USF Libraries
USF Digital Collections

Bathymetric alterations due to urbanization and their effects on residual salinity, flow field, and transport time for T...

MISSING IMAGE

Material Information

Title:
Bathymetric alterations due to urbanization and their effects on residual salinity, flow field, and transport time for Tampa Bay, Florida
Physical Description:
Book
Language:
English
Creator:
Linville, Amanda J
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Model
Residence time
Estuary
Navigational channel
Bridges
Dissertations, Academic -- Marine Science -- Masters -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Growth and development over the past one hundred years has resulted in the construction of causeways and navigational channels in Tampa Bay. Urbanization has lead to bathymetric alterations of the bay, and thus has effectively changed the residual salinity, flow fields, and transport time in Tampa Bay. In this study a numerical ocean model of Tampa Bay was first used to simulate the circulation for 2001-2003 using present day bathymetry. Then, an identical simulation was performed using the bathymetry generated from 1879 depth soundings. The residual (30-day time average) circulation fields and salinity, along with transport times was intercompared for different freshwater inflow conditions to investigate the impacts of these physical alteration. The salinity for the "present" simulation is about 3 psu higher than the "1879" simulation in the areas of upper Middle Tampa Bay, Lower Hillsborough Bay and Old Tampa Bay, a result of a stronger axial pressure gradient associated with the deeper more extensive channels. Velocities are up to 10 cm/s higher in the "present" run than in the "1879" simulation in the areas where water must converge and diverge through the narrow openings of the bridges/causeways. Transport time is short (~10 days) during strong residual circulation, and long (~90 days) during weak residual circulation. Bridges and causeways are associated with longer transport times (~90 days), except in the area North of the Courtney Campbell Causeway. The navigational channel is associated with long transport times during dry periods and short transport time during wet periods.
Thesis:
Thesis (M.S.)--University of South Florida, 2007.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Amanda J. Linville.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 98 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001920059
oclc - 187080158
usfldc doi - E14-SFE0002092
usfldc handle - e14.2092
System ID:
SFS0026410:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001920059
003 fts
005 20080103134526.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 080103s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002092
035
(OCoLC)187080158
040
FHM
c FHM
043
n-us-fl
049
FHMM
090
GC11.2 (ONLINE)
1 100
Linville, Amanda J.
0 245
Bathymetric alterations due to urbanization and their effects on residual salinity, flow field, and transport time for Tampa Bay, Florida
h [electronic resource] /
by Amanda J. Linville.
260
[Tampa, Fla.] :
b University of South Florida,
2007.
3 520
ABSTRACT: Growth and development over the past one hundred years has resulted in the construction of causeways and navigational channels in Tampa Bay. Urbanization has lead to bathymetric alterations of the bay, and thus has effectively changed the residual salinity, flow fields, and transport time in Tampa Bay. In this study a numerical ocean model of Tampa Bay was first used to simulate the circulation for 2001-2003 using present day bathymetry. Then, an identical simulation was performed using the bathymetry generated from 1879 depth soundings. The residual (30-day time average) circulation fields and salinity, along with transport times was intercompared for different freshwater inflow conditions to investigate the impacts of these physical alteration. The salinity for the "present" simulation is about 3 psu higher than the "1879" simulation in the areas of upper Middle Tampa Bay, Lower Hillsborough Bay and Old Tampa Bay, a result of a stronger axial pressure gradient associated with the deeper more extensive channels. Velocities are up to 10 cm/s higher in the "present" run than in the "1879" simulation in the areas where water must converge and diverge through the narrow openings of the bridges/causeways. Transport time is short (~10 days) during strong residual circulation, and long (~90 days) during weak residual circulation. Bridges and causeways are associated with longer transport times (~90 days), except in the area North of the Courtney Campbell Causeway. The navigational channel is associated with long transport times during dry periods and short transport time during wet periods.
502
Thesis (M.S.)--University of South Florida, 2007.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 98 pages.
590
Advisor: Mark E. Luther, Ph.D.
653
Model.
Residence time.
Estuary.
Navigational channel.
Bridges.
690
Dissertations, Academic
z USF
x Marine Science
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2092



PAGE 1

Bathymetric Alterations Due to Urbanization and Their Effects on Residual Salinity, Flow Field and Transport Time for Tampa Bay, Florida by Amanda J. Linville A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: Mark E. Luther, Ph.D. Steven Meyers, Ph.D. Boris Galperin, Ph D. Date of Approval: May 16, 2007 Keywords: model, residence time, estuary, navigational channel, bridges Copyright 2007, Amanda J. Linville

PAGE 2

ACKNOWLEDGEMENTS I would like to thank my friends and family for all their support, without them this thesis would not have been possible.

PAGE 3

i TABLE OF CONTENTS List of Tables................................................................................................................. .....ii List of Figures....................................................................................................................iii Abstract....................................................................................................................... .....viii Introduction.........................................................................................................................1 Tampa Bay History And Significance....................................................................3 Recent studies in estu ary circulation.......................................................................7 The Tampa Bay Numerical Circulation Model....................................................10 Materials and Methods......................................................................................................14 Bathymetry............................................................................................................14 Model Simulations................................................................................................15 Results...............................................................................................................................18 Salinity..................................................................................................................18 Surface......................................................................................................18 Bottom.......................................................................................................19 Velocity.................................................................................................................22 Surface......................................................................................................22 Bottom.......................................................................................................23 Transport Time......................................................................................................24 Flushing Time.......................................................................................................24 Residence Time.....................................................................................................24 Discussion.........................................................................................................................27 Salinity and Velocity.............................................................................................27 Transport Time......................................................................................................29 Caveats........................................................................................................................ ......32 Future Work.................................................................................................................... ..33 References.........................................................................................................................94

PAGE 4

ii LIST OF TABLES Table 1 The conditions during the model time periods examined. Given are the start date and duration of the time period, the mean total freshwater, rms wind speed, and mean wind during the duration of the time period indicated................39

PAGE 5

iii LIST OF FIGURES Figure 1. Bridge and causeways spanning Tampa Bay. ...................................................34 Figure 2. Present bathymetric grid for Tampa Bay, FL. ...................................................35 Figure 3. 1879 Bathymetric grid for Tampa Bay, FL. ......................................................36 Figure 4. Bathymetric grid difference, Present 1879. .....................................................37 Figure 5. Differenced residual RMS speed, grey strips indicate the three 30 day periods selected for analysis. .......................................................38 Figure 6. Tampa Bay Freshwater In flow graph, boxes indicate the three 30 day periods selected for analysis, T1, T2, and T3 respectively. ................................................................................................40 Figure 7. Tampa Bay Model wind speed, boxes indicate the three 30 day periods selected for the analysis, T1, T2, and T3 respectively. ......................................................................................................41 Figure 8. Tampa Bay Model dir ection speed, boxes indicate the three 30 day periods selected for the analysis, T1, T2, and T3 respectively. .........................................................................................42 Figure 9. Tampa Bay Freshwater In flow graph, boxes indicate the three 90 day periods selected for the residence time analysis, T1r, T2r, and T3r respectively. ..........................................................43 Figure 10. Tampa Bay Model wind speed, boxes indicate the three

PAGE 6

iv 90 day periods selected for the residence time analysis, T1r, T2r, and T3r, respectively. ........................................................................44 Figure 11. Tampa Bay Model wind speed, boxes indicate the three 90 day periods selected for the residence time analysis, T1r, T2r, and T3r, respectively. ........................................................................45 Figure 12. Residual surface salinity during T1 for the present run..................................46 Figure 13. Residual surface salinity during T1 for the 1879 run......................................47 Figure 14. Residual surface salinity difference (p-1879) during T1. .....................................................................................................................48 Figure 15. Residual surface salinity during T2 for the present run..................................49 Figure 16. Residual surface salinity during T2 for the 1879 run......................................50 Figure 17. Residual surface salinity difference (p-1879) during T2. .....................................................................................................................51 Figure 18. Residual surface salinity during T3 for the present run..................................52 Figure 19. Residual surface salinity, OTB, during T3 for the present run. .......................................................................................................53 Figure 20. Residual surface salinity during T3 for the 1879 run......................................54 Figure 21. Residual surface salinity difference (p-1879) during T3. .....................................................................................................................55 Figure 22. Residual bottom salinity dur ing T1 for the present run...................................56 Figure 23. Residual bottom salinit y during T1 for the 1879 run......................................57 Figure 24. Residual bottom salin ity difference (p-1879), during T1. .....................................................................................................................58

PAGE 7

v Figure 25. Residual bottom salinity dur ing T2 for the present run...................................59 Figure 26. Residual bottom salinity, OTB, during T2 for the present run. .......................................................................................................60 Figure 27. Residual bottom salinit y during T2 for the 1879 run......................................61 Figure 28. Residual bottom salinity, OTB, during T2 for the 1879 run. ....................................................................................................................62 Figure 29. Residual bottom salin ity difference (p-1879) during T2. .....................................................................................................................63 Figure 30. Residual bottom salinity dur ing T3 for the present run...................................64 Figure 31. Residual bottom salinit y during T3 for the 1879 run......................................65 Figure 32. Residual bottom salin ity difference (p-1879), during T3. .....................................................................................................................66 Figure 33. Residual bottom salin ity difference (p-1879), OTB, during T3. .........................................................................................................67 Figure 34. Residual surface velocity during T1 for the present run.................................68 Figure 35. Residual surface velocity during T1 for the 1879 run.....................................69 Figure 36. Residual surface velo city difference (P-1879), during T1. .....................................................................................................................70 Figure 37. Residual surface velocity during T2 for the present run.................................71 Figure 38. Residual surface velocity during T2 for the 1879 run.....................................72 Figure 39. Residual surface velocity difference (p-1879) for T2.....................................73 Figure 40. Residual surface velocity during T3 for the present run.................................74 Figure 41. Residual surface velocity during T3 for the 1879 run.....................................75

PAGE 8

vi Figure 42. Residual surface velocity difference (p-1879) for T3.....................................76 Figure 43. Residual bottom velocity during T1 for the present run..................................77 Figure 44. Residual bottom velocity during T1 for the 1879 run.....................................78 Figure 45. Residual bottom veloc ity difference (p-1879) during T1. .....................................................................................................................79 Figure 46. Residual bottom velocity during T2 for the present run..................................80 Figure 47. Residual bottom velocity during T2 for the 1879 run.....................................81 Figure 48. Residual bottom veloc ity difference (p-1879) during T2. .....................................................................................................................82 Figure 49. Residual bottom velocity during T3 for the present run..................................83 Figure 50. Residual bottom velocity during T3 for the 1879 run.....................................84 Figure 51. Residual bottom veloc ity difference (p-1879) during T3. .....................................................................................................................85 Figure 52. Flushing time in days during T1 (bold solid line) 187 days, T2 (solid line) 97 days, and T3 (dotted line) 36 days for the Present run. ...................................................................................86 Figure 53. Flushing time in days during T1 (bold solid line) 144 days, T2 (solid line) 85 days, and T3 (dotted line) 32 days for the 1879 run. .......................................................................................87 Figure 54. Residence time during T1r for the Present run................................................88 Figure 55. Residence time during T1r for the 1879 run. ..................................................89 Figure 56. Residence time during T2r for the Present run................................................90 Figure 57. Residence time during T2r for the 1879 run. ..................................................91

PAGE 9

vii Figure 58. Residence time during T3r for Present run. .....................................................92 Figure 59. Residence time during T3r for the 1879 run. ....................................................93

PAGE 10

viii Bathymetric Alterations Due to Urbanizati on and Their Effects on Residual Salinity, Flow Field and Transport Time for Tampa Bay, Florida Amanda J. Linville ABSTRACT Growth and development over the past one hundred years has resulted in the construction of causeways and navigational channels in Tampa Bay. Urbanization has lead to bathymetric alterations of the bay, and thus has effectively changed the residual salinity, flow fields, and transport time in Tampa Bay. In this study a numerical ocean model of Tampa Bay was first used to simulate the circulation for 2001-2003 using present day bathymetry. Then, an identical simulation was performed using the bathymetry generated from 1879 depth sounding s. The residual (30-day time average) circulation fields and salinity, along with tr ansport times was intercompared for different freshwater inflow conditions to investigate the impacts of these phys ical alteration. The salinity for the present simulation is about 3 psu higher than the simulation in the areas of upper Middle Tampa Bay, Lower Hillsborough Bay and Old Tampa Bay, a result of a stronger axial pressure gradient associated with the d eeper more extensive channels. Velocities are up to 10 cm/s higher in the present run than in the simulation in the areas where water must converge and diverge through the narrow openings of the bridges/causeways. Trans port time is short (~10days) during strong residual circulation, and long (~90 days) during weak residua l circulation. Bridges and causeways are associated with longer trans port times (~90days), except in the area North of the Courtney Campbell Causeway. The navi gational channel is associated with long transport times during dry periods and s hort transport time during wet periods.

PAGE 11

1 INTRODUCTION Over the past one hundred years many areas surrounding waterways have experienced rapid human population growth. Coas tal counties represent just 17 percent of the total acreage of the contiguous U.S., yet in 2002 they were home to more than half of the nations people (Beach, 2002). This high pop ulation density has generated a need for all types of infrastructure including roads, causeways, bridges and navigational channels. Many of these are built within semi-enclosed waterways such as bays and estuaries directly altering their bathymetry. Over the coming decades, as population increases, so will the need for additional infrastructure (Beach, 2002). Estuaries, places where rivers meet the sea, are among the most unique and ecologically important ecosystems on our planet (Cloern, 2001). Thousands of species of birds, fish, mammals, and other wildlife depend on estuarine habitat as places to f eed, reproduce, and to live. Identifying and assessing the environmental im pacts on estuarine systems due to urbanization is essential to managing their ecological health. Bathym etric alterations can significantly impact hydrodynamic circulation within an estuary. The hydrodynamics of the estuary controls the flux of salt and other biological, chemi cal and geological materials in estuarine environments (Kim, 2005). The second is a nontidal baroclinic circ ulation generated by a head to mouth salinity gradient (Galpe rin et al., 1991; Weisberg and Zheng, 2006; Meyers et al., 2007), with typical speeds on the order of 10 cm/s. The salinity gradient is regulated by the freshwater discharge into the bay and the baroclinic circulation is

PAGE 12

2 modified by the surface wind stress. This study will use a numerical ocean model to examine the hydrodynamic effects of bathymet ric alterations due to urbanization and their effects on residual salinity, flow field a nd transport time in Tampa Bay, Florida. The bathymetry of Tampa Bay has been si gnificantly altered in the past century due to urbanization. With an average dept h of 3.7 meters at Mean Lower Low Water (MLLW), Tampa Bay is a drowned river bed w ith channels that have been deepened by dredging (Zervas, 1993). A major ship channe l has been dredged from the mouth to the upper reaches of Lower Tampa Bay and into Mi ddle Tampa Bay, where it splits into two branches, one entering Old Tampa Bay and the other going into Hillsborough Bay. This dredging has increased the main channel de pths from 10 meters to 15 meters (Vincent 2001). In addition to the dredging, four major bridges now span the Bay: the Courtney Campbell Causeway (CCC), the Howard Frankl in Bridge (HFB), the Gandy Bridge (GB), and the Sunshine Skyway Bridge (SSB) (Fig 1). Physical alterations like these are expected to influence circulation and transp ort time in estuaries, which in turn can profoundly affect the Bays ecosystems. In Kingston Harbour, Jamaica a hydrodynamic modeling study found contrary to what was suspected, stagnant conditions in the bay would occur even without the presence of a major causeway (Andrew et al 1999). A study conducted in Sheepscot River Estuary, Maine found that fo llowing the actual removal of a causeway, tidal flows in th e main channel increased by almost 50% (McAlice and Jaeger, 1983). These studies sugges t that the effect of an infrastructure on an estuary largely depends upon the type of an infrastructu re and the local hydrodynamics of the bay.

PAGE 13

3 In this study a numerical ocean model of Tampa Bay will first be used to simulate the circulation for 2001-2003 using present day bathymetry (Fig 2). Then, an identical simulation will be performed using the bathymetry generated from the 1879 depth soundings (Fig 3).The residual (30-day time av erage) circulation fi elds and salinity, along with transport times will be intercompared for different freshwater inflow conditions to investigate the impacts of these physical alteration. Tampa Bay History And Significance The Tampa Bay estuary is located near 82.5 degrees longitude West and 27.60 degrees latitude north on the west-central coast of peninsular Florida, and extends approximately 60 km to the northeast and is about 11 km wide(Vincent 2001). With a surface area of 1032 km 2 it is one of the largest estu aries in the U.S. (Donahue 2003). Tampa Bay consists of four major water bodies: Old Tampa Bay (OTB), Hillsborough Bay (HB), Middle Tampa Bay (MTB), Lower Tampa Bay (LTB) (Lewis &Whitman 1985) (Fig. 1). Tampa Bay is an essential sanctuary for waterfowl, marine life, plants, invertebrates and mammals. The Bay supports th e most diverse colonies of shorebirds in North America (TBEP, 2005a) and a resident population of more than 500 dolphins (TBEP, 2005b). More than 200 species of fish can be found in Tampa Bay and it is home to the endangered Manatee. Tampa Bay encompasses an environment that also provides aesthetic enjoyment, with its miles of blue waterways, white s horelines, and a near-perfect average annual temperature of 72 degrees (The Greater Tampa Chamber of Commerce, 2005). People

PAGE 14

4 flock to Tampa Bay to enjoy activities on the water like swimming, kayaking, boating, and fishing. Besides having an ecological and recreatio nal value, Tampa Bay has harbors and ports vital for shipping, transportation, and industry. The bay contributes more than 5 billion dollars annually from trade, tourism, development and fishing, and boasts three major seaports (TBEP, 2005b). The three por ts that border Tampa Bay support 130,000 jobs and contribute more than $15 billion per year to the local economy (TBEP, 2005b). The Port of Tampa handles nearly half of all sea borne commerce in Florida (Bay Soundings, 2003). Floridas population has increased by 600 % in the last 50 years (U.S. Census). Most of the urbanization is taking place along side the estuary and placing stress on the quality and quantity of this valuable habitat. Today, about 2.5 million people live in the four-county metropolitan statistical area (MSA ) that surrounds the bay, making it the second-largest MSA in the Southeast and the 21st-largest MSA in the U.S. It is important to realize the critic al role Tampa Bay has in preser ving the abundance of wildlife and health of the environment and economy and to minimize adverse effects on it. Egmont Channel to the north of Egmont Key, at the mouth of the bay, is the major connection between the Gulf of Mexico and Tampa Bay. The Southwest Channel and Passage Key Inlet are channels located on th e south side of the bay. The original navigational charts of 1879 reveal a natural central channel running from Egmont Key to the area south of Gadsen point (Kehring, 1985). Natural depths range from 6 to 10 meters and in one area at the Bays mouth, the na tural depth is 27 meters (NOS, 1990). Dredging of ship channels has increased the main ch annel depths to 15 meters (NOS, 1990) in

PAGE 15

5 some areas (Fig. 4). As shown in the st udy by the New South Whales Department of Natural Resources (NWNR, 2005), dredging can have a number of harmful effects on estuarine habitats and ecosystems. The seagrass beds may be destroyed or degraded by mining the underlying sediment causing increased levels of turbidity and sedimentation. Dredged channels or any deep hole can also cause flow to become stagnant or may adversely affect current patterns also harming seagrass beds. Bridge and causeway infrastructure as we ll as dredging can have an adverse effect on an estuary (NWNR, 2005). In particular, causeways across an estuary can severely restrict tidal flows and reduce the tidal pr ism (NWNR, 2005). The reduced fluctuation in tidal water levels and velocities, together with highly reduced salinity levels, can lead to a progressive, but ultimately marked change to upstream habitat areas, including the loss of mangroves and the smothering of any upstream seagrass beds (NWNR, 2005). These tidal restrictions can also affect transport time scales, or the amount of time a material spends in the bay before exiting. Much work has been done to understand transport time scales because they affect the dilution of nutrient and contaminants inputs. In a transport time study conducted in Tampa Bay by Burwell et al (2001), it was found that in areas restricted by causeways, time scales exceeded 140 days, while areas adjacent to the main channel were 10 to 20 days or less. Predicti ng the transport and fate of pollutants in the estuaries and the coastal zone appears as one of the most important challenges of the environmental sciences. Tampa Bay tidal waters are categorized as mixed, semi-diurnal (Goodwin, 1987). Tides of this classification are composed of two high and two low unequal tides during a

PAGE 16

6 one day period. Mean range of tide and gr eat diurnal range are 0.44 meters and 0.74 meters respectively for Tampa Bay (NOS, 1990). Tidal analysis indicates that the tide travels from the mouth of the bay to the h ead of Old Tampa Bay and Hillsborough Bay in approximately 4.6 hours and 3.2 hours respectively. The residual water level signal is define d here as the observed water level minus the harmonic water level. The meteorologica l forcing responsible for this non-tidal contribution ranges from seasonal steric hea ting to synoptic scale frontal passages and local thunderstorms (Vincent, 2001). Currents in Tampa Bay are also categorized as mixed predominately semi-diurnal. Maximum current speeds on the order of 1. 0-1.5 m/s are observed in the Egmont Channel, the channel near the Skyway Bridge, and in the channel leading to Old Tampa Bay (Vincent, 2001). Like the bay water levels, the currents also contain a large non-tidal component. Zervas (1993) found that the ratio of the standard deviation of the residual currents to the standard deviation of the obser ved currents ranged from 17.2 % to 88.2%. The salinity of the bay is governed by th at of the Gulf of Mexico, freshwater inflow from rivers, streams, groundwater, non-point sources and rainfall. Evaporation also removes a large volume of water from th e bay. Initial bulk aerodynamic calculations suggest that the annual evaporation is approxi mately the same magnitude as the annual precipitation (Vincent, 2001). The major rivers include the Alafia, H illsborough, Manatee and Little Manatee. The Alafia and Hillsborough rivers drain into the bay from the Northeast, near the head of the bay; little Manatee enters on the Eastern side; and the Manatee at the South, near the mouth of the bay. The mean value of the total freshwater inflow is 9.4 million m 3 d -1

PAGE 17

7 and ranges from 0.02 to 226 million m 3 d -1 Forty percent of the av erage fresh water input is in the form of precipi tation (Meyers et al., 2007). Small freshwater inflow rates combined with shallow depth, and strong tidal mixing lead to a vertical salinity structure th at is well mixed (Galperin, 1991). Vertically quasihomogenous estuaries with strong horizon tal salinity gradients have been shown to have developed baroclinic (density driven ) subtidal flows (Wong 1994: Pritchard 1956), which that contribute to the re sidual circulation of the bay. Recent studies in estuary circulation In 1987 Carl Goodwin of the U.S. Geological Survey conducted a study similar to this one only he used a two-dimensional, finite difference, hydrodynamic model to simulate flood, ebb, and residual transport of both water and dissolved constituent for the physical conditions that existed in Tampa Bay during 1880 and 1972 and for the conditions that were likely to exist in 1985 (Goodwin, 1987). The model employed by Goodwin assumes that the density driven mode of circulation is not important due to the vertical mixing and the shallowness of the ba y. The goal of the project was to try to assess the impact of dredge and fill projects on water circulation. The model predicted the tidal stages at different locations in the bay reasonably well; however, it could not model the density-driven (baroclinic) residual circul ation in Tampa Bay (Li, 1993) and therefore is not useful for examining issues related to long-term transport within the bay. Weisberg and Williams (1991) concluded that although Tampa Bay is vertically well-mixed in salinity, it does have a dynamica lly significant horizonta l salinity gradient due to the distribution of inflowing freshwater. These horizontal gradients and surface

PAGE 18

8 wind forcing maintain a fully three-dimensional circulation in the Bay. Galperin et al (1991) used a three-dimensional, time-depende nt model of circula tion in Tampa Bay, they have found that the circulation in the ba y, is complex and three-dimensional, it is influenced by the horizontal dens ity gradients throughout the estuary. The force generated by the horizontal salinity gradient in Tampa Bay drives a non-tidal, baroclinic circulation (Galperin et al 1992a, Li 1993, Weisberg et al 2006). This baroclinic circulation, also known as estuar ine circulation, is characteri zed as a two layer flow, with more saline water flowing landw ard beneath a top layer of fre sher water flowing seaward (Pritchard, 1956). The strength of the baroclinic currents varies with the magnitude of the salinity gradient (Meyers et al., 2007). The numerical model used here is based on the three dimensional Estuarine and Coastal Ocean Model (ECOM-3D), a variation of the Princeton Ocean Model (Blumberg and Mellor, 1987). The Tampa Bay model is a three-dimensional, time-dependent model of the hydrodynamics of circulation in Tampa Bay (Galperin et al., 1992a, b; Vincent et al., 1997) The need to calculate transport time scales (ex. flushing, residence time) arises in many studies of estuaries. Transport time s cales affect the dilution of nutrient and contaminants inputs as well as a variety of eco logical processes, and can be used to assess the ecological implications of changes in the ci rculation of an estuary. Better descriptors of estuarine and coastal dynamics will enable coastal managers to more effectively make decisions. There is a historical lack of consistency in the definition of transport time scales. Many terms are used in the literature and ar e often inconsistent and sometimes imprecise.

PAGE 19

9 Care must be exercised to determine the meaning of terms being used, to avoid misinterpretation or incorrect comparisons of data. Recent works identify three different concepts of transport time; flushing time/ret ention time, age and residence time (Monsen et al., 2001). Flushing time/retention time is de fined as time to replace the volume of the estuary by the total freshwater input influx (O fficer and Kester, 1991). Age is defined as the time elapsed since the material entere d the system (Zimmerman, 1988). Residence time has been defined as the time it takes a water parcel to leave the system (Dronkers and Zimmerman, 1982). Flushing time is an integral system measure (baywide calculation) of transport time, whereas both residence time and age are local measures (spatial calculations) (Monsen, 2001). There is no single transport time scale for a system that is applicable to all time periods, locations, and constituents, and no one time scale describes all transport processes (Monsen et al., 2001). Therefore, identifying the unde rlying assumptions associated with the application of that concept and its validity will determine which concept to adopt. Here, the calculation of both system, specifically flushing time, and local parameters, specifically residence time, are performed to help give a more complete characterization of tran sport time scales for Tampa Bay. Calculating flushing time and/or residen ce time itself can be estimated using many different methods (Pilson,1985, Geyer,1997, Solis and Powell, 1999, Callaway, 1981, Dettmann et al., 1989, Dettmann et al ., 1992, Hagy et al., 2000, and Miller and McPherson, 1991). In this study a numerical computer model will be used calculate a global flushing time similar to the method us ed by Monsen (2001), and Oliveira and Baptista (1997), where the number of particles is fit to an exponential decay. The same

PAGE 20

10 model will also be used to calculate a lo cal residence time. The method for calculating residence time will be the same Lagrangian method used by Burwell in 2001 on Tampa Bay, it is a method similar to the ones used by Brooks et al. (1999), Signell (1992), Signell and Butman (1992). Burwell (2001) conducted a numerical study in Tampa Bay that compared two distinct numerical model methodologies (E ulerian and Lagranian) for determining spatiotemporal variation in estuarine residence times. The Eulerian approach was based on the concentration equation, using advect ion and diffusion of a passive tracer in the model domain, while the Lagranian method is based on a particle tracking approach, where neutrally buoyant dimensionless particle s are advected by the model velocity and a small random Markovian displacement (Bur well 2001). Burwell concluded that the Eulerian method is sensitive to the value of the horizontal diffustivity whereas the Lagrangian approach provides a more realis tic sub-grid scale mo tion (Burwell 2001). He also concluded that the Lagrangian spatial distributions show more detail in each subregion of the bay, and thus indicates how the relative residence time varies spatially within these regions. The Lagrangian approach is therefore the method used in this study for calculating the residence time. The Tampa Bay Numerical Circulation Model The Tampa Bay model solves equations fo r the conservation of mass, momentum, salt, heat and turbulence quantities in an incompressible hydrostatic fluid (BlumbergMellor, 1987; Galperin et al. 1992a, 1992b, Vincent et al, 19 97, Meyers et al., 2007). The

PAGE 21

11 model uses a grid with 70 by 100 cells in the horizontal and 11 sigma levels in the vertical. The dimensions of active cells range from 2240 meters to 307.67 meters, with a mean of 668.34 meters. The mean cell area is 0.425 square kilometers. The maximum, mean and minimum depths of the grid cells are 13.93, 3.62 and 1.3 meters mean lower low water (MLLW) respectively. Turbulence closure is provided by an embedded Mellor Yamada 2.5 level closure submodel (Mellor and Yamada, 1982) as modified by Galperin et al (1988). A mode splitting technique is used that separates the fast external gravity waves and the slow moving internal grav ity waves (Blumberg and Mellor, 1987). In Tampa Bay, the model is forced with hourly water levels, and daily winds, freshwater inflows, precipitation, evaporation, and m onthly salinity at the mouth. Along the lateral boundaries, rivers are prescribed by temperatur e (set to 25 C), salinity (set to zero), and flow rates as set by Meyers et al. (2007) Among the important parameters computed from the model are free surface height, magnitude and direction of currents, and the fields of temperature and salinity. Several of the model boundary conditions are obtained from PORTS (Physical Oceanographic Real-Time System) and CO MPS (Costal Ocean Monitoring and Prediction System) data. Tampa Bay PORTS includes the measurement of real-time currents, water levels, winds, wave height, visibility, air and water temperatures, and barometric pressure at multiple locations with a data dissemination system. COMPS is a network of instruments both along the coast of Tampa Bay and offshore of Florida that measure current, temperature, salinity, and meteorological parameters, which satellite telemetry of the data to the shore. These data streams are automatically updated every 6

PAGE 22

12 minutes and can be viewed from the OMPL Web site (ht tp://ompl.marine.usf.edu) and accessed via a DODS (Distributed Ocean Data System) server. River flow data used to run the model is either from the United States Geological Survey or interpolated using an algorithm based on drainage areas River flow includes estimates of underground seepage into rivers of about 8% of the mean flow (Vincent, 2001). Direct precipitation into the bay is estimated from regional rain gauges. Observational evaporation was provided by the Southwest Florida Water Management District (SWFWMD). Details of the boundary conditions are available in Meyers et al. (2007). The modeled channel is shallower and wider than the actual channels. However the model has been extensively evaluated ag ainst observations and has been shown to capture the hydrodynamics of Tampa Bay realis tically. For more information about the evaluations see Vincent et al (1991) and Meyers et al. (2007). Near Egmont Key, at the mouth of the Ta mpa bay, a naturally scoured hole exists with a depth of about 27 meters. The depth of the area surrounding this hole is around 10 meters in the 1879 bathymetric grid, and about 15 meters in the pres ent bathymetric grid. This area of steep topography may have potential to create a vorticity error in the model. Studies have indicated that the use of si gma coordinates may cause numerical errors when dealing with steep topography (Haney 1991, Mellor et al 1993, Mellor et al 1997).More specifically the st udy by Mellor et al (1997) f ound that three-dimensional sigma models contain a vorticity error the au thors call a sigma coordinate error, that unlike in two-dimensional models, the erro r does not decay advectively. The error however is described by the authors of that study as being tolerably small. The potential

PAGE 23

13 numerical error produced by steep topography in Tampa Bay was therefore not taken into account for this study.

PAGE 24

14 MATERIALS AND METHODS Two model runs are performed for this st udy. The first utilizes the model output variables from Meyers et al. (2007). In pa rticular, model salinit y and velocity are recorded every model hour for the years 2001-03 using realistic boundary conditions and modern bathymetry (present day) (Fig 2). Th e second model run is id entical to the first except the bathymetry was based on depth s oundings from 1879 () (Fig 3), whose development is described below. By only ch anging the bathymetric grid, differences in model output between each simulation will solely derive from changes due to construction and dredging. This study is not an attempt to recreat e the circulation in Tampa Bay in 1879. Bathymetry Bay bathymetry used for the present day grid scheme was obtained from the National Geophysical Data Center GEODAS version 3.3 and 4.0 data set. For the present grid scheme, the bathymetry at the center of each cell (i,j) was computed from the mean of all data points within a ra dius of (m1(i,j) + m2(i,j))/4, which was converted to units of degrees. M1 (i,j) and m2 (i,j) are the model gr id cell matrices. A plot of the present day grid bathymetry is provided in Figure 2. For the 1879 grid scheme, depth soundings were obtained from the United States Geological Survey of St. Petersburg Florida. To construct the 1879 bathymetric grid, the 1879 de pth sounding locations were compared to the present day bathymetric grid. The depth sound ings that fell within each grid cell were

PAGE 25

15 averaged and used as the depth for that par ticular cell. Since the model requires that the grid cells surrounding freshwater inflow to be land, it was necessary to add a few cells of land to the 1879 bathymetry. The cells of la nd were added to Hillsborough Bay where the Hillsborough power plant is now located. Figure 3 is a plot of the reconstructed 1879 bathymetric grid. Some of the major differen ces in bathymetry can be seen in HB. The ship channel has been extended to reach upper Hillsborough bay, increasing the depth by 3-4 meters, and depths around the port of Ta mpa have also been deepened by up to 9 meters. Eight meters of dredged soil ha s been added to areas in HB. In OTB bridges/causeways now span the bay resulti ng in a 9 meter bathymetric gain in some areas. The natural channel in OTB has been dredged, increasing depth by 4 meters. MTB and LTB have also been dredged, resulting in a depth increase of about 2-3 meters. A four meter gain, in areas of LTB, is the result of the Sunshine Skyway Bridge. Model Simulations Thirty day averages of model output variables were computed in order to filter out the tidal signal and reveal the non-tidal, or residual, circulation. Three time periods were selected to represent minimal, moderate, and maximal differences between the present and 1879 runs. The difference between the averaged model velocities from each simulation (Present-1879) was calculated for ea ch grid point at every time. Then rootmean square (RMS) of the difference was calculated over the 3D grid at each time, yielding a time series of rms velocity differe nce. A thirty-day running mean was applied to each time series. The first time period sele cted for analysis, T1, corresponds to model days 150-180, the second, T2, corresponds to days 370-400, and third, T3, to model days

PAGE 26

16 974-1004 (Fig. 5). The mean freshwater inflow for T1, T2, and T3 is 5.4 million m3 d-1, 1.6 million m3 d-1, and 31.2 million m3 d-1, respectively (Table 1). However, a very high freshwater impulse occurred about 100 days before the start of T2 (Fig. 6). During T1 mean wind speed is 1.1 m s-1 and the mean direction is -197.4 (deg r) (Table 1) ( Figs 7 & 8, resp.) For T2, the mean wind speed is 0.6 m s-1 and the mean direction is 60.5 (degr) (Table1) (Figs 7 & 8, resp). For T3, the mean wind is 2.68101 m s-1 and the mean wind direction is 77.2 (degr) (Table1, Figs 7 & 8). In order to capture the baroclinic structure of the residual circulation the averaged model fields of salinity and velocity are exam ined across the model su rface (level 1 of the model) and near-bottom (level 9 of the model). Differences in the variables (salinity and velocity) between the two simulations (pre sent and 1879) are examined for T1, T2.and T3. For estimating transport time each model gr id cell was seeded with 100 particles and then run for three, 90 day time periods, T1r, T2r, and T3r. The 90 day time periods begin at the same start time as T1, T2, and T3 The mean freshwater inflow for T1r, T2r, and T3r is 5.5 million m3 d-1, 2.7million m3 d-1 and, 22.9621 million m3 d-1, respectively (Table 1) (Fig. 9). During T1r mean wind speed is 0.462738 m s-1 and the mean direction is -194.4 (degr) (Table 1) (Figs 10 & 11, resp). For T2r, the mean wind speed is 1.1 m s-1 and the mean direction is 47.0 (degr) (Table 1) (Figs. 10 & 11, resp.). For T3r, the mean wind is 3.2 m s-1 and the mean wind direction is 68.0961 (degr) (Table1, Figs 10 & 11). In this study, transport time is define d as the amount of time for the number of particles N, to drop below the e-folding thresh old N0/e, where N0 is the initial number of particles and e=2.71828 Two calculations are performed. The first is the flushing time.

PAGE 27

17 In this case N is only a function of time, and N=N(t), is the total number of particles in the bay. The particle count N(t), is fit w ith an exponential function Aexp(t/F), where A=1. The value of FT is the flushing tim e. The second calculation is residence time where N is a function of both space and time, and N=N(x,y,t) is the total number of particles in the water column of each grid cell. For this calcula tion N does not fit an exponential curve. Residence time (RT) is defi ned as the last time, within the period of the calculation, where N drops below N0/e. Also twenty five hour averages of N are used in order to filter out the tidal signal.

PAGE 28

18 RESULTS Salinity Surface Overall surface salinity in both (present &1879) simulations is higher at the mouth of the bay and lower at th e head of the bay. This gradient varies over time, with T1 having the weakest gradient (34 to 33psu) a nd T3 having the strongest (27 to 17psu). The biggest salinity differences between the pres ent and the 1879 run occurs during T3, up to 5 psu near the head of the bay. The present surface salinity during T1 is around 34 psu at the mouth of the bay, and about 33 psu at the head of the bay (Fig 12). The surface salinity for 1879 run is almost the same as the present (Fig 13). Fo r the present run, bay salinity is < 0.5 psu lower throughout most of the bay than in the 1879 run (Fig 14). In Lower Tampa Bay (LTB) around the Northern section of the Suns hine Skyway Bridge (SSB), the salinity is <0.5 psu higher than in the 1879 run (Fig 14). During T2, for the present salinity run, a salinity tongue stretches from the mouth of the bay into Middle Tampa Bay (MTB), the mouth of the bay is around 33psu and about 27psu at the head of the bay (Fig 15). In the 1879 run the salinity tongue is much wider, and there is a weaker salinity grad ient in Old Tampa Bay (OTB) than in the present run (Fig 16). The present run is <1 psu higher in upper Tampa Bay, and <1 psu lower in LTB than in the 1879 run (Fig 17) The present salinity run is also higher by

PAGE 29

19 about 2 psu near river input s in Hillsborough Bay (HB), the Howard Franklin Bridge (HFB) and the Gandy Bridge (GB) (Fig 17) The present salinity run is around 2 psu lower near the Northwestern part of the SSB, and about 3 psu lower around the Northeastern part of the Courtney Campbell Causeway (CCC) (Fig 17). For T3, the present run is about 27 psu at the mouth of the bay where a salinity tongue stretches past the SSB, decreasing in salinity to about 17 psu in OTB (Fig. 18). Nearly half of these gradients occur within OTB (Fig 19). The salinity structure in the 1879 run is similar to the pres ent, except in 1879, the salinity tongue is much wider at the mouth of the bay and in HB th e salinity gradient is weaker (Fig 20). For the present run, salinity is about 3-5 psu higher in the Eastern part of HB and <1.5 higher in OTB as compared to the 1879 run (Fig 21).). The pres ent run is lower by a bout 2.5psu near the Northeastern side of the CCC, and <1-3 ps u lower around the Northwestern part of the SSB (Fig 21). Bottom Baywide bottom salinity in both (present & 1879) simulations is higher at the mouth of the bay and lower at the head of the bay. These sa linity gradients vary over time with the strongest gradient (28-17psu) in T3 and the weakest (34-33psu) in T1. In both runs, a salinity tongue stretches from the mouth of the Bay into MTB. This tongue is narrower in width and stretches further into MTB in the present run. The tongue stretches furthest in the pres ent run during T3. Present bottom salinity for T1 is abou t 34psu at the mouth of the bay and around 33psu at the head of the bay (Fig 22). The horizontal salinity structure for the 1879 run is

PAGE 30

20 almost the same as the present run (Fig 23). The salinity over the entire bay is <0.5 psu lower in the present run than in the 1879 r un, except the area around the North portion of the SSB, where it is <0.5 psu hi gher in the present (Fig 24). During T2, for the present bottom salinity, a saline tongue stretches from the mouth of the bay and narrows into MTB (Fig 25). Salinity at the mouth of the bay, for the present run, is about 33psu and around 27 psu at the head, with more than half of the salinity gradient occurr ing in OTB (Fig 26). In the 1879 r un, the salinity tongue is wider and does not stretch as far into MTB as in the present run (Fig 27). For the 1879 run, the salinity gradient in OTB is weaker than in the present run (Fig 28) Compared to the 1879 run, the present run is 2-3 psu higher in HB near the river inputs and 1.5-2.5 psu higher around the HFB and the GB (Fig 29). For the present run, salinity is 1-3 psu lower around the Northeastern section of the CCC and 1.5-2 psu lower ar ound the Northwestern part of the SSB (Fig 29). Bottom salinity during T3 for the present r un, has a salinity tongue that stretches from the mouth of the bay, and narrows into strips in MTB (Fig 30). The salinity at the mouth of the bay for the present run is about 28 psu and 17 psu at the head (Fig 30). The bottom salinity structure for the 1879 run is sim ilar to the present, except the tongue is wider and does not narrow into pronounced strips (Fig 31). Compared to the 1879 run, the present run is 2-3 psu highe r in upper HB, and 2.5-3 psu hi gher in the areas just west of the land spoils and around ri ver inflows (Fig 32) In MTB the present salinity run is higher by about 2-3 psu (Fig 32). In OTB the present run is 1-2 psu higher, except around the Northeastern portion of the CCC where salini ty is lower by 1.5-2 psu. (Fig 33) The area around the Northwestern section of the SSB is 1.5-2.5 psu lower in the present run

PAGE 31

21 than in the 1879 run (Fig 32).

PAGE 32

22 Velocity Surface In both simulations, residual surface veloci ty is seaward and is strongest in HB around the dredge spoils, in OTB and through the gaps in the CCC, the HFB, the GB, and in LTB around the SSB. The highest velocitie s seen are during T3 for both runs. During T1, surface residual flow in Ta mpa Bay is seaward at about 2-4 cm/s, except the southern part of OT B where velocity is about 4-8 cm/s, and near the mouth of the bay where velocities range up to 20 cm/s (Fig 34). The difference in residual velocity between present and 1879 for T1 is 1cm/s or less except near the mouth of the bay where the difference is about 3-5cm/s (Figs. 35 & 36). During T2, surface velocity is <4cm/s, except near the bridges/causeways, the channels and the mouth of the bay, where velocities are around 4-8 cm/s (Fig. 37). Surface velocity for the 1879 run has a simila r pattern (Fig. 38). Su rface velocity in Hillsborough bay is 4-6cm/s stronger in the present run than in the 1879 run (Fig. 39). Also during T2, areas around the openings of the CCC, the HFB, and the GB, have a surface velocity that is around 4-10 cm/s strong er in the present run than in the 1879 run (Fig. 39). As compared to the 1879 run, the surf ace velocity in the pr esent run is about 46 cm/s stronger in a strip from middle Tamp a Bay down to mouth of the Bay and around the SSB (Fig. 39). During T3, residual surface velocity is seaward from about 4-8 cm/s, with velocities ranging from to 10-20 cm/s near th e openings of the causeways/bridges, and

PAGE 33

23 over the channels (Fig. 40). Compared to the 1879 run, the present run, during T3 has a surface velocity up to 10 cm/s stronger around th e spoils in and near river inputs in HB (Figs. 41 & 42). The present run is also around 3-6cm/s stronger th roughout the rest of HB (Fig 42). In OTB, the surface velocity in the present run is about 10cm/s stronger around the openings of the CCC, the Howard HFB, the GB, and the SSB (Fig. 42). In MTB the surface velocity is 4-6 cm/s stronger in the present run, and 8-10 cm/s stronger in LTB (Fig. 42). Bottom In the present run, bottom velocity in the bay is generally a landward flow of about 2cm/s. During T1, there is little cha nge in velocity between the two runs, only a 1cm/s or less difference, near the mouth of the bay, and near the opening of the SSB (Figs. 43, 44, & 45). During T2, flow is gene rally landward by about 2cm/s or less, 4cm/s or less in the channel. The difference in velocity is about 1-4cm/s stronger for the present run in the channels, the dredge spoils in HB, and through the openings of causeways/bridges, except the CCC where the di fference ranges up to about 8cm/s (Figs. 46, 47, & 48). Flow during T3 is almost the same as T2 with the increased flow associated with the channels being more extensive (Figs. 49, 50, 51).

PAGE 34

24 Transport Time Flushing Time Flushing time for the present run, as co mpared to the 1879 run, is longer in both T1r and T2r, with the longest residence time occurring during T1r (Figs. 52 & 53). During T1r, for the present run, residence time is 187 days and is 144 days for 1879 run (Figs. 52 & 53). Residence time during T2r, fo r the present run is 97 days and 85 days for 1879. T3r residence time for the present is 27 days and is 32 days for the 1879 run (Figs 52 & 53). Residence Time Residence time across the bay varies locally from a few days to 90 days (the limit of this studys computation). In HB residenc e times are relatively short compared to the rest of the bay. In OTB, the area north of the CCC, and the lower area in the western region of OTB both have short residence ti mes as well. MTB near/in the channels, residence time is locally long during T1 a nd T2, but locally short during T3. On the Eastern side of OTB between the bridges, re sidence time is locally long, in the present run but short in the 1879. LTB has locally long residence times at the mouth of the bay and locally short residence times to the North of the SSB, in the present run, which is the opposite of the 1879 run. During T1r, the Present day residence time is around 45 days near the dredging spoils in Hillsborough Bay, which is about 25 days longer than the 1879 simulation (Figs. 54 & 55). The local residence time, north of the CCC, is around 20days for the present run and around 70days for the 1879 run (Figs. 54 & 55). Between the HFB and the GB on

PAGE 35

25 the Eastern side, local residence time for pres ent run is about 75days and about 20days in the 1879 run (Figs. 54 & 55). MTB around where the channels run, residence time is about 60-75days in the present and 50-70days in the 1879 run. On the Eastern flanks of MTB, local residence time for the presen t run is approximately 45days and is approximately 20days for the 1879 run (Figs. 54 & 55). Local residence time near the Eastern portions of the SSB for the pres ent run is 35-70days and 5-20days for the 1879 run (Figs. 54 & 55). The Western portion of the SSB is 5-35days for the present run and 5-10days for the 1879 run (Figs. 54 & 55). Residence time during T2r in HB around th e dredge spoils, is about 60 days for the present run and is about 20 days for the 1879 run (Figs. 56 & 57). In OTB, North of the CCC, residence time is about 5 days for the present run and about 10 days for the 1879 run (Figs. 56 & 57). On the Eastern side of Old Tampa Bay (OTB), between the HFB and the GB, the local residence time fo r the present run is approximately 45days and is approximately 20days for the 1879 run (Figs. 56 & 57). During T2r there is not a visible area of residence time th at is associated with the channels. In MTB on the Eastern side, the present run is about 80days and is about 40days for the 1879 run (Figs. 56 & 57). The Western flank of MTB is around 40days in the present and around 20days in 1879 run (Figs 56 & 57). The local residence tim e near the Eastern portions of the SSB for the present run is 45-75days and 20-35da ys for the 1879 (Figs. 56 & 57). The Western portion of the SSB for the pr esent is 5-40days and 30-35days for the 1879 run (Figs. 56 & 57). During T3r, local residence time in H illsborough Bay is about 10 days for the present run, and about 5 days for the 1879 run (Figs. 58 & 59). In OTB North of the

PAGE 36

26 CCC, local residence time is approximately 5 days for the present run and approximately 7 days for the 1879 ( Figs. 58 & 59). The ch annels in MTB have a residence time of about 15-25days for both runs, except the part of the channel near the SSB where during the present run it is about 35-65daysand about 10-35days in the 1879 run (Figs. 58 & 59).

PAGE 37

27 DISCUSSION Salinity and Velocity The flow pattern observed in Tampa Bay is a typical two-layered estuarine flow, a process first described by Pritchard (1956). Sa linity contrast from the head to the mouth of the bay, sets up an axial pressure gradient that drives a non-tidal gravitational circulation in the bay (Galperi n et al 1992a,b, Weisberg 2005). Internal pressure gradient force drives flow into the bay in the deeper channels (Galperin et al 1992a,b, Weisberg 2005). External pressure gradient force drives flow out of the bay at shallower depths (Galperin 1992a,b, Weisberg 2005). Residual circulation is greatly modulated by freshwater influx. T1 and T2 both have low freshwater inflow (5.5 million m 3 d -1 or less). A large fres hwater inflow pulse occurred about 100 days before the start day of T2. This impulse caused the density gradient to be stronger in T2 than in T1, de spite the fact that T2 had the lowest mean freshwater inflow. This higher density gradient resulted in T2 having a stronger residual circulation, compared to T1. T3 features the greatest amount of freshwater inflow and corresponds to the time period with the greatest salinity gradient and the strongest residual circulation in both the present and the 1879 simulations. During T3, the salinities in upper MTB, lower HB and OTB are higher in the present run than in the 1879. This is a resu lt of more extensive channels and a greater pressure gradient associated with the added depth of the channels. Salty Gulf of Mexico

PAGE 38

28 (GOM) water is funneled at depth up Ta mpa Bay through the channels, which is a baroclinic consequence of incr easing depth. The baroclinic contribution to the pressure gradient force is due to the vertical integral of the horizontal salinity gradient, and this integral is largest in the channel, see e.g., Galperin 1992a,b, and Weisberg 2005). The relatively high surface velocities seen in lower Western HB, in both runs, are the result of water convergence due to bay geometry. Higher surface velocities in HB during the present run are the result of water converging and diverging around dredge spoils (which are not in the 1879 run). Higher bottom velocities in eastern HB are also the result of a stronger in ternal pressure gradient that ex ists in the present run due to a more extensive, deeper channel. Three bridges/causeways exist in OTB for the present run. In OTB, the cells adjacent to the Northeastern portion of the CCC has a lower salinity in th e present run during T3 because the CCC impedes circulati on by trapping freshwater input from upper OTB. The strong velocities seen in OTB during T2 & T3 for the present run result from water converging and diverging around th e bridges/causeways, through their narrow openings. These infrastructures do not exis t in the 1879 run and th ese strong velocities are not seen in that run. Of the bridges/ causeways, the CCC has the strongest velocities through its opening and it also has the narrowest opening. In the present, the salinity tongue at the mouth of the bay must converge and squeeze through the SSB, resulting in a narrowe r tongue in the present run than the 1879 run, as the SSB does not exist in the 1879 run. This constriction associated with the SSB also causes velocities to be hi gher, as water converges and diverges in order to moves through the narrow opening. In MTB, the internal pressure gradient force is larger in the

PAGE 39

29 present run due to a deeper channel. This causes more saline Gulf of Mexico (GOM) water to be funneled into MTB, thus cont ributing to the higher salinities and stronger velocities seen in the present run. This feature is most prominent during T3, when salinity gradients are the largest, and thus th e pressure gradient is the strongest. Transport Time T1r has the longest flushing time, and is the period with th e weakest salinity gradient and residual circulation. T3r has the shortest flushing time and is the period with the strongest density gradient and residual ci rculation. Flushing time is longer in the present run than the 1879 run, except during T3r, where the 1879 run is 5 days longer. The biggest difference between the runs is dur ing the time when the residual circulation is the weakest, suggesting that this is time period when the bathymetric alterations have the greatest effect on circulation. Regions of distinct geometry /flow regime tend to behave as units with regard to residence. Here, channels are associated with long residence times (60-90days) during T1r and T2r (relatively dry periods) and short residence times (5-30days) during T3 (relatively wet period). In T2r, residence times associated with th e channels is not distinguishable as in the other time periods. During T2r, residence times in the main segment of the bay are short on the west side and long on the east side. This could be because there was a strong eastward wind in T2r that occu rred during the last 10 days of the simulation period. The eastward wind possibly moved particles to the eastern section of the channels and to the

PAGE 40

30 shallow eastern flanks of MTB. Then th e simulation period ended, possibly before mixing could take place, and the particles could be moved. The SSB, which only exists in the present run, is also associated with longer residence times as it impedes bay inflow and outflow and traps particles in areas immediately surrounding the infrastructure. So uth of the CCC in OTB, particles become trapped and restricted by th e bridges/causeways, causing a longer residence time in the present run, with the longest residence time occurring in T1r, the period with the weakest residual circulation. This effect is also most prominent on th e eastern side of OTB, south of the CCC, because the flow is weaker here than in the western side. The western side has stronger velocitie s that are associated with na tural bay geometry. These longer residence times associated with flow constr ictions do not occur in the 1879 run as these infrastructures do not exist. Upper OTB, North of the CCC, has a loca lly short residence time due to tidal rectification, which is most pronounced dur ing T3r, when residual circulation is the strongest. This effect is exaggerated further in the present run by th e magnified velocity through the opening of the CCC, a result of th e water constriction by the infrastructure The calculation of flushing time is an e xponential curve, so the introduced mass never completely leaves the system and flus hing is never complete. It only reflects the average amount of time the mass spends in a system Flushing time in the present run during T1r, and T2r is probably longer than in the 1879 because of the impediment of circul ation caused by the bridges/causeways, and from particle entrapment associated with th e channels during weak residual circulation. For T3r, flushing is probably shorter in the 1879 run as compared to the present run

PAGE 41

31 because it has a strong residual ci rculation. It is possible that the channels do not entrap particles when residual circul ation is strong. Strong residual circulation enhances flow in the channels, and because the present run has deeper more extensive channels, flow is stronger and thus flushing time is shor ter than in the 1879 run. Also, the causeways/bridges do not appear to impede ci rculation as much during a strong residual circulation. These effects may be why the biggest difference in flushing time between the present and 1879 runs is during T1r, a time with the weakest residual circulation. T3r has the strongest residual circulation and the diffe rence in flushing time between runs is the smallest, demonstrating that these infrastruc tures are not greatly effecting flushing during this time.

PAGE 42

32 CAVEATS 1) The model output is not data, the results are numerical approximations. 2) This study did not take into account potentia l numerical vorticity errors that could have resulted due to the steep topography at the mouth of the bay. 3) This study did not attempt to reconstruct the circulation of the bay in 1879, as it did not use/recreate boundary conditions that would have existed. 3) When calculating transport time, the ope n boundary condition at the mouth of the bay did not allow for particle re-entry or for non-zero concentration input to the domain, which could possibly effect transport time. It is important to emphasize that any mode ling effort that attempts to stimulate realistic flow field dynamics is greatly in fluenced by the quality of boundary condition data used to force the model and by the da ta used to validate the flow fields characteristics. This thesis attempted to si mulate the dynamics of the bay in a realistic manner by using high resolution environmental data inputs to the model, when available. Even when great care is taken, there are limita tions of all modeling exercises that must be borne in mind when interpreting the results, especially when applying them to environmental effects.

PAGE 43

33 FUTURE WORK Possible future research includes: 1) A finer bathymetric grid should be adde d in order to better resolve the channels. 2) Since the model open boundary concentration is set at zero at the mouth of the bay, it does not allow for particles or concentration to exit the bay and then re-enter. A larger model domain is needed to test the in fluence of re-entry on residence time. 3) Bay geometry should be altered in the mode l to better determine which features play the most significant role in altering transport time (e.g. remove just the bridges and/or causeways). 4) Various bridge/causeway configurations s hould be modeled to asses their impact on bay circulation and transport time (e.g. double the size of bridge/causeway openings, and/or add more openings). 5) Specific wind fields should be modeled with specific bathymetric alterations to determine specifically how each combination ef fects bay circulation and transport time.

PAGE 44

Howard Franklin Courtne y Cam p bell Sunshine Skyway Bi d Gandy Figure 1. Bridge and causeways spanning Tampa Bay. 34

PAGE 45

Figure 2. Present bathymetric grid for Tampa Bay, FL. 35

PAGE 46

Figure 3. 1879 Bathymetric grid for Tampa Bay, FL. 36

PAGE 47

Figure 4. Bathymetric grid difference, Present 1879. 37

PAGE 48

Figure 5. Differenced residual RMS speed, grey strips indicate the three 30day periods selected for analysis. 38

PAGE 49

39 Conditions for Model Experiments Start Time Mean RMS Wind Mean Wind Mean Wind Name Day Duration Freshwater Speed Speed Direction (10 6 m 3 d -1 ) (m s -1 ) (m s -1 ) (deg True) T1 150 30 5.4 3.0 1.0 -197.3 T2 370 30 1.6 3.4 0.55 60.5 T3 974 30 31.2 4.5 2.6 77.1 T1r 150 90 0.58 2.8 0.46 -194.2 T2r 370 90 2.6 4.4 1.1 47.0 T3r 974 90 22.9 5.1 3.1 68.1 Table 1. The conditions during the model time p eriods examined. Given are the start date and duration of the time period, the mean total fres hwater, rms wind speed, and mean wind during the duration of the time period indicated.

PAGE 50

Figure 6. Tampa Bay Freshwater Inflow graph, boxes indicate the three 30day periods selected for analysis, T1, T2, and T3 respectively. 40

PAGE 51

Figure 7. Tampa Bay Model wind speed, boxes indicate the three 30day periods selected for the analysis, T1, T2, and T3 respectively. 41

PAGE 52

Figure 8. Tampa Bay Model direction speed, boxes indicate the three 30day periods selected for the analysis, T1, T2, and T3 respectively. 42

PAGE 53

Figure 9. Tampa Bay Freshwater Inflow graph, boxes indicate the three 90day periods selected for the residence time analysis, T1r, T2r, and T3r respectively. 43

PAGE 54

Figure 10. Tampa Bay Model wind speed, boxes indicate the three 90day periods selected for the residence time analysis, T1r, T2r, and T3r, respectively. 44

PAGE 55

Figure 11. Tampa Bay Model wind speed, boxes indicate the three 90day periods selected for the residence time analysis, T1r, T2r, and T3r, respectively. 45

PAGE 56

Figure 12. Residual surface salinity during T1 for the present run. 46

PAGE 57

Figure 13. Residual surface salinity during T1 for the 1879 run. 47

PAGE 58

Figure 14. Residual surface salinity difference (p-1879) during T1. 48

PAGE 59

Figure 15. Residual surface salinity during T2 for the present run. 49

PAGE 60

Figure 16. Residual surface salinity during T2 for the 1879 run. 50

PAGE 61

Figure 17. Residual surface salinity difference (p-1879) during T2. 51

PAGE 62

Figure 18. Residual surface salinity during T3 for the present run. 52

PAGE 63

Figure 19. Residual surface salinity, OTB, during T3 for the present run. 53

PAGE 64

Figure 20. Residual surface salinity during T3 for the 1879 run. 54

PAGE 65

Figure 21. Residual surface salinity difference (p-1879) during T3. 55

PAGE 66

Figure 22. Residual bottom salinity during T1 for the present run. 56

PAGE 67

Figure 23. Residual bottom salinity during T1 for the 1879 run. 57

PAGE 68

Figure 24. Residual bottom salinity difference (p-1879), during T1. 58

PAGE 69

Figure 25. Residual bottom salinity during T2 for the present run. 59

PAGE 70

Figure 26. Residual bottom salinity, OTB, during T2 for the present run. 60

PAGE 71

Figure 27. Residual bottom salinity during T2 for the 1879 run. 61

PAGE 72

Figure 28. Residual bottom salinity, OTB, during T2 for the 1879 run. 62

PAGE 73

Figure 29. Residual bottom salinity difference (p-1879) during T2. 63

PAGE 74

Figure 30. Residual bottom salinity during T3 for the present run. 64

PAGE 75

Figure 31. Residual bottom salinity during T3 for the 1879 run. 65

PAGE 76

Figure 32. Residual bottom salinity difference (p-1879), during T3. 66

PAGE 77

Figure 33. Residual bottom salinity difference (p-1879), OTB, during T3. 67

PAGE 78

Figure 34. Residual surface velocity during T1 for the present run. 68

PAGE 79

Figure 35. Residual surface velocity during T1 for the 1879 run. 69

PAGE 80

Figure 36. Residual surface velocity difference (P-1879), during T1. 70

PAGE 81

Figure 37. Residual surface velocity during T2 for the present run. 71

PAGE 82

Figure 38. Residual surface velocity during T2 for the 1879 run. 72

PAGE 83

Figure 39. Residual surface velocity difference (p-1879) for T2. 73

PAGE 84

Figure 40. Residual surface velocity during T3 for the present run. 74

PAGE 85

Figure 41. Residual surface velocity during T3 for the 1879 run. 75

PAGE 86

Figure 42. Residual surface velocity difference (p-1879) for T3. 76

PAGE 87

Figure 43. Residual bottom velocity during T1 for the present run. 77

PAGE 88

Figure 44. Residual bottom velocity during T1 for the 1879 run. 78

PAGE 89

Figure 45. Residual bottom velocity difference (p-1879) during T1. 79

PAGE 90

Figure 46. Residual bottom velocity during T2 for the present run. 80

PAGE 91

Figure 47. Residual bottom velocity during T2 for the 1879 run. 81

PAGE 92

Figure 48. Residual bottom velocity difference (p-1879) during T2. 82

PAGE 93

Figure 49. Residual bottom velocity during T3 for the present run. 83

PAGE 94

Figure 50. Residual bottom velocity during T3 for the 1879 run. 84

PAGE 95

Figure 51. Residual bottom velocity difference (p-1879) during T3. 85

PAGE 96

Figure 52. Flushing time in days during T1 (bold solid line) 187 days, T2 (solid line) 97 days, and T3 (dotted line) 36 days for the Present run. 86

PAGE 97

Figure 53. Flushing time in days during T1 (bold solid line) 144 days, T2 (solid line) 85 days, and T3 (dotted line) 32 days for the 1879 run. 87

PAGE 98

Figure 54. Residence time during T1r for the Present run. 88

PAGE 99

Figure 55. Residence time during T1r for the 1879 run. 89

PAGE 100

Figure 56. Residence time during T2r for the Present run. 90

PAGE 101

Figure 57. Residence time during T2r for the 1879 run. 91

PAGE 102

. Figure 58. Residence time during T3r for Present run. 92

PAGE 103

Figure 59 Residence time during T3r for the 1879 run. 93

PAGE 104

94 REFERENCES Bay Soundings. 2004. And So it Flows: New Bridge at Fort Desoto Restores Tidal Flow Eliminated by Causeway. Pinellas Park, FL. Bay Soundings. 2003. Business of the Bay: Tampa Bay Fuels Regions Economy. Spring 2003 2:2 Pinellas Park, FL. Beach, D. 2002. Coastal Sprawl: Effects of Urban Design on Aquatic Ecosystems and in the United States. Pew Oceans Commissions, Arlington, Virginia. B.T. Donahue, A.C. Hine, S. Tebbens, S.D. Locker, D.C. Twichell .2003 .Late Holocene estuarine inner shelf interactions; Is there evidence of an estuarine retreat path for Tampa Bay, Florida? Ma rine Geology 200:219-241. Blumberg, A.F., and G.L. Mellor. 1987. A de scription of a three-dimensional coastal ocean circulation model. Pp. 1-16 in Three-Dimensional Coastal Models, (ed. Heaps), American Geophysical Union, Washington D.C. Brooks, DA, MW Baca, Y-T Lo. 1999. Tidal circulation and residence time in a macrotidal estuary: Cobscook Bay, Ma ine. Estuarine Coas tal Shelf Science 49:647-665. Brown, WS, E Arellano. 1980. The application of a segmented tidal mixing model to the Great Bay Estuary, NH. Estuaries 3(4):248-257. Burwell, D. 2001. Modeling Eulerian and Lagrangian Estuarine Residence Times. Dissertation, College of Marine Science, University of Florida, St. Petersburg, Florida. Burwell D., Vincent M., Luther M., M. Galperin. 2000. Modeling Residence Times: Eurlerian vs. Lagranian. In: Estuarine a nd Coastal Modeling, M. L. Spaulding and H. L. Butler, eds., ASCE Reston, VA, 995-1009. Callaway, RJ. 1981. Flushing study of South Beach Marina, Oregon. J Waterway, Port, Coastal and Ocean Div. ASCE 107(WW2):47-58. Cloern, J.E. 2001. Our evolving conceptual mo del of the coastal eutrophication problem. Marine Ecology Progress Series 210: 223-253.

PAGE 105

95 Community Overview. The Greater Tampa Chamber of Commerce. June 10 2005 < http://www.tampachamber.com/ed_overview.asp > Dettmann, EH, JF Paul, JS Rosen, CJ Strobel. 1989. Transport, fate and toxic effects of a sewage treatment plant effluent in a R hode Island estuary. U.S. Environmental Protection Agency. Narragansett, RI. EPA/600/X-89/062. Dettmann, EH, WA Brown, WM Warren, MF Fox, DR Kester. 1992. Final report on application of a wasteload allocation mode l to multiple discharge sources into an estuary US Environmental Protection Agency Report. Narragansett, RI: Environmental Protection Agency. EPA/600/X-92/026. Dronkers, J. & Zimmerman, J. T. F. 1982. So me principles of mixing in tidal lagoons. Oceanologica Acta SP, 107. Galperin, B., A. Blumberg, and R. Weisberg. 1992a. A Time-Dependent Three Dimensional Model of Circulation in Tampa Bay, In: Proceedings, Tampa Bay Area Scientific Information Symposium 2, Tampa, FL, February 27-March 1, 1991, 77-98 Galperin, B., A. Blumberg, and R. Weisberg. 1992b. The Importance of Density-Driven Circulation in Well-Mixed Estuarie s: The Tampa Bay Experience. In: Proceedings of the 2nd International Conference on Estuarine and Coastal Modeling, Tampa, FL, November 13-15, 1991, 332-343. Geyer, W.R., 1997. Influence of wind on dyna mics and flushing of shallow estuaries. Estuarine Coastal Shelf Science 44:713-722. Goodwin, C.R. 1987. Tidal Flow, Circulation, and Flushing Changes Caused by Dredge and Fill in Tampa Bay, Florida. U.S. Geological Survey, Water Supply Paper, 2282, Tampa FL. 88pp. Hagy, J.D., Sanford, L. P., Boynton, W. R ., 2000. Estimation of Net Physical Transport and Hydraulic Residence Times for a Co astal Plain Estuary Using Box Models, Estuaries V. 23, 3, 328-340. Hansen, D.V. Rattray, M., 1965. Gravitational circulation in straits a nd estuaries. Journal of Marine Science 23, 104-122. Kim, Yong H. and Voulgaris, George, 2005. Effect of Channel Bifurcation on Residual Estuarine Circulation: Winyah Bay, South Carolina. Estuarine Coastal and Shelf Science, 65, 671-686. Lewis, P.R., III. and Whitman, R.L., Jr., 1985. A New Geographic Description of the Boundaries and Subdivisions of Tampa Bay. Proceedings, Tampa Bay Area

PAGE 106

96 Scientific Information Symposium. Treat S.F., et al., eds. Bellwether Press, Tampa FL. Pp. 19-26. Li, Zhen. 1993. Circulation of Tampa Bay in Relation to Tides, Winds and Buoyancy Driven Convection. Masters Thesis, Colle ge of Marine Science, University of South Florida. Tampa, Florida. Meyers, S. et al. 2007. A Numerical Model of the Residual Circulation in Tampa Bay. Part I: Low-Frequency Temporal Variations. Estuaries In Press. McLusky, D. S.1981. The Estuarine Ecosystem. New York. John Wiley & Sons, 150p. Mellor, G.L., Oey, L., and Ezer, T. 1997. Sigm a Coordinate Pressure Gradient Errors and the Seamount Problem. Journal of Atmospheric and Oceanic Technology, 15: 1122-1131. Mellor, G.L., Oey, L., and Ezer, T.1993. The Pressure Gradient Conundrum of Sigma Coordinate Ocean Models. Journal of Atmospheri c and Oceanic Technology 11:1126-1134. Mellor, G.L. and Yamada. T. 1982. Devel opment of Turbulence Closure Model for Geophysical Fluid Problems. Rev. Geophys. And Space Phys ., 20, No. 4, 851-875. Miller, RL, BF McPherson. 1991. Estimating estuarine flushing and residence times in Charlotte Harbor, Florida, via sa lt balance and a box model. Limnology Oceanography 36(3):602-612. Monsen, N.E. 2001. A study of sub-tidal tr ansport in Susan Bay and Sacramento-San Joaquin Delta, California, PhD. Thesis. Stanford Univ. NOS. 1990. Unpublished Recapitulation of Ha rmonic Constant Reductions. National Ocean Service, NOAA. Officer, C.B. and D. R. Kester. 1991. On Es timating the Non-advective Tidal Exchange and Advective Gravitational Circulation Exchange in an estuary. Estuarine, Coastal and Shelf Science 32:99-103. Oliveira, A. & A.M. Baptista, 1997. Di agnostic Modeling of Residence Times in Estuaries. Water Resources Research. 33(8)1935-1946. Pilson M. 1985. On the Residence Ti me of Water in Narragansett Bay. Estuaries. 8: 2 14. Pritchard, D. W. 1956. The Dynamic Structure of a Coastal Plane Estuary. Journal of Marine Research, 33-42.

PAGE 107

97 Railroad and Shipping Timeline. Tampa History Center. 10 June 2005 "Road and bridge construction." New South Whales Department of Natural Resources (NSW). 18 July 2005 . Signell, RP, B Butman. 1992. Modeling tidal exchange and dispersion in Boston Harbor. Journal of Geophys Research 97(C10):15591-15606. Signell, RP. 1992. Tideand wind-driven flus hing of Boston Harbor, Massachusetts. In. ML Spaulding, K Bedford, A Blumberg, R Cheng, C Swanson, eds. Estuarine and Coastal Modeling: Proceedings of the 2nd International. Conference, Tampa, FL, American Society of Civ il Engineers. pp. 594-606. Solis, R.S. and G.L. Powell.1999. Hydrodynamics Mixing Characteristics, and residence times of Gulf of Mexico estuaries, p29-61. In T.S. Bianchi, J.R. Pennock, and R.R. Twilley (eds.), Biogeochemistry of Gu lf of Mexico Estuaries. John Wiley & Sons, New York. Tampa Bay Estuary Program (TBEP).2005a. A Po rtrait of the Tampa Bay Estuary. Fast Facts about Tampa Bay. FYI. St. Petersburg, FL. Tampa Bay Estuary Program (TBEP). 2005b. The State of the Bay. Tampa Bay: On The Road To Recovery. FYI. St. Petersburg, FL. Vincent, M., Burwell, D., Luther, M., Galperin, B. 1997. Real-Time Data Acquisition and Modeling in Tampa Bay, p 427-440. In Proceedings of th e 5th International Conference on Estuarine and Coastal Mode ling, Alexandria, Virginia, 427-440. Vincent, Mark S. 2001. Development, Implementation and Analysis of the Tampa Bay Coastal Prediction System. Dissertation, Co llege of Engineering, University of South Florida. Tampa, Florida. Weisberg, R.H. and R.G. Williams. 1991. Initial findings on the circulation of Tampa Bay. Proceedings, Tampa Bay Area Scientif ic Information Synposium 2. Treat, S. F. and P. A. Clark, eds. TEXT, Tampa, FL. pp. 49-66. Wong, K.C., 1994. On the Nature of Transverse Variability in a Coastal Plain Estuary. Journal of Geophysical Research 99 (C7), 14209-14222.

PAGE 108

98 Zervas, C. E. (Editor), 1993. Tampa Bay Oceanography Project: Physical Oceanographic Synthesis. NOAA National Ocean Service, Office of Ocean and Earth Sciences, Silver Spring, MD. 175 pp. Zimmerman, J. T. F. 1988 Estuarine residenc e times: Volume I Estuarine physics. In Hydrodynamics of Estuaries (Kjerfve, B ., ed.). CRC Press, Boca Raton, pp. 75 84. Zimmerman, J. T. F. 1976. Mixing and Flushi ng of Tidal Embayments in the Western Dutch Wadden Sea. Part I: Distribution of Salinity and Calculation of Mixing Time Scales. Netherlands Journal of Sea Research, 10, 149-191.