idealized scenarios described by canonical systems of analytical equations to produce a catalog of datasets. Research scientists will be able to quickly retrieve data from this event/regime archive for use in circulation model simulations/predictions, hydrological modeling, natural systems restoration, and biological impact studies . We plan to analyze the most significant storms to affect Florida Bay over the past century. In so doing we will attempt to show a variety of wind field affects on Florida Bay which could comprise an archive of the types of extreme events that can be expec ted over the next century. Data will be processed to achieve a consistent framework in terms of averaging time, height and exposure. For storms with few data available the wind profile fit currently used in the SLOSH storm surge model will be adapted to construct a background field. These data will then be objectively analyzed using the Spectral Application of Finite-Element Representation (SAFER) method (Ooyama 1987, Franklin et al. 1993). This method uses cubic B-splines to minimize the difference betw een the input observations and the analysis. The methods used to standardize the input data for analysis are described in Powell et al., (1996), and Powell and Houston (1996). The scale of the analysis is controlled by the analyst depending on the features th at need to be resolved. Snap shots of the streamline and isotach fields for each storm will be generated for a 2 x 2 degree lat-lon domain centered on Florida Bay at 6-12 h intervals, depending on the translation speed of the storm. This domain should also be suitable for those interested in studying extreme wind affects on the southern end of the Everglades. Images of the wind fields will be archived on a World Wide Web site for access by other Florida Bay researchers and gridded fields will be genera ted upon request. These fields should be capable of import to geographical information systems for correlation studies with other geo-referenced fields such as mangroves, reefs, and turbidity plumes.

(2) Mesoscale Atmospheric Modeling -- A major component of this dataset construction effort is the use of the Center for Analysis and Prediction of Storms' Advanced Regional Prediction System (ARPS) cloud-/mesoscale atmospheric numerical weather predictio n model which can simulate/predict surface winds, rainfall and thermodynamic fields relevant to Florida Bay at high resolution. These fields can be used as boundary conditions/forcing for bay and ocean circulation models. The atmospheric model can also be used to study the effect of specific processes on the freshwater input to Florida Bay, such as evaporation/precipitation under different weather regimes, as well as the transport/rainout of toxic atmospheric substances in the South Florida ecosystem.

Accomplishment: In addition to analyses of Hurricane Andrew of 1992 (Figure 1) and Hurricane Gordon of 1994 (Figure 2). We have focused on reconstructing the wind field of Hurricane Donna (1960). With the help of a student intern we have evaluated the archived radar center fixes and aircraft reconnaissance wind center fixes to compile a new track. Flight-level data have been adjusted to the surface and combined with data from lighthouses, ships, NWS offices, and Cooperative Hurricane Network (CHURN) stations. These data were extracted from HRD's and NHC's archives, as well as newspaper accounts. An initial field was constructed that is representative of Donna's winds while offshore from the Florida Keys (Figure 3). This field will be used as a background and combined with maximum surface wind observations during Donna's passage over the middle Ke ys and the western portion of Florida Bay.

Numerical simulations of seasonal weather regimes have also begun. The ARPS model has been configured with a horizontal grid mesh of 9 km to include the Florida peninsula and surrounding waters, and 42 hyperbolic tangentially stretched vertical levels (50 meter vertical spacing in lowest 500 meters) to resolve the planetary boundary layer. A clay-loam soil with grass/shrub vegetation is used over land in the two soil-layer force/restore surface energy budget. ARPS was then initialized with a homogeneous base state from a morning August Miami sounding. The horizontal motion fields, depicted by flow streamlines at the Earth's surface (Figure 4) show that, at 1 PM local time (elapsed simulation time = 5 hours), the surface heating (temperatures in the low-mid 90's F) induces a fairly strong sea breeze (SB) front. There is also a classic cold imprint from Lake Okeechobee - a divergent surface outflow toward the west. Full solenoidal circulations develop at both coasts, and a very intense convective "cell" explodes along the SB front further north. The realism of the atmospheric simulations will be improved by initializing the model with non-homogeneous 3-D fields from the operational National Weather Service Eta/Meso mo del, incorporating rain/ice microphysics packages, and using high-resolution land cover/use, soil/vegetation databases from the South Florida Water Management District (SFWMD).

Key references:
South Florida Atmospheric Modeling Workshop (NOAA, Nov. 1994)

Droegemeier, K.K., M. Xue, P.V. Reid, J. Brandley and R. Lindsay, 1991: Development of the CAPS Advanced Regional Prediction System (ARPS): An adaptive, massively parallel, multi-scale prediction model, Ninth Conference on Numerical Weather Prediction , Denver, Co., American Meteorological Society, Boston, 289-292.

Franklin, J. L., S. J. Lord, S. E. Feuer, and F. D. Marks, 1993: The kinematic structure of Hurricane Gloria (1985), Mon. Wea. Rev., 121, 2433-2451.

Ooyama, K. V., 1987: Scale controlled objective analysis. Mon. Wea. Rev., 115, 2479-2506.

Powell, M. D., S. H. Houston, and T. Reinhold, 1996: Hurricane Andrew's landfall in south Florida. Part I: Standardizing measurements for documentation of surface wind fields. Weather and Forecasting, 11, 304-328.

Powell, M. D., and S. H. Houston, 1996: Hurricane Andrew's landfall in south Florida. Part II: Surface wind fields and potential real-time applications. Weather and Forecasting, 11, 329-349.

Last modified: 10/21/96