USF Libraries
USF Digital Collections

Saharan dust and phosphatic fidelity

MISSING IMAGE

Material Information

Title:
Saharan dust and phosphatic fidelity a three dimensional biogeochemical model of Trichodesmium on the West Florida shelf
Physical Description:
Book
Language:
English
Creator:
Lenes, Jason M
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Nitrogen fixation
Phosphorus
Iron
Ecosystem
Modeling
Dissertations, Academic -- Marine Science -- Doctoral -- USF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The availability of iron within the surface waters of the broad, oligotrophic West Florida shelf (WFS) controls periodic blooms of the pelagic marine cyanobacterium Trichodesmium. Summer delivery of iron (Fe), in the form of Saharan dust, alleviates this growth constraint, shifting limitation to the efficiency of phosphorus (P) cycles. Florida's rivers drain Miocene phosphorus deposits to supply the WFS with freshwater nutrient supplies at molar dissolved inorganic nitrogen/phosphate (DIN/PO4) ratios of less than 6. These diazotrophs draw upon ubiquitous stocks of dissolved nitrogen gas, once stimulated by Fe-deposition within P-replete waters of the West Florida shelf. An extensive in situ data set collected between 1998-2001 (NEGOM / ECOHAB / HyCODE) provided plankton taxonomy, hydrographic, nutrient, DOM, pigment, and optical properties on the shelf. A three-dimensional numerical model was constructed to analyze the impact of iron fertilization of the diazotroph Trichodesmium and the resultant effect upon the elemental cycles of N, P, and Fe. Based on the results of the coupled physical and ecological models, wet deposition of Fe-rich Saharan dust was necessary to stimulate enough nitrogen fixation to support the toxic red tide (Karenia brevis) of ~20 micrograms chl per liter found in October 1999. Ultimately, the magnitude and longevity of the Trichodesmium population, and therefore 'new' nitrogen production, was controlled by both phosphorus and iron availability.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2006.
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 Jason M. Lenes.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 109 pages.
General Note:
Includes vita.

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 - 001789847
oclc - 141189179
usfldc doi - E14-SFE0001476
usfldc handle - e14.1476
System ID:
SFS0025795:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Saharan Dust and Phosphatic Fidelity: A Three Dimensional Biogeochemical Model of Trichodesmium on the West Florida Shelf by Jason M. Lenes A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Marine Science University of South Florida Major Professor: John J. Walsh, Ph.D. Gabriel A. Vargo, Ph.D. Kendall L. Carder, Ph.D. Robert H. Weisberg, Ph.D. Joseph M. Prospero, Ph.D. Date of Approval: March 23, 2006 Keywords: Nitrogen Fixation, Phosphorus, Iron, Ecosystem, Modeling Copyright 2006, Jason M. Lenes

PAGE 2

Acknowledgments This analysis was funded by grants R-ESSF/03-0000-0039 to JML from the National Aeronautics and Space Administration as part of the Earth Systems Science Fellowship program, NNG04GL55G to KLC and NNG04GG04G to FMK from the National Aeronautics and Space Administration, OCE 0095970 to CAH from the National Science Foundation as part of the DOTGOM program, N00014-96-1-5024 to KAF and JJW, and N00014-98-1-0158 to RHW from the Office of Naval Research as part of the HyCODE and FSLE programs, 1435-01-97-CT-30851 to AEJ from the Minerals Management Service for the NEGOM program. I’d also like to thank Wachovia Bank for the Wachovia Fellowship to JML and the Tropical Rainfall Measurement Mission and National Center of Environmental Prediction for precipitation data. While working towards my degree, I’ve been quite fortunate to receive assistance from many people. First of all, I’d like to thank my advisor, John J. Walsh, for endless support and timely wit. I am grateful to Gabriel A. Vargo, Kendall L. Carder, Robert H. Weisberg, Joseph M. Prospero, and Cynthia A. Heil for the intellectual challenge and access to data. I am eternally beholden to Dwight Dieterle and Brian Darrow for a thousand favors each. I’d also like to thank my friends for good times and always being there when needed. Finally, I am most indebted to my family for always providing everything one would hope for their own; and making me spit milk out my nose.

PAGE 3

Note to Reader: The original of this document contains color that is necessary for understanding the data. The original dissertation is on file with the USF library in Tampa, Florida.

PAGE 4

i Table of Contents List of Table………………………………………………………………………………. ii List of Figures……………………………………………………………………………. iii Abstract….………………………………………………………………………………. vii Introduction……………………………………………………………………………….. 1 Methods………….……………………………………………………………………….24 1. Physical model…..……………...…………………………………………… 24 2. Equations of motion…...…………………………………….………………. 24 2.1. Optics……...…………………………………….……………………… 26 3. Atmospheric model..………………………....……………………………… 31 3.1. Dust deposition…….………..……………………………….…………. 31 3.2. Iron concentrations…………………………………..…………………..37 4. Biology………………………………………………………………………. 42 4.1. Primary producers…...………………………………………….………. 42 4.2. Loss terms……………………...……………………………….………. 45 4.3. Vertical migration………………...…………………………….………. 46 4.4. Microbial loop…………………..…………….………………………… 46 5. Chemistry……………………………………………………………………. 47 5.1. Carbon…………………………………………………………………... 47 5.2. Nitrogen…………...………..…..……………………………....………. 49 5.3. Phosphorus…………………………………….………..………………. 50 6. Boundary and initial conditions……………………………………………... 53 Results and Discussion…………………………………………………………………...57 1. Atmospheric deposition…………..……..…………………………………... 57 2. Water column iron………………………..…………………………………. 66 3. Initiation of Trichodesmium ………………...……………………………….. 72 3.1. Iron response……………………………………………………………. 72 3.2. Phosphorus acquisition………….……………………………………… 76 4. Growth and maintenance………………………………..…………………... 80 Conclusions……………………………………………………………………………… 97 References Cited…..………………………………………………….……………… ...100 About the Author………………………………………………………………… End Page

PAGE 5

ii List of Tables Table 1. Model parameters and values…..……………………………………………...27 Table 2. Nutrient and chlorophyll ratios for each functional group……………..……...51 Table 3. Initial conditions applied uniformly over the West Florida shelf.…...……….. 55 Table 4. Open boundary conditions…….……..……………………………………….. 56 Table 5. The dates of significant dust events (Saharan and Continental) based on mean atmospheric aerosol concentrations >10 g dust m-3 (averaged over the grid) in relation to the dates of concurrent rain events (>5 mm d-1 averaged over the grid), the cumulative precipitation during these rain events (mm), and the dust deposition during these events (g dust m-2) as calculated by the model.………….………………………………………....……64 Table 6. Variation in the environmental parameters leading to the transfer of nutrients to red tides on the northern (NS) and southern (SS) continental shelf in the eastern Gulf of Mexico in the summer/fall of 1999 (+ = yes, = no)..………………………………….…………………………………… ……67

PAGE 6

iii List of Figures Figure 1. Location of ECOHAB: Florida ( + ), ECOHAB middle grounds ( ), and NEGOM () data sets, relative to the 20 and 200 m isobaths………..…….……...2 Figure 2. The 1999 surface distributions of total dissolved iron (nmol kg-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 68 August, (e) 7-9 September, and (f) 5-7 October...…...………………… ...……..4 Figure 3. The 1999 surface distributions of nitrate+nitrite ( mol kg-1) across the West Florida shelf during (a) 5-8 June, (b) 5-7 July, (c) 6-8 August, (d) 79 September, and (e) 5-7 October…..……………………………………………………………….... ……...5 Figure 4. Mineral dust concentration (solid line) at Miami from 1998-2001 in relation to the offshore surface dissolved iron concentration (hollow squares) on the West Florida shelf averaged between the 50-200 m isobaths…………………………………………………………………… ………8 Figure 5. The 1999 surface distributions of phosphate ( mol kg-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, (e) 7-9 September, and (f) 5-7 October…………….…………… ……...10 Figure 6. The 1999 surface distributions of DOP ( mol kg-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, (e) 7-9 September, and (f) 5-7 October..……………………... ………...11 Figure 7. The molar DIN/PO4 ratios (mol/mol) in the eastern Gulf of Mexico at the a) surface and b) bottom in May 1999…………….…………………… ……14 Figure 8. The molar DIN/PO4 ratios (mol/mol) in the eastern Gulf of Mexico at the a) surface and b) bottom in August 1999…………………….……….. ……..15 Figure 9. The bottom distributions of dissolved inorganic nitrogen ( mol DIN kg1) across the northeastern Gulf of Mexico during a) May and b) August 1999………………………………………………………………………………16 Figure 10. The bottom distributions of phosphate ( mol PO4 kg-1) across the northeastern Gulf of Mexico during a) May and b) August 1999...……...... ……17

PAGE 7

iv Figure 11. The 1999 surface distributions of Trichodesmium (colonies l-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 68 August, (e) 7-9 September, and (f) 5-7 October, as sampled by bottles…. …...19 Figure 12. The 1999 surface distributions of DON ( mol kg-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, and (e) 7-9 September...………………………………………... ………20 Figure 13. a) The surface chlorophyll pigment concentration ( g chl l-1) and b) the surface salinity measured during the NEGOM (Northeastern Gulf of Mexico) project in August 1999……….…………………………………... ……21 Figure 14. A SeaWiFS image of the West Florida shelf on 17 September 2001, at 1-km resolution, using a combination of the visible bands: 555 nm (red), 490 nm (green), and 443 nm (blue)………………………………………... ……23 Figure 15. a) Schematic of the biochemical model pathways. …....………………. ……25 Figure 16. Daily dust aerosol concentration ( g m-3) from May-October 1999 over the model grid as interpolated from mineral dust collected at Miami (courtesy of Joseph Prospero, RSMAS-MAC).……………………………. ……33 Figure 17. a) Tropical Rainfall Measurement Mission (TRMM) 3-hr precipitation estimate using the 3b42 algorithm [eight 3-hr images were totaled to get daily precipitation (mm d-1) over the eastern Gulf of Mexico……..…….. ……...35 Figure 18. Diel assessment of dissolved iron stocks (nmol kg-1) on the West Florida shelf (40-50 m isobaths) during DOTGOM III on 10-16 July 2002. ……39 Figure 19. Total monthly dry deposition as calculated by the model at each grid point (mg dust m-2 month-1) during 1999…………………………………. .…….58 Figure 20. Total monthly wet deposition of dust as calculated by the model at each grid point (mg dust m-2 month-1) for the months of a) May, b) June, c) July, d) August, e) September, and f) October 1999...…………………. .……59 Figure 21. Total monthly precipitation as calculated by the model at each grid point (mm month-1) for the months of a) May, b) June, c) July, d) August, e) September, and f) October 1999………………………………………… ……60 Figure 22. a) Total wet deposition (g dust m-2) and b) total dust deposition (g dust m-2) as calculated by the model at each grid point during the 6-month simulation……………………………………………………………….. ……….61

PAGE 8

v Figure 23. a) Total Fe-deposition (mg Fe m-2) and b) total Fe-dissolution ( mol Fe m-2) as calculated by the model at each grid point during the 6-month simulation…………………………………………………………………... ……62 Figure 24. Total monthly dust deposition as calculated by the model at each grid point (mg dust m-2 month-1) for the months of a) May, b) June, c) July, d) August, e) September, and f) October 1999……………………………………...65 Figure 25. The simulated total dissolved iron (dFe) in the surface sigma layer (nmol Fe kg-1) on a) 4 May, b) 7 June, c) 6 July, d) 7 August, e) 8 September, and f) 6 October 1999.………………………………………… ……68 Figure 26. The null case (no atmospheric dust input) of simulated total dissolved iron (dFe) in the surface sigma layer (nmol Fe kg-1) on a) 4 May, b) 7 June, c) 6 July, d) 7 August, e) 8 September, and f) 6 October 1999……… ……71 Figure 27. The simulated Trichodesmium a) depth-integrated chlorophyll biomass (mg chl m-2) and b) 15-m chlorophyll biomass ( g chl l-1) on 4 May 1999. …….73 Figure 28. The simulated a) bottom nitrate + nitrite ( mol NO3 kg-1) and b) bottom phosphate ( mol PO4 kg-1) on 4 May 1999……………….……….. ……74 Figure 29. The simulated a) bottom dissolved iron (nmol dFe kg-1) and b) bottom colloidal iron (nmol cFe kg-1) on 4 May 1999………….………..……… ………75 Figure 30. The simulated circulation (cm s-1) calculated at the a) surface and b) near bottom on 4 May 1999………………………………….…….….. ………...79 Figure 31. The simulated molar DIN/PO4 ratios (mol/mol) calculated at the a) surface and b) bottom on 4 May 1999……………………………………... ……81 Figure 32. The simulated DIN/PO4 molar ratios calculated at the a) surface and b) bottom on 7 June 1999………………………………..…………………..... ……83 Figure 33. The simulated a) depth-integrated Trichodesmium chlorophyll biomass (mg m-2) and b) the bottom colloidal iron (nmol cFe kg-1) on 7 June 1999.. ...….84 Figure 34. The simulated surface distributions of phosphate ( mol kg-1) on (a) 4 May, (b) 7 June, (c) 6 July, (d) 7 August, and (e) 8 September 1999……………85 Figure 35. The simulated DIN/PO4 molar ratios (mol/mol) calculated at the a) surface and b) bottom on 6 July 1999……………………………………… ……87 Figure 36. The simulated a) depth-integrated Trichodesmium chlorophyll biomass (mg chl m-2) and b) the surface ammonia ( mol NH4 kg-1) on 6 July 1999.. ……88

PAGE 9

vi Figure 37. The simulated a) surface dissolved organic nitrogen ( mol DON kg-1) and b) surface ammonifying bacteria ( mol C kg-1) on 6 July 1999………. ……89 Figure 38. The simulated a) surface DIN/PO4 molar ratios (mol/mol) and b) depth-integrated Trichodesmium chlorophyll biomass (mg chl m-2) on 7 August 1999……………………………………………………………… ……...91 Figure 39. The simulated a) depth-integrated nitrogen fixation rates ( mol N m-2 d-1) by Trichodesmium and b) surface nitrate + nitrite ( mol NO3 kg-1) on 7 August 1999…………………………………………………………….. ……..92 Figure 40. The simulated a) surface diatom chlorophyll biomass ( g chl l-1) and b) the Trichodesmium /diatom surface chlorophyll ratio (chl/chl) on 8 August 1999………………………………………………………………………………93 Figure 41. The simulated a) surface DIN/PO4 molar ratios (mol/mol) and b) depth-integrated Trichodesmium chlorophyll biomass (mg chl m-2) on 8 September 1999…………………………………………………………… …….95

PAGE 10

vii Saharan Dust and Phosphatic Fidelity: A Three Dimensional Biogeochemical Model of Trichodesmium on the West Florida Shelf Jason M. Lenes ABSTRACT The availability of iron within the surface waters of the broad, oligotrophic West Florida shelf (WFS) controls periodic blooms of the pelagic marine cyanobacterium Trichodesmium Summer delivery of iron (Fe), in the form of Saharan dust, alleviates this growth constraint, shifting limitation to the efficiency of phosphorus (P) cycles. Florida’s rivers drain Miocene phosphorus deposits to supply the WFS with freshwater nutrient supplies at molar dissolved inorganic nitrogen/phosphate (DIN/PO4) ratios of <6. These diazotrophs draw upon ubiquitous stocks of dissolved nitrogen gas, once stimulated by Fe-deposition within P-replete waters of the West Florida shelf. An extensive in situ data set collected between 1998-2001 (NEGOM / ECOHAB / HyCODE) provided plankton taxonomy, hydrographic, nutrient, DOM, pigment, and optical properties on the shelf. A three-dimensional numerical model was constructed to analyze the impact of iron fertilization of the diazotroph Trichodesmium and the resultant effect upon the elemental cycles of N, P, and Fe. Based on the results of the coupled

PAGE 11

viii physical and ecological models, wet deposition of Fe-rich Saharan dust was necessary to stimulate enough nitrogen fixation to support the toxic red tide ( Karenia brevis ) of ~20 g chl l-1 found in October 1999. Ultimately, the magnitude and longevity of the Trichodesmium population, and therefore ‘new’ nitrogen production, was controlled by both phosphorus and iron availability.

PAGE 12

1 Introduction The concept of a limiting nutrient to describe plant growth is the cornerstone of marine primary production (Liebig, 1840). The availability of nitrogen (N) is thought to control production over the majority of the world’s oceans, though increased awareness of the limiting potential of phosphorus (P) on various time scales has forced a reinvestigation of its role (Broecker, 1982; Froelich, 1982). To further complicate the issue, over the last decade there has been an invigorated investigation into the role of the trace metal iron (Fe). In situ studies, such as the IronEx and SOIREE experiments, have conclusively shown that iron is critical in limiting primary production over much of the world’s oceans (Martin and Fitzwater, 1988; Barber and Chavez, 1991; Martin et al., 1991; Coale et al., 1996; Gordon et al., 1998; Boyd et al., 2000). This shift in oceanographic dogma forces constant revisions of our previous paradigm of marine carbon cycling (Walsh, 1996; Falkowski, 1997). The oligotrophic West Florida shelf (WFS) ecosystem, which ranges from 100150 km in width along the eastern Gulf of Mexico (Fig. 1), remains relatively devoid of iron most of the year despite extensive discharge from the Mississippi River and approximately 25 other major rivers along the Florida peninsula. Summer iron stocks on the shelf should be controlled by two dominant factors: (1) Local riverine inputs (Kim and Martin, 1974; U.S.G.S., 1976-1981) and (2) far-field aerosol dust (Carder et al.,

PAGE 13

2 Figure 1. Location of ECOHAB: Florida ( + ), ECOHAB middle grounds ( ), and NEGOM ( ) data sets, relative to the 50 and 200 m isobaths. -92-91-90-89-88-87-86-85-84-83-82-81 25 26 27 28 29 30 31 32 20 m 200 m

PAGE 14

3 1991; Prospero, 1999a). High iron content of the West Florida river systems (>20 nmol kg-1 at the mouth of Florida Bay; Caccia and Millero, 2003) has led to elevated iron concentrations of 1-2 nmol kg-1 at the 10-m isobath throughout most summer months (Lenes et al., 2001). Although, in the absence of Saharan dust input, mean background iron levels of <0.1 nmol kg-1 were measured between the 50and 200-m isobaths (Fig. 2a,f). Recent investigations on the WFS suggest that a combination of limiting factors (N, P, Fe, light) can drive seasonal control of the phytoplankton community and nutrient cycles (Walsh et al., 2003). Along the shelf, the toxic red-tide dinoflagellate Karenia brevis forms episodic blooms >100 g chl L-1 (Walsh and Steidinger, 2001), despite ambient nitrate concentrations of <0.50 mol kg-1 within 5 km of the Florida coast (Steidinger et al., 1998). Offshore, background nitrate concentrations are <0.05 mol kg-1 (Fig. 3; Masserini and Fanning, 2000). Therefore, an alternative source of nitrogen must be found in order to support the elevated red tide biomass, usually first observed near shore between Tampa Bay and Charlotte Harbor (Walsh et al., 2006). At 18-25 C, thermal impacts on gas solubility suggest a change in dinitrogen stocks of 383-429 mol N2 kg-1 (Weiss, 1970) at the same salinity of 35.0 within surface waters of the WFS. Thus, one possible source of ‘new’ nitrogen, nitrogen fixation by the colonial diazotroph Trichodesmium erythraeum appears to play an important role in the nitrogen economy of K. brevis along the WFS (Walsh and Steidinger, 2001). Blooms of T. erythraeum have been observed within 75 km of the west coast of Florida for more than 50 yr (King, 1950; Walsh and Steidinger, 2001), while

PAGE 15

4 -84-83.5-83-825-82-81.5 26 26.5 27 27.5 28 28.5 0 0.1 0.2 0.3 -84-835-83-825-82-815 26 265 27 275 28 285 0 1 2 3 4 5 6 7 8 9 10 11 12 13 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 -840-835-830-825-820-815 260 265 270 275 280 285 0 05 1 15 2 25 3 35 4 45 5 -84-835-83-825-82-815 26 265 27 275 28 285 0 05 1 15 2 25 3 35 4 45 5 -84-835-83-825-82-815 26 265 27 275 28 285 0 05 1 15 2 25 3 35 4 45 5Figure 2. The 1999 surface distributions of total dissolved iron (nmol kg1 ) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, (e) 7-9 September, and (f) 5-7 October. a b c d e f

PAGE 16

5 -8 4 -83.5-8 3 -82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 0 0.1 0.2 0.3 0.4 0.5 -8 4 -83.5-8 3 -82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 0 0.1 0.2 0.3 0.4 0.5 -84-835-83-825-82-815 26 265 27 275 28 285 0 01 02 03 04 05Figure 3. The 1999 surface distributions of nitrate + nitrite ( mol kg-1) across the West Florida shelf during (a) 5-8 June, (b) 5-7 July, (c) 6-8 August, (d) 7-9 September, and (e) 5-7 October. a b c d e

PAGE 17

6 surface stocks of these diazotrophs have reached >200 g chl l-1, as found off St. Pete Beach during May 2000 (Walsh et al., 2005). A population of this magnitude is unlikely from physical aggregation due to onshore flows (Walsh et al., 2006), and the positive buoyancy expressed by Trichodesmium (Walsby, 1992), without active growth of the diazotrophs as well. Examination of the impact of iron on the fecundity of Trichodesmium spp. (Rueter, 1988; Rueter, 1990; Paerl et al., 1994; Berman-Frank et al., 2001) has indicated a strong dependence on metabolic pathways responsible for nitrogen fixation. The large (2.2 x 10-3) Fe/N ratio for T. erythraem (Rueter et al., 1992) compared, for example to less than 2 x 10-4 for diatoms and flagellates (Sunda and Huntsman, 1995) is a reflection of the high iron levels required for optimal nitrogenase enzyme activity. Furthermore, Orcutt et al. (2001) measured a variable Fe/N ratio (1.9 x 10-4 – 2.9 x 10-3) dependent upon iron availability, implying a direct effect on cell health. Therefore, a significant source of iron is required for these diazotrophs to successfully compete against other phytoplankton. Duce (1986) and Martin and Gordon (1988) indicated that a majority of phytoplankton Fe-requirements might be supplied by atmospheric deposition. Direct uptake of particulate iron is of negligible significance in most phytoplankton (Rich and Morel, 1990), whereupon iron bioavailability to algal assemblages is controlled by dissolved iron species. In contrast to this generalization for phytoplankton, Rueter et al (1990, 1992) suggested that Trichodesmium may take up both dissolved and particulate iron associated with dust.

PAGE 18

7 Saharan dust in vast quantities is swept into air streams over west Africa during the dry summer months, and is transported thousands of kilometers across the Atlantic Ocean to the Caribbean and eastern United States (Schtz et al., 1981; Prospero and Nees, 1986). Desert dust often contains iron oxides. As the winds transport the dust across the Atlantic, a large fraction settles, thereby reducing the amount that eventually reached the West Florida shelf. Recent investigations (Carder et al., 1991; Young et al., 1991, Fung et al., 2000) suggest that aerosols are an important source of iron to phytoplankton in several oligotrophic oceanic regions. Indeed, aeolian input is believed to be responsible for 30-96% of the dissolved iron in the photic zone of the Sargasso Sea and 16-76% in the central North Pacific gyre (Duce, 1986). Prospero (1999b) recently showed from a 23 yr dust record at Miami that large amounts of aeolian mineral dust are periodically carried into Florida every summer yielding daily atmospheric concentrations in the range of 10-100 g m-3 above Miami (Fig. 4). Based on atmospheric nss-nitrate stocks, Saharan dust events can be distinguished from local dust of US continental origin. Maximum concentrations of Saharan dust generally occur between June-August (Prospero et al., 2001). Lenes et al. (2001) traced the history of a June 1999 Saharan dust event with AVHRR imagery, airmass trajectories, and optical depth measurements. With usual mean summer wind velocities of 3-5 m s-1, the path for a SE wind from Miami to St. Petersburg is such that direct air mass transit (~400 km) between the two locations should take between 20 and 32 h. This was verified by concurrent measurements made at Ft. Myers and Miami in 1995-96, where the timing and magnitude of the dust peaks closely matched this estimate of transit time (Prospero et al., 2001).

PAGE 19

8 0 20 40 60 80 100 120 J-98A-98J-98O-98D-98A-99J-99S-99D-99 DateMineral Dust (ug m-3)0 0.5 1 1.5 2 2.5 3 3.5Dissolved Iron (nM) 0 20 40 60 80 100 120 J-00A-00J-00S-00D-00M-01J-01S-01D-01 DateMineral Dust (ug m-3)0 0.5 1 1.5 2 2.5 3 3.5Dissolved Iron (nM) Figure 4. Mineral dust concentration (solid line) at Miami from 1998-2001 in relation to the offshore surface dissolved iron concentration (hollow squares) on the West Florida shelf averaged between the 50-200 m isobaths. Saharan dust events of low nss-nitrate are denoted by arrows, with filled arrows indicating concurrent wet deposition (i.e., 24-48 hr delay) at the Tampa Airport of >1.0 mm rain. Data from Joe Prospero and Jason Lenes.

PAGE 20

9 Previous assessments of iron inputs to the West Florida shelf have been based on time series dust records at Fort Myers and Miami, dust composition, and relative wet (75%) and dry (25%) mineral deposition rates (Walsh and Steidinger, 2001). With a combined wet and dry deposition rate of ~1.25 g m-2 yr-1 (Prospero et al. 1987; Landing et al. 1995) and a ~3.5% mass fraction of iron in mineral aerosols (Duce et al. 1991; Zhu et al. 1997), about 80% of the estimated annual loading (0.6 mM Fe m-2 yr-1) may be deposited in one month on the West Florida shelf. Total dissolved iron concentrations measured on the 5-7 July 1999 ECOHAB (Ecology and Oceanography of Harmful Algal Blooms) cruise following a 26 June-4 July Saharan dust event yielded a July mean of 3.0 nmol kg-1 within near surface waters at salinities >35.0 (stations above the 50-200 m isobaths) remote from riverine supplies (Lenes et al., 2001). In contrast, the iron found on the WFS during the August 1998, May 1999, and May 2000 cruises, in the absence of Saharan dust input, represent mean background levels <0.3 nmol kg-1. Summer delivery of iron, in the form of Saharan dust, thus increased surface dissolved iron concentrations 3-30 times, leading to a subsequent 100-fold increase of Trichodesmium stocks at mid-shelf (Lenes et al., 2001). A concurrent depletion of midshelf stocks of inorganic (Fig. 5) and organic (Fig. 6) phosphorus between June-July suggests that the potential size of the bloom during the dust season might not be regulated by iron availability, but by P-limitation within the offshore system. Subsequent return of P was observed in September and October as Trichodesmium stocks diminished. Phosphorus limitation was also reported by Sanudo-Wilhelmy (2001) for Trichodesmium

PAGE 21

10 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-83-825-82-815 26 265 27 275 28 285 0 01 02 03 04 05 06 07 08 09 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-83-825-82-815 26 265 27 275 28 285 0 01 02 03 04 05 06 07 08 09 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-83-825-82-815 26 265 27 275 28 285 0 01 02 03 04 05 06 07 08 09 Figure 5. The 1999 surface distributions of phosphate ( mol kg-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, (e) 7-9 September, and (f) 5-7 October. a b c d e f

PAGE 22

11 -84-835-83-825-82-815 26 265 27 275 28 285 -84-835-83-825-82-815 26 265 27 275 28 285 0 02 04 06 08 1 12 14 16 18 2 22 24 26 28 3 -84-835-83-825-82-815 26 265 27 275 28 285 -84-835-83-825-82-815 26 265 27 275 28 285 0 02 04 06 08 1 12 14 16 18 2 22 24 26 28 3 -84-835-83-825-82-815 26 265 27 275 28 285 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Figure 6. The 1999 surface distributions of DOP ( mol kg-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, (e) 7-9 September, and (f) 5-7 October. a b e d c f

PAGE 23

12 populations at the Bermuda Atlantic Time Series (BATS) site, as did Karl et al., (1997) at the Hawaii Ocean Time-Series (HOTS) station. Off the west coast of Barbados, Trichodesmium chlorophyll reaches a maximum observed depth-integrated stock of ~10 mg m-2 (Borstad, 1978; Borstad, 1982), compared to >100 mg chl m-2 off Tampa Bay (Walsh and Steidinger, 2001). In the absence of much local land drainage, the inorganic phosphate concentrations of the euphotic zone remain <0.1 mol kg-1 throughout the year off Barbados (Stevens et al., 1970), where the atmospheric dust concentrations are ~2.5-fold those found over South Florida (Prospero et al., 1987). In contrast, phosphorus-rich waters are usually found near the West Florida coast (0.2-0.5 mol P kg-1; Walsh and Steidinger, 2001; Fig. 5). The near shore supplies of nutrients along the West Florida coast are found in a low molar N/P ratio (Walsh et al., 2003). These waters can intersect with offshore populations of Trichodesmium (Walsh and Steidinger, 2001), given shoreward transport by wind-driven circulation. Dissolved inorganic nitrogen (DIN) to phosphate (PO4) ratios of <6 for the Caloosahatchee River, Charlotte Harbor, Sarasota and Tampa Bays emphasize the fossil Hawthorne formation (Dragovich et al., 1968) nutrient loading from the southern estuaries, compared to >40 for the nutrient supplies of the northern estuaries and ~16 for deep-sea intrusions on the shelf (Walsh et al., 2006). Since the molar N:P ratio of fertilizers reaching drainage basins of the Gulf of Mexico was ~8 (Turner and Rabalais, 1999), the remaining phosphorus found in the central west Florida region could be attributed to both fossil phosphorus (Dragovich et al., 1968) and agriculture (Hammet, 1988).

PAGE 24

13 Surface and bottom measurements of the molar DIN/PO4 ratios in the eastern Gulf of Mexico during May are >15 over most of the northern shelf and <15 south of the Big Bend region and on the outer shelf (Fig. 7). This demonstrates the disparity in DIN and PO4 concentrations supplied to each part of the shelf by different river systems (Walsh et al., 2006). In August, DIN/PO4 ratios had increased along the outer and mid-shelf regions south of the Big Bend (Fig. 8). Bottom DIN concentrations increased on the outer shelf between May and August (Fig. 9), while bottom phosphate had not (Fig. 10). Therefore, a non deep-sea source of nitrogen must have caused the elevation in DIN/PO4 ratios. Diazotrophs are a source of ammonium during population growth (Prufert-Bebout et al. 1993) and, as well, after bloom collapse (Devassy et al. 1978). Also, studies have shown that up to 50% of the N-fixed by Trichodesmium is excreted as DON, with 50100% of this released as amino acids (Capone et al. 1994; Glibert and Bronk 1994) and ammonia (Mulholland and Capone, 2001). Mulholland (personal communication) found preliminary evidence that K. brevis is capable of growth from N (NH4 + and DON) directly derived from continuous cultures of Trichodesmium Thus we postulated that large pools of released N might accumulate in the water column during periods of Feavailability on the outer shelf, eliminating or mitigating nitrogen limitation for other components of the phytoplankton community and the microbial loop. Therefore, the WFS, with phosphorus-rich estuarine outwelling at a molar N:P ratio of <1 (Vargo et al., 2001) and a summer source of aeolian iron, is an ideal laboratory for analyzing variations in Trichodesmium biomass. High organic N and P

PAGE 25

14 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 33 36 39 0 3 6 9 12 15 18 21 24 27 30 33 36 39 -92-9 0 -88-86-8 4 -82 25 26 27 28 29 30 31 32 Figure 7. The molar DIN/PO4 ratios (mol/mol) in the eastern Gulf of Mexico at the a) surface and b) bottom in May 1999.a b

PAGE 26

15 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 33 36 39 -92-91-90-89-88-87-86-85-84-83-82-81 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 33 36 39 Figure 8. The molar DIN/PO4 ratios (mol/mol) in the eastern Gulf of Mexico at the a) surface and b) bottom in August 1999. a b

PAGE 27

16 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Figure 9. The bottom distributions of dissolved inorganic nitrogen ( mol DIN kg-1) across the northeastern Gulf of Mexico during a) May and b) August 1999. a b

PAGE 28

17 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 Figure 10. The bottom distributions of phosphate ( mol PO4 kg-1) across the northeastern Gulf of Mexico during a) May and b) August 1999. a b

PAGE 29

18 concentrations have been measured with the beginning of K. brevis blooms, but no clear relation to estuarine sources could be drawn (Lester et al., 2001). During 1999, a Trichodesmium increase of two orders of magnitude from background (Fig. 11) led to elevated DON concentrations (Fig.12; Lenes et al., 2001) from ~7-8 mol kg-1 to ~15 mol kg-1 within the peak Trichodesmium population. This provides enough ‘new’ nitrogen to yield a red tide of >20 g chl l-1 of the toxic dinoflagellate K. brevis (Walsh and Steidinger, 2001). Elevated chlorophyll concentrations measured along the northern shelf were attributed to a diatom population in the Mississippi plume (Fig. 13). Unlike the southern shelf, slow growing dinoflagellates such as K. brevis would be at disadvantage to the established diatom population when competing for any ‘new’ nitrogen provided by Trichodesmium Comparison of the in situ response to three distinct case years (1998, 1999, 2001) helped detail the control factors that lead to an onset of red tide (Walsh et al., 2006). In 1998, the WFS received minimal dust/iron loading, but received a significant amount of rainfall (i.e. high riverine P-supplies). Yet little increment in Trichodesmium and K. brevis stocks were seen (Walsh et al., 2003). In contrast, in 1999, the WFS was subject to above average summer dust/iron concentrations while precipitation was moderate around periods of dust deposition (Fig. 4). A subsequent 10-fold increase in Trichodesmium biomass was observed between June and July (Fig. 11), leading to an October bloom of K. brevis of ~20 g chl l-1 (Lenes et al., 2001; Walsh and Steidinger, 2001). A third case (2001) saw one of the most significant red tides along the WFS in recent memory, starting in September and continuing for 8 months. Not only was this an above average

PAGE 30

19 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-83-825-82-815 26 265 27 275 28 285 0 2 4 6 8 10 12 14 16 18 20 22 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-8 3 -825-8 2 -815 26 265 27 275 28 285 0 2 4 6 8 10 12 14 16 18 20 22 -8 4 -83.5-8 3 -82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-83-825-82-815 26 265 27 275 28 285 0 2 4 6 8 10 12 14 16 18 20 22 a b c dFigure 11. The 1999 surface distributions of Trichodesmium (colonies l1 ) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, (e) 7-9 Se p tember and ( f ) 5-7 October as sam p led b y bottles.e f

PAGE 31

20 -84-835-83-825-82-815 26 265 27 275 28 285 -84-835-83-825-82-815 26 265 27 275 28 285 2 4 6 8 10 12 14 16 18 -8 4 -83.5-8 3 -82.5-82-81.5 26 26.5 27 27.5 28 28.5 -84-835-83-825-82-815 26 265 27 275 28 285 2 4 6 8 10 12 14 16 18 -84-83.5-83-82.5-82-81.5 26 26.5 27 27.5 28 28.5 Figure 12. The 1999 surface distributions of DON ( mol kg-1) across the West Florida shelf during (a) 2-5 May, (b) 5-8 June, (c) 5-7 July, (d) 6-8 August, (e) 7-9 September. a b c d e

PAGE 32

21 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 2 0 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Figure 13. a) The surface chlorophyll pigment concentration ( g chl l-1) and b) the surface salinity measured during the NEGOM (Northeastern Gulf of Mexico) project in August 1999.

PAGE 33

22 summer for dust/iron deposition, Florida recorded an anomalous amount of rainfall (Heil, personal communication). Satellite imagery taken on 17 September 2001 estimated the position of Trichodesmium along the outer shelf south of Tampa Bay in relation to that of the red tide (Fig. 14). Previous simple models have had some success analyzing the role of Trichodesmium in nutrient dynamics (Hood et al., 2001; Fennel et al., 2002; Moore et al., 2002; Lenes et al., 2005). But an increase in new production by the nitrogen-fixing cyanobacterium Trichodesmium in the oceanic North Pacific (Karl et al., 2001) and at Bermuda (Orcutt et al., 2001) over long time scales has shown a shift in the dissolved molar N:P ratio towards a more phosphorus limited system. The formation of ‘new’ nitrogen is associated with the sequestration of CO2 as a carbon source (Capone et al., 1997), therefore creating a possible atmospheric carbon sink.

PAGE 34

23 Figure 14. A SeaWiFS image of the West Florida shelf on 17 September 2001, at 1-km resolution, using a combination of the visible bands: 555 nm (red), 490 nm (green), and 443 nm (blue). Populations of Trichodesmium spp. [pale blue] and Karenia brevis [reddish-black] are inferred from their different backscattering properties (courtesy of Kendall Carder and Robert Chen, USF). Trichodesmium K.brevis

PAGE 35

24 Methods The coupled 3-dimensional model was composed of 4 submodels: 1) physical (water circulation), 2) bio-optical, 3) atmospheric (dust input), and 4) biochemical (Fig. 15a). The solutions for 24 partial differential equations describing the spatio-temporal fields of: [temperature, salinity, u, v, w, kz, spectral light, atmospheric dust, diatoms (P1), Trichodesmium (P2), ammonifying bacteria (B1), nitrifying bacteria (B2), dissolved iron (dFe), colloidal iron (cFe), dissolved organic carbon (DOC), dissolved organic nitrogen (DON), dissolved organic phosphorus (DOP), dissolved inorganic carbon (DIC), nitrate + nitrite (NO3), ammonia (NH4), phosphate (PO4), siliceous detritus (D1), non-siliceous detritus (D2), and siliceous fecal pellets (Z1)] embedded within the submodels were solved over an orthogonal curvilinear grid (Fig. 15b) of the WFS at a resolution of ~2-6 km and 21 vertical sigma levels (He and Weisberg, 2002a; 2003). 1. Physical model 1.1. Equations of motion The advective transport (Tra) term was described by: 2 1 2 1 1 2) ( h h A h h vdA h udA h dA Tra (1)

PAGE 36

25 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 Figure 15. a) Schematic of the biochemical model pathways. b) The orthogonal curvilinear grid at a resolution of ~2-6 km. a b Light Saharan Dust CO2~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dFe cFe DIC P1 P2 DON NH4 NO3 D1 D2 DOC Z1 DOP B1 B2 PO4 Light Saharan Dust CO2~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dFe cFe DIC P1 P2 DON NH4 NO3 D1 D2 DOC Z1 DOP B1 B2 PO4

PAGE 37

26 where A was any of the above state variables to which the transport applied, was the horizontal curvilinear coordinate in the cross shelf direction, was the horizontal curvilinear coordinate in the alongshelf direction, h1 and h2 were the length and width of the grid box in the x and y direction respectively, and u, v and were the velocities in the x, y, and z directions from the circulation model results (Table 1). Due to the strong horizontal diffusion in the numerical algorithm for advective transport, explicit horizontal turbulent mixing was ignored and modeled only in the vertical component, such that diffusive transport (Trd) was: A d k dA Trh d) ( (1a) where kh was the coefficient of vertical eddy diffusivity, derived from a second moment turbulence closure submodel, embedded within the USF adaptation (Weisberg and He, 2003) of the (POM) Princeton Ocean Model (Mellor and Yamada, 1982). 1.2. Optics A simple spectral solar irradiance model (Gregg and Carder, 1990) was employed to derive solar irradiance at 30 visible spectral wavelengths (10 nm bins) below the ocean surface (I0()) to maximize the species-specific response to the varying light field without compromising computational efficiency. In the marine environment, the intensity and spectral quality of light is influenced by several variables. Light energy is absorbed and

PAGE 38

27 Table 1. Symbol description and values. Symbol Description Units Value ____________________________________________________________________________________________________________ b Rate of phototransformation of DOM to inorganic matter s-1 1.98E-9 B Total bacteria mmol m-3 B1 Ammonifying bacteria mmol m-3 B2 Nitrifying bacteria mmol m-3 cFe Concentration of colloidal iron mol m-3 Cair Aerosol concentration in the air g dust (kg air)-1 Crain Concentration of dust in rainwater g dust (kg rain)-1 CDOM Colored dissolved organic matter mmol m-3 d Depth of sigma level m dFe Concentration of dissolved iron mol m-3 dFei Iron input into the upper sigma level mol m-3 D Total detritus mmol m-3 D1 Diatom detritus mmol m-3 D2 Trichodesmium detritus mmol m-3 DIC Concentration of dissolved inorganic carbon mmol m-3 DOC Concentration of dissolved organic carbon mmol m-3 DON Concentration of dissolved organic nitrogen mmol m-3 DOP Concentration of dissolved organic phosphorus mmol m-3 f Molecular weight of iron g mole-1 55.847 Fem Mass fraction of iron in mineral aerosols % 3.5 Fed Iron dissolution rate % 5.0 g Gravitational acceleration cm s-1 980 h1 Transformation coefficient from the Cartesian (x) to the orthogonal ( ) h2 Transformation coefficient from the Cartesian (y) to the orthogonal ( ) I0( ) Solar irradiance at wavelength, just below the ocean surface E m-2 s-1 Io Saturation intensity effecting photoreduction of colloidal iron E m-2 s-1 2500 Isat(1) Saturation intensity effecting growth for diatoms E m-2 s-1 250 Isat(2) Saturation intensity effecting growth for a Trichodesmium E m-2 s-1 400 Iz Total incident radiation at depth, z, over all wavelengths (400-700 nm) E m-2 s-1 Iz( ) Incident radiation at depth, z, and wavelength, E m-2 s-1 k Fe Half-saturation constant for iron scavenging by particles mol m-3 1.5 kB( ) Specific attenuation coefficient of bacteria at wavelength, m-1 kC( ) Specific attenuation coefficient of CDOM at wavelength, m-1 kd Detritus remineralization rate s-1 5.7E-7 kdFe(1) Half-saturation constant for dFe uptake for diatoms mol m-3 0.2 kdFe(2) Half-saturation constant for dFe uptake for Trichodesmium mol m-3 0.5 kDOC(3) Half-saturation constant for DOC uptake for ammonifying bacteria mmol m-3 45 kDON(3) Half-saturation constant for DON uptake for ammonifying bacteria mmol m-3 6.5 kDOP(2) Half-saturation constant for DOP uptake for Trichodesmium mmol m-3 0.2 kDOP(3) Half-saturation constant for DOP uptake for ammonifying bacteria mmol m-3 0.25 kfp Fecal pellet remineralization rate s-1 5.7E-7 kh Coefficient of vertical eddy diffusivity cm s-1 khvC Rate constant for photoreduction of colloidal iron s-1 2.33E-4 kNH4(1) Half-saturation constant for ammonia uptake for diatoms mmol m-3 1.5 kNH4(4) Half-saturation constant for ammonia uptake for nitrifying bacteria mmol m-3 0.2 kNO3(1) Half-saturation constant for nitrate uptake for diatoms mmol m-3 1.0 kP1( ) Specific attenuation coefficient of diatoms at wavelength, m-1 kP2( ) Specific attenuation coefficient of Trichodesmium at wavelength, m-1 kphoto Rate of photoreduction of colloidal iron mol m-3 s-1

PAGE 39

28 Table 1 (Continued). Symbol Description Units Value ____________________________________________________________________________________________________________ kPO4(1) Half-saturation constant for phosphate uptake for diatoms mmol m-3 0.1 kPO4(2) Half-saturation constant for phosphate uptake for Trichodesmium mmol m-3 0.1 kPO4(4) Half-saturation constant for phosphate uptake for nitrifying bacteria mmol m-3 0.05 kw( ) Specific attenuation coefficient of water at wavelength, m-1 k Total attenuation coefficient at wavelength, m-1 kzoo Maximum grazing rate on diatoms mmol m-3 s-1 1.15E-5 m3 Ammonifying bacterial mortality s-1 2.77E-6 m4 Nitrifying bacterial mortality s-1 2.77E-6 NH4 Concentration of ammonium mmol m-3 NO3 Concentration of nitrate + nitrite mmol m-3 pCO2 CO2 concentration in water mmol m-3 pCO2air CO2 concentration in air atm 383 P1 Concentration of diatoms mmol m-3 P2 Concentration of Trichodesmium mmol m-3 PO4 Concentration of inorganic phosphate mmol m-3 Pref (1) Background population for diatoms mmol m-3 0.5 Pref (2) Background population for Trichodesmium mmol m-3 0.01 r Radius of dust particles m 2.0 r1 Coefficient for diatom grazer respiration 0.2 r2 Coefficient for Trichodesmium grazer respiration 0.2 r3 Ingestion coefficient for grazing on diatoms 0.8 r4 Ingestion coefficient for grazing on Trichodesmium 0.8 t Time step s 360 T Temperature C TDP Total dissolved phosphorus mmol m-3 Tra Transport due to advection cm s-1 Trb Transport due to diffusion cm s-1 Trt Total transport cm s-1 u Velocity in the direction cm s-1 v Velocity in the direction cm s-1 va Air viscosity cm s g-1 1.78E-4 vstk Stokes settling velocity for dust cm s-1 0.115 w1 Sinking rate of diatoms m s-1 5.7E-6 w2 Vertical migration rate of Trichodesmium mm s-1 +/0.9 x Rate of dFe loss to higher trophic levels s-1 8.68E-5 z Depth interval within the water-column m Z Scavenging ratio of dust in rainwater 132 Z1 Diatomaceous fecal pellets mmol m-3 Particle scavenging of iron s-1 1 Diatom lysis rate % 3.0 2 Trichodesmium lysis rate mmol m-3 s-1 1 Diatom respiration rate % 10.0 2 Trichodesmium respiration rate % 12.0 3 Ammonifying bacteria respiration rate % 45.0 4 Nitrifying bacteria respiration rate % 45.0 1 Wet deposition of mineral aerosols g m-3 d-1 2 Dry deposition of mineral aerosols g m-3 d-1 1 Diatom grazing rate mmol m-3 s-1 2 Trichodesmium grazing rate % 1.0

PAGE 40

29 Table 1 (Continued). Symbol Description Units Value ____________________________________________________________________________________________________________ Wavelength of radiation nm 1 Realized net growth rate for diatoms s-1 2 Realized net growth rate for Trichodesmium s-1 3 Realized net growth rate for ammonifying bacteria s-1 4 Realized net growth rate for nitrifying bacteria s-1 max1 Maximum gross growth rate for diatoms s-1 1.45E-5 max2 Maximum gross growth rate for Trichodesmium s-1 8.1E-6 max3 Maximum gross growth rate for ammonifying bacteria s-1 1.5E-5 max4 Maximum gross growth rate for nitrifying bacteria s-1 1.5E-5 T max1 Maximum gross growth rate for diatoms adjusted for temp. s-1 T max2 Maximum gross growth rate for Trichodesmium adjusted for temp. s-1 Dust particle density g cm-3 2.5 Sigma coordinate 1 Fraction of maximum nutrient uptake by diatoms mmol m-3 2 Fraction of maximum nutrient uptake by Trichodesmium mmol m-3 3 Fraction of maximum nutrient uptake by ammonifying bacteria mmol m-3 4 Fraction of maximum nutrient uptake by nitrifying bacteria mmol m-3 Velocity component in the direction cm s-1 Cross-shelf curvilinear coordinates Excretion rate of diatoms Excretion rate of Trichodesmium % 25.0 Alongshore curvilinear coordinates

PAGE 41

30 scattered by the water, the organic component, i.e. algae, colored dissolved organic matter (Kirk, 1994), bacteria (Nelson and Robertson, 1993), and the non-living particulate matter, i.e. detritus and suspended sediments. The underwater light field not only determines the photosynthetic efficiency of the photoautotrophs, but the position of Trichodesmium in the water column due to buoyancy regulation in relation to the ambient light field (Kromkamp and Walsby, 1992). Therefore, spectral irradiance at depth, Iz(), was calculated according to Beer’s law of exponential decay: z k ze I I 0 (2) where represented a specific wavelength. k was the attenuation coefficient for that wavelength defined as: z dz z A k k kz A w 0 (2a) where A denoted each optically significant functional group, i.e. diatoms (P1), cyanophytes (P2), bacteria (B = B1 + B2), detritus (D = D1 + D2), and colored dissolved organic matter (CDOM). Their specific attenuation coefficients (kA()) were represented by kP1(), kP2(), kB(), and kC(); while kw() was the total attenuation coefficient for water (Smith and Baker, 1981; Pope and Fry, 1997). Attenuation coefficients for phytoplankton (Subramanian et al., 1999), bacteria (Morel and Ahn, 1990), and detritus (Roesler et al.,

PAGE 42

31 1989) were equal to the sum of the wavelength specific absorption coefficients and the associated scattering coefficients. As for CDOM, backscattering was ignored due to its dissolved nature and generally high absorption values (Gordon et al., 1988). The CDOM attenuation coefficients were equal to published CDOM absorption coefficients in the eastern Gulf of Mexico (Carder et al., 1989). 2. Atmospheric model 2.1. Dust deposition The deposition of dust aerosols required several parameterizations to simulate concentrations within to the surface mixed layer of the ocean. Gravitational sedimentation, turbulent mixing, and wet deposition by rain have all been shown to greatly affect the fallout of dust, with varying importance based on precipitation rates and particle size (Tegen and Fung, 1994). A prior dust transport model (Tegen and Fung, 1994) considered dust particles of 4 size categories by diameter: clay (<2 m), small silt (2-20 m), large silt (20-50 m), and sand (>50 m). Prospero (1999b) reported that the majority of dust arriving at Miami during the summer was less than 10 m in diameter, with 33-50% <2.5 m (PM 2.5) due to the distance from the source (~5000 km). Since most particles are within the smaller range, only one size class was considered in this model (clay/silt <10 m). Aerosols collected at Ft. Myers in 1995 and 1996 showed similar concentrations at almost no lag time when compared to the Miami data (Prospero et al., 2001). Therefore, it was assumed that atmospheric concentrations of dust aerosols collected at Miami represented

PAGE 43

32 concentrations over the WFS during 1999 (Fig. 16). Tegen and Fung (1994) expressed dry deposition by gravitational settling and turbulent mixing as a function of Stokes law (Genthon, 1992). For smaller particles (clay/silt), the particle sizes appeared to follow a lognormal size distribution (U.S. Department of Agriculture, 1975), where the corresponding mass was distributed equally over logarithmic size intervals. They found that the atmospheric lifetimes for clay-sized particles with a radius of 0.7 m (1.4 m diameter) to be 275 days by dry deposition, while the same size particles were removed after 14 days through wet deposition. This suggests that wet deposition was a more important removal process, given significant precipitation. To calculate dry deposition, we estimated a mean particle diameter and calculated the settling velocity (vstk) using Stokes law: g v r va stk9 22 (3) where was the particle density (2.5 g cm-3 for clay; Tegen and Fung, 1994), g, the gravitational acceleration (980 cm s-2), and va, the air viscosity (1.78E-4 cm s g-1). From this velocity, a net deposition was derived from atmospheric concentrations of dust at Miami and applied uniformly over the WFS (Fig. 16), assuming homogeneity of dust up to 4 km altitude. Since this model did not employ a multi-layered atmospheric sub-model, dry deposition was a function of the settling velocity from an atmospheric height of 30-m above the sea-surface, the height of the collection tower in Miami (Prospero, 1999a).

PAGE 44

33 0 10 20 30 40 50 60 70 5/15/155/296/126/267/107/248/78/219/49/1810/210/1610/30 DateMineral Dust ( g m-3) Figure 16. Daily dust aerosol concentration (g m-3) during May-October 1999 applied uniformly over the model grid as extrapolated from mineral dust collected at Miami (courtesy of Joseph Prospero, RSMAS-MAC).

PAGE 45

34 In order to calculate a mean particle diameter, a total annual dry deposition rate (312.5 mg dust m-2 yr-1) was derived from the total annual dust deposition at Miami of 1.25 g m-2 yr-1 (Prospero et al., 1987) and a wet/dry deposition ratio of ~3 (Prospero et al., 1987). The total annual dry deposition was converted to daily deposition (856.2 g dust m-2 d-1) and divided by the four year daily dust average (8.64 g dust m-2 d-1) to get a sinking rate of 99.1 m d-1 (vstk = 0.115 cm s-1). Therefore, a final mean particle diameter of 4 m was back calculated using Stokes law. The mean particle size was similar to the mass mean diameter of mineral dust over the oceans (2-3 m; Duce, 1991). Wet deposition of aerosols was temporally and spatially dependent upon precipitation on the WFS. Therefore, choice of a data product was crucial to the reproduction of deposition patterns. Tropical Rainfall Measurement Mission (TRMM) precipitation estimates were obtained (http://trmm.gsfc.nasa.gov) for 1999. The 3b42 algorithm (adjusted merged-infrared precipitation) produced 3-hr gridded estimates at 0.25-degree by 0.25-degree spatial resolution (Huffman et al., 1995). These data were converted to daily accumulation (mm d-1) by summing the eight images. Since these data were based on a “snapshot” once every 3-hr period, the daily accumulation over a 2dimensional field appeared banded, i.e. areas receiving rainfall were missed as fronts propagated (Fig. 17a). While these data were not directly usable, they were used to develop a daily accumulation index (DAI) relative to our study area. The TRMM data were averaged over our grid region yielding relative rainfall per day, and then divided by the average monthly total accumulation to get the DAI (percent of monthly precipitation per day). The DAI was applied to monthly mean National Centers for Environmental

PAGE 46

35 -9 2 -9 0 -88-86-8 4 -8 2 -8 0 24 25 26 27 28 29 30 31 32 33 0 2 4 6 8 10 12 14 16 18 -92-9 0 -88-86-8 4 -82 24 25 26 27 28 29 30 31 32 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Figure 17. a) Tropical Rainfall Measurement Mission (TRMM) 3-hr precipitation estimate using the 3b42 algorithm [eight 3-hr images were totaled to get daily precipitation (mm d-1) over the eastern Gulf of Mexico. b) Mean monthly (NCEP) reanalysis data corrected to daily (mm d-1) using the daily accumulation index (DAI) over the eastern Gulf of Mexico. a b

PAGE 47

36 Prediction (NCEP) Reanalysis data provided by the NOAA-CIRES ESRL/PSD Climate Diagnostics Branch, Boulder, Colorado, USA, from their website at http://www.cdc.noaa.gov. A subset of these data (Fig. 17b) were trimmed to our study area and interpolated to the model grid structure, giving spatial, time-dependent precipitation estimates at each grid box. In the tropics, we can assume most summer rain events are convective (Tegen and Fung, 1994), with rain cloud heights of ~10 km. Dust storms generally range from the surface to ~3-4 km in altitude (Prospero et al., 2001). Therefore, the vertical rainfall structure was ignored. Although many rain events were on the order of minutes to hours, all the wet deposition events were distributed equally over a one-day time step at each grid point (Lenes et al., 2005). Tegen and Fung (1994) calculated the efficiency of aerosol removal by rain as a function of the scavenging ratio Z, defined as: air rainC C Z (4) where Crain was the concentration in rain in units of grams of dust per kilogram of rainwater and Cair, the aerosol concentration in air in units of grams of dust per kilogram of air. Scavenging ratios of 500-1000 were reported for clay-sized mineral particles in the tropical Pacific Ocean (Buat-Menard and Duce, 1986). Tegen and Fung (1994) found little sensitivity to selection of a scavenging ratio within this range. Since Cair was a

PAGE 48

37 constant source term within the model, i.e. it does not decrease with precipitation or settling, a much lower scavenging ratio (Z = 132) was selected to compensate, compared to the ratio (Z = 750) selected for the dust transport model (Tegen and Fung, 1994). Atmospheric transport of dust was not modeled, just deposition. Therefore, wet deposition was a function of the spatio-temporal precipitation in relation to a homogeneous atmospheric dust concentration as collected at Miami. 2.2. Iron concentrations The biological and chemical interactions that determine iron cycling in marine systems are poorly understood, preventing complex mathematical formulations. Several assumptions were be made in order to simulate water column concentrations. Sources include atmospheric input, riverine outflow, and sediment resuspension, while losses are derived from biological uptake, organic complexation, precipitation as iron oxides, and particulate scavenging. In an attempt to simplify non-atmospheric iron sources, both riverine and sediment resuspension were accounted for using boundary conditions. A salinity to iron ratio was determined for near shore grid points ( 4 m depth): 0074 0 0002 0 sal dFe (5) over the salinity range 22-34 {sal > 34, dFe = 0.0004; sal < 22, dFe = 0.003}, allowing for increased iron concentrations during periods of increased river outflow. This

PAGE 49

38 boundary equation was regressed from measurements of total dissolved iron (Lenes et al., 2001) collected near shore during ECOHAB:Florida (1998-2001). Since the model was closed to the sediment, remineralization of particulate organic iron in the bottom sigma layer served as the bottom boundary source. Two pools of iron (mol Fe m-3) were incorporated into the model. The dissolved iron pool (dFe) was bioavailable to phytoplankton and nitrifying bacteria. The second pool, colloidal iron (cFe), represented an organically bound fraction, bioavailable to only the ammonifying bacteria. Since particulate inorganic iron is not bioavailable, it was ignored in the model. Rue and Bruland (1995, 1997) found that at low iron concentrations (<0.6 nmol kg-1) >99% of the dissolved iron was organically bound. Therefore, scavenging at lower concentrations is minimal (Johnson et al., 1997). This model incorporates a Michaelis-Menten type kinetic relationship to calculate particle scavenging (), similar to the work of Moore et al. (2002). They assumed a scavenging loss of 1% per year at iron concentrations below 0.6 nmol kg-1, with increased scavenging as iron concentrations increased (maximum = 2.74% per day with a half-saturation of 2.5 nmol kg-1). This accounted for increased particle reactivity as dissolved iron concentrations exceed concentrations of organic ligands, and losses due to precipitation of iron oxides (Moore et al., 2002). While their model focused on a global scale and multiyear time frame, our regional spatial scale and sub-yearly cases require faster scavenging to balance the photoreduction of colloidal iron as measured during DOTGOM III [Daughters of Trichodesmium in the Gulf of Mexico] (Fig. 18). A maximum scavenging rate of 4.32%

PAGE 50

39 0 0.5 1 1.5 2 2.5 6:008:0010:0012:0014:0016:0018:0020:00Time (EST)dFe (nmol kg-1) 11-Jul 12-Jul 15-Jul Figure 18. Diel assessment of dissolved iron stocks (nmol kg-1) on the West Florida shelf (40-50 m isobaths) during DOTGOM III on 10-16 July 2002. Samples were analyzed after Lenes et al. (2001).

PAGE 51

40 per day and half-saturation constant of 1.5 nmol dFe kg-1 was applied to [dFe] >0.3 nmol kg-1. A higher minimum scavenging rate (0.432% per day) offset the daily light/dark cycling of dFe concentrations. Photolytic reduction of colloidal iron by ultraviolet radiation has been shown to provide a mechanism for conversion of Fe(III) oxides and organically bound iron to the bioavailable dissolved pool (Wells et al., 1983; Wells et al., 1991; Zhu et al., 1992; Johnson et al., 1994; Miller et al., 1995). It has been demonstrated that photoreduction of dissolved inorganic Fe(III) is not significant in waters with a pH above 6.5 (King et al., 1993). Therefore, Fe(III) photoreduction was attributed to organically complexed or colloidal iron (Miller et al., 1995). Hong and Kester (1986) observed Fe(II) surface concentrations up to 12 nmol kg-1 off the coast of Peru with photoinduced variations. Similar concentrations (~15 nmol Fe kg-1) were observed in Narragansett Bay (King et al., 1991). Miller (1990) reported diel variations in Fe(II) from Narragansett Bay of <0.2 to 1.5 nmol kg-1. In July 2002 during DOTGOM III, O’Neil and Lenes (unpublished) measured diel variations of total dissolved Fe within offshore surface waters of <0.1 to 2.3 nmol kg-1 (Fig. 18), with the rate of reduction and maximum concentration varying with the intensity of ultraviolet radiation, i.e. complete cloud cover on 11-12th of July and no cloud cover on 15th of July. In the model, we attempt to account for diel fluctuations in the bioavailable iron pool by selecting a single “consumable” organic Fe(III) ligand pool (cFe) as a source for photoreduction (Miller et al., 1995). The equation to calculate the rate of photoreduction (kphoto) was after Johnson et al. (1994):

PAGE 52

41 o hvC photoI I cFe k k (6) where cFe was the concentration of colloidal iron and I, the irradiance. The maximum irradiance was Io = 2500 E m-2 s-1, and the rate constant khvC = 2.33 x 10-4 s-1. This equation ignores the wavelength dependence since only bulk rates are known. The state equation for dissolved iron (dFe) was: ) ( xdfe B P P r k B P d ddFe Tr t dFen n n n n n n photo n n n n t (7) where the first term represented advective and diffusive fluxes, the second and third term, uptake (n) by phytoplankton (Pn) and nitrifying bacteria (B2) respectively, and the fourth term (), particle scavenging. The fifth term, kphoto, was remineralization of colloidal iron to the dissolved pool by photolysis. The 6th – 8th terms represented the percent of phytoplankton-bound iron returned to the dissolved pool (r1,2) during grazing (n), as well as during phytoplankton and bacteria respiration (n). The last term was a generic loss term for transfer to the higher trophic levels. The dissolved iron state equation was subject to iron input into the upper sigma level (dFei) based on wet (1) and dry (2) deposition of dust: f Fe Fe dFed m i 2 1 (7a)

PAGE 53

42 where Fem (3.5%; Zhu et al., 1997) was the average mass fraction of iron in mineral aerosols for our size class and f, the molecular weight of iron. The iron dissolution rate, Fed (5%), was a safe estimate given soluble iron rates of <1% for dry deposition and ~14% for wet deposition over the oceans (Jickells and Spokes, 2001). The state equation for colloidal iron (cFe) was: ) ) 1 ( ( ) (4 4 1 n n fp n d n n n n n n n photo tB m B z k d k P P P r k C Fe d dcFe Tr t cFe (8) where loss was due to photolysis and uptake by ammonifying bacteria (B1). Phytoplankton and nitrifying bacteria utilized only dFe. Accumulation of colloidal iron was dependent upon the release of phytoplankton-bound iron during lysis (n) and grazing, remineralization of detritus (kd) and fecal pellets (kfp), as well as excretion (n) by phytoplankton and bacterial mortality (mn). 3. Biology 3.1. Primary producers The dynamics of a general pelagic diatom pool (P1) and the cyanobacterium Trichodesmium spp. (P2) were modeled, with the rate of change in biomass (mmol C m-3) over time (t) described by:

PAGE 54

43 ) ( ) ( ) (n n n n n n n n n n n n n n t nP P r P P P d P w dP Tr t P (9) where w1 was the sinking rate of pelagic diatoms and w2, the vertical migration rate of Trichodesmium. The terms n, n, n, n, and n represented the respective growth, respiration, excretion, grazing, and lysis rates. Sloppy grazing for both phytoplankton populations was accounted for via an ingestion term (r3,4). The realized net growth rate for Trichodesmium, 2, was determined by the least available resource of light or nutrients: !" !" # dFe k dFe or DOP k DOP PO k PO e I In dFe n DOP n PO I I sat z Tsat z) ( ) ( 4 ) ( 4 4 ) 1 ( 2 max 2, min (10) where T max2 was the maximum gross growth rate adjusted for temperature-stress, as Trichodesmium growth rate was maximum at temperatures >20 C (Carpenter et al., 1983). The Michaelis-Menten half-saturation constants for uptake of phosphate (kPO4), dissolved organic phosphorus (kDOP), and iron (kdFe) were specific to Trichodesmium Phosphorus availability was calculated as total dissolved phosphorus (TDP=DOP+PO4), where preferential uptake was given to phosphate over DOP through use of a smaller half-saturation constant (Table 1). High levels of alkaline phosphatase activity have been measured in association with phosphate-depleted colonies (Yentsch et al., 1972; Stihl et al., 2001), indicating Trichodesmium ’s can assimilate DOP. Isat was the saturation light

PAGE 55

44 intensity for Trichodesmium Photo-inhibition was ignored due to high saturation light intensities (Carpenter and Roenneberg, 1995) and the ability to control their buoyancy in relation to the light field (Walsby, 1978; Villareal and Carpenter, 1990). Subramanian et al. (1999) also showed that Trichodesmium releases excess energy as fluorescence and possesses high levels of photoprotective pigments. The realized net growth rate of pelagic diatoms, 1, was determined by the same method, though nitrogen species were included: !" !" !" # dFe k dFe or PO k PO NH k NH NO k NO e I In dFe n PO n NH n NO I I sat z Tsat z) ( 4 ) ( 4 4 4 ) ( 4 4 3 ) ( 3 3 ) 1 ( 1 max 1, , min (11) where each half-saturation constant was specific to diatoms (Walsh et al., 2003). Diatoms were selected as a second phytoplankton species to directly compete with Trichodesmium for nutrient supplies (A red tide state variable was not included as it is the focus of Scott Milroy’s dissertation). The biochemical model was formulated to investigate the pathways for growth and decay of a Trichodesmium population, and the subsequent effect on water column nutrient ratios. Any anticipated nutrient feedback from K. brevis was ignored. The maximum phytoplankton gross growth rate varies with changes in temperature (Eppley, 1972):

PAGE 56

45 2706330 )max( )max(expb#tT n T n (12) where T was the temperature obtained from POM and max(n) was the functional group specific maximum growth rate (Table 1). 3.2. Loss terms Grazing rates are low on populations of Trichodesmium, since only a few species of copepods can deal with the toxins (ONeil and Roman, 1992; Sellner, 1997). Thus Trichodesmium grazing (2) was calculated as a linear function of biomass (Hood et al., 2001), removing 1% of the population per day. Diatom grazing (1) was calculated as a Michaelis-Menten function after Mullin et al. (1975): ))(/)((11 11 max 11 ref zoo refPPkPP b b#t (13) where 1 max was the maximum grazing rate and kzoo, the half saturation constant. P1 ref represented the background reference population for Trichodesmium (0.01 mmol C m-3) and diatoms (0.5 mmol C m-3), below which grazing was set to zero. Viruses can be a significant mode of mortality for single-cell marine cyanobacteria (Proctor and Furman, 1990; 1991) and Trichodesmium spp. 0.3 6.5% of trichomes per day (Hewson et al., 2004). Berman-Frank et al. (2004) measured programmed cell death in Trichodesmium at a rate of >45% in 24 hr. with increased rates under nutrient, light, or oxidative stress. An exponential non-grazing mortality term (2)

PAGE 57

46 was formulated as a negative function of phytoplankton limitation (Lenes, 2002): 25413exp15.0$bt (14) where 2 was the fraction of maximum uptake by Trichodesmium of Fe and P. Thus, as limitation of Trichodesmium growth becomes larger, their non-grazing mortality increases exponentially, reflecting other losses. These lysed materials are returned to the water column mainly as DOC, DON, and DOP, as well as dFe. 3.3. Vertical migration Trichodesmium has been shown to vertically migrate throughout the euphotic zone in response to light levels, with significant populations observed at depths >100 m oligotrophic, clear waters (Borstad, 1982). As significant light becomes available, ballast is added through photosynthesis, causing the organism to sink. When light is no longer available, the ballast is respired, allowing the organism to float back to the surface. Previous simulations (Kromkamp and Walsby, 1992) suggest that trichomes and colonies may sink to these depths. Walsby (1992) indicated that sinking and floating velocities are often in excess 1 mm s-1, depending upon colony size and form resistance. In this model, a constant migration rate of 0.9 mm s-1 operated as an on/off switch. The population sank when Iz >25 E m-2 s-1 and floated when Iz <25 E m-2 s-1. This underestimated the delay observed in migrating populations at the surface, but corrected for continued sinking below the light compensation depth. Diatoms had a constant sinking rate (Table 1).

PAGE 58

47 3.4. Microbial loop A large community of heterotrophic bacteria was often associated with populations of Trichodesmium (Carpenter and Price, 1977; Borstad, 1978; Capone et al., 1994). Sheridan et al. (2002) found bacterial enrichment of Trichodesmium colonies 2-5 times that of the water column. Increases in bacterial biomass were also found for unhealthy versus healthy colonies, verifying bacterial dependence on Trichodesmium’s dissolved organic exudates as a growth media (Sellner, 1997). Therefore, heterotrophic bacteria of the model both remineralize organic matter to inorganic matter, and nitrify ammonium to nitrate. The bacterial population was broken into two functional groups: the ammonifying (B1) and nitrifying (B2) bacteria. These were described by: ) ( ) (n n n n n n n t nB m B B d dB Tr t dB (15) where 3,4 represented the bacterial respiration rates (del Giorgio and Cole, 2000) and mn, the bacterial mortality rates (Table 1). The realized bacterial growth rates (3,4) were calculated as a Michaelis-Menten function of the least available nutrient, where ammonifying bacteria were limited by organic nutrients (DOC, DON, DOP) and nitrifying bacteria by inorganic nutrients (NH4, PO4). 4. Chemistry 4.1. Carbon

PAGE 59

48 Peng et al. (1987) described the net CO2 flux across the air-sea interface by: airpCO pCO CO E CO F2 2 2 2% # (16) where E(CO2) is the CO2 gas exchange coefficient and %pCO2, the difference between pCO2 in seawater and air. Utilization and remineralization of dissolved inorganic carbon (DIC) was then described as: ) ( ) (2 2DOC k P r B P B P d dDIC Tr t DICphoto n n n n n n n n t (17) where DIC was a carbon source for Trichodesmium, diatoms, and the nitrifying bacteria, but not the ammonifying bacteria. DIC was respired by Trichodesmium, diatoms, and bacteria, and since grazers are not explicitly modeled, a small portion of grazed diazotrophs (r1) and diatoms (r2) were converted to DIC to compensate for grazer respiration. Photoconversion of dissolved organic carbon was the final source of DIC. At the surface boundary (h=0), CO2 flux is calculated from eq. (16). Dissolved organic carbon (DOC) was described as:

PAGE 60

49 ) ) 1 ( ( ) (1 4 3 1 1bDOC z k d k P P r P B m B d dDOC Tr t DOCfp n d n n n n n n n n t (18) where DOC was a carbon source for ammonifying bacteria. A fraction of grazed phytoplankton-bound carbon was released as DOC, representing sloppy grazing. Additional sources were bacterial death, phytoplankton excretion, phytoplankton lysis, and remineralization of fecal pellets and detritus. 4.2. Nitrogen As the underlying focus of this model, the nitrogen dynamics were critical in determining whether, under various scenarios, the WFS can produce a diazotrophic population large enough to support the observed red tide blooms of >10 ug chl l-1 and ~2 g C m-2 d-1 (Vargo et al., 1987). Typical measurements of nitrogen fixation in the western Atlantic of ~200 mmol N m-2 yr-1 (Carpenter and Roman, 1991), or ~0.5 mmol N m-2 d-1, suggests a potential production by Trichodesmium of only 0.4 g C m-2 d-1. Recent studies have shown western Atlantic nitrogen fixation is actually 4-5 times greater than previously thought (Capone et al., 2005; Lenes et al., 2005). This raises “new” production estimates to 1.6-2.0 g C m-2 d-1, similar to those measured for red tide blooms in the Gulf of Mexico. Yet, Trichodesmium accumulations of >100 ug chl l-1 have been found in the WFS (Walsh et al., 2003), suggesting either much higher rates of new production or mass accumulation along fronts or beaches. Mulholland et al. (2001) demonstrated that Trichodesmium could assimilate

PAGE 61

50 inorganic nutrients (ammonium [NH4 +] and nitrate [NO3 -]) and amino acids (glutamate and glutamine) without significant suppression of nitrogen fixation. They suggest that this may be a possible mechanism for cell clusters containing nitrogenase (diazocysts) to provide N to non-nitrogen fixing regions of the trichomes. For purposes of this model, all required N will be supplied via N2 fixation (function of primary production). Since Trichodesmium releases up to 50% of fixed nitrogen as DON (Glibert and Bronk, 1994), much of which in the form of amino acids (Capone et al., 1994), 25% of fixed-N will be released as DON via excretion. The DON state equation mirrors that of DOC, with differences found only in the variation of N:C ratios (Table 2). Therefore, the state equation for ammonium was: ) ( ) (4 3 3 3 4 4 1 1 4 4bDON P r B P B P C N d dNH Tr t NHn n n n t (19) where ammonium was a nitrogen source for nitrifying bacteria and diatoms. Ammonium sources included respiration of phytoplankton and ammonifying bacteria, phytoplankton grazing, and photolytic conversion of DON. Nitrate + nitrite was described in the watercolumn by: ) ( ) (1 1 4 4 3 3P B C N d dNO Tr t NOt (20) where nitrate was released by respiration of nitrifying bacteria and taken up by diatoms.

PAGE 62

51 Table 2. Nutrient and chlorophyll ratios for each functional group. Functional Group Carbon:Nitrogen Carbon:Phosphorus Carbon:Iron Carbon:Chlorophyll ____________________________________________________________________________________________________________ Diatoms (P1) 6.625 106 7000 45 Trichodesmium (P2) 6 144 1500 220 Bacteria (B) 6.625 200 15000

PAGE 63

52 4.3. Phosphorus The availability of phosphorus is generally considered in terms of orthophosphate. However, Trichodesmium exhibits high levels of alkaline phosphatase activity when inorganic P-stocks are minimal (Yentsch et al., 1972; Stihl et al., 2001). Concentrations of DOP and PO4 are also enriched in and around blooms of Trichodesmium (Karl et al., 1992). During 1999, Lenes et al. (2001) measured increased surface DOP concentrations on the WFS at the onset of an offshore Trichodesmium bloom. When the population reached its maximum biomass (Fig. 11), the water-column was devoid of both inorganic (Fig. 5) and organic P stocks (Fig. 6). Further model investigation of P dynamics suggested that available phosphorus was quickly assimilated by the diazotrophs during exponential growth phase, only to be re-released as organic P by excretion, grazing, and lysis (Lenes et al., 2005). Therefore, the state equation for inorganic phosphorus (PO4) was described by: ) ( ) (4 4 4 3 4 4bDOP B B P P r P C P d dPO Tr t POn n n n n n n n t (21) where phosphate was taken up by phytoplankton and nitrifying bacteria, and released during phytoplankton grazing, as well as by phytoplankton and both functional groups of bacteria during respiration. The DOP state equation varies from that of DOC and DON due to Trichodesmium’s ability to utilize organic P for growth:

PAGE 64

53 ) ) 1 ( ( ) (1 3 3 4 3 2 2bDOP z k d k B m B P P r P P C P d dDOP Tr t DOPfp n d n n n n n n n n t (22) where utilization by Trichodesmium and ammonifying bacteria were the major sinks. Source terms were excretion, grazing, and lysis of phytoplankton, bacterial cell death, and remineralization of fecal pellets and detritus. 5. Boundary and initial conditions Weekly and monthly data were collected during cross-shelf sections off Tampa Bay, Sarasota, and Charlotte Harbor during March 1998-November 2001 (Fig. 1) as part of the NOAA/EPA ECOHAB: Florida study. Temperature, salinity, nutrients (NO3, NO2, NH4, PO4, Fe, DOP, DON), chlorophyll a, phaeopigmnents, PON, POC, POP, 15PON, and abundances of the dominant phytoplankton and zooplankton species were obtained. Similar measurements were collected during 2001-02 as part of the NSF funded program DOTGOM in an attempt to examine Trichodesmium and K. brevis nutrient pathways on the WFS. Additionally, on some of these cruises, optical measurements were made (K. Carder – USF Ocean Optics Lab) of the spectral dependence of absorption, backscattering, water-leaving radiance, and light attenuation as part of the ONR HyCODE [Hyperspectral Coastal Ocean Dynamics Experiment] program. Surface seawater samples were collected at discrete optical stations and processes for absorption spectra by total particles (ap()), detritus (ad()), and CDOM (aCDOM()). Remote sensing

PAGE 65

54 reflectance spectra were measured at daylight stations to validate atmospherically corrected satellite data. Quarterly surveys of the MMS NEGOM [Northeastern Gulf of Mexico] project (http://www.gomr.mms.gov) provided the upstream conditions across 11 other sections during spring, summer, and fall of 1998-2000 (Fig. 1). This third set of observations includes underway ADCP, salinity, temperature, chlorophyll, CDOM data, and discrete observations of nutrients (NO3, NO2, NH4, urea, PO4), HPLC pigments, POC, PON, turbidity, and the same bioptical properties. These data sets provided initial (Table 3) and open boundary (Table 4) conditions for the model predictions of the availability of nutrients to primary producers via both deep-sea and fluvial nutrient supplies. Aeolian dust concentrations collected at Miami (Fig. 16) provided the atmospheric boundary conditions. Finally, salinity to nutrient ratios were set for near shore grid points ( 4 m) to mimic riverine input using the same formulation as for dFe (eq. 5). PO4 was calculated between salinities of 22-34 {sal > 34, PO4 = 0.025; sal < 22, PO4 = 1.225}: 425 3 1 04 sal PO (23) for all coastal boundaries. A deviation in molar N:P ratios were observed along the Gulf Coast (16-30 for most northern rivers; 2-6 for most southern rivers; Walsh et al., 2006). This was simulated by adding inorganic nutrients at a NO3/PO4 ratio of ~20 north of 28.5 N latitude and ~3 south of this line. DIC was fixed at 2100 mmol m-3 for all river influenced grid points. CDOM input did not vary with source since molar N:P ratios were accounted for using the inorganic pools.

PAGE 66

55 Table 3. Initial conditions applied uniformly over the West Florida shelf. Depth P1 P2 NH4 NO3 PO4 DON DOP DOC m mmol C m-3 mmol C m-3 mmol m-3 mmol m-3 mmol m-3 mmol m-3 mmol m-3mmol m-31 0.125 0.02 0.1 0.05 0.05 6.5 0.1 55 10 0.125 0.02 0.1 0.05 0.05 6.5 0.1 55 50 0.125 0.02 0.1 0.05 0.05 6.5 0.1 55 100 0.125 0.01 0.1 0.05 0.05 6.5 0.1 55 150 0.125 0 0.1 0.05 0.05 6.2 0.1 54 200 0 0 0.1 5.04 0.24 6 0.1 53 250 0 0 0.1 7.55 0.375 6 0.1 52 300 0 0 0.1 9.06 0.5 6 0.1 51 350 0 0 0.1 10.57 0.595 6 0.1 50 400 0 0 0.1 12.08 0.68 6 0.1 49 450 0 0 0.1 13.59 0.765 6 0.1 48 500 0 0 0.1 15.1 0.85 6 0.1 47 600 0 0 0.1 18.12 1.02 6 0.1 46 700 0 0 0.1 21.14 1.19 6 0.1 45 800 0 0 0.1 24.16 1.36 6 0.1 45 900 0 0 0.1 27.18 1.53 6 0.1 45 1000 0 0 0.1 27.5 1.75 6 0.1 45 1500 0 0 0.1 30 1.75 6 0.1 45 2000 0 0 0.1 35 1.75 6 0.1 45 2500 0 0 0.1 40 1.75 6 0.1 45 Depth DIC Z1 B1 B2 D1 D2 dFE cFe m mmol m-3 mmol C m-3 mmol C m-3mmol C m-3mmol C m-3mmol C m-3 mmol m-3mmol m-31 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 10 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 50 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 100 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 150 2100 0 0.05 0.05 0.005 0.005 0.0004 0.0004 200 2120 0 0.05 0.05 0.005 0.005 0.0005 0.0005 250 2125 0 0.05 0.05 0.005 0.005 0.0005 0.0005 300 2130 0 0.05 0.05 0.005 0.005 0.0005 0.0005 350 2135 0 0.05 0.05 0.005 0.005 0.0006 0.0006 400 2140 0 0.05 0.05 0.005 0.005 0.0006 0.0006 450 2145 0 0.05 0.05 0.005 0.005 0.0006 0.0006 500 2150 0 0.05 0.05 0.005 0.005 0.0007 0.0007 600 2160 0 0.05 0.05 0.005 0.005 0.0007 0.0007 700 2170 0 0.05 0.05 0.005 0.005 0.0008 0.0008 800 2180 0 0.05 0.05 0.005 0.005 0.0008 0.0008 900 2190 0 0.05 0.05 0.005 0.005 0.0009 0.0009 1000 2200 0 0.05 0.05 0.005 0.005 0.0009 0.0009 1500 2200 0 0.05 0.05 0.005 0.005 0.001 0.001 2000 2200 0 0.05 0.05 0.005 0.005 0.001 0.001 2500 2200 0 0.05 0.05 0.005 0.005 0.001 0.001

PAGE 67

56 Table 4. Open boundary conditions. Depth P1 P2 NH4 NO3 PO4 DON DOP DOC m mmol C m-3 mmol C m-3 mmol m-3 mmol m-3 mmol m-3 mmol m-3 mmol m-3mmol m-31 0.125 0.02 0.1 0.05 0.05 6 0.05 55 10 0.125 0.02 0.1 0.05 0.05 6 0.05 55 50 0.125 0.01 0.1 0.05 0.05 6 0.05 55 100 0.125 0 0.1 0.05 0.05 6 0.05 55 150 0.125 0 0.1 0.05 0.05 6 0.05 54 200 0 0 0.1 5.04 0.24 6 0.05 53 250 0 0 0.1 7.55 0.375 6 0.05 52 300 0 0 0.1 9.06 0.5 6 0.05 51 350 0 0 0.1 10.57 0.595 6 0.05 50 400 0 0 0.1 12.08 0.68 6 0.05 49 450 0 0 0.1 13.59 0.765 6 0.05 48 500 0 0 0.1 15.1 0.85 6 0.05 47 600 0 0 0.1 18.12 1.02 6 0.05 46 700 0 0 0.1 21.14 1.19 6 0.05 45 800 0 0 0.1 24.16 1.36 6 0.05 45 900 0 0 0.1 27.18 1.53 6 0.05 45 1000 0 0 0.1 27.5 1.75 6 0.05 45 1500 0 0 0.1 30 1.75 6 0.05 45 2000 0 0 0.1 35 1.75 6 0.05 45 2500 0 0 0.1 40 1.75 6 0.05 45 Depth DIC Z1 B1 B2 D1 D2 dFE cFe m mmol m-3 mmol C m-3 mmol C m-3mmol C m-3mmol C m-3mmol C m-3 mmol m-3mmol m-31 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 10 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 50 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 100 2100 0 0.05 0.05 0.005 0.005 0.0003 0.0003 150 2100 0 0.05 0.05 0.005 0.005 0.0004 0.0004 200 2120 0 0.05 0.05 0.005 0.005 0.0005 0.0005 250 2125 0 0.05 0.05 0.005 0.005 0.0005 0.0005 300 2130 0 0.05 0.05 0.005 0.005 0.0005 0.0005 350 2135 0 0.05 0.05 0.005 0.005 0.0006 0.0006 400 2140 0 0.05 0.05 0.005 0.005 0.0006 0.0006 450 2145 0 0.05 0.05 0.005 0.005 0.0006 0.0006 500 2150 0 0.05 0.05 0.005 0.005 0.0007 0.0007 600 2160 0 0.05 0.05 0.005 0.005 0.0007 0.0007 700 2170 0 0.05 0.05 0.005 0.005 0.0008 0.0008 800 2180 0 0.05 0.05 0.005 0.005 0.0008 0.0008 900 2190 0 0.05 0.05 0.005 0.005 0.0009 0.0009 1000 2200 0 0.05 0.05 0.005 0.005 0.0009 0.0009 1500 2200 0 0.05 0.05 0.005 0.005 0.001 0.001 2000 2200 0 0.05 0.05 0.005 0.005 0.001 0.001 2500 2200 0 0.05 0.05 0.005 0.005 0.001 0.001

PAGE 68

57 Results and Discussion 1. Atmospheric deposition Total dust deposition was defined as wet deposition + dry deposition calculated daily over each grid box. While monthly dry deposition (Fig. 19) was assumed homogeneous over the model domain, monthly wet deposition (Fig. 20) was directly dependent upon precipitation (Fig. 21). The model calculated that 82-93% of deposition on the WFS resulted from wet processes. During the 6-month simulation (May-October, 1999), total wet (Fig. 22a), total dust (Fig. 22b), and total Fe (Fig. 23a) deposition ranged from 0.7-2.0 g dust m-2, 0.9-2.2 g dust m-2, and 27.9-78.2 mg Fe m-2, respectively. Given the 5.0% dissolution utilized in the model, the total Fe dissolved within the surface sigma layer over the 6-month simulation was 25.0-70.0 mol m-2 (Fig. 23b). The range of modeled deposition estimates were similar to those measured at Miami (~1.25 g dust m-2 yr-1; Prospero et al., 1987) and over the length of Florida (0.78-1.9 dust g m-2 yr-1; Landing et al., 1995), as was the Fe-deposition in the SE Unites States and NW Caribbean (10-100 mg Fe m-2 yr-1; Duce et al., 1991). Also, Lenes et al. (2001) estimated the response to the 1999 summer aeolian influx on the WFS at ~18 nmol kg-1 of total dissolved iron in the 5-m surface mixed layer. This translated to a surface deposition of ~101 mg Fe m-2 and a surface dissolution of ~90 mol Fe m-2, only slightly higher than the modeled calculations. During 1999, there were 8 dust events (average concentration >10 g dust m-3).

PAGE 69

58 0 10 20 30 40 50 60 70 80 May JuneJulyAugustSeptemberOctober MonthDust Deposition (mg m-2 month-1) Figure 19. Total monthly dry deposition as calculated by the model at each grid point (mg dust m-2 month-1) during 1999.

PAGE 70

59 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 10 20 30 40 50 60 70 80 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 100 200 300 400 500 600 700 800 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 50 100 150 200 250 300 350 400 450 500 550 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 40 80 120 160 200 240 280 320 360 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 20 40 60 80 100 120 140 160 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 20 40 60 80 100 120 140 160 180 200 Figure 20. Total monthly wet deposition of dust as calculated by the model at each grid point (mg dust m-2 month-1) for the months of a) May, b) June, c) July, d) August, e) September, and f) October 1999. a b c d e f

PAGE 71

60 -92-9 0 -88-86-84-82 25 26 27 28 29 30 31 32 0 20 40 60 80 100 120 140 160 180 200 220 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 50 100 150 200 250 300 350 400 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 50 100 150 200 250 300 350 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 50 100 150 200 250 300 350 400 -9 2 -90-8 8 -86-8 4 -8 2 25 26 27 28 29 30 31 3 2 0 25 50 75 100 125 150 175 200 225 250 275 300 325 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 25 50 75 100 125 150 175 200 225 250 275 Figure 21. Total monthly precipitation as calculated by the model at each grid point (mm month-1) for the months of a) May, b) June, c) July, d) August, e) September, and f) October 1999. a b c d e f

PAGE 72

61 -92-9 0 -88-86-8 4 -82 25 26 27 28 29 30 31 32 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 -92-9 0 -88-86-8 4 -82 25 26 27 28 29 30 31 32 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 Figure 22. a) Total wet deposition (g dust m-2) and b) total dust deposition (g dust m-2) as calculated by the model at each grid point during the 6-month simulation. b a

PAGE 73

62 -92-9 0 -88-86-8 4 -8 2 25 26 27 28 29 30 31 32 30 35 40 45 50 55 60 65 70 75 -9 2 -9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 26 30 34 38 42 46 50 54 58 62 66 70 Figure 23. a) Total Fe-deposition (mg Fe m-2) and b) total Fe-dissolution ( mol Fe m-2) as calculated by the model at each grid point during the 6-month simulation. b a

PAGE 74

63 Five of those dust events coincided with precipitation events (>5 mm d-1 averaged over the grid) between May-October (Table 5). This led to a total deposition of 0.4-1.4 g dust m-2 (mean = 0.8 g dust m-2) during these 5 events, with the most significant occurring on June 25th-27th (Table 5). Subsequently, 14.0-49.1 mg Fe m-2 (mean = 28.1 mg Fe m-2) was deposited into the surface sigma layer over these 5 wet deposition episodes. Then 4464% of the annual total dust deposition took place on these high precipitation/high dust days. The summer/fall (July-October) is generally considered the rainy season in Florida. A significant percentage of summer rainfall occurs when convective storms move onshore. In 1999, the majority of precipitation fell during June-September over land (Fig. 21). Also, most Saharan dust events occurred during June-August (Fig. 4). Once the position of the Bermuda High moved NE in late spring of that year, the westerly trade winds shifted north carrying Fe-rich Saharan dust (Prospero, 1999b). Thereafter, the spatial variability of precipitation controlled the level of dust deposition on the WFS. In terms of the total monthly dust deposition in 1999 (Fig. 24), wet deposition dominated total deposition during summer months. The spatial variability of wet deposition was controlled by convective storms at the coast. A shift in maximum precipitation from the southern most point of Florida during spring (Fig. 21a) to central Florida during the summer (Fig. 21b,c,d) and back south during the fall (Fig. 21e,f) coincided with observed patterns in direct solar radiation and seasonal water temperature (Weisberg et al., 1996). Finally, the timing of the dust events with precipitation events (Table 5) dominated total dust deposition. Peak total deposition (Fig. 24) occurred in

PAGE 75

64 Table 5. The dates of significant dust events (Saharan and Continental) based on mean atmospheric aerosol concentrations >10 g dust m-3 (averaged over the grid) in relation to the dates of concurrent rain events (>10 mm d-1 averaged over the grid), the cumulative precipitation during these rain events (mm), and the dust deposition during these events (g dust m-2) as calculated by the model. Dust Events Mean Aerosol Concurrent Cumulative Deposition Mean Deposition Concentration Rain Events Precipitation Per Event Per Event (Date) ( g dust m-2) (Date) (mm) (g dust m-2) (g dust m-2) 16-17 Jun 12.0 25 Jun-1Jul 21.1 25-27 Jun 97 0.20-0.76 0.39 8-13 Jul 42.8 8-9 Jul 28 0.08-0.23 0.15 17-22 Jul 35.7 17-20 Jul 18 0.05-0.11 0.07 28 Jul-2 Aug 18.6 5-11 Aug 12.1 28 Aug –1 Sept 13.0 28 Aug-1 Sept 109 0.07-0.23 0.15 3-7 Oct 14.2 7-Oct 44 0.02-0.08 0.04

PAGE 76

65 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 10 20 30 40 50 60 70 80 90 100 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 100 200 300 400 500 600 700 800 900 -9 2 -9 0 -8 8 -8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 0 100 200 300 400 500 600 700 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 50 100 150 200 250 300 350 400 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 20 40 60 80 100 120 140 160 -9 2 -90-8 8 -86-8 4 -8 2 25 26 27 28 29 30 31 3 2 0 20 40 60 80 100 120 140 160 180 200 Figure 24. Total monthly dust deposition as calculated by the model at each grid point (mg dust m-2 month-1) for the months of a) May, b) June, c) July, d) August, e) September, and f) October 1999. e f c d a b

PAGE 77

66 June, when a large Saharan dust event intersected with days of heavy rainfall (25-27 June). In contrast, peak dry deposition occurred in July (Fig. 19), directly correlating to maximum atmospheric dust concentrations. The relationship of these three variables dictated overall patterns of total dust deposition within our region (Fig. 24), with the highest concentrations of dust deposited along the west Florida coast (Fig. 22b). Specifically, maximum concentrations were calculated between Tampa Bay and Charlotte Harbor (Table 6). A breakdown of the spatiality of monthly dust deposition showed similar patterns to precipitation, with maximum deposition occurring at the southern most tip of Florida in May (Fig. 24a). This was followed by a shift to west central Florida during June-August (Fig. 24b,c,d) and back south in the fall. Seasonal deposition varied greatly, with the maximum precipitation and highest atmospheric dust concentrations occurring during the summer. While interannual variability was not explicitly modeled, increased desertification due to long-term drought in the Sahel region of Africa has led to increased atmospheric dust concentrations measured at Barbados (Prospero et al., 1987). Therefore, shifts in precipitation due to the North Atlantic Oscillation (NAO) or El Nino events can cause significant variability in total dust deposition. 2. Water column iron The model’s surface total dissolved iron (dFe) concentrations (Fig. 25) followed the general observed pattern from data collected during monthly ECOHAB cruises in 1999 (Fig. 2). Measured dFe concentrations were <0.5 nmol kg-1 during 2-5 May

PAGE 78

67 Table 6. Variation in the environmental parameters leading to the transfer of nutrients to red tides on the northern (NS) and southern (SS) continental shelf in the eastern Gulf of Mexico in the summer/fall of 1999 (+ = yes, = no). Variables NS SS N:P Ratios (<6) + Dissolved Iron (>1 nmol l-1) + + Trichodesmium (>5 mg chl m-2) + + Silica (>1 mol kg-1) + Diatoms (>2 g chl l-1) + Red Tides (~20 g chl l-1) +

PAGE 79

68 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3 3.3 3.6 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3 3.3 3.6 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.5 1 1.5 2 2.5 3 3.5 4 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 1 2 3 4 5 6 7 8 9 10 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 1 2 3 4 5 6 7 8 9 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 Figure 25. The simulated total dissolved iron (dFe) in the surface sigma layer (nmol Fe kg-1) on a) 4 May, b) 7 June, c) 6 July, d) 7 August, e) 8 September, and f) 6 October 1999. e f c d a b

PAGE 80

69 throughout the ECOHAB control volume (Fig. 2a). The model mimicked this pattern within the area of the measurements. Modeled concentrations were >1 nmol Fe kg-1 near the mouth of the Mississippi plume and along the coast (Big Bend region and Naples). By 7 June (Fig. 25b), the plume of dFe near Naples was north and slightly farther offshore. Elevated Fe-stocks were both measured (Fig. 2b) and simulated (Fig. 25b) near shore. The wet deposition event on 25-27 June provided 125-476 nmol Fe kg-1 over the model domain with the highest input along the Florida coast. After 5% dissolution, 6.323.8 nmol Fe kg-1 was added to the dFe pool. Ten days later on 5-7 July, dFe concentrations were >2 nmol kg-1 inside the 50-m isobath (Fig. 2c), though a small zone of depletion was measured along the 30-m isobath between Tampa Bay and just north of Charlotte Harbor at salinities >36.0 (Lenes et al., 2001). Maximum coastal concentrations of dFe were measured (Fig. 2c) along the 20-m isobath offshore of Charlotte Harbor (~4 nmol kg-1). This increased homogeneity along the full length of the WFS was reproduced in the model (Fig. 25c), with peak concentrations (~4 nmol Fe kg-1) SSE of Charlotte Harbor between the 20and 30-m isobaths. Maximum dFe concentrations (>5 nmol kg-1) measured along the shelf break (200-m isobath) at the end of the Sarasota Transect was assumed to have resulted from elevated wet deposition during the 25-27 June event. Since the model partitioned precipitation estimates from monthly to daily using the DAI, spatially anomalous storms were smoothed to the general pattern of the monthly rain field. On 6-8 August, the continued increase in measured coastal dFe concentrations (Fig. 2d) coincided with the maximum monthly dry deposition (Fig. 19) and several wet

PAGE 81

70 deposition events during July (Table 5). The measured and simulated (Fig. 25d) values were >2 nmol Fe kg-1 inside the 50-m isobath. The depleted zone in the center of the ECOHAB control volume was still apparent, but had shifted farther offshore (Fig. 2d). By 7-9 September, both measured (Fig. 2e) and simulated (Fig. 25e) dFe concentrations began to decrease to <1.5 nmol kg-1 inside the 50-m isobath. Elevated concentrations were still observed at the mouth of the major Florida river systems. The dFe pool had returned to background (<0.5 nmol kg-1) by 5-7 October (Fig. 2f; Fig.25f). A null case of the model was run to test the impact of riverine versus atmospheric dust input. In the null case, there was no atmospheric deposition of iron so only the deepsea, riverine, and bottom boundaries supplied iron to the biochemical model. Since the contribution of both the deep-sea component and bottom boundary were minimal due to strong stratification, the surface dFe concentration was driven mainly by fluvial inputs (Fig. 26). These results indicated that both atmospheric deposition and riverine supplies were important sources of iron to the WFS. In 1999, fluvial stocks drifted primarily along the coast (Fig. 26) in response to the seasonal wind direction (Weisberg and He, 2003). Since dFe concentrations were below background over most of the model domain in the null case, the primary producers were consistently Fe-limited. Therefore, the simulated dFe concentrations in the atmospheric deposition case (Fig. 25) more closely matched those observed during the 1999 monthly ECOHAB cruises (Fig. 2).

PAGE 82

71 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 Figure 26. The null case of simulated total dissolved iron (dFe) in the surface sigma layer (nmol Fe kg-1) on a) 4 May, b) 7 June, c) 6 July, d) 7 August, e) 8 September, and f) 6 October 1999.

PAGE 83

72 3. Initiation of Trichodesmium 3.1. Iron response In the model, the depth-integrated Trichodesmium chlorophyll stocks on 4 May 1999 demonstrate a maximum biomass along the full length of the outer shelf (Fig. 27a). Maximum Trichodesmium chlorophyll concentrations at a depth of 15-m follow the same pattern (Fig. 27b). Bottom nitrate (Fig. 28a) and phosphate (Fig. 28b) did not invade due to the constraints of the Taylor-Proudman Theorem (Weisberg and He, 2003). In 1998, a spring loop current intrusion forced deep-sea water through the DeSoto Canyon at the northeastern corner of the shelf (Weisberg and He, 2003). The infusion of nutrients led to a domination of diatoms through the spring and summer months on the shelf (Walsh et al., 2003). Without a 1999 infusion of deep-sea nutrients on the WFS, the lack of a nongaseous nitrogen source shifted competitive advantage to the diazotrophs (Table 6). In early May, atmospheric iron supply was minimal (Fig. 24a). Bottom concentrations did not represent an appreciable dFe source to Trichodesmium (Fig. 29a), though bottom cFe stocks were transported up the shelf break and onto the outer WFS (Fig. 29b) by vertical mixing and diffusion. Since cFe was not bioavailable for phytoplankton species, it remained untouched until it reached the euphotic zone were photoconversion to the dFe pool allowed uptake by primary producers. The spatial extent of the cFe intrusion (Fig. 29b) overlapped with the band of maximal chlorophyll stocks (Fig. 27). While this source provided enough iron to supply the calculated Trichodesmium biomass in May of ~1.5 mg chl m-2 along the outer shelf, phosphorus acquisition still remained a problem.

PAGE 84

73 -92-9 0 -88-86-8 4 -82 25 26 27 28 29 30 31 32 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 -9 2 -91-90-8 9 -88-87-86-8 5 -84-83-8 2 -81 25 26 27 28 29 30 31 32 0 0.004 0.008 0.012 0.016 0.02 Figure 27. The simulated Trichodesmium a) depth-integrated chlorophyll biomass (mg chl m-2) and b) 15-m chlorophyll biomass ( g chl l-1) on 4 May 1999. a b

PAGE 85

74 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 5 10 15 20 25 30 35 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 -92-9 0 -88-8 6 -8 4 -82 25 26 27 28 29 30 31 32 Figure 28. The simulated a) bottom nitrate + nitrite ( mol NO3 kg-1) and b) bottom phosphate ( mol PO4 kg-1) on 4 May 1999. a b

PAGE 86

75 -92-9 0 -88-8 6 -8 4 -82 25 26 27 28 29 30 31 32 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 -92-9 0 -88-8 6 -8 4 -82 25 26 27 28 29 30 31 32 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Figure 29. The simulated a) bottom dissolved iron (nmol dFe kg-1) and b) bottom colloidal iron (nmol cFe kg-1) on 4 May 1999. a b

PAGE 87

76 3.2. Phosphorus acquisition Multiple physical transport scenarios exist for populations of Trichodesmium in relation to observed vertical migration. Thermohaline and wind driven circulation, eddies, fronts, and Langmuir cells all drive varying mechanisms by which Trichodesmium can change location. In the oligotrophic open ocean near Barbados, cells are isolated in surface water masses with depths of >100 m (Borstad, 1982). Given the ability to fix atmospheric nitrogen (Devassy, 1978), it is generally accepted that limitation of Trichodesmium growth is a function of phosphorus and iron stocks (SanudoWilhelmy et al., 2001; Mills et al., 2004), given sufficient light. Trichodesmium has also demonstrated the ability to vertically migrate to depths >100 m (Walsby, 1992). Therefore, growth potential of a population isolated within a specific water mass is limited by that population’s ability to acquire P and Fe from multiple depths within that mass. Unfortunately, the transport scenario differs in coastal regions. When examining physical transport on continental shelves, it is frequently observed that multiple water masses occupy the euphotic zone (Weisberg and He, 2003). This not only provides access for the vertically migrating Trichodesmium to more than one water mass, but also directly affects the population’s horizontal movement. The WFS circulation is dominated by differential heating and wind driven Ekman transport, with direction of the undercurrents traveling counter to surface currents (He and Weisberg, 2002a). Therefore, a vertically migrating population is subject to transport in several directions over the course of a migration cycle, depending upon the depth of the water column and migration speed. For

PAGE 88

77 example, a population might be transported south within the surface Ekman layer, but return to the north as it migrates to depth. The movement could mimic tidal pulsing, though tides are minimal on the WFS (He and Weisberg, 2002b). The theory of ‘phosphorus mining’ applies the addition and subtraction of ballast by Trichodesmium in response to the light field, allowing for cellular access to higher concentrations of P from depth (Karl et al., 1992). While vertical mixing and diffusion of nutrients across the thermocline can supply additional P, open ocean regions that contain high Trichodesmium biomass (subtropical/tropical) generally demonstrate minimal upward flux of nutrients due to strong thermal stratification (Hood, 2001). This would limit the overall P-availability to the population unless they migrated and returned from below the thermocline. Given sufficient supply of Fe from atmospheric sources (Lenes et al., 2001), maximal biomass constraints are dependent upon P supply and the diazotrophs ability to vary cellular N:P ratios (Vargo et al., 2005) as organic nutrients are excreted to other co-existing organisms (Sellner, 1997). Within the framework of this theory, prediction of a maximum Trichodesmium biomass, and therefore nitrogen fixation, is possible with knowledge of P concentrations, Fe supply, and vertical transport. A second scenario is needed for physical transport mechanisms in coastal waters. Postulation of a new theory, ‘phosphorus combing’, expands upon ‘phosphorus mining’ to account for multiple water masses. As the Trichodesmium population migrates through density gradients in a shallow water column, access to nutrients changes as concentrations differ within the masses. The theories deviate due to directional variations of the overlying water masses, i.e. the population might not return to the original water

PAGE 89

78 mass. P depleted masses are replaced as integrated water column horizontal displacement of the Trichodesmium population differs from the layered horizontal flux (Fig. 30). Therefore, the population can access additional phosphorus by ‘combing’ the water masses. This assumes patchiness in P distribution at the same depth. Phosphorus combing could also apply to warm core, anti-cyclonic eddies (downwelling) in open ocean scenarios. The positive buoyancy of Trichodesmium would offset the downwelling effect of the eddy circulation, preventing both loss of colonies from the sinking entrained water and maintenance of their position within the euphotic zone. Mulholland and Capone (2000) reported C-based doubling times of ~2-14 days (specific growth rate: 0.05-0.32 d-1) for cultures of Trichodesmium under varying nutrient and light levels. Ambient stocks of <0.1 mol PO4 kg-1 (Stevens et al., 1970) could explain low, but consistent in situ growth rates (<0.12 d-1) measured in the subtropical/tropical Atlantic (Carpenter, 1983). Based on a PO4 half-saturation constant of 0.1 mol PO4 kg-1 (Walsh and Steidinger, 2001), a 40% respiration/excretion rate (Lenes et al., 2005), and a maximum growth rate of 0.7 d-1 (Walsh and Steidinger, 2001), a Trichodesmium population would need ~0.04 mol P kg-1 to maintain a realized growth rate of 0.12 d-1. Therefore, at this P concentration a depth integrated population biomass on the order of 1-10 mg chl m-2 (Lenes et al., 2005) would fix between 0.4-3.7 mmol N m-2 d-1. As much as half of this ‘new’ N (Capone et al, 1994; Glibert and Bronk, 1994) would be available to co-existing organisms or export. This represents a potential carbon sink of 111 mmol m-2 d-1 within the warm core eddy. Calculation of carbon loss for eddies and

PAGE 90

79 Figure 30. The simulated circulation (cm s-1) calculated at the a) surface and b) near bottom on 4 May 1999. a b

PAGE 91

80 rings shed by the North Brazil Current (NBC), given an average diameter of ~100 km and residence time of ~115 days (Goni and Johns, 2001) was 0.01-0.13 x106 tons C eddy-1. This ignored the initial P load supplied by the Amazon source region to the NBC, inclusion of which would drastically increase production estimates. 4. Growth and maintenance In May 1999, surface PO4 distributions were almost undetectable >20 km from the Florida coast (Fig. 5a). The molar DIN/ PO4 ratios (Fig. 7) were >15 along the northern shelf, while ratios were <15 south of 28 latitude and along the shelf break. The simulated molar DIN/ PO4 ratios (Fig. 31) reproduce this pattern, though the spatial influence of the northern river systems were underestimated. The low N:P regions along the WFS gave a competitive advantage to nitrogen fixers (Table 6). By June, surface concentrations ranged from 0.0-0.8 mol PO4 kg-1 across the ECOHAB control volume (Fig. 5b), with the maximum found near shore. As much as 0.2 mol PO4 kg-1 was measured at the end of the Ft. Myers transect above the 50-m isobath. Trichodesmium biomass showed a small decrement between May and June (Fig. 11) since P stocks were incorporated into particulate matter by May (Lenes et al., 2001). Increased surface PO4 concentrations measured inside the 50-m isobath (Fig. 5b) coincided with increased precipitation near the coast in June (Fig. 21b), i.e. fluvial supplied PO4. The observed decrement in maximum Trichodesmium biomass (Fig. 11) was related to increased DOP (Fig. 6) and DON (Fig. 12) stocks along the outer shelf. These two pools of phosphorus and an influx of dust in June (Fig. 24b) fueled the large

PAGE 92

81 -92-9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 33 36 39 -92-9 0 -88-86-8 4 -8 2 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 33 36 39 Figure 31. The simulated molar DIN/PO4 ratios (mol/mol) calculated at the a) surface and b) bottom on 4 May 1999.

PAGE 93

82 Trichodesmium population increase measured in July (Fig. 11c). The simulated surface and bottom DIN/PO4 ratios of <15 on 7 June (Fig. 32) indicated nitrogen limitation out to the 200-m isobath. The 7 June peak in depthintegrated chlorophyll concentrations (Fig. 33a) once again overlaid the bottom cFe signal (Fig. 33b), though bottom cFe did not encroach as far onto the shelf. Also, the Trichodesmium biomass had decreased from May. During June, increased atmospheric dust had led to increased surface dFe concentrations inside the 100-m isobath (Fig. 25b). As Trichodesmium biomass increased after Fe-fertilization (Fig. 11c), a subsequent removal of PO4 and DOP was observed in the measured distributions between June and July (Fig. 5; Fig. 6). In the model, PO4 distributions >20 km from the West Florida coast rarely exceeded 0.1 mol kg-1 (Fig. 34). Lenes et al. (2001) reported a maximum Trichodesmium biomass of ~20 colonies l-1 over the 100-m isobath in July 1999 (Fig. 11c), a 100-fold increase from background concentrations of 0.1-0.2 colonies l-1 found in May 1999 (Fig. 11a). This amounted to a pigment biomass of 0.7 g chl l-1, representing ~50% of the observed chlorophyll stocks. A September maximum of ~10 colonies l-1 measured above the 10-m isobath off Charlotte Harbor had disappeared by October (Fig. 11). Walsh et al. (2006) assumed a mature colony size of ~3x104 cells (Carpenter et al., 1983), a PC/Chl weight ratio of 220 (Carpenter et al., 1983), a PC/PN molar ratio of 6.1 (McCarthy and Carpenter, 1979), and a cellular chlorophyll content of 1.2x10-6 (Borstad, 1982) to calculate the potential particulate nutrient stocks (10.9 mol PN kg-1 and 0.5 mol PP kg-1) associated with the observed decrement in population size. Since a majority of Trichodesmium loss is due to

PAGE 94

83 -9 2 -9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 1 4 7 10 13 16 19 22 25 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 1 4 7 10 13 16 19 22 Figure 32. The simulated molar DIN/PO4 ratios (mol/mol) calculated at the a) surface and b) bottom on 7 June 1999.a b

PAGE 95

84 -9 2 -9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 0 0.2 0.4 0.6 0.8 -92-9 0 -88-86-8 4 -82 25 26 27 28 29 30 31 32 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Figure 33. The simulated a) depth-integrated Trichodesmium chlorophyll biomass (mg chl m-2) and b) the bottom colloidal iron (nmol cFe kg-1) on 7 June 1999. a b

PAGE 96

85 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Figure 34. The simulated surface distributions of phosphate ( mol PO4 kg-1) on (a) 4 May, (b) 7 June, (c) 6 July, (d) 7 August, and (e) 8 September 1999. a e d c b

PAGE 97

86 cell lysis near the surface, the nutrient supply was released in organic form (Lenes et al., 2005). These values were consistent with nutrient concentrations measured within a decaying Trichodesmium bloom off St. Pete Beach, FL during May 2000 of 36.9 mol DON kg-1, 15.9 mol NH4 kg-1, and 0.2 mol PO4 kg-1. By 6 July, the simulated DIN/PO4 ratios (Fig. 35) began to increase along the southern outer shelf from June (Fig. 31). This coincided with a patch of the Trichodesmium population reaching from the outer shelf into the 50-m isobath (Fig. 36a). The nutrient ratio shift resulted from elevated surface NH4 (Fig. 36b). No noticeable increase in DON was observed due to high rates of bacterial remineralization, as seen in the river effluent of the southern rivers (Fig. 37). A coastal DIN/PO4 signal >6 was observed between Charlotte Harbor and Naples as DON was stripped to ~5 mol kg-1 (Fig. 37a). The gradual shift to higher DIN/PO4 ratios continued between July and August along the outer shelf and coast (Fig. 38a). The simulated Trichodesmium biomass had reached a 6-month maximum (Fig. 38b) in August, a few weeks after the measured maximum (Fig. 11c). In contrast, the null case reached a maximum in May after exhaustion of Fe-stocks (Fig. 26). The increase in DIN/PO4 ratios (Fig. 38a) away from the coastal nutrient supplies directly coincided with areas of high Trichodesmium biomass (Fig. 38b). Several remnant low DIN:PO4 areas were found away from both the coast and high cyanophyte biomass (Fig. 38a). Similar patterns were observed in the validation data (Fig. 8), with encroachment of higher DIN:PO4 waters from the outer to middle WFS (Table 6).

PAGE 98

87 -9 2 -9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 -9 2 -9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 Figure 35. The simulated DIN/PO4 molar ratios (mol/mol) calculated at the a) surface and b) bottom on 6 July 1999.a b

PAGE 99

88 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 1 2 3 4 5 6 -9 2 -9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Figure 36. The simulated a) depth-integrated Trichodesmium chlorophyll biomass (mg chl m-2) and b) the surface ammonium ( mol NH4 kg-1) on 6 July 1999. a b

PAGE 100

89 -92-90-88-8 6 -8 4 -82 25 26 27 28 29 30 31 32 5 5.3 5.6 5.9 6.2 6.5 6.8 7.1 -9 2 -9 0 -88-86-84-8 2 25 26 27 28 29 30 31 32 0 1 2 3 4 5 6 7 8 Figure 37. The simulated a) surface dissolved organic nitrogen ( mol DON kg-1) and b) surface ammonifying bacteria ( mol C kg-1) on 6 July 1999. a b

PAGE 101

90 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 1 2 3 4 5 6 7 8 9 10 Figure 38. The simulated a) surface DIN/PO4 molar ratios (mol/mol) and b) depth-integrated Trichodesmium chlorophyll biomass (mg chl m-2) on 7 August 1999. a b

PAGE 102

91 Nitrogen fixation rates were up to ~700 mol N m-2 d-1 in August (Fig. 39a). The ‘new’ N source supplied DON and NH4 to the water column during excretion/respiration and cell death. The rapid nutrient turnover times for the bacterial terms quickly converted these pools to NO3. A substantial build up of NO3 (Fig. 39b) was not observed in the model due to the quick growth response of diatoms (Fig. 40a), although August concentrations were >1 mol NO3 kg-1 (Fig. 39b) in the area of maximum Trichodesmium (Fig. 38b) and diatom (Fig. 40a) biomass. The northern shelf supported >2 g chl l-1 in the outflow of the Mississippi River (Fig. 13). The ‘new’ nitrogen provided via atmospheric N2 fixation forced the simulated offshore elevation of DIN/PO4 ratios. Nutrient inputs associated with Trichodesmium do not include a source of silica (Si), so the available Si in the water column would limit the transfer of fixed nitrogen to diatoms. Si was not included as a state variable in the model, so diatoms were free to utilize this offshore nitrogen source, doubling biomass (Fig. 40a). Non-siliceous organisms such as dinoflagellates would have a competitive advantage over the faster growing diatoms for this N pool. Surface Si concentrations measured on the WFS were <1 mol kg-1 in September along the outer shelf (Walsh et al., 2006). Therefore, diatoms would exhaust the deep-sea Si supply at roughly the same rate as the deep-sea nitrate supply (Fig. 28a), reducing the simulated shelf break diatom population. This would especially be true on the southern shelf away from the Mississippi influence (Table 6). The model results suggested that the maximum Trichodesmium /diatom surface chlorophyll ratio was restricted to the edge of the southern WFS. Thus, the simulated

PAGE 103

92 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 100 200 300 400 500 600 700 800 900 1000 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 Figure 39. The simulated a) depth-integrated nitrogen fixation rates ( mol N m-2 d-1) by Trichodesmium and b) surface nitrate + nitrite ( mol NO3 kg-1) on 7 August 1999. a b

PAGE 104

93 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 -92-90-88-86-84-82 25 26 27 28 29 30 31 32 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Figure 40. The simulated a) surface diatom chlorophyll biomass ( g chl l-1) and b) the Trichodesmium /diatom surface chlorophyll ratio (chl/chl) on 8 August 1999. a b

PAGE 105

94 dominance of >50% diazotrophs expanded shoreward as far as the 40-m isobath (Fig. 40b), the edge of the predicted K. brevis initialization isobaths (Walsh et al., 2006). Further analysis of Trichodesmium migration and decay were needed to accurately mimic observed K. brevis locations. Diatoms instead dominated the Mississippi river plume (Fig. 40a), with <30% diazotrophs at the northern shelf break despite depth-integrated Trichodesmium concentrations >6 mg chl m-2 (Fig. 38b). The established diatom population along the edge of the northern shelf assimilated ‘new’ nitrogen released by Trichodesmium limiting it’s potential as a source for non-siliceous phytoplankton assemblages. Therefore, the model reproduced a scenario in which nutrient transfer to non-siliceous phytoplankton blooms was possible on the WFS. A further increment in DIN/PO4 ratios had occurred by September (Fig. 41a). The simulated rate of cell death increased exponentially as nutrient supply decreased (Eq. 14). Therefore, as atmospheric Fe decreased, particulate N and P stocks were released to the water column. Depth-integrated Trichodesmium biomass was reduced to <2.7 g chl l-1 by September (Fig. 41b), a 73% decline. A transition to southwesterly winds in the fall forced cross-shelf bottom Ekman transport toward the coast (coastal upwelling) (Weisberg and He, 2003). Therefore, the ~7.3 g chl l-1 reduction in Trichodesmium biomass represented a potentially transferable particulate nutrient pool of ~22.3 mol N kg-1 and ~0.9 mol P kg-1 and another ~8.3 mol N kg-1 and ~0.3 mol P kg-1 between September and October (Table 6). Walsh et al. (2006) predicted the inoculation of K. brevis blooms on the WFS was near bottom along the 30-m isobath between Tampa Bay and Naples, within ~40 km

PAGE 106

95 -92-90-88-86-8 4 -82 25 26 27 28 29 30 31 32 0 3 6 9 12 15 18 21 24 27 30 -9 2 -9 0 -88-8 6 -8 4 -8 2 25 26 27 28 29 30 31 32 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 Figure 41. The simulated a) surface DIN/PO4 molar ratios (mol/mol) and b) depthintegrated Trichodesmium chlorophyll biomass (mg chl m-2) on 8 September 1999. a b

PAGE 107

96 from the major fluvial sources of phosphorus. In October 1999, ~20 g chl l-1 of K. brevis was measured within near-shore waters above the 10-m isobath between Tampa Bay and Charlotte Harbor. This would require access to a minimum dissolved nitrogen supply of ~12.5 mol N kg-1. Therefore, 56% nitrogen transfer efficiency was needed during the simulated Trichodesmium decline to support observed K. brevis biomass. This does not account for the nitrogen previously released by Trichodesmium during growth. Given a molar N:P ratio of ~20 for K. brevis ~0.6 mol P kg-1 or 69% phosphorus transfer efficiency was required indicating an additional demand for phosphorus. High rates of P cycling have been measured in conjunction with K. brevis blooms (Heil, personal communication). In order to reproduce observed K. brevis stocks, a high P transfer efficiency or an additional P source was required. Therefore, proximity to the coast (fluvial P supply) could satisfy the remaining P demand.

PAGE 108

97 Conclusions The transport of Saharan dust across the N. Atlantic delivered ‘new’ iron to the Gulf of Mexico (Fig. 4). The model results demonstrated the spatial and temporal dependence upon both aeolian dust and precipitation. Wet deposition estimates were 4-13 times greater than dry deposition, with a majority of iron input occurring in the summer between Tampa Bay and Charlotte Harbor (Fig. 23). In 1999, five co-occurring dust/precipitation events supplied more than half of the simulated iron fertilization. Enrichment of surface waters above the WFS temporarily removed Fe-limitation for the marine diazotroph, Trichodesmium spp. The growth rate became a function of phosphorus acquisition. Late spring/early summer phosphate concentrations on the oligotrophic WFS were <0.1 mol kg-1, except near the coast where river P sources maintained elevated P stocks of >0.2 mol kg-1 inside the 10-m isobath over the simulation (Fig. 34). In the model, the WFS water column had molar N:P ratios <5 in late spring (Fig. 31). Even without interspecies competition for nutrients, not enough fluvial nitrogen makes it onto the shelf to support the observed increment in biomass of the toxic dinoflagellate, K. brevis A nitrogen source was needed. According to the model, nitrogen fixation gave Trichodesmium a competitive advantage over other phytoplankton species (Table 6). Since phosphorus availability

PAGE 109

98 controlled growth rates and maximum biomass once Fe-limitation was alleviated, a near shore population increase would be expected. This was not observed in the model, though a small in situ increment occurred during 1999. While the fluvial derived nutrients helped support the Trichodesmium population, not enough P was available to reproduce the observed biomass. The model calculated a Trichodesmium population slowly forming over the shelf break in June (Fig. 33a) and July (Fig. 36a) in response to uptake of ambient P stocks, with the peak depth-integrated biomass found in August (~10 mg chl m-2) between the 50and 100-m isobaths (Fig. 38b). A variation of the P transport hypothesis (Karl et al., 1992), P combing, was postulated to describe the affects of three-dimensional circulation on the continental shelf. Vertical migration of Trichodesmium in the model not only provided access to total integrated water column P stocks, but also altered the population’s position on the shelf relative to variations in advective transport at different depths. The diazotrophs utilized P from the initial water mass. As they descended in response to changes in light levels, additional P became available. A shift in current speed and direction with depth due to Ekman dynamics (Fig. 30) transported the population away from the initial water mass. Ascension back to the surface provided access to a new water mass. Trichodesmium acted as a sieve, combing the outer shelf for phosphorus. It was able to utilize riverine P and Fe as advective transport carried a portion of the population toward the coast between Tampa Bay and Charlotte Harbor at the end of the summer. Nitrogen fixation rates vary as a function of internal P more than internal Fe (Sanudo-Wilhelmy et al., 2001). Therefore, P-combing increased maximum biomass and net nitrogen fixation rates.

PAGE 110

99 The simulated decrement in Trichodesmium biomass (Fig. 38b; Fig. 41b) as Fefertilization decreased during the fall transition (Fig. 24) provided adequate particular nutrient stocks to support the ~20 g chl l-1 found between Tampa Bay and Charlotte Harbor in October. Surface molar N:P ratios in this area had increased to near Redfield (Fig. 41a). Without an additional Si source, this pool of nutrients would be preferentially available to non-siliceous organisms, such as red tide. In contrast, the high molar N:P ratio waters of the Mississippi River supported elevated diatom biomass on the northern shelf (Fig. 40a). The diatoms quickly assimilated any Trichodesmium derived nitrogen. Therefore, the WFS combined low molar N:P river systems and far-field aeolian dust to create a ecosystem capable of supplying enough ‘new’ nitrogen via N2 fixation to support blooms of K. brevis (Table 6). Additionally, the physiological dynamics of Trichodesmium allowed biomass increments and nitrogen fixation rates far greater than predicted from surrounding nutrient stocks.

PAGE 111

100 References Cited Barber, R.T., Chavez, F.P., 1991. Regulation of primary productivity rate in the equatorial Pacific. Limnol. Oceanogr. 36, 1803-1815. Berman-Frank, I., Cullen, J.T., Shaked, Y., Sherrell, R.M., Falkowski, P.G., 2001. Iron availability, cellular iron quotas, and nitrogen fixation in Trichodesmium Limnol. Oceanogr. 46 (6), 1249-1260. Berman-Frank, I., Bidle, K.D., Haramity, L., Falkowski, P.G., 2004. The demise of the marine cyanobacterium, Trichodesmium spp., via an autocatalyzed cell death pathway. Limnol. Oceanogr. 49 (4), 997-1005. Borstad, G.A., 1978. Some aspects of the occurrence and biology of Trichodesmium (Cyanophyta) in the Western Tropical Atlantic near Barbados, West Indies. Ph.D. Thesis, McGill Univ., Montreal, Canada, unpublished. Borstad, G.A., 1982. The influence of the meandering Guiana Current on the surface conditions near Barbados temporal variations of Trichodesmium (Cyanophyta) and other plankton. J. Mar. Res. 40, 435-452. Boyd, P.W., Watson A.J., Law C.S., Abraham E.R., Trull T., Murdoch R., Bakker D.C.E., Bowie A.R., Buesseler K.O., Chang H., Charette M., Croot P., Downing K., Frew R., Gall M., Hadfield M., Hall J., Harvey M., Jameson G., LaRoche J., Liddicoat M., Ling R., Maldonado M.T., McKay R.M., Nodder S., Pickmere S., Pridmore R., Rintoul S., Safi K., Sutton P., Strzepek R., Tanneberger K., Turner S., Waite A., Zeldis J., 2000. A mesoscale phytoplankton bloom in the polar Southern Ocean stimulated by iron fertilization. Nature 407 (6805), 695-702. Broecker, W.S., 1982. Glacial to interglacial changes in ocean chemistry. Prog. Oceanogr. 11 (2), 151-197. Buat-Menard, P., Duce, R.A., 1986. Precipitation scavenging of aerosol-particles over remote regions. Nature 321 (6069), 508-510. Caccia, V.G., Millero, F.J. 2003. The distribution and seasonal variation of dissolved trace metals in Florida Bay and adjacent waters, Aquatic Geochem. 9(2), 111-144. Capone, D.G., Ferrier, M.D., Carpenter, E.J., 1994. Amino acid cycling in colonies of the planktonic marine cyanobacterium Trichodesmium thiebautii Appl. Environ. Microbiol. 60, 3989-3995.

PAGE 112

101 Capone, D.G., Zehr, J.P., Paerl, H.W., Bergman, B., Carpenter, E.J., 1997. Trichodesmium a globally significant marine cyanobacterium. Science 276, 1221-1229. Capone, D.G., Burns, J.A., Montoya, J.P., Subramaniam, A., Mahaffey, C., Gunderson, T., Michaels, A.F., Carpenter, E.J., 2005. Nitrogen fixation by Trichodesmium spp.: An important source of new nitrogen to the tropical and subtropical North Atlantic Ocean. Glob. Biogeochem. Cyc. 19, GB2024. Carder, K.L. Steward, R.G., Harvey, G.R., Ortner, P.B., 1989. Marine humic and fluvic acids: Their effects on remote sensing of ocean chlorophyll. Limn. Oceanogr. 34, 68-81. Carder, K.L., Gregg, W.W., Costello, D.K., Haddad, K., Prospero, J.M., 1991. Determination of Saharan dust radiance and chlorophyll from CZCS imagery. J. Geophys. Res. 96, 53695378. Carpenter, E.J., Price, C.C., 1977. Nitrogen fixation, distribution, and production of Oscillatoria (Trichodesmium) spp. in the western Sargasso and Caribbean seas. Limnol. Oceanogr. 22, 60-72. Carpenter, E.J., 1983. Nitrogen fixation by marine Oscillatoria ( Trichodesmium ) in the world's oceans. In: Carpenter, E.J., Capone, D.G. (Eds.), Nitrogen in the Marine Environment. Academic Press, New York, pp. 65-103. Carpenter, E.J., Roman, K., 1991. Major role of the cyanobacterium Trichodesmium in nutrient cycling in the North Atlantic Ocean. Science 254 (5036), 1356-1358. Carpenter, E.J., Roenneberg, T., 1995. The marine planktonic cyanobacteria Trichodesmium spp.: photosynthetic rate measurements in the SW Atlantic Ocean. Mar. Ecol. Progr. Ser. 118, 267-273. Coale, K.H., Johnson, K.S., Fitzwater, S.E., Gordon R.M., Tanner S., Chavez F.P., Ferioli L., Sakamoto C., Rogers P., Millero F., Steinberg P., Nightingale P., Cooper D., Cochlan W.P., Landry M.R., Constantinou J., Rollwagen G., Trasvina A., Kudela R., 1996. A massive phytoplankton bloom induced by an ecosystem-scale iron fertilization experiment in the equatorial Pacific Ocean. Nature 383 (6600), 495-501. del Giorgio, P.A., Cole, J.J., 2000. Bacterial growth energetics and growth efficiency. In: Microbial Ecology of the Oceans, ed. Kirchman, D.L., Wiley-Liss, New York, NY, USA. 289-325. Devassy, V.P., Bhattathiri, P.M., Qasim, S.Z., 1978. Trichodesmium phenomena. Ind. J. Mar. Sci. 7, 168-186. Dragovich, A., Kelly, Jr., J.A. 1968. Hydrological and biological characteristics of Florida's west coast tributaries. Fisheries Bulletin, 66(3), 463-477. Duce, R.A. 1986. The impact of atmospheric nitrogen, phosphorous, and iron species on marine biological productivity. In: P. Buat-Menard (eds.), The Role of Air-Sea Exchange in Geochemical Cycling. Reidel, Dordrecht, pp. 497-529.

PAGE 113

102 Duce, R.A., Liss, P.S., Merrill, J.T., Atlas, E.L., Buat-Menard, P., Hicks, B.B., Miller, J.M., Prospero, J.M., Arimoto, R., Church, T.M., Ellis, W., Galloway, J.N., Hansen, L., Jickells, T.D., Knap, A.H., Reinhardt, K.H., Schneider, B., Soudine, A., Tokos, J.J., Tsunogai, S., Wollast, R., Zhou, M., 1991. The atmospheric input of trace species to the ocean. Glob. Biogeochem. Cycles 5, 193-259. Eppley, R. W., 1972. Temperature and phytoplankton growth in the sea. Fishery Bulletin 70, 1063-1085. Falkowski, P.G., 1997. Photosynthesis: The paradox carbon dioxide efflux. Curr. Biol. 7 (10), R637-R639. Fennel, K., Spitz, Y.H., Letelier, R.M., Abbott, M.R., Karl, D.M., 2002. A deterministic model of N2 fixation at stn. ALOHA in the subtropical North Pacific Ocean. Deep-Sea Research II 49, 149-174. Froelich, P.N., Bender, M.L., Luedtke, N.A., Heath, G.R., Devries, T., 1982. The marine phosphorus cycle. Am. J. Sci. 282 (4), 474-511. Fung I.Y., Meyn S.K., Tegen I., Doney S.C., John J.G., Bishop J.K.B., 2000. Iron supply and demand in the upper ocean. Glob. Biogeochem. Cyc. 14 (1), 281-295. Genthon, C., 1992. Simulations of desert dust and sea salt aerosols in Antarctica with a general circulation model of the atmosphere. Tellus 44, 371-389. Glibert, P.M., Bronk, D.A., 1994. Release of dissolved organic nitrogen by marine diazotrophic cyanobacteria, Trichodesmium spp. Appl. Environ. Microbiol. 60, 3996-4000. Gregg, W.W., Carder, K.L., 1990. A simple spectral solar irradiance model for cloudless maritime atmospheres. Limnol. Oceanogr. 35 (8), 1657-1675. Goni, G., Johns, W., 2001. Census of Warm Rings and Eddies in the North Brazil Current Retroflection Region from 1992 through 1998 Using TOPEX/POSEIDON Altimeter Data. Geophys. Res. Let., 28, 1-4. Gordon, H.R., Brown, O.B., Evans, R.H., Brown, J.W., Smith, R.C., Baker, K.S., Clark, D.K., 1988. A semianalytical radiance model of ocean color. J. Geophys. Res. 93, 1090910924. Gordon R.M., Johnson K.S., Coale K.H., 1998. The behaviour of iron and other trace elements during the IronEx-I and PlumEx experiments in the Equatorial Pacific. Deep-Sea Research II 45 (6), 995-1041. He R.Y., Weisberg R.H., 2002a. West Florida shelf circulation and temperature budget for the 1999 spring transition. Cont. Shelf Res. 22, 719-748. He R.Y., Weisberg R.H., 2002b. Tides on the West Florida shelf. J. Phys. Ocean., 32 (12), 3455– 3473.

PAGE 114

103 He, R ., Weisberg, R.H., 2003. A loop current intrusion case study on the west Florida shelf. J. Phys. Ocean., 33 (2), 465-477. Hewson, I., Govil, S.R., Capone, D.G., Carpenter, E.J., Furman, J.A., 2004. Evidence of Trichodesmium viral lysis and potential significance for biogeochemical cycling in the oligotrophic ocean. Aquat. Microb. Ecol. 36, 1-8. Hong, H.S., Kester, D.R., 1986. Redox state of iron in the offshore waters of Peru. Limnol. Oceanogr. 31 (3), 512-524. Hood, R.R., Bates, N.R., Capone, D.G., Olson, D.B., 2001. Modeling the effect of nitrogen fixation on carbon and nitrogen fluxes at BATS. Deep-Sea Research II 48, 1609-1648. Huffman, G.J., Adler, R.F., Rudolf, B., Schneider, U., Keehn, P.R., 1995. Global Precipitation Estimates Based on a Technique for Combining Satellite-Based Estimates, Rain Gauge Analysis, and NWP Model Precipitation Information. J. Climate, 8, 1284-1295. Jickells, T.D., Spokes, L.J., 2001. Atmospheric Iron Inputs to the Oceans. In: Turner, D.R., Hunter, K., The Biogeochemistry of Iron in Seawater Wiley, Chistester, UK, pp. 85-121. Johnson, K.S., Coale, K.H., Elrod, V.A., Tindale, N.W., 1994. Iron photochemistry in seawater from the equatorial Pacific. Mar. Chem. 46 (4), 319-334. Johnson, K.S., Gordon, R.M., Coale, K.H., 1997. What controls dissolved iron concentrations in the world ocean? Mar. Chem. 57, 137-161. Karl, D.M., Letelier, R., Hebel, D.V., Bird, D.F., Winn, C.D., 1992. Trichodesmium blooms and new nitrogen in the North Pacific Gyre. In: Carpenter, E.J., Capone, D.A., Rueter, J.G., (Eds.), Marine Pelagic Cyanobacteria: Trichodesmium and other diazotrophs. Kluwer Academic Publishers, Netherlands, pp. 219-237. Karl, D.M., Letelier, R., Tupas, L., Dore, J., Christian, J., Hebel, D., 1997. The role of nitrogen fixation in biogeochemical cycling in the subtropical North Pacific Ocean. Nature 388, 533-538. Karl, D.M., Bjorkman, K.M., Dore, J.E., Fujieki, L., Hebel, D.V., Houlihan, T., Letelier, R.M., Tupas, L.M., 2001. Ecological nitrogen-to-phosphorus stoichiometry at station ALOHA. Deep-Sea Research II 48, 1529-1566. Kim, Y.S., Martin, D.F., 1974. Interrelationship of Peace River parameters as a basis of the iron index: a predictive guide to the Florida red tide. Wat. Res. 8, 607-616. King, J.E. 1950. A preliminary report on the plankton of the west coast of Florida. Quart. J. Fla. Acad. Sci. 12, 109-137. King, D.W., Lin, J., Kester, D.R., 1991. Determination of Fe(II) in seawater at nanomolar concentrations. Anal. Chim. Acta 247, 125-132. King, D.W., Aldrich, R.A., Charnecki, S.E., 1993. Photochemical redox cycling of iron in NaCl solutions. Mar. Chem. 44, 105-120.

PAGE 115

104 Kirk, J.T.O., 1994. Light and photosynthesis in aquatic systems. University Press, Cambridge, 509 pp. Kromkamp, J., Walsby, A.E., 1992. Buoyancy regulation and vertical migration of Trichodesmium : a computer model prediction. In: Carpenter, E.J., Capone, D.A., Rueter, J.G., (Eds.), Marine Pelagic Cyanobacteria: Trichodesmium and other diazotrophs. Kluwer Academic Publishers, Netherlands, pp. 239-248. Landing, W.M., Perry, J.J., Guentzel, J.L., Gill, G.A., Pollman, C.D., 1995. Relationships between the atmospheric deposition of trace elements, major ions, and mercury in Florida: the FAMS project (1992-1993). Water, Air, and Soil Poll. 80, 343-352. Lenes, J.M., Darrow, B.P., Cattrall, C., Heil, C.A., Callahan, M., Vargo, G.A., Byrne, R.H., Prospero, J.M., Bates, D.E., Fanning, K.A., Walsh, J.J., 2001. Iron fertilization and the Trichodesmium response on the West Florida shelf. Limnol. Oceanogr. 46 (6), 12611277. Lenes, J.M., 2002. Iron fertilization of Trichodesmium off the west coast of Barbados: A onedimensional numerical model. Master’s Thesis, Univ. of S. Florida, St. Petersburg, FL. Lenes, J.M., Walsh, J.J., Otis, D.B., Carder, K. L., 2005. Iron fertilization of Trichodesmium off the west coast of Barbados: A one-dimensional numerical model. Deep-Sea Res. I 52, 1021-1041. Lester, K.M., Merkt, R., Heil, C.A., Vargo, G.A., Neely, M.B., Spence, D.N., Melahn, L., Walsh, J.J., 2001. Evolution of a Gymnodinium breve (Gymnodiniales, Dinophyceae) red tide on the West Florida shelf: Relationship with organic N and P. In: Harmful Algal Blooms, ed. Hallegraeff, G.H., Blackburn, S.I., Bolch, C.J., Lewis, R.J., Hobart, Australia, pp. 161164. Liebig, J. 1840. Chemistry and its application to agriculture and physiology. Taylor and Walton, London. Martin, J.H., Gordon, R.M., 1988. Northeast Pacific iron distributions in relation to phytoplankton productivity. Deep-Sea Res. 35, 177-196. Martin, J.H., Fitzwater, S.E., 1988. Iron deficiency limits phytoplankton growth in the northeast Pacific subarctic. Nature 331, 341-343. Martin, J.H., Gordon, R.M., Fitzwater, S.E., 1991. The case for iron. Limnol. Oceanogr. 36 (8), 1793-1802. Masserini R.T., Fanning K.A., 2000. A sensor package for the simultaneous determination of nanomolar concentrations of nitrite, nitrate, and ammonia in seawater by fluorescence detection. Mar. Chem. 68 (4), 323-333. McCarthy, J.J., Carpenter, E.J., 1979. Oscillatoria (Trichodesmium) thiebautii (Cyanophyta) in the central North Atlantic Ocean. J. Phycol. 15, 75-82.

PAGE 116

105 Mellor, G.L., Yamada, T., 1982. Development of a turbulence closure model for geophysical fluid problems. Review of geophysical Space Physics 20, 851-875. Miller, W.L., 1990. An investigation of peroxide, iron, and iron bioavailability in irradiated marine waters. Ph.D. thesis. Univ. Rhode Island, Kingston, RI, 400 pp. Miller, W.L., King, D.W., Lin, J., Kester, D.R., 1995. Photochemical redox cycling of iron in coastal waters. Mar. Chem. 50, 63-77. Mills, M.M., Ridame, C., Davey, M., LaRoche, J., Geider, R.J., 2004. Iron and phosphorus colimit nitrogen fixation in the eastern tropical North Atlantic. Nature 429, 292-294. Moore, J.K., Doney, S.C., Kleypas, J.A., Glover, D.M., Fung, I.Y., 2002. An intermediate complexity marine ecosystem model for the global domain. Deep-Sea Research II 49, 403-462. Morel, A., Ahn, Y.-H., 1990. Optical efficiency of free-living marine bacteria: Influence of bacterioplankton upon the optical properties and particulate organic carbon in oceanic waters. J. Mar. res. 48, 145-175. Mulholland, M.R., Capone, D.G., 2000. The nitrogen physiology of the marine N2-fixing cyanobacteria Trichodesmium spp. Trends Plant Sci. 5, 148-153. Mulholland, M.R., Ohki, K., Capone, D.G., 2001. Nutrient controls on nitrogen uptake and metabolism by natural populations and cultures of Trichodesmium (Cyanobacteria). J. Phycol. 37, 1001-1009. Mullin, M.M., Stewart, E.F., Fuglister, F.J., 1975 Ingestion by planktonic grazers as a function of concentration of food. Limnol. Oceanogr. 20, 259-262. Nelson, J.R., Robertson, C.Y., 1993. Detrital spectral absorption: Laboratory studies of visible light effects on phytodetritus absorption, bacterial spectral signal, and comparison to field measurements. J. Mar. Res. 51, 181-207. O'Neil, J.M., Roman, M.R., 1992. Grazers and associated organisms of Trichodesmium In: Carpenter, E.J., Capone, D.A., Rueter, J.G., (Eds.), Marine Pelagic Cyanobacteria: Trichodesmium and other diazotrophs. Kluwer Academic Publishers, Netherlands, pp. 239-248. Orcutt, K.M., Lipschultz, F.J., Gundersen, K., Arimoto, R., Michaels, A.F., Knap, A.H., Gallon, J.R., 2001. A seasonal study of the significance of N2 fixation by Trichodesmium at the Bermuda Atlantic Time-series (BATS) site. Deep-Sea Research II 48, 1583-1608. Paerl, H.W., Prufert-Bebout, L.E., Guo, C., 1994. Iron-stimulated N2 fixation and growth in natural and cultured populations of the planktonic marine cyanobacterium Trichodesmium Appl. Envir. Microbiol. 60, 1044-1047. Peng, T.H., Takahashi, T., Broecker, W.S., Olafsson, J., 1987. Seasonal variability of carbon dioxide, nutrients and oxygen in the northern North Atlantic surface water:

PAGE 117

106 Obsevations and a model. Tellus, Ser. B 39, 439-458. Pope, R.M., Fry, E.S., 1997. Absorption spectrum (380-700 nm) of pure water. 2. Integrating cavity measurements. Appl. Optics 36 (33), 8710-8723. Proctor, L.M., Fuhrman, J.A., 1990. Viral mortality of marine bacteria and cyanobacteria. Nature 343, 60-62. Proctor, L.M., Fuhrman, J.A., 1991. Roles of viral-infection in organic particle-flux. Mar. Ecol.Prog. Ser. 69 (1-2), 133-142. Prospero, J.M., Nees, R.T., 1986. Impact of the North African drought and El Nino on mineral dust in Barbados trade winds. Nature 320, 735-738. Prospero, J.M., Nees, R.T., Uematsu, M., 1987. Deposition rate of particulate and dissolved aluminum derived from Saharan dust in precipitation at Miami, Florida. J. Geophys. Res. 92, 14,723-14,731. Prospero, J.M., 1999a. Long-range transport of mineral dust in the global atmosphere: Impact of African dust on the environment of the southeastern United States. P. Natl. Acad. Sci. USA 96 (7), 3396-3403. Prospero, J.M., 1999b. Long-term measurement of the transport of African mineral dust to the southeastern United States: Implications for regional air quality. J. Geophys. Res.104, 15,917-15,927. Prospero, J.M., Olmez, I., Ames, M., 2001. Al and Fe in PM 2.5 and PM 10 suspended particles in south-central Florida: The impact of long-range transport of African mineral dust. Water, Air, and Soil Poll. 125 (1-4), 291-317. Prufert-Bebout, L., Paerl, H.W., Lassen, C., 1993.Growth, nitrogen-fixation, and spectral attenuation in cultivated Trichodesmium species. Appl. Environ. Microbiol. 59, 13671375. Rich, H.W., Morel, F.M., 1990. Availability of well-defined colloids to the marine diatom Thalassiosira weissflogii Limnol. Oceanogr. 35, 652-662. Roesler, C.S., Perry, M.J., Carder, K.L., 1989. Modeling in situ phytoplankton absorption from total absorption spectra in productive inland marine waters. Limnol. Oceanogr. 34(8), 1510-1523. Rue, E.L., Bruland, K.W., 1995. Complexation of iron(III) by natural organic-ligands in the central north Pacific as determined by a new competitive ligand equilibration adsorptive cathodic stripping voltammetric method. Mar. Chem. 50 (1-4), 117-138. Rue, E.L., Bruland, K.W., 1997. The role of organic complexation on ambient iron chemistry in the equatorial Pacific Ocean and the response of a mesoscale iron addition experiment. Limnol. Oceanogr. 42 (5), 901-910. Rueter, J.G., 1988. Iron stimulation of photosynthesis and nitrogen fixation in Anabena 7120 and Trichodesmium (Cyanophyceae). J. Phycol. 24, 249-254.

PAGE 118

107 Rueter, J.G., Ohki, K., Fujita, Y., 1990. The effect of iron nutrition on photosynthesis and nitrogen fixation in cultures of Trichodesmium (Cyanophyceae). J. Phycol. 26, 30-35. Rueter, J.G., Hutchins, D.A., Smith, R.W., Unsworth, N.L., 1992. Iron nutrition of Trichodesmium In: Carpenter, E.J., Capone, D.A., Rueter, J.G., (Eds.), Marine Pelagic Cyanobacteria: Trichodesmium and other diazotrophs. Kluwer Academic Publishers, Netherlands, pp. 289-306. Sanudo-Wilhelmy, S.A., Kustka, A.B., Gobler, C.J., Hutchins, D.A., Yang, M., Lwiza, K., Burns, J., Capone, D.G., Carpenter, E.J., 2001. Phosphorus limitation of nitrogen fixation by Trichodesmium in the central Atlantic Ocean. Nature 411, 66-69. Schtz, L., Jaenicke, R., Petrer, H., 1981. Saharan dust transport over the North Atlantic Ocean. In: T. L. Pw (ed.), Desert Dust: Origin, Characteristics, and Effects on Man. Geol. Soc. Amer., Boulder. Sellner, K.G., 1997. Physiology, ecology, and toxic properties of marine cyanobacteria blooms. Limnol. Oceanogr. 42, 1089-1104. Sheridan, C.C., Steinberg, D.K., Kling, G.W., 2002. The microbial and metazoan community associated with colonies of Trichodesmium spp.: a quantitative survey. J. Plank. Res. 24, 913-922. Smith, R.C., Baker, K.S., 1981. Optical properties of the clearest natural waters (200-800 nm). Applied Optics 20, 177-184. Steidinger, K.A., G.A. Vargo, P.A. Tester and C.R. Tomas. 1998. Bloom dynamics and physiology of Gymnodinium breve with emphasis on the Gulf of Mexico. In: D.M. Anderson, A.D. Cembella and G.H. Hallegraeff (eds.), Physiological ecology of harmful algal blooms. Springer Verlag, Berlin pp. 135-153. Stevens, D.M., Brooks, A.L., Moore, E.A., 1970. Primary and secondary production in the tropical Atlantic. ONR report contract N00014-67-A-0432-0001. Bermuda Biol. Stn. Spec. Publ., 124 p. Stihl, A., Sommer, U., Post, A.F., 2001. Alkaline phosphatase activities among populations of the colony-forming diazotrophic cyanobacterium Trichodesmium spp. (cyanobacteria) in the Red Sea. J. Phycol. 37 (2), 310-317. Subramaniam, A., Carpenter, E.J., Falkowski, P.G., 1999. Bio-optical properties of the marine diazotrophic cyanobacteria Trichodesmium spp. I. A reflectance model for remote sensing. Limnol. Oceanogr. 44, 618-627. Sunda, W.G., Huntsman, S.A., 1995. Iron uptake and growth limitation in oceanic and coastal phytoplankton. Mar. Chem. 50, 189-206. Tegen, I., Fung, I., 1994. Modeling of mineral dust in the atmosphere – Sources, transport, and optical thickness. J. Geophys. Res. 99 (D11), 22897-22914.

PAGE 119

108 Turner, J.T., Tester, P.A., 1997. Toxic marine phytoplankton, zooplankton grazers, and pelagic food webs. Limnol. Oceanogr. 42 (5), 1203-1214. U.S. Department of Agriculture, 1975. Soil Taxonomy, Agr. Handb. Soil Conservation Service, 436 pp. U.S.G.S. 1976-1981. Water resources data for Florida. U.S. Geological Survey water data reports. Vargo, G.A., Carder, K.L., Gregg, W., Shanley, E., Heil, C., Steidinger, K.A., Haddad, K., 1987. The potential contribution of primary production of red tides to the West Florida shelf ecosystem. 32 (3), 762-767. Vargo, G.A., Heil, C.A., Spence, D., Neely, M.B., Merkt, R., Lester, K., Weisberg, R.H., Walsh, J.J., Fanning, K., 2001. The hydrographic regime, nutrient requirements, and transport of a Gymnodinium breve Davis red tide on the West Florida shelf. In: Harmful Algal Blooms, ed. Hallegraeff, G.H., Blackburn, S.I., Bolch, C.J., Lewis, R.J., Hobart, Australia, pp. 157-160. Vargo, G.A., Heil, C.A., Ault, D.N., Neely, M.B., Murasko, S., Havens, J. Lester, K.M., Dixon, K., Merkt, R., Walsh, J.J., Weisberg, R.H., Steidinger, K.A. 2004. Four Karenia brevis blooms: a comparative analysis. Steidinger, K.A., Landsberg, J.H., Tomas, C.R., Vargo, G.A. (Eds.). Harmful Algae 2002. Florida Fish and Wildlife Conservation Commission, Florida Institute of Oceanography, and Intergovernmental Oceanographic Commission of UNESCO. Villareal, T.A., Carpenter, E.J., 1990. Diel buoyancy regulation in the marine diazotrophic cyanobacterium Trichodesmium thiebautii. Limnol. Oceanogr. 35, 1832-1837. Walsby, A.E., 1978. The properties and buoyancy providing role of gas vesicles in Trichodesmium ehrenberg Br. Phycol. J. 12, 103-116. Walsby, A.E., 1992. The gas vesicles and buoyancy of Trichodesmium In: Carpenter, E.J., Capone, D.A., Rueter, J.G., (Eds.), Marine Pelagic Cyanobacteria: Trichodesmium and other diazotrophs. Kluwer Academic Publishers, Netherlands, pp. 141-161. Walsh, J.J., Dieterle, D.A., 1994. CO2 cycling in the coastal ocean. I. A numerical analysis of the southeastern Bering Sea, with applications to the Chukchi Sea and the northern Gulf of Mexico. Progr. Oceanogr. 34, 335-392. Walsh, J.J., 1996. Nitrogen-fixation within a tropical upwelling ecosystem: Evidence for a Redfield budget of carbon/nitrogen cycling by the total phytoplankton community. J. Geophys. Res. 101, 20,607-20,616. Walsh, J.J., Steidinger, K.A., 2001. Saharan dust and Florida red tides: the cyanophyte connection. J. Geophys. Res.-Oceans 106, 11,597-11,612. Walsh, J.J., Weisberg, R.H., Dieterle, D.A., He, R., Darrow, B.P., Jolliff, J.K., Lester, K.M., Vargo, G.A., Kirkpatrick, G.J., Fanning, K.A., Sutton, T.T., Jochens, A.E., Biggs, D.C.,

PAGE 120

109 Nababan, B., Hu, C., Muller-Karger, F.E., 2003. A phytoplankton response to intrusions of slope water on the West Florida shelf: models and observations. J. Geophys. Res. 108: 3190. Walsh, J.J., Kirkpatrick, G.J., Darrow, B.P., Vargo, G.A., Fanning, K.A., Peebles, E.B., Heil, C.A., Steidinger, K.A., 2005. Fish farming by Karenia brevis : a nutrient source for large red tides. Cont. Shelf Res. (in press). Walsh, J.J., Jolliff, J.K., Darrow, B.P., Lenes, J.M., Milroy, S.P., Dieterle, D.A., Carder, K.L., Chen, F.R., Vargo, G.A., Weisberg, R.H., Fanning, K.A., Muller-Karger, F.E., Steidinger, K.A., Heil, C.A., Tomas, C.R., Prospero, J.S., Lee, T.N., Kirkpartick, G.J., Whitledge, T.E., Stockwell, D.A., Villareal., T.A., Jochens, A.E., Bontempi, P.S., 2006. Red tides: where, when, why? J. Geophys. Res. (in review). Weisberg, R.H., Black, B.D., Yang, H., 1996. Seasonal modulation of the west Florida continental shelf circulation. Geophys. Res. Let. 23, 2247-2250. Weisberg, R.H., He, R ., 2003. Local and deep-ocean forcing contributions to anomalous water properties on the West Florida Shelf. J. Geophys. Res. 108 (C6), 3184. Weiss, R.F., 1970. The Solubility of Nitrogen, Oxygen and Argon in Water and Seawater. DeepSea Res. 17, 721-735. Wells, M.L., Zorkin, N.G., Lewis, A.G., 1983. The role of colloidal chemistry in providing a source of iron to phytoplankton. J. Mar. res. 41, 731-746. Wells, M.L., Mayer, L.M., 1991. The photoconversion of colloidal iron oxyhydroxides in seawater. Deep-Sea Res. 38, 1379-1395. Yentsch, C.M., Yentsch, C.S., Perras, J.P. 1972. Alkaline phosphatase activity in the tropical marine blue-green alga Oscillatoria erythaea (" Trichodesmium "). Limnol. Oceanogr. 17, 772-774. Young, R.W., Carder, K.L., Betzer, P.R., Costello, D.K., Duce, R.A., Ditullio, G.R., Tindale, N.W., Laws, E.A., Uematsu, M., Merrill, J.T., Feely, R.A., 1991. Atmospheric iron inputs and primary productivity: phytoplankton responses in the North Pacific. Glob. Biochem. Cycles. 5, 119-134. Zhu, X.R., Prospero, J.M., Millero, F.J., Savoie, D.L., Brass, G.W., 1992. The solubility of ferric ion in marine mineral aerosol solutions at ambient relative humidities. Mar. Chem. 38, 91-107. Zhu, X.R., Prospero, J.M., Millero, F.J., 1997. Diel variability of soluble Fe(II) and soluble total Fe in North African dust in the trade winds at Barbados. J. Geophys. Res. 102, 21,29721,306.

PAGE 121

About The Author Jason Michael Lenes received a Bachelor’s Degree in Biology from Bucknell University in 1996 and a M.S. in Marine Science from University of South Florida in 2002. His research interests have varied from the population genetics of wolf spiders to ecosystem modeling of toxic algae, while his specialization consists of the biogeochemical response to atmospheric iron. While in the Marine Science Ph.D. program at University of South Florida, Mr. Lenes received several awards and fellowships including: Wachovia Endowed Fellowship in Marine Science and NASA Earth System Science (ESS) Fellowship. He has also coauthored seven peerreviewed publications, two of which he was lead author. Several more publications are in preparation from his dissertation. Mr. Lenes received international press from his first lead authorship published in Limnology and Oceanography, 2001.


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 001789847
003 fts
005 20070608110726.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 070608s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001476
040
FHM
c FHM
035
(OCoLC)141189179
049
FHMM
090
GC11.2 (ONLINE)
1 100
Lenes, Jason M.
0 245
Saharan dust and phosphatic fidelity :
b a three dimensional biogeochemical model of Trichodesmium on the West Florida shelf
h [electronic resource] /
by Jason M. Lenes.
260
[Tampa, Fla] :
University of South Florida,
2006.
3 520
ABSTRACT: The availability of iron within the surface waters of the broad, oligotrophic West Florida shelf (WFS) controls periodic blooms of the pelagic marine cyanobacterium Trichodesmium. Summer delivery of iron (Fe), in the form of Saharan dust, alleviates this growth constraint, shifting limitation to the efficiency of phosphorus (P) cycles. Florida's rivers drain Miocene phosphorus deposits to supply the WFS with freshwater nutrient supplies at molar dissolved inorganic nitrogen/phosphate (DIN/PO4) ratios of less than 6. These diazotrophs draw upon ubiquitous stocks of dissolved nitrogen gas, once stimulated by Fe-deposition within P-replete waters of the West Florida shelf. An extensive in situ data set collected between 1998-2001 (NEGOM / ECOHAB / HyCODE) provided plankton taxonomy, hydrographic, nutrient, DOM, pigment, and optical properties on the shelf. A three-dimensional numerical model was constructed to analyze the impact of iron fertilization of the diazotroph Trichodesmium and the resultant effect upon the elemental cycles of N, P, and Fe. Based on the results of the coupled physical and ecological models, wet deposition of Fe-rich Saharan dust was necessary to stimulate enough nitrogen fixation to support the toxic red tide (Karenia brevis) of ~20 micrograms chl per liter found in October 1999. Ultimately, the magnitude and longevity of the Trichodesmium population, and therefore 'new' nitrogen production, was controlled by both phosphorus and iron availability.
502
Dissertation (Ph.D.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic dissertation) 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 109 pages.
Includes vita.
590
Adviser: John J. Walsh, Ph.D.
653
Nitrogen fixation.
Phosphorus.
Iron.
Ecosystem.
Modeling.
690
Dissertations, Academic
z USF
x Marine Science
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1476