USF Libraries
USF Digital Collections

Effects of nutrients from the water column on the growth of benthic microalgae in permeable sediments

MISSING IMAGE

Material Information

Title:
Effects of nutrients from the water column on the growth of benthic microalgae in permeable sediments
Physical Description:
Book
Language:
English
Creator:
Darrow, Brian P
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Marine ecology
Diatoms
Numerical modeling
Nutrient cycles
Phytoplankton
Dissertations, Academic -- Marine Science -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: In some continental shelf sediments integrated benthic microalgal biomass is greater than the integrated phytoplankton biomass in the overlying water column. In addition, benthic microalgae may account for up to 10% of the primary production responsible for the coastal fishery yield of the eastern United States. A three-dimensional model of the eastern Gulf of Mexico examines the effects of water-column nutrient sources on the growth of benthic microalgae. To parameterize the exchange of nutrients across the sediment/water interface in these permeable sediments, a non-local exchange submodel was constructed and tested within the framework of the model's grid. Based on the results of the three dimensional simulations, the growth of benthic microalgae from water-column nutrients is highly dependent on the light limitation of overlying phytoplankton. When light is available to phytoplankton in high enough quantities, water-column nutrients are used up before reaching the sediments. When the overlying phytoplankton are light limited, nutrients are able to reach the sediments where the shade adapted benthic microalgae can grow.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Brian P. Darrow.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 118 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 - 001994148
oclc - 317403110
usfldc doi - E14-SFE0002431
usfldc handle - e14.2431
System ID:
SFS0026748:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001994148
003 fts
005 20090330124103.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090330s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002431
035
(OCoLC)317403110
040
FHM
c FHM
049
FHMM
090
GC11.2 (Online)
1 100
Darrow, Brian P.
0 245
Effects of nutrients from the water column on the growth of benthic microalgae in permeable sediments
h [electronic resource] /
by Brian P. Darrow.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 118 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
3 520
ABSTRACT: In some continental shelf sediments integrated benthic microalgal biomass is greater than the integrated phytoplankton biomass in the overlying water column. In addition, benthic microalgae may account for up to 10% of the primary production responsible for the coastal fishery yield of the eastern United States. A three-dimensional model of the eastern Gulf of Mexico examines the effects of water-column nutrient sources on the growth of benthic microalgae. To parameterize the exchange of nutrients across the sediment/water interface in these permeable sediments, a non-local exchange submodel was constructed and tested within the framework of the model's grid. Based on the results of the three dimensional simulations, the growth of benthic microalgae from water-column nutrients is highly dependent on the light limitation of overlying phytoplankton. When light is available to phytoplankton in high enough quantities, water-column nutrients are used up before reaching the sediments. When the overlying phytoplankton are light limited, nutrients are able to reach the sediments where the shade adapted benthic microalgae can grow.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: John J. Walsh, Ph.D.
653
Marine ecology
Diatoms
Numerical modeling
Nutrient cycles
Phytoplankton
690
Dissertations, Academic
z USF
x Marine Science
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2431



PAGE 1

Effects of Nutrients From the Water Column on the Growth of Benthic Microalgae in Permeable Sediments by Brian P. Darrow A dissertation submitted in partial fulfillment of the requirement s 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. Kent A. Fanning, Ph.D. Robert H. Weisberg, Ph.D. Richard A. Jahnke, Ph.D. Date of Approval: November 12, 2007 Keywords: Marine Ecology, Diatoms, Nume rical Modeling, Nutrient Cycles, Phytoplankton Copyright 2008, Brian P. Darrow

PAGE 2

This work is dedicated to my wife Raffi and my two daughters, Alice and Wendy. I cannot think of three better women with whom to share my life. I just finished my dissertation. Im going to Disney World.

PAGE 3

Acknowledgments This work was funded by various grants to John Walsh, Kent Fanning, Bob Weisberg and Gabe Vargo from the National Aeronautics and Space Administration, the National Science Foundation, and the Office of Naval Research. Additional funding was also pr ovided by research fellowships from the Tampa Bay Parrotheads in Paradise, the Sanibel Captiva Shell Club, St. Petersburg Progress and the Unit ed States Geological Survey. Id like to thank my major advisor, J ohn Walsh for his challenging tutelage, stimulating conversations and fatherly adv ice over the last 9 years. I am indebted to my committee members, Bob Weisberg, Kent Fanning, Gabe Vargo, and Rick Jahnke for suggestions that have greatly improved this manuscript as well as my graduate school career. This work would not have been possible without the technical assistance of Dwight Dieterle. His patient guidance has been invaluable in discovering silly coding errors and getting things to work throughout my graduate school career. Finally, Id like to thank Jason Lenes for his friendship and inspiration. Many creative solutions to complex problems were hatched while shooting the breeze on the third floor landing of the Kn ight Oceanographic Research Center.

PAGE 4

Table of Contents List of Tables iii List of Figures iv Abstract x Introduction 1 Methods 24 1. Physical Model 24 1.1. Pelagic Transport 24 1.2. Benthic Transport 25 1.3. Optics 35 2. Biological Model 37 2.1. Primary Producers 37 2.2. Secondary Producers 39 2.3. Microbial Loop 40 3. Nutrient Model 42 3.1. Non-living particles 42 3.2. Dissolved Organic Matter 45 3.3. Dissolved Inorganic Matter 49 4. Boundary Conditions 52 i

PAGE 5

Results 64 1. Case 1 64 1.1 Spring 1998 62 1.2 Summer 1998 82 2. Case 2 87 Discussion 102 Conclusions 108 References 110 About the Author End Page ii

PAGE 6

List of Tables Table 1. The average benthic chlorophyll stocks int egrated over the top 0.5 cm on the West Flor ida shelf during 2000-2001 18 Table 2. The depth dependent water-column conditions applied to the open boundaries over the entire m odel run, and applied to the entire model grid in the initial time step 55 Table 3. The depth based initial conditions applied to the sediments over the entire model grid 58 Table 4. Model parameters 61 Table 5. Comparison of the model's si mulated chlorophyll stocks during the two cases of the model 100 iii

PAGE 7

List of Figures Figure 1. The surface distributions of A) NO 3 + NO 2 and B) PO 4 off the coasts of Louisiana and Alabama during the May 1999 NEGOM cruise 5 Figure 2. The surface distri butions of A) SiO 4 abd B) salinity off the coast of Louisiana and Alabama dur ing the May 1999 NEGOM cruise 6 Figure 3. A) August 1999 surface phosphat e concentrations in the eastern Gulf of Mexico during the NEGOM a nd ECOHAB cruises and B) the flow of the Peace river near C harlotte Harbour 7 Figure 4. The near-bottom A) chlorophyll and B) NO 2 + NO 3 observations during the NEGOM cruise May 4-15 1998 9 Figure 5. The near-bottom A) PO 4 and B) SiO 4 observations during the NEGOM cruise May 4-15 1998 10 Figure 6. The near-bottom observations of A) chlorophyll and B) NO 2 + NO 3 during the May 1999 NEGOM and EC OHAB cruises 11 Figure 7. The near-bottom observations of A) PO 4 and B) SiO 4 during the May 1999 NEGOM and ECOHAB cruises 12 iv

PAGE 8

Figure 8. A) The land and marine extent of the near-surface phosphate rich Miocene strata based on well cores and sediment phosphorite percentrages relative to the 20m and 200 m isobaths, the NEGOM, ECOHAB, and ECOHAB Middle Grounds stations in the eastern Gulf of mexico. B) The model grid 13 Figure 9. The observed A) percentage of surface PAR and B) total PAR reaching the ocean floor during the NEGOM cruise July 25-August 9, 1998 19 Figure 10. Observed benthic chlorophyll stocks integrated over the top 0.5 cm of sediment in A) Ju ly 1992, and B) Ocober 1992 during the Coastal Production cruises 20 Figure 11. Observed benthic chlorophyll sto cks integrated over the top 0.5 cm of sediment in A) April 1993 and B) August 1993 during the Coastal Production cruises 21 Figure 12. Geographical locations of A) site 1, site 2, and site 3 of the tracer case of the model, and the B) flow at those locations over time 29 Figure 13. Tracer concentrations in the near-bottom water during 17 March 1998 (day 18) of t he NLE tracer case of the model 30 Figure 14. Sediment profiles of tracer at st ations 1, 2, and 3 during 17 March 1998 (day 18) of A) the diffusion only tracer case and B) the NLE tracer case of t he model 31 v

PAGE 9

Figure 15. A) Concentration of tracer in the near-bottom water and B) sediment profiles of tracer at stations 1,2, and 3 during 22 April 1998 (day 54) of t he NLE tracer case of the model 33 Figure 16. A) Concentration of tracer in the near-bottom water and B) sediment profiles of tracer at stations 1,2, and 3 during 28 May 1998 (day 90) of the NLE tracer ca se of the model 34 Figure 17. The simulated near-bottom A) NO 3 and B) NH 4 during 9 May 1998 in the standard case of the model 65 Figure 18. The simulated near-bottom A) PO 4 and B) SiO 4 during 9 May 1998 in the standard case of the model 66 Figure 19. The simulated concentrations of A) NO 3 and B) NH 4 in the surface sediment layer during 9 May 1998 in case 1 of the model 68 Figure 20. The simulated concentrations of A) PO 4 and B) SiO 4 in the surface sediment layer during 9 May 1998 in the case 1 of the model 69 Figure 21. The simulated flux of A) tota l inorganic nitrogen and B) inorganic phosphorus across the sediment water interface during 9 May 1998 in the case 1 of the model 70 Figure 22. The A) simulated and B) observ ed surface chlorophyll concentrations in the eastern Gulf of Me xico during 9 May, 1998 72 Figure 23. The A) simulated and B) observ ed near-bottom chlorophyll concentrations in the eastern Gulf of Me xico during 9 May, 1998 73 Figure 24. The simulated limiting factors in the growth of near-bottom A) diatoms and B) flagellates during 9 May, 1998 74 vi

PAGE 10

Figure 25. The simulated A) benthic chlorophy ll stocks integrated over the top 2 mm of sediment on 9 May, 1998 and B) their limiting factors 76 Figure 26. The simulated growth rate and limiting factors affecting the growth rate of benthic diatom s in the Florida Middle Grounds during 9 May, 1998 77 Figure 27. Observations of A) NO3 and B) NH4 in the near-bottom waters of the eastern Gulf of Me xico during August 1998 from the NEGOM and ECOHAB Florida pr ograms 78 Figure 28. Observations of A) PO4 and B) SiO4 in the near-bottom waters of the eastern Gulf of Me xico during August 1998 from the NEGOM and ECOHAB Florida pr ograms 79 Figure 29. Simulated near-bottom concentrations of A) NO 3 and B) NH 4 during 7 August, 1998 80 Figure 30. Simulated nar-bottom concentrations of A) PO 4 and B) SiO 4 during 7 August, 1998 81 Figure 31. Simulated porewater concentrations of A) NO 3 and B) NH 4 during 7 August, 1998 83 Figure 32. Simulated porewater concentrations of A) PO 4 and B) SiO 4 during 7 August, 1998 84 Figure 33. The simulated flux of A) dissolved inorganic nitrogen and B) dissolved inorganic phosphorus across the sediment water interface during 7 August, 1998 85 vii

PAGE 11

Figure 34. Cumulative surface chlorophyll concentrations A) from diatoms and flagellates si mulated in the standard case of the model and B) observed during the NEGOM and ECOHAB cruises of August 1998 88 Figure 35. Cumulative near-bottom chlorophyll concentrations A) from diatoms and flagellates simu lated in case 1 of the model and B) observed during t he NEGOM and ECOHAB cruises of August 1998 89 Figure 36. The simulated factors limiting the growth of A) diatoms and B) flagellates in the near-bottom waters of case 1 of the model during 7 August, 1998 90 Figure 37. Simulated A) benthic chlorophyll stocks integrated over the top 0.5 cm of sediment during 7 August, 1998 and B) the factors limiting their gr owth 91 Figure 38. Simulated total surface chloroph yll from diatoms and flagellates during 1 April, 1998 in A) case 2 and B) case 1 of the model 93 Figure 39. Simulated total surface chloroph yll from diatoms and flagellates during 15 April, 1998 in the A) case 2 and B) case 1 of the model 94 Figure 40. Simulated near-bottom chlorophyll from diatoms and flagella tes during 15 April 1998 in A) ca se 2 and B) case 1 of the model 95 viii

PAGE 12

Figure 41. Simulated total near-bottom chlorophyll from diatoms and flagellates during 23 April 1998 in A) case 2 and B) case 1 of the model 96 Figure 42. Simulated phytoplankton dominance in the near-bottom waters of the West Florida Shel f during 23 April 1998 in the A) case 2 and B) case 1 of t he model 97 Figure 43. The simulated factors limiting the growth of A) diatoms and B) flagellates in the near-bottom waters of the West Florida Shelf during 23 April, 1998 in case 2 of the model 98 Figure 44. The simulated benthic chlorophy ll stocks integrated over the top 2 mm of sediment during 23 April, 1998 in A) case 2 and B) case 1 of the model 99 ix

PAGE 13

Effects of Nutrients from the Water Column on the Growth of Benthic Microalgae in Permeable Sediments Brian P. Darrow ABSTRACT In some continental shelf sediment s integrated benthic microalgal biomass is greater than the integr ated phytoplankton biomass in the overlying water column. In addition, benthic microalgae may account for up to 10% of the primary production responsible for the coas tal fishery yield of the eastern United States. A three-dimensional model of the easte rn Gulf of Mexico examines the effects of water-column nutrient sources on the growth of benth ic microalgae. To parameterize the exchange of nutrients across the sediment/water interface in these permeable sediments, a non-loca l exchange submodel was constructed and tested within the framewor k of the models grid. Based on the results of the three dim ensional simulations, the growth of benthic microalgae from water-column nutrients is highly dependent on the light limitation of overlying phyt oplankton. When light is av ailable to phytoplankton in high enough quantities, wate r-column nutrients are used up before reaching the x

PAGE 14

sediments. When the overlying phytoplankton are light limited, nutrients are able to reach the sediments where the shade adapted benthic microalgae can grow. xi

PAGE 15

Introduction The principle of competitive excl usion states that no two species competing for exactly the same res ources can stably coexist. The apparent violation of this principle by the many species of freshwater and marine phytoplankton has been the impetus for countless experiment s, discussion and research papers in the field of ecology since George Evelyn Hutchinson (1961) first articulated the paradox of the plankton. Hutchinson realized that a body of water is not as homogeneous as it seems on the surface. Physical, chemic al and biological conditions change on every possible time scale. This conti nual change means that no one species can be adapted to the conditions at a single point in space at all times. Likewise, conditions at any spatial point in a body of water are likely to be different from those at other spatial points at any given ti me. If the water is in constant motion, as bodies of water tend to be, conditi ons at one space may move to another space, or mix with conditions at other spaces creating completely new conditions. It is in this very heterogeneous and cont inually changing envir onment that many species of phytoplankton coexist. 1

PAGE 16

Every species of phytoplankton uses the suns energy through photosynthesis to convert inorganic car bon, to organic carbon. Phytoplankton are a significant sink of CO 2 because of the vast size of the Earths oceans relative to its landmass. It may, therefore, seem that all phytoplankton are one functional group with the purpose of keepi ng Earths carbon budget in balance. In reality, each phytoplankton spec ies plays a role in defining the environmental conditions in which it thri ves. Some species support productive fisheries while others kill fish. Some species rapidly sink to the depths, transporting their carbon away from t he surface layer, while others become buoyant when they die. Clearly, the success of one species of phytoplankton over another in a given environment can have serious social, environmental, economic and political impacts. Such wide ranging impacts have recently led to extensive studies of the factors that lead to the success of the harmful dinoflagellate, Karenia brevis in the Gulf of Mexico. Blooms of Karenia brevis occur regularly off the west coast of Florida on the shallow West Florida shelf. The first documentation of such a bloom was recorded in 1884 (Walker, 1884), and the causative organism, then known as Gymnodinium brevis was first identified in 1948 (Davis, 1948). Throughout time, several causes have been suggested, ranging from volcanic heating to agricultural pollution. Recent studies have shown that the causes of the toxic blooms are complex, such that a very specific succession of conditions is required for a large bloom to develop and persist (Walsh et al., 2006). 2

PAGE 17

An important condition for the initiation of Karenia brevis blooms is a phosphorus-rich nutrient source (Walsh et al, 2006). Such a nutrient source favors the initiation of nitrogen fixing organisms, which then undergo cell lysis, releasing their nitrogen to the water-c olumn for use by other species, like Karenia brevis For this reason, the West Florida shelf is an ideal location for Karenia brevis blooms because of a rich source of fossil phosphorus near the surface of the landmass and underlying the continental shelf. The eastern Gulf of Mexico comp rises a complex nutrient habitat for algae. The region is dominated by t he West Florida shelf, which extends approximately 700 km along the west coast of Florida and ranges 100-150 km in width. In the northeastern portion of the Gulf of Mexico, the Louisiana, Mississippi and Alabama shelves are narrowe r, ranging from 56 km in width near DeSoto canyon to about 10 km off the Mi ssissippi River delta. The Mississippi River, itself, encompasses the worlds second largest drainage basin, draining approximately 40% of the continental United States, with an average discharge of over 18,500 m 3 s -1 to the Gulf of Mexico (Mor gan & Dale, 2007). In total, approximately 25 other rivers discharge to the Gulf of Mexico between the Mississippi river delta and the Florida Keys. Most of these are in Florida and drain relatively small, coastal regi ons surrounding the Gulf (Nordlie, 1990). Despite such riverine discharge, nutri ent concentrations in the eastern Gulf of Mexico are gener ally low, often below detection levels (Masserini and Fanning, 2000), except in the most near shore areas. There are, however, 3

PAGE 18

several seasonal and regional features, which help to bring about the ecological complexity of the region. The most obv ious of these is the Mississippi River. In September 1991, near-surface nitrat e concentrations greater than 30 M were measured in the low salinity pl ume of the Mississippi River above the Louisiana shelf. Silicate concentrations greater than 40 M with phosphate concentrations larger than 1 M (Smith and Hitchcock, 1994) were also observed. Relatively high nutrient c oncentrations were again measured during May 1999 (Figs. 1,2) within a similar low salinity plume. Like other years (Gilbes et al., 1996; Muller-Karger et al. 1991) the May 1999 low salinity plume was carried southeastward by oceanic current s to the West Florida Shelf, thus impacting a large area of the Gulf. The Mobile, Apalachicola and Suwannee rivers (Del Castillo et al. 2000) have been observed to similarly interact with oceanic currents (Gilbes et al. 1996), episodically produci ng low salinity, nutrient rich plumes that may deliver nutrients to the continental shelf directly or through the degradation of algal biomass and gr azer byproducts within the plume. In addition to the intermittent peaks in discharge of the various rivers, further complexity is added because of r egional differences in characteristics of riverine waters. Each low salinity plum e associated with a river flowing into the eastern Gulf of Mexico exhi bits a different relationship between dissolved organic matter (DOM) and salinity (Conmy and C oble, 2002). Furthermore, enhanced surface phosphate concentrations (Fig. 3) in the southern portion of the shelf are likely the result of the seasonal outflow of southern rivers (Nordlie, 1990) draining the regions of the shallow Bone Valle y and Hawthorn phos phate deposits. 4

PAGE 19

Figure 1. The surface distribution of (A) NO 3 +NO 2 and (B) PO 4 off the coast of Louisiana during the May 1999 NEGOM cruise. 5

PAGE 20

Figure 2. The surface distribution of (A) SiO 4 and (B) salinity off the coast of Louisiana during the May 1999 NEGOM cruise. 6

PAGE 21

Figure 3. A) August 1999 surface phosphate concentrations (mol/kg) in the eastern Gulf of Mexico during the NE GOM () and ECOHAB(+) cruises and B) the flow of the Peace river near Charlotte Harbour. 7

PAGE 22

Another source of nutrients to the euphot ic zone of the eastern Gulf of Mexico is the introduction of slope water due to upwelling events induced either by the intrusion of the Loop Current onto the continent al shelf, and/or by local wind forcing. This phenomenon was obser ved in April 1982 (Paluskiewicz, et al ., 1983) during a Loop Current frontal eddy in trusion onto the West Florida shelf where the 1.0 mol kg -1 isopleth of nitrate+nitrite intersected the 60 m isobath. Similar observations were made in 1998 (Walsh et al. 2003) after Loop Current water intruded (He and Weisberg, 2003) onto the outer shelf, coincident with upwelling favorable winds. The 1.0 mol kg -1 isopleth of nitrate+nitrite and the 3.0 mol kg -1 isopleth of silicate reached the 20 m isobath over much of the shelf (Figs. 4,5), in contrast to 1999 (Figs. 6 and 7), when the prevailing winds were not favorable for upwelling, and these is opleths were located in much deeper waters. The Miocene facies of the Hawthor ne phosphate deposit (Fig. 8) extend out onto the continental shelf approxima tely between Tarpon Springs, FL and Port Charlotte, FL (Ryder, 1985; Sinclair et al. 1985), where they lie within a few meters of the sediment surface (Birds all, 1977). Sediments off St. Petersburg Beach, FL have 3.4% phosphorite, while those off Venice, FL contain as much as 4.6% phosphorite (Bates, 1963). Su ch phosphorite percentages have been found in the sediments on the West Florida shelf past the 18 m isobath, with no significant amounts being found beyond appr oximately the 30 m isobath, thus providing a potential additional source of phosphate, to the shelf in that region. 8

PAGE 23

Figure 4. Near-bottom A) chlorophyll ( g l -1 ) and B) NO 2 + NO 3 ( mol kg -1 ) observations during the NEGOM ( ) cruise May 4-15 1998. ECOHAB cruise stations (+) are included for reference, but there were no ECOHAB observations during May 1998. 9

PAGE 24

Figure 5. Near-bottom A) PO 4 ( mol kg -1 ) and B) SiO 4 ( mol kg -1 ) observations during the NEGOM ( ) cruise May 4-15 1998. ECOHAB cruise stations (+) are included for reference, but there we re no ECOHAB observations during May 1998. 10

PAGE 25

Figure 6. Near-bottom observations of A) chlorophyll ( g l -1 ) and B) N0 2 +NO 3 ( mol kg -1 ) during the May 1999 NEGOM ( ) and ECOHAB (+) cruises. 11

PAGE 26

Figure 7. Near-bottom observations of A)PO 4 ( mol kg -1 ) and B)SiO 4 ( mol kg -1 ) during the May 1999 NEGOM ( ) and ECOHAB (+) cruises. 12

PAGE 27

Figure 8. A) The land (shaded) and marine ( ) extent of near-surface phosphate rich Miocene strata based on well cores (Bates, 1963) and sediment phosphorite percentages (Birdsall, 1977) relative to the 20 m and 200 m isobaths, the NEGOM (), ECOHAB(+), and ECOHAB Middle Grounds () stations in the eastern Gulf of Mexico. B) The model grid. 13

PAGE 28

Increased phosphorus, in relation to nitrogen, may thus be delivered to the euphotic zone in the eastern Gulf of Mexi co in a number of ways. Phosphorus rich waters may be upwelled from outsi de the euphotic zone in the deep water of the Gulf of Mexico. Rivers, eroding the fossil phosphorus deposits of peninsular Florida may deliver phosphate in greater quantities, relative to nitrogen and silica, to the continental shelf than rivers al ong the northern Gulf coast. Likewise, porewaters overlying the underwater phosphate deposits may be a source of phosphate to the overlying water column. Finally, the faster turnover rate of phosporus relative to nitrogen (Darrow et al 2003) may lead to decreased nearbottom N/P ratios following the decline of surface phytoplankton blooms. An understanding of these sediment relat ed phosphorus delivery methods requires a more detailed consideration of the sediments than has been attempted in previous simulation analyses. The ocean sediments have traditionally been regarded as an area where diffusion rules the transport of matter over minute spatial scales. While this idea still holds true in many parts of the ocean, studies primarily in continental shelf regions (Riedl et al., 1972; Riedl and Machan, 1972) have shown that large areas of sediment, especially on the continental shelves, are highly permeable. Simply put, sediment permeability is the capacit y of sediments to transmit fluid. Exchange of materials in the surface of high permeability sediments is typically dominated by advective processes in t he pore-waters (Huettel and Gust, 1992) that often results in the upwelling of nut rient rich pore-water to the sediment surface (Huettel et al. 1996). The coarse-grained carbonate and quartz sands 14

PAGE 29

that dominate the eastern Gulf of Mexico sediments (Brooks, 1973) are characterized by such high permeabilities. In addition to physical complexities nutrient dynamics near the poorly understood sediment/water interface are further affected by benthic microalgae. Though previous simulation analyses (Darrow et al. 2003) have included the microphytobenthos, a large scale study of their role in production and nutrient cycling in the eastern Gulf of Mexico has not yet been attempted. Although sediments in the South Atlantic Bight (SAB) are primarily quartz sediments, compared to mostly carbonate sediments on the WFS, the sedi ments of the two regions are characterized by similar permeabilities. Studies in the SAB have clearly shown that microphytobenthos are widespread across the continental shelf (Cahoon and Cooke, 1992; Cahoon et al. 1990; Nelson et al. 1999). The microphytobenthos have been estima ted to provide at least 10% of the primary production leading to the coastal fishery yield of the Atlantic coast of the United States from Ne w York to Georgia (Mallin et al. 1992). Benthic microalgae are a known food source fo r many different types of benthic macrofauna in shallow waters, including shrimp and other crustaceans (Miller et al. 1996), some of which are of economic importance in the eastern Gulf of Mexico. Interactions of benthic microflora with the water column trophic structure have been demonstrated through multiple pathways (Thomas and Cahoon, 1993). The isotopic signatur e of benthic microalgae ha s been traced to fishes such as the tomtate (a type of grunt) and red porgy through demersal zooplankton (Bolden, 1990; Cahoon and Tronzo, 1992), and benthic 15

PAGE 30

invertebrates (Manooch, 1977), respectively Tomtate, specifically, have been found in the gut of the econom ically important Red Grouper in the Gulf of Mexico (Schirripa et al. 1999) along with several species of crabs and other crustaceans (Moe, 1969). In addition to their potentially im portant ecological role as primary producers in economically significant components of shelf ecosystems, benthic microalgae may play an important role as a particulate stor age pool of nutrients on the continental shelf. Alteration of nutrient and oxygen fluxes across the sediment water interface due to bent hic microalgae has been shown in the laboratory (Sundback and Gr aneli, 1988), in microecosystems (Admiraal, 1977), and in the field (Jahnke et al. 2000). Previous simulation analyses (Darrow et al. 2003) suggested that a mechanism for the reduction of nutrient fl ux across the sediment/water interface might be the uptake of pore-water nut rients by the microphytobenthos. Consequently, lower concentrations of nut rients in the pore-waters, and thus, a smaller gradient across the interface, lead to less diffusion, assuming that bioturbation and other mixing processes re main the same. Benthic microalgae are also known to create biofilms (Miller et al. 1996), which may trap nutrients in the sediments. On t he other hand, increased benthic macrofaunal activity (Marinelli, 1992), due to the presence of t he microalgal food source, may lead to increased bioturbation of the sediments, and therefore increased flux. There may also be a profound impact on the water column upon the death of benthic flora, through grazing or th rough the mechanism of resuspension. 16

PAGE 31

Experiments on estuarine sediments (De Jonge and Bergs, 1987; De Jonge and Beusekom, 1995) have shown significant wind and tidal resuspension of benthic diatoms. Furthermore, such sand movem ent can effectively pulverize the algal cells (Delgado et al. 1991), causing them to lyse, and release their organic contents to the water column. Simple sediment resuspension models (Darrow, 2002) have shown that sediment suspens ion is possible on the West Florida shelf at the 30 m isobath, when signific ant wave height exceeds 3.5 m. The critical shear stress of the sediment layer may be exceeded even as deep as 70 m, under even rougher conditions. The euphotic zone, designated by irr adiance greater than 1% of the surface irradiance, extends to the bottom over much of the continental shelf in the eastern Gulf of Mexico (F ig. 9). It is greatest on the oligotrophic West Florida shelf where the percentage of surface irr adiance reaching the bottom is greater than 25% in some locations (Fig. 9) and the 1% light level can extend past the 100 m isobath (Joyce and Williams, 1969). Though no extensive studies have occurred, benthic microflora have been observed on the West Florida shelf in 1992 and 1993 (Figs. 10,11). They were studied more recently during ECOHAB mi ddle grounds cruises in July and August, 2000, and by Jim Nelson and Charle s Robertson off central Florida in November 2001 (Table 1). 17

PAGE 32

Table 1. Average chlorophyll (mg m -2 ) stocks integrated over the top 0.5 cm of sediment on the West Florida shel f during 2000-2001 (Nov-01 benthic data courtesy of Jim Nelson and Charles Robertson) Date Water-column Sediments # Stations Jul-00 14.6 5.4 15 Aug-00 15.3 21.0 15 Nov-01 16.3 15.8 3 18

PAGE 33

Figure 9. The observed A) percentage of surface PAR and B) total PAR ( E m -2 s -1 ) reaching the ocean floor during the NEGOM cruise July 25 August 9, 1998. 19

PAGE 34

Figure 10. Observed benthic ch lorophyll stocks (mg m -2 ) integrated over the top 0.5 cm of sediment in A) July 1992, B) October 1992 during the Coastal Production ( ) cruises. 20

PAGE 35

Figure 11. Observed benthic ch lorophyll stocks (mg m -2 ) integrated over the top 0.5 cm of sediment in A) April 1993, B) August 1993 during the Coastal Production ( ) cruises. 21

PAGE 36

Given that average observed benthic ch lorophyll stocks vary from 5.4 mg m -2 to 30 mg m -2 during the seven months of av ailable measurements, one might wonder what factors lead to high chlorophyll concentrations vs. low ones. Light is certainly a factor. Previous analyses (Darrow et al. 2003) showed that benthic photosynthesis is reduced on cloudy days. Field observations (Jahnke et al. 2000) in benthic chambers indicate a signifi cant correlation between light flux to the bottom and benthic prim ary production when average benthic PAR flux is less than 83.3 E m -2 s -1 In some areas of the easte rn Gulf of Mexico, however, the benthic PAR flux exceeds 83.3 E m -2 s -1 (Fig. 9). Given the generally oligotrophic conditions (Masserini & Fanning, 2000) and episodic nutrient enhancements (Gilbes et al, 1996; Fig. 3, Fig. 4), one might expect nutrient limitation to also play a role in benthic primary production. Benthic macroalgae and seagrasses can also be important producers on the continental shelf. Although growing specimens of macroalgae have been found at depths as great as 400 m (H umm, 1957), these pr oducers are most dominant nearshore in waters of less than 10 m depth (El-Sayed et al. 1972; Iverson and Bittaker, 1986). Although the growth and production of bent hic microflora is important in terms of total community production and fis hery yield on continental shelves, the mechanisms of their death and decline des erve equal attention in terms of nutrient recycling to the water column. Pr evious studies in the northeastern Gulf of Mexico (Fanning et al. 1982) indicate that the resu spension of as little as 1 22

PAGE 37

mm of organic rich sediment could enhance productivity in the water-column by as much as 200% in areas influenced by the Mississippi and M obile rivers. A simple simulation analysis (Darrow, 2002) calculated that integrated water column primary production could be briefly enhanced by ~225 mg C m -2 day -1 upon suspension and lysis of a population of benthic diatoms living in carbonate sands of otherwise poor organic c ontent on the West Florida shelf. During fall overturn, such resuspension and lysis of benthic microflora could be a nutrient source for phytoplankton in the water column like K. brevis. If the biotic remineralization of or ganic matter from the suspended benthos outpaces the dissolution of silica, t he slow growing dinoflagellates could outcompete the faster diatoms. Suspended organi c and lithogenic particles would also attenuate li ght for the shade adapted K. brevis (Millie et al. 1995), giving it a further competitive advantage. This three-dimensional simulation anal ysis assesses the role of two watercolumn nutrient sources (upwelled deepwater nutrients and recycled nutrients from a dying phytoplankton bloom) in the growth of benthic microalgae living in the high permeability sediments of the easte rn Gulf of Mexico. Using two cases of the simulation, the hypothesis that wa ter column nutrients play a significant role in the growth of benthic microalgae is tested. The role of upwelled nutrients with and without the addition of fallout from a surfac e phytoplankton bloom is tested within the framework of the spring and summer 1998 physical conditions in the eastern Gulf of Mexico. 23

PAGE 38

Methods 1. Transport Model 1.1 Pelagic Transport Advective transport (Tr a ) in the water column is described by: (1a) hh Q hh vdQ h udQ h dQ Tr21 21 1 2 a where Q represents any particular state variable to which the transport applies, is the horizontal curvilinear coordinate in the cross shore direction, is the horizontal curvilinear coordinate in the alongshore direction, is the depth dependent vertical coordinate, h 1 and h 2 are the length of the grid box in the x and y directions respectively, and u, v and represent the velocities in the x,y and z directions taken from a circulat ion model of the West Florida Shelf (Weisberg & He, 2003) adapted from the (POM) Princeton Ocean Model (Blumberg & Mellor, 1987). 24

PAGE 39

Strong horizontal diffusion is implicit in the numerical algorithm for advective transport (Blumberg & Mellor, 1987), such that explicit horizontal turbulent mixing is ignored and only the vert ical component of diffusive transport (Tr d ) is modeled: Q d K dQ Trh d (1b) where K h is the coefficient of vertical eddy diffusivity derived from a second moment turbulence closure submodel (Mellor and Yamada, 1982) embedded within the circulation model. Total transport (Tr b ) is then the sum of Tr a and Tr d in the water column. 1.2 Benthic Transport Processes in the sediment occur ov er millimeter scales. The present application of these processes to a kilo meter scale model thus requires some approximations. Nonlocal exch ange models (Boudreau, 1997) have been employed to approximate enhanced exchange of dissolved species due to porewater irrigation by benthic macrofauna: )(0CCNLE (2) 25

PAGE 40

where NLE represent s non-local exchange (mol m -3 ), is the non-local exchange parameter (Boudreau, 1997), is the sediment porosity, C 0 is the concentration of the chemical species bei ng exchanged in t he overlying water, and C is pore water the concentration of that species in the sediment layer in question. The non-local exchange formulation is used in the present simulation to approximate advective porew ater exchange withi n the sediments. Advective porewater exchange is the result of pr essure perturbations caused by the irregular flow of water over obstructions (modeled as sediment ripples) on the sediment surface. The maxi mum pressure perturbation (P max ) can be determined as: 8/3 2 max) 34.0 (14.0 Hu P (3) where is the density of the water, u is the mean current velocity over the ripple, is the ripple height and H is the depth of the water (Huettel & Gust, 1992). The mean current velocity is calculated fr om the near-bottom fl ow fields of a circulation model of the eastern Gulf of Mexico (He & Weisberg, 2003). The density of the water is calculated from the near-bottom tem perature and salinity calculations of the same circulation m odel. The depth of the water column is dynamically calculated based on the bat hymetry and the sea surface height of the circulation model at each time step. The ripple heig ht is calculated from a sediment transport model (Li & Amos, 1995). Wave data for the sediment 26

PAGE 41

transport model was obtained from 1998 NOAA buoy data in the eastern Gulf of Mexico. From the maximum pressure pertu rbation, the porewater flow (W 0 ) can be calculated by: P L WD max 0) k2 ( (4) where is the kinematic viscosity of the fluid (obtained from the circulation model), L D is the ripple wavelength (calculated by the sediment transport model), and k is the sediment permeability. The sediment permeability is calculated by: 1 106.52 2 3 3dx ks (5) where is again the sediment porosity and d s is the mean sediment grain diameter (Boudreau, 1997). Bo th parameters were obtained from Gulf of Mexico observations in the usSEABED (Reid, et al., 2001) database. Permeabilities were calculated for the sediments underly ing each point on the horizontal grid (Fig. 8B). Equation 5 is known to be inaccurate in areas where porosity is greater than 0.8 (Boudreau, 1997). Therefor e, at points where sediment porosity was greater than 0.8 according to the usSEABED database, sediment permeability was assumed to be 1.0 x 10 -14 cm 2 27

PAGE 42

The non-local exc hange parameter () at the surface sediment layer can then be determined from: h W0 0 02 (6) Where W 0 is again the porewater flow, and h 0 is the thickness of the surface sediment layer, assumed to be 2 mm in this model. The non-local exchange parameter decays exponentia lly in the sediment on the length scale of the wavelength of the sedim ent ripple (Boudreau, 1997). In order to test the non-local exchange approximation for porewater advection in the sediments without interference from biochemical processes, a simulation case considered only the physical factors affecting a single tracer in the sediments and the overlying water-colum n. Concentrations of the tracer in the bottom layer of sediments in this tracer case were set to 1000 mmol kg -1 at randomly chosen points within three pre-det ermined regions of the model grid the northern Gulf coast, t he West Florida shelf and t he continental slope off the west coast of Florida (Fig. 12). These c oncentrations were held constant over a 90 day model run from February 27 to Ma y 28 1998. The model was run twice. One run considered diffusion only and a second run used diffusion and the nonlocal exchange (NLE) approximation. Under these conditions, molecular di ffusion would tend to produce a linear profile of the tracer in the sediments (Fig 13A) since the concentration of the deepest layer is held constant at 1000 mmol m -2 The concentration of tracer in 28

PAGE 43

Figure 12. Geographical locations of (A) site 1( ), site 2( ) and site 3( )of the tracer case of the model, along with the flow (B) at those locations over time. 29

PAGE 44

Figure 13. Sediment profiles of tr acer at stations 1 ( ), 2( ), and 3( ) during 17 March 1998 (day 18) of A) the diffusion only tracer case and B) the NLE tracer case of the model. 30

PAGE 45

Figure 14. Tracer concentration (mmol kg -1 ) in the near-bottom water during 17 March 1998 (day 18) of the NLE tracer case of the model. tracer at sites 1 and 2 (Fig. 14B) 31

PAGE 46

the water column is almost constant at near zero because the tracer diffusing out of the sediment is flushed away by the cu rrents. At site 3, where the simulated near-bottom flow was relatively slow compared to sites 1 and 2 (Fig. 12b), the profile of tracer in the s ediment was nearly linear by day 18 of the simulation (Fig 13b). In contrast, comparatively swift near-bottom flow at site 1 resulted in greater flushing of the surfac e sediments at sites 1 and 2. The vertical profile of tracer at sites 1 and 2 (Fig. 14B) is sim ilar to those observed by Jahnke et al. (2000) under similar flow condition s on the South Atlantic Bight. In the days leading up to day 54 of t he simulation in April 1998, flow over site 1 was somewhat reduced (Fig. 12B), rela tive to early in the simulation. As a result, the surface sediments were not fl ushed to the extent they were on day 18 (Fig. 15B). Flow at site 2 had in creased somewhat, however leading to increased flushing of the surface sediments there (Fig 15B). The near-bottom tracer at the offshore site 2 moved shoreward, while the near-bottom tracer at sites 1 and 3 moved cyclonically around the eastern Gulf of Mexico against the coastline (Fig. 15A). By day 90, flow over all three si tes had slowed to less than 2 cm s -1 (Fig 12B). As a result, sediment profiles at all three sites were similar (Fig 16). Flushing of the surface sediment was mini mal compared to that at station 1 on day 18 (Fig. 14B). Calculated non-local exchange rates at the sediment/water interface in the model ranged from 0 day -1 to 3 day -1 and were consistent with previous estimates (Jahnke et al 2005) of non-local exchange rates required to produce 32

PAGE 47

Figure 15. A) Concentration of tracer in the near-bottom water (mmol kg -1 ) and B) sediment profiles of tracer at stations 1 ( ), 2( ), and 3( ) during 22 April 1998 (day 54) of NLE trac er case of the model. 33

PAGE 48

Figure 16. A) Concentration of tracer in the near-bottom water (mmol kg -1 ) and B) sediment profiles of tracer at stations 1 ( ), 2( ), and 3( ) during 28 May 1998 (day 90) of the NLE tr acer case of the model. 34

PAGE 49

the sediment profiles of observed nutrients on the SAB. As such, the sediment transport model is assumed to be a su itable approximation of pore-water advection processes in permeable continent al shelf sediment environments such as those on the South Atlantic Bight and the West Florida shelf. 1.3 Optics In almost every ecological system, light plays a key role in primary production. Particularly in light limi ted benthic habitats, small changes in PAR flux through the water column can alter benthic light flux si gnificantly and can therefore determine the growth of algae liv ing there. Thus, a detailed light model is employed to simulate the dynamics of the benthic and pelagic microalgae in the ecological simulation. In this case, a spectral solar irradiance model (Gregg and Carder, 1990) is corrected for cloud cover, and used to derive solar irradiance at 8 visible spectral wavelengt h integrals just below the ocean surface (I 0() ), thus representing photosyn thetically active radiation (PAR) in distinct 25 nm bands. Although smaller bands might resu lt in a more detailed determination of the light field, the 25 nm bands r epresent a compromise between a detailed determination of PAR at depth vs. the computation time of the model. Within the water column, light ener gy is absorbed and scattered by the water itself, algae, and colored dissolved organic matter (Kirk, 1995) as well as non-living particulate matter (detritus) and bacteria (Nelson and Robertson, 1993). In addition, to the above factor s, light is scattered and absorbed by 35

PAGE 50

sediment particles as it travels through the benthos. These processes account for the attenuation of spectral irradianc e as it travels through the ecosystem. Spectral irradiance at a given depth, z (I z() ), was calculated according to Beers law of exponential decay: e zk II)(0)(z (7) where denotes the wavelength of interest, and k is an attenuation coefficient described by, h }dh)h(Qk{ k z }dz)z(Qk{ k kh z )(Q )(s z 0 )(Q )(w (7a) The second term of equation 7a applies only to the benthos and Q now represents pelagic diatoms (P 1 ), and pelagic microflagellates (P 2 ) in the first term, benthic diatoms (P 3 ) in the second term, and detritus (D), bacteria (B), and CDOM in both terms. Their specif ic attenuation coefficients (k Q() ) are then represented by k P1( ) k P2( ) k P3( ) k D( ) k B( ) and k C( ) respectively. The specific attenuation coefficients of water and wet sandy sediment are respectively k w( ) and k s( ) The specific attenuation coefficients at each wavelength for each parameter are determined from published values. In the case of the wet, sandy sediments (Kuhl and Jorgensen, 1994) and water (Pope and Fry, 1997; Smith 36

PAGE 51

and Baker, 1981), total attenuation coefficien ts at each spectral wavelength are known. Due to the dissolved nature of CDOM, backscattering is ignored in the present model (Gordon et al. 1988) and attenuation coefficients for CDOM are equal to published CDOM absorption c oefficients in the eastern Gulf of Mexico(Carder et al. 1989). Attenuation coefficient s for phytoplankton, benthic microalgae (Bidigare et al. 1990), bacteria (Morel and Ahn, 1990) and detritus (Roesler et al. 1989) are, instead, equal to the sum of the wavelength specific absorption coefficients and the corre sponding scattering coefficients. 2. Biological Model 2.1 Primary Producers Fueled by light energy, and nutrient fert ilization the dynamics of the three functional groups of microalgae are modeled. The processes affecting the rate of change with time (t) for the pelagic diatoms (P 1 ) and microflagellates (P 2 ) are described by: P dl P d P d P d P dg) P w ()dP( Tr t dP1 1 1 1 1 1 1 1 1 1 1 1 1 t 1 (8) P dl P d P d P d P dg)dP( Tr t dP2 2 2 2 2 2 2 2 2 2 2 t 2 (9) 37

PAGE 52

where w 1 represents the sinking rate of t he pelagic diatoms. The terms, g n n n n and l n represent the respective growth, re spiration, excretion, grazing and lysis rates of the pelagic microalgae. The benthic diatoms (P 3 ) are described by: l g hPhPTr t hP3 33 3 3 33t 3 (10) where g 3 3 3 3, and l 3 are, again, the growth, resp iration, excretion, grazing and lysis rates of the benthic diatoms. The specific biological rates, g n expand to non-linear, time dependent expressions, SiO k / SiO PO k / PO NH k / NH or NO k / NO e I / Imin g4 )3,1(4sio 4 4 )3,1(4PO 4 4 )3,1(4NH 4 3 )3,1(3NO 3 3,1(Isat/Iz1 satz 3,13,1 (8a,10a) for benthic and pelagic diatoms, and PO k / PO NH k / NH or NO k / NO e I / Imin g4 )2(4PO 4 4 )2(4NH 4 3 )2(3NO 3 2(Isat/Iz1 satz 22 (9a) 38

PAGE 53

for pelagic microflagellates, which are not dependent on silicate as a nutrient source. In the above equations, kNO3(n), kNH4(n), kPO4(n), and kSiO4(n) represent the Michaelis-Menten half-saturation const ants, with respect to each microalgal functional group, for uptake of nitrate, ammonia, phosphate and silicate respectively. Isat(n) is the saturation light intens ity for each functional group effecting photo-inhibition. The maximal microalgal gross growth rate, n is a function of temperature (Eppley, 1972): e)27T(06330 )n(tn (8b,9b,10b) where T is the temperature ( C) obtained from the POM and t(n) is a functional group specific maximum growth rate normalized to a 24 h period at 27 C (Eppley, 1972). 2.2 Secondary Producers The secondary producers (micro/mesozooplankton and multiple benthic grazers) are not explic itly simulated but their activities do, instead, provide closure for this ecological simulation. The proposed simulations grazing rates are determined as Michaelis-Menten functi ons of the microal gal biomass (Mullin et al., 1975): PPk / PP' iizi ii mii (11) 39

PAGE 54

where i is the functional group of microalgae (1,2 or 3), mi is the maximum grazing rate for the herbivore grazing the corresponding microalgal functional group, and k zi is the corresponding half-saturation constant for grazing. Each microalgal functional group is subj ect to a refuge population, P below which grazing does not occur. 2.3 Microbial Loop The role of heterotrophic bacteria in t he recycling of nutrients is significant (Azam et al., 1983), and measurements of bacter ial production in the Gulf of Mexico have been as high as 156 g C l -1 day -1 (Pakulski et al. 2000). Though these microbes are quite diverse, their dynamics in the marine environment are still difficult to discern and they are often treated as a homogenous assemblage (Ducklow, 2000). With respect to bacteria, the primary concern of this model is to simulate the remineralization of organi c matter to inorganic matter that can be used by the simulated microalgae as well as the nitrification of ammonia to nitrate to be used by benthic diatoms in the sedi ments. Such nitrification occurs whether or not light is pr esent (Eriksson, P.G. & S.E.B. Weisner,1999). Thus, the models bacterioplankton are divided into two distinct functional groups the ammonifying (B a ) and nitrifying (B n ) bacteria. In the wa ter-column, these are described by: 40

PAGE 55

dmB B d B dg)dB( Tr t dB1a 1a 1a 1a 4 1a t 1a (12a) dmB B d B dg)dB( Tr t dB1n 1n 1n 1n 5 1n t 1n (13a) and in the sediments, they are described by: hmBB h B hg)hB( Tr t hB2a 2a 2a 2a 6 2a t 2a (12b) hmBB h B hg)hB( Tr t hB2n 2n 2n 2n 7 2n t 2n (13b) where represents published bacterial respirat ion rates (del Giorgio and Cole, 2000) and m represents the bacterial mortality rates. Like the microalgae, bacterial growth rates (g4-7) are a function of temper ature (Walsh and Dieterle, 1994) such that: e008.0 gT0920 74 (12c,13c) where T is again the temperature obtained from the circulation model and the temperature in the sediments is assum ed to be the same as the near-bottom sigma level of the circulati on model (Weisberg & He, 2003). 41

PAGE 56

3. Nutrient model 3.1 Non-living particles The fecal pellets of the models grazers are explicit ly described state variables, Z 1 Z 2 Z 3 representing the egested produc ts of organisms feeding on pelagic diatoms, pelagic microflagellate s, and benthic diatoms respectively. They are described in the model by: ZkP 1d Z w dZ 1 Tr dZ1fp1 1 11 1 1z t 1 (14a) ZkP 1d Z w dZ 2 Tr dZ2fp2 2 2 2 2z t 2 (14b) ZkP 1 h hZ 3 Tr hZ3fp3 3 33 t 3 (14c) where w z1,2 are the fecal pellet sinking rates. Note that the fecal pellets of the benthic grazers (and those of the pelagic graz ers) are not allowed to sink through the sediments and are theref ore assumed to remain in the sediment layer in which the matter was originally ingested. The parameters, 1,2,3 represent the fraction of grazed microalgae which are actually ingeste d, ie. not lost due to sloppy feeding (Banse, 1992; Jumars et al. 1989). Grazer respiration is represented by 1,2,3 such that the term [(11,2,3 ) 1,2,3 1,2,3 P 1,2 ,M] is the grazer assimilated carbon, all of which is convert ed to fecal pellets to provide closure in 42

PAGE 57

the model since the grazer biomass is not explicitly repr esented. Finally, k fp is the fecal pellet degradation rate. The other form of non-living particula te organic carbon in the model is detritus. It primarily cons ists of particulate cell material from dead or dying microalgae. Its dynamics are descr ibed in the water column by: Dk P l 1 d D w dD Tr dD1D 1 1 1D 1 t 1 (15a) Dk P l 2 d D w dD Tr dD2D 2 2 2D 2 t 2 (15b) and in the sediments by: Dk P l 3 hhDTr hD3D 3 3 t 3 (15c) where w D1,2 are the sinking rates of the detri tal products of the pelagic diatoms and microflagellates respectively. Like the sediment fecal pellets, sediment detritus is assumed to remain in the s ediment layer where it was generated. The parameters k fp and k D represent the degradation rates of the fecal pellets and detritus respectively. Becaus e the fecal pellets and detritus are algal products, they are assumed to have the same carbon to nitrogen and carbon to phosphorus ratios as the proposed m odels microalgae. As the non-living particulate organic material is degr aded, DOC, DON, and DOP are thus 43

PAGE 58

proportionally produced. The inorganic parti culate silica contained within some of the detritus and fecal pellets, howev er, is not subject to the same degradational processes as the organic materi als. Due to the differential sinking rates of fecal pellets (100 m day -1 ) and detritus (5 m day -1 ), the particulate silica in the water-column is separated into two distinct groups: detrital silica (Si D ) and fecal pellet silica (Si fp ). These are described in the model by: Sid PlP 1 C Si dSiw dSi Tr t dSiD 111 1 111 r DD D D (16a) Sid P 1 C Si dSiw dSi Tr t dSifp 1 1 11 r fpfp fp fp (16b) where is the dissolution rate of particulate silic a. Note that particulate silica is added to the models detrital pool thr ough the sloppy feeding and respiration of grazers in the third term of equation 16a in order to maintain a mass balance with the organic matter. The fe cal pellets and detritus in the sediments are not subject to a sinking flux, such that t he particulate silica dynamics there can be described by one equation: Sih PlP C Si h hSi Tr t hSis 333 3 r s s (16c) 44

PAGE 59

3.2 Dissolved Organic Matter Dissolved organic matter is important in marine ecosystem studies of phytoplankton dynamics for two reasons. Organic matter colors the water, thus affecting the attenuation of light in the water column and the sediments, and it also acts as a reservoir of carbon, nitrogen and phosphorus which have varying levels of reactivity. Despite extensive study of the properties of dissolved organic matter, however, only 4-11% of DOC and 7-14% of DON have been characterized in the surface ocean (Benner, 2002). Given the range of properties of mari ne DOM (Benner, 2002), it is difficult to adequately model the dynam ics of these substances within the ecosystem given just one pool. These properties, however, fall across a wide spectrum, making it difficult to separate the DOM in to distinct pools. In much of the literature (Hedges, 2002), marine DOM is separated into two sets of two DOM classes. One set separates DOM acco rding to size (Benner, 2002) into high molecular weight (HMW) and low molecula r weight (LMW) fractions. The other separates DOM according to its reactivity (Carlson, 2002) into refractory and labile components. Unfortunately, these groups are not mutually exclusive, ie. HMW DOM can be both refractory and l abile. Furthermore, in terms of ecosystem models, a separation by size is useful for light attenuation since HMW compounds are typically the colored com ponent of marine DOM (Kirk, 1995), but a separation by reactivity is far more us eful when calculating the relation of DOM to nutrient cycling. 45

PAGE 60

For the purposes of this model, two r eactively distinct pools of DOM are used to calculate the models nutrient cycl es. A mean value of 25% of the total DOM will be considered to be high molecular weight (Benner, 2002), and therefore CDOM affecting light attenuation (Kirk, 1995). Labile DOM (LDOC) is thus described in terms of carbon by: bLDOC B g DkDk ZkZkP1 P P 1 mBmB d )dLDOC(Tr t dLDOC1a 4 2D1D 2fp1fp1 1 1 2 2 1 1n 1a (17a) where m is equal to the mortality rate of all functional groups of bacteria and is the fraction of the dying bacterial biomass which becomes labile (Ogawa et al. 2001). In contrast, the fecal pellet and detri tal material are fully labile (Amon and Benner, 1996) and are calculated as a function of available substrate and the respective coefficients k fp and k d The coefficient, b, is the fraction of dissolved organic matter undergoing photodegradatio n to inorganic products. The sediment pool of labile DOC is then: bLDOC B g DkZkP1 P 3 mBmB d )hLDOC(Tr t dLDOC2a 6 3D3fp3 3 3 3 2n 2a (17b) Based on available data (Mopper et al. 1991), b is usually zero in the sediments and much of the lower porti on of the water-column. 46

PAGE 61

The models refractory pool of DO M (RDOC) is then described in the water-column by: bRDOC mB)1(mB)1( d )dRDOC(Tr t dRDOC1n 1a (18a) and in the sediments by: bRDOC mB)1(mB)1( h )hRDOC(Tr t dRDOC2n 2a (18b) Since refractory organic matter exists in the model only due to initial conditions and production by bacteria, re fractory dissolved organic nitrogen (RDON) and dissolved organic phosphorus ( RDOP) are assumed to be related to RDOC by the respective bacterial C/N and C/P ratios. The models labile organic matter, however, has bacterial and microal gal origins such that labile dissolved organic nitrogen (LDON) is described in the models water-column by: bLDON B g mBmB C/N d DDkZZk P 1 PP C/N d )dLDON(Tr t dLDON1a 4 1n 1a b 21D21fp 1 1 1 2 2 1 1 r (19a) and in the sediments by: 47

PAGE 62

bLDON B g mBmB C/N h DkZk P 1 P C/N h )hLDON(Tr t hLDON2a 6 2n 2a b 3D3fp 3 3 3 3 3 r (19b) Labile dissolved organic phosphorus (LDOP) is described in the watercolumn by: bLDOP B g mBmB C/P d DDkZZk P 1 PP C/P d )dLDOP(Tr t dLDOP1a 4 1n 1a b 21D21fp 1 1 1 2 2 1 1 r (20a) while the pore-water LDOP is described by: bLDOP B g mBmB C/P h DkZk P 1 P C/P h )hLDOP(Tr t hLDOP2a 6 2n 2a b 3D3fp 3 3 3 3 3 r (20b) 48

PAGE 63

3.3 Dissolved Inorganic matter Remineralization, in terms of carbon, is described in the water-column by: B g P g P g P P bLDOC bRDOC BBPP d )dDIC(Tr t dDIC1n 5 2 2 1 1 2 2 2 1 1 11 1n 1n 1a 1a 2 2 1 1 (21a) where carbon is respired by pelagic microalgae, bacteria and grazers, and is further produced by the photolysis of disso lved organic carbon. The DIC is then a carbon source for the phytoplankton and t he nitrifying bacteria (del Giorgio and Cole, 2000), but not for the ammonifying bacteria, which make use of the labile organic carbon. Similarly, pore-water DIC is described by: B g P g P bLDOC bRDOC BBP h )hDIC(Tr t hDIC2n 7 3 3 3 3 33 2n 2n 2a 2a 3 3 (21b) The only form of new nitrogen in the wate r-column of the model is described by: B C/N d P g P g C/N ddNO Tr t dNO1n 1n b 2 3NO2 1 3NO1 r 3 t 3 (22a) 49

PAGE 64

where (N/C)r is a Redfield nitrogen to carbon ratio for the models phytoplankton and (N/C)b is the nitrogen to carbon ratio of t he models bacteria. The simulated nitrification in the model occurs thr ough the nitrogen equivalent of respiration (n1) of the models nitrifying bacteria (Bn1). New nitrogen in the sediments is similarly described by: B C/N h P g C/N h hNO Tr t hNO2n 2n b 3 3NO3 r 3 t 3 (22b) Regenerated nitrogen is descr ibed in the water-column by: bLDON B g B C/N d PP P g PP g P C/N ddNH Tr dNH1n 4NH5 1a 1a b 2 2 2 1 1 11 2 4NH2 2 2 1 4NH1 1 1 r 4 t t 4 (23a) where the pelagic nitrifying bacteria are in competition with the pelagic diatom s and microflagellates for ammonium. In addi tion to ammonification, which occurs through the respiration (a1) of the models ammonifying bacteria (Ba1), ammonium is produced through the nitrogen equivalent of respiration of the models pelagic grazers. The fraction of carbon respired is represented by Regenerated nitrogen in the sediments is described by: bLDON B g B C/N h P P g P C/N h hNH Tr hNH2n 4NH7 2a 2a b 3 3 33 3 4NH3 3 3 r 4 t t 4 (23b) 50

PAGE 65

The nitrifying bacteria are also in competition with the microalgae for phosphate, which is described in the models water-column by: bLDOP B g BB C/P d PP P g PP g P C/P ddPO Tr dPO1n 4PO5 1n 1n 1a 1a b 2 2 2 1 1 11 2 4PO2 2 2 1 4PO1 1 1 r 4 t t 4 (24a) where (P/C)r and (P/C)b are the phosphorus to carbon ratios of the phytoplankton and bacteria, respectively. Note that phosphate is taken up by the nitrifying bacteria, but is regenerated through the re spiration of both the nitrifying and ammonifying bacteria (del Giorgio & Cole, 2000). The functional group of ammonifying bacteria, instead, uses di ssolved organic phosphorus (DOP) as a source of phosphorus during t he break down of organic matter. Sediment phosphate dynamics are described by: bLDOP B g BB C/P d P P g P C/P h hPO Tr hPO2n 4PO7 2n 2n 2a 2a b 3 3 33 3 4PO3 3 3 r 4 t t 4 (24b) Silicate dynamics are simply descri bed in the models water-column by: d PP g C/Si ddSiO Tr dSiO1 1 1 4SiO1 r 4 t t 4 (25a) 51

PAGE 66

where the sole silicate loss term is due to uptake by the pelagic diatoms. Silicate is then regenerated through the dissolution () of particulate silica upon the death of the diatoms. A small amount of silicate is also regenerated as dissolution of silica frustules from living diatoms to ma intain a simple mass balance within the model upon excretion of or ganic material. Likewise, silicate dynamics in the sediments are described by: d PP g CSi hhSiO Tr dSiOSiO r t t 3 1 3 43 4 4/ (25b) 4. Boundary conditions The model employs a no-flux bou ndary condition along solid coastal boundaries, ignoring estuaries. Cross s helf open boundaries exist off Mobile Bay and Charlotte Harbor, and an along shelf open boundary exists at the 200 m isobath. At these boundaries, time-dependent nitrate, amm onium, and silicate values are prescribed at inflow poin ts using the NEGOM and ECOHAB data sets (Table 2). Estuarine nutrients and DO M are added to the model as boundary conditions at various points according to observations made during the NEGOM and ECOHAB cruises in 1998. Nutrient c oncentrations are linearly interpolated spatially for grid points between stations and temporally between observations. 52

PAGE 67

At the bottom of the sediment la yer, a constant boundary condition is employed (Table 3), ignoring groundwater input of nutrients in this examination of water column nutrients. Atmospheric depo sition of chemical species are ignored in this model such that fluxes for all c hemical species across the air-sea interface are zero, with the exception of DIC for which the diffusive flux at the air-sea interface is described by: pCO pCO W 10x11.1 DIC d k2 0 2 air 5 h 0 (26) where W is the wind speed, and is the solubility of CO 2 in seawater. In the upper layer of the water-column, the partial pressure of CO 2 (pCO 2 ) is calculated (Peng et al. 1987) as a function of POMs temperature and salinity, using alkalinity = 520 + 51.2 salinity (Millero et al. 1998), while (pCO 2 ) air is assumed to be 365 atm. All simulations are forced by daily av erage flow fields from a 1998 case of a circulation model of the We st Florida Shelf (Weisberg & He, 2003). Model runs simulate the spring and summer of 1998 in two cases: 1) The standard case, considering only the available observations during the spring and summer 1998 from the NEGOM and ECOHAB:Florida pr ograms (Fig. 8a), and 2) a case of different initial conditions, considering t he fallout of an initialized spring diatom bloom off the coast of the Big Bend regi on of Florida, such as those that are produced in low salinity plumes of the Apalachicola and Mississippi rivers (Gilbes et al., 1996) and similar to that analyzed in a previous 1-D simulation (Darrow et 53

PAGE 68

al 2003). Case 1 tests whether upwelled nu trients on their own are sufficient to grow benthic diatoms. Case 2 tests the ability of upwelled nutrients in conjunction with recycled nutrients from a dying phytoplankton bloom to grow benthic microalgae. Model parameters are su mmarized in table 4. 54

PAGE 69

Table 2. The depth dependent water-column conditions applied at the open boundaries over the entire model run, and a pplied over the entire model grid in the initial time step. Depth (m) Diatoms ( mol C l -1 ) Flagellates ( mol C l -1 ) NH 4 ( mol kg -1 ) NO 3 ( mol kg -1 ) PO 4 ( mol kg -1 ) SiO 4 ( mol kg -1 ) 1 0.55 0.3125 0.2 0.05 0.025 1 10 0.55 0.3125 0.2 0.05 0.025 1 50 0.55 0.3125 0.2 0.75 0.1 1 100 0.55 0.3125 0.2 2.5 0.125 1 150 0.55 0.3125 0.2 5.64 0.225 1 200 0 0 0.2 10 0.34 6 250 0 0 0.2 10.5 0.425 7.5 300 0 0 0.2 10.6 0.51 9 350 0 0 0.2 11 0.595 10.5 400 0 0 0.2 12.08 0.68 12 450 0 0 0.2 13.59 0.765 13.5 500 0 0 0.2 15.1 0.85 15 600 0 0 0.2 18.12 1.02 18 700 0 0 0.2 21.14 1.19 21 800 0 0 0.2 24.16 1.36 24 900 0 0 0.2 27.18 1.53 27 1000 0 0 0.2 27.5 1.75 27.5 1500 0 0 0.2 27.5 1.75 27.5 2000 0 0 0.2 27.5 1.75 27.5 2500 0 0 0.2 27.5 1.75 27.5 55

PAGE 70

Table 2 (cont) Depth (m) DON ( mol kg -1 ) DOP ( mol kg -1 ) DOC ( mol kg -1 ) DIC ( mol kg -1 ) Siliceous Fecal Pellets ( mol C kg -1 ) 1 11.3 0.7 75 2100 0 10 11.3 0.7 75 2100 0 50 11.3 0.7 75 2100 0 100 11.3 0.7 75 2100 0 150 11.3 0.7 75 2100 0 200 10.3 0.64 68 2120 0 250 10.05 0.625 66.25 2125 0 300 9.8 0.61 64.5 2130 0 350 9.55 0.595 62.75 2135 0 400 9.3 0.58 61 2140 0 450 9.05 0.565 59.25 2145 0 500 8.8 0.55 57.5 2150 0 600 8.3 0.52 54 2160 0 700 7.8 0.49 50.5 2170 0 800 7.3 0.46 47 2180 0 900 6.8 0.4 45 2190 0 1000 6.8 0.4 45 2200 0 1500 6.8 0.4 45 2200 0 2000 6.8 0.4 45 2200 0 2500 6.8 0.4 45 2200 0 56

PAGE 71

Table 2 (cont) Depth (m) Non-Siliceous Fecal Pellets ( mol C kg -1 ) Nitrifying Bacteria ( mol C kg -1 ) Ammonifying Bacteria ( mol C kg -1 ) Siliceous Detritus ( mol C kg -1 ) Non-Siliceous Detritus ( mol C kg -1 ) 1 0 0.05 0.35 0.0075 0.005 10 0 0.05 0.35 0.0075 0.005 50 0 0.05 0.35 0.0075 0.005 100 0 0.05 0.35 0.0075 0.005 150 0 0.05 0.35 0.0075 0.005 200 0 0.05 0.35 0.0075 0.005 250 0 0.05 0.35 0.0075 0.005 300 0 0.05 0.35 0.0075 0.005 350 0 0.05 0.35 0.0075 0.005 400 0 0.05 0.35 0.0075 0.005 450 0 0.05 0.35 0.0075 0.005 500 0 0.05 0.35 0.0075 0.005 600 0 0.05 0.35 0.0075 0.005 700 0 0.05 0.35 0.0075 0.005 800 0 0.05 0.35 0.0075 0.005 900 0 0.05 0.35 0.0075 0.005 1000 0 0.05 0.35 0.0075 0.005 1500 0 0.05 0.35 0.0075 0.005 2000 0 0.05 0.35 0.0075 0.005 2500 0 0.05 0.35 0.0075 0.005 57

PAGE 72

Table 3. The depth based initial conditions appl ied to the sediments over the entire model grid based on measurements in the South At lantic Bight (Marinelli et al. 1998). Depth (mm) Diatoms (mmol C m 3 ) NH 4 ( mol kg -1 ) NO 3 ( mol kg -1 ) PO 4 ( mol kg -1 ) SiO 4 ( mol kg -1 ) DON ( mol kg -1 ) 2 4000 10 2 0.1 1.5 3.773585 4 0 10 2 0.1 1.5 22.64151 6 0 10 2 0.1 1.5 45.28302 8 0 10 2 0.1 1.5 75.4717 10 0 10 2 0.1 1.5 113.2075 12 0 10 2 0.2 1.5 135.8491 14 0 10 1 0.3 1.5 166.0377 16 0 10 1 0.4 1.5 181.1321 18 0 10 1 0.5 1.5 196.2264 20 0 10 1 0.6 1.5 211.3208 22 0 10 1 0.7 1.5 226.4151 24 0 10 1 0.8 1.5 241.5094 26 0 10 1 0.9 1.5 256.6038 28 0 10 1 1 1.5 271.6981 30 0 10 1 1.1 1.5 279.2453 32 0 10 1 1.2 1.5 283.0189 34 0 10 0 1.3 1.5 286.7925 36 0 10 0 1.4 1.5 292.8302 38 0 10 0 1.5 1.5 295.8491 40 0 10 0 1.6 1.5 301.8868 58

PAGE 73

Table 3 (cont) Depth (mm) DOP ( mol kg -1 ) DOC ( mol kg -1 ) DIC ( mol kg -1 ) Siliceous Fecal Pellets ( mol C kg -1 ) Non-Sliceous Fecal Pellets ( mol C kg -1 ) 2 0.235849 25 2200 0 0 4 1.415094 150 2200 0 0 6 2.830189 300 2200 0 0 8 4.716981 500 2200 0 0 10 7.075472 750 2200 0 0 12 8.490566 900 2200 0 0 14 10.37736 1100 2200 0 0 16 11.32075 1200 2200 0 0 18 12.26415 1300 2200 0 0 20 13.20755 1400 2200 0 0 22 14.15094 1500 2200 0 0 24 15.09434 1600 2200 0 0 26 16.03774 1700 2200 0 0 28 16.98113 1800 2200 0 0 30 17.45283 1850 2200 0 0 32 17.68868 1875 2200 0 0 34 17.92453 1900 2200 0 0 36 18.30189 1940 2200 0 0 38 18.49057 1960 2200 0 0 40 18.86792 2000 2200 0 0 59

PAGE 74

Table 3 (cont) Depth (mm) Siliceous Detritus ( mol C kg -1 ) NonSiliceous Detritus ( mol C kg -1 ) Nitrifying Bacteria ( mol C kg -1 ) Ammonifying Bacteria ( mol C kg -1 ) Denitrifying Bacteria ( mol C kg -1 ) 2 0.0075 0.005 0.05 0.35 0 4 0.007125 0.00475 0.05 0.35 0 6 0.00675 0.0045 0.05 0.35 0 8 0.006375 0.00425 0.05 0.35 0 10 0.006 0.004 0.05 0.35 0 12 0.005625 0.00375 0.05 0.35 0 14 0.00525 0.0035 0.05 0.35 0.05 16 0.004875 0.00325 0.05 0.35 0.05 18 0.0045 0.003 0.05 0.35 0.075 20 0.004125 0.00275 0.05 0.35 0.1 22 0.00375 0.0025 0.05 0.35 0.125 24 0.003375 0.00225 0.05 0.35 0.15 26 0.003 0.002 0.05 0.35 0.175 28 0.002625 0.00175 0.05 0.35 0.2 30 0.00225 0.0015 0.05 0.35 0.225 32 0.001875 0.00125 0.05 0.35 0.25 34 0.0015 0.001 0.05 0.35 0.275 36 0.001125 0.00075 0.05 0.35 0.3 38 0.00075 0.0005 0.05 0.35 0.325 40 0.000375 0.00025 0.05 0.35 0.35 60

PAGE 75

Table 4. Model parameters. Symbol Description Units B Bacteria mmol m -3 B a1 Pelagic ammonifying bacteria mmol m -3 B a2 Benthic ammonifying bactera mmol m -3 B n1 Pelagic nitrifying bacteria mmol m -3 B n2 Benthic nitrifying bacteria mmol m -3 b 1.9 x 10 -9 Coefficient of phototransformation of DOM s -1 CDOM Colored dissolved organic matter mmol m -3 D Detritus mmol m -3 d Depth of level m g 1 Realized growth rate of pelagic diatoms s -1 g 2 Realized growth rate of pelagic microflagellates s -1 g 3 Realized growth rate of benthic diatoms s -1 g 4 Realized growth rate of pelagic ammonifyign bacteria s -1 g 5 Realized growth rate of pelagic nitrifying bacteria s -1 g 6 Realized growth rate of benthic ammonifying bacteria s -1 g 7 Realized growth rate of benthic nitrifying bacteria s -1 h 0.002 depth of each sediment level m I 0( ) Solar irradiance at wavelength, just below the surface E m -2 s -1 I sat Saturation intensity effecting photoinhibiti on for a given microal gal functional group E m -2 s -1 I z( ) Incident irradiance at depth, z and wavelength, E m -2 s -1 l 1 0.03 Pelagic diatom lysis rate s -1 l 2 0.03 Pelagic microflagellate lysis rate s -1 l 3 0.03 Benthic diatom lysis rate s -1 k B( ) Specific attenuation coefficient of bacteria at wavelength, m -1 k C( ) Specific attenuation coefficient of CDOM at wavelength, m -1 k D( ) Specific attenuation coefficient of detritus at wavelength, m -1 k d 5.7 x 10 -7 Degredation rate of detritus s -1 k fp 5.7 x 10 -7 Degredation rate of fecal pellets s -1 k h Coefficient of vertical eddy diffusivity cm s -1 k M( ) Specific attenuation coefficient of benthic diatoms at wavelength, m -1 k NH4 Michaelis-Menten half-saturation constant fo r Ammonium uptake fo r a given functional group mmol m -3 k NO3 Michaelis-Menten half-saturation c onstant for Nitrate uptake for a given functional group mmol m -3 k PO4 Michaelis-Menten half-saturation c onstant for Phosphate uptake for a given functional group mmol m -3 k P1( ) Specific attenuation coefficient of pelagic diatoms at wavelength, m -1 k P2( ) Specific attenuation coefficient of pelagic microflagellates at wavelength, m -1 k s( ) Specific attenuation coefficient of we t, sandy sediment at wavelength, m -1 k SiO4 Michaelis-Menten half-saturation c onstant for Silicate uptake for a given functional group mmol m -3 k w( ) Specific attenuation coefficient of water at wavelength, m -1 k Total attenuation coefficient at wavelength, m -1 61

PAGE 76

Table 4 (cont) Symbol Description Units LDOC Concentration of labile DOC mmol m -3 LDON Concentration of labile DON mmol m -3 LDOP Concentration of labile DOP mmol m -3 M Benthic diatoms mmol m -3 (N/C) b 0.15 Carbon to Nitrogen ratio for bacteria (N/C) r 0.15 Redfieldian Carbon to Nitrogen ratio for microalgae NH 4 Ammonium concentration mmol m -3 NO 3 Nitrate concentration mmol m -3 (P/C) b 0.0188 Carbon to phosphorus ratio for bacteria (P/C) r 0.009 Redfieldian carbon to phosphorus ratio for microalgae P 1 Pelagic diatoms mmol m -3 P 2 Pelagic microflagellates mmol m -3 PO 4 Phosphate concentration mmol m -3 RDOC Concentration of refractory DOC mmol m -3 RDON Concentration of refractory DON mmol m -3 RDOP Concentration of refractory DOP mmol m -3 Si Particulate silica mmol m -3 (Si/C) r 0.15 Silica to carbon ratio for benthic and pelagic diatoms SiO 4 Silicate concentration mmol m -3 T Temperature C Tr a Transport due to advection cm s -1 Tr b Transport across the sediment/water interface cm s -1 Tr d Transport due to diffusion cm s -1 Tr s Trasport within the sediments cm s -1 Tr t Total transport cm s -1 t Time step s -1 u Velocity component in the direction cm s -1 v Velocity component in the direction cm s -1 W Wind speed m s -1 w 1 5.7 x 10 -6 Sinking rate of pelagic diatoms m s -1 w D1 5.7 x 10 -6 Sinking rate of detritus of diatom origin m s -1 w D2 0 Sinking rate of detritus of flagellate origin m s -1 w Z1 1.16 x 10 -3 Sinking rate of siliceous fecal pellets m s -1 w Z2 1.16 x 10 -3 Sinking rate of non-siliceous fecal pellets m s -1 z Depth interval within the water-column h Rate of non-local exchange at depth, h, in the sediments s -1 Fraction of material grazed by copepods which is ingested Fraction of material grazed by protozoans which is ingested Fraction of material grazed by ben thic grazers which is ingested 62

PAGE 77

Table 4 (cont) Symbol Description Units a1 0.45 Respiration rate of pelagic ammonifying bacteria n1 0.45 Respiration rate of pelagic nitrifying bacteria 1 0.20 Respiration rate of pelgaic copepods grazing pelagic diatoms 2 0.20 Respiration rate of protozoans grazing pelagic microflagellates Cross-shelf curvilinear coordinate Respiration rate of pelagic diatoms Respiration rate of pelagic microflagellates Respiration rate of benthic diatoms Grazing rate of copepods on pelagic diatoms Grazing rate of protozoans microflagellates Grazing rate of benthic grazers on benthic diatoms Wavelength of radiation nm 1.45 x 10 Maximal growth rate for pelagic diatoms s -1 1.16 x 10 Maximal growth rate for pelagic microflagellates s -1 1.45 x 10 Maximal growth rate for benthic diatoms s -1 Functional group specific maximum growth rate normalized to a 24 h period at 27 C s -1 Alongshore curvilinear coordinate 6.9 x 10 -6 Dissolution of silica mmol m -3 s -1 Sigma coordinate Fraction of dying bacterial biomass which is labile 0.04 Pelagic diatom excretion rate 0.04 Pelagic microflagellate excretion rate Benthic diatom excretion rate Velocity component in the direction 63

PAGE 78

Results 1. Case 1 1.1 Spring 1998 Compared to the previous 20 year clim atological mean, the wind field over the eastern Gulf of Mexico during March-May 1998 was anomalous, featuring persistent eastward, upwelli ng favorable winds along the Panhandle of Florida (Walsh et al, 2003). Although the 1 mol kg -1 isopleth of nitrate was initially located at the shelf break in the eastern Gulf of Mexico due to the models initial conditions (Table 2), the circulation models flow fields transported nitrate to the continental shelf during the fi rst week of the simulation. In the absence of loop current forcing in this simulated scenario, the 1 mol kg -1 isopleth of nitrate penetrated to the 35 m isobath along the northern gulf coast and the southern West Florida Shelf by May 9 (Fig. 17A). In the real world, the combined effect of strong coastal upwelling, loop current im pingement at the shelf break, and light limitation of microalgae (Walsh et al., 2003) drove the 1 mol kg -1 isopleth of nitrate all the way to the 20 m isobath (Fig. 4B). 64

PAGE 79

Figure 17. The simulated near-bottom A) NO 3 (mol kg -1 ), B) NH 4 (mol kg -1 ), during 9 May 1998 in case 1 of the model. 65

PAGE 80

Figure 18. The simulated near-bottom A) PO 4 (mol kg -1 ) and B) SiO 4 (mol kg -1 ), during 9 May 1998 in case 1 of the model. 66

PAGE 81

The recycled NH 4 was elevated along the shelf-break, reaching concentrations as high as 0.8 mol kg -1 at 1 m off the bottom (Fig. 17B). These are similar to near-bottom concentrati ons measured during the NEGOM program (Walsh et al, 2003), with the exception of the Big Bend region where the model did not match elevated concentrations. Phosphate concentrations in the present simulation (Fig. 18A) were similar to the real world (Fig. 7A), where the 0.25 mol kg -1 isopleth of phosphate penetrated the 20 m isobath across the entire Eastern Gulf of Mexico. Like the nitrate, silica did not penetrate as far onto the continental shelf in the simulation as it did in the real world. Here, the simulated 2.5 mol kg -1 isopleth of silica did not penetrate to the 20 m isobat h over most of the shelf (Fig. 18B), while it made it all the way to the shore in most r egions in the real world (Fig. 5B). Unlike, previous one-dimensional si mulations of the West Florida shelf (Darrow et al, 2003) in which advective porewater exchange was not considered, the present simulation paramet erized stronger flushing of the interstitial waters of the top 4 cm of sediments. Porewater ammonium concentrations (Fig. 19B) were now less than those of the previous analysis. They were also less than those observed by Marinelli et al (1998) in the South Atlantic Bight. Phosphate (Fig. 20A) and nitrate (Fig. 19A) concentrations of the surficial sediments on the continental shelf were, however, similar to those observed at the same depths in the South Atlantic Bight. Porewater s ilicate concentrations (Fig. 20D) in the 67

PAGE 82

Figure 19. The simulated concentrations of A) NO 3 (mol kg -1 ) and B) NH 4 (mol kg -1 ) in the surface sediment layer dur ing 9 May 1998 in case 1 of the model. 68

PAGE 83

Figure 20. The simulated concentrations of A) PO 4 (mol kg -1 ) and B) SiO 4 (mol kg -1 ) in the surface sediment layer dur ing 9 May 1998 in case 1 of the model. 69

PAGE 84

Figure 21. The simulated flux (mmol m -2 day -1 ) of A) total inorganic nitrogen and B) inorganic phosphorus across the sediment water interface during 9 May 1998 in case 1 of the model. 70

PAGE 85

present simulation were significantly less than those of the previous analysis, but were consistent with observations at the 27 m isobath in the South Atlantic Bight where similar sediment permeabilities ar e found. Overall, simulated nutrient concentrations in the surface sediment pore waters (Figs. 19,20) were nearly equal to those of the near-bottom water (Fig s. 17,18) due to the rapid flushing of these highly permeable sediments. Al though flushing of the sediments was rapid, net fluxes of nutrients across the sediment/water interface were small and predominantly into the sediment (Fig. 21) rather than out of the sediments as calculated in the one-dimensional analysis and observed in the South Atlantic Bight. Large influxes of nitrogen and phosph orus were simulated near the mouth of the Mississippi river. In the real wo rld, particles are transported deeper than 4 cm into the sediments where they ar e eventually broken down to dissolved organic and inorganic nutrients through digenetic processes. This cycle, absent from the present model, typically leads to high concentrations of dissolved nutrients in the deep sediments relative to the water column and therefore an efflux of nutrients from the sedi ments to the bottom waters. Simulated surface chlorophyll concentra tions in the water column were approximately 0.5 g l -1 over most of the West Flori da Shelf (Fig. 22A). This is higher than concentrations observed dur ing the NEGOM program on the West Florida Shelf (Fig. 22B). Simula ted and observed surface chlorophyll concentrations were both elevated along the northern Gulf coast. Near-bottom chlorophyll concentrations in the model were similar to most of the real world observations, but t he simulation missed some of the enhanced 71

PAGE 86

Figure 22. The A) simulated and B) observed surface chlorophyll concentrations (g l -1 ) in the eastern Gulf of Mexico during 9 May, 1998. 72

PAGE 87

Figure 23. The A) simulated and B) observed near-bottom chlorophyll concentrations (g l -1 ) in the eastern Gulf of Mexico during 9 May, 1998. 73

PAGE 88

Figure 24. Simulated limiting factors in the growth of near-bottom A) diatoms and B) flagellates during 9 May, 1998. Wh ite represents light, grey represents nitrogen and black represents phosphorus. 74

PAGE 89

near-bottom concentrations along the nor thern Gulf coast. Although the simulated near-bottom chlorophyll concent rations were highest near the northern coast, reaching 1.4 g l -1 (Fig. 23A), the observed concentrations there were as high as 2 g l (Fig. 23B). While light was a limit ing factor for much of the day, the diatoms were limited mostly by nitrogen availability when light levels were at their highest (Fig. 24A). The diatoms were limited by phosphorus in some areas along the northern Gulf coast. Flagella tes were limited by phosphorus where they reached their highest concentrations on the West Florida Shelf (Fig. 24B). In lower concentrations, the fl agellates were limited by nitrogen. Simulated benthic chlorophyll on ly reached maximum stocks of approximately 5 mg m -2 over the top 2 mm of sedim ent on the West Florida Shelf and Gulf Coast by May 9 (Fig. 25A), compared to similar stocks observed by G. Vargo in April 1993 (Fig. 11A). Benthic chlorophyll stocks reached as high as 10 mg m -2 near the mouth of the Mississippi River. In the narrow band on the West Florida shelf, where benthic chlorophyll stocks were greater than 5 g l -1 the benthic diatoms were limited by light for the first 3 hours after sunrise. As light levels continued to increase, the benthic diatom growth became nitrogen limited and remained so for the next 6.5 hours (Fig 26). As light levels decreased, t he diatoms were again limited by light beginning about 3 hours before sunset. Similar patterns were observed elsewhere across the model domain, although phosphorus was the mid-day limiting factor along the northern Gulf coast. 75

PAGE 90

Figure 25. Simulated A) benthic chlorophyll stocks (mg m -2 ) integrated over the top 2 mm of sediment on 9 May, 1998 and B) their lim iting factors. White represents light, grey represents ni trogen and black represents phosphorus. 76

PAGE 91

Figure 26. Simulated growth rate (day -1 ) and limiting factors affecting the growth rate of benthic diatoms in the Flor ida Middle Grounds during 9 May, 1998. 77

PAGE 92

Figure 27. Observations of A) NO 3 (mol kg -1 ) and B) NH 4 (mol kg -1 ) in the near bottom waters of the eastern Gulf of Mexico during August 1998 from the NEGOM ( ) and ECOHAB Florida (+) programs. 78

PAGE 93

Figure 28. Observations of A) PO 4 (mol kg -1 ) and B) SiO 4 (mol kg -1 ) in the near bottom waters of the eastern Gulf of Mexico during August 1998 from the NEGOM ( ) and ECOHAB Florida (+) programs. 79

PAGE 94

Figure 29. Simulated near-bottom concentrations of A) NO 3 (mol kg -1 ) and B) NH 4 (mol kg -1 ) during 6 August, 1998. 80

PAGE 95

Figure 30. Simulated near-bottom concentrations of A) PO 4 (mol kg -1 ) and B) SiO 4 (mol kg -1 ) during 6 August, 1998. 81

PAGE 96

1.2 Summer 1998 As observed during the NEGOM and EC OHAB:Florida cruises (Fig. 27A), the simulated near-bottom algae (Fig. 35) had depleted near-bottom nitrate stocks over the continental shel f by early August 1998, with the 1 mol kg -1 isopleth now penetrating only to the 50 m isobath (Fig. 29A). On the southern West Florida Shelf, where the near-bo ttom algae was limited primarily by phosphorus (Fig. 36), the 1 mol kg -1 isopleth of nitrate still penetrated as far as the 20 m isobath (Fig. 29). Near-bottom NH 4 concentrations were elevated only in isolated areas near the mouth of the Mississippi river and Mobile Bay (Fig. 29B). The locations of elevated NH 4 coincide with field measurements of increased near-bottom ammonium during the July 1998 NEGOM crui se (Fig. 27B), but the observed NH 4 occupied a much wider area. Simulated near-bottom PO 4 concentrations (Fig. 30A) were very similar to those observed (Fig. 28A), with the 0.25 mol kg -1 isopleth penetrating nearly all the way to coast near the Alabama/Flor ida border. Elsewhere, the 0.25 mol kg -1 isopleth of PO 4 penetrated only to the 50 m isobath. In August 1998, elevated near-bottom silicate concentrations of 7.5 mol SiO 4 kg -1 (Fig. 28B) were measured nearsh ore on the West Florida Shelf, especially near the mouth of Tampa Bay. Along the northern Gulf of Mexico coast, silicate concentrations as high as 10 mol kg -1 reached the shore. The simulation reflected elevated silica concent rations from the deep water source 82

PAGE 97

Figure 31. Simulated porewater concentrations (mol kg -1 ) of A) NO 3 and B) NH 4 during 7 August, 1998. 83

PAGE 98

Figure 32. Simulated porewater concentrations (mol kg -1 ) of A) PO 4 and B) SiO 4 during 7 August, 1998. 84

PAGE 99

Figure 33. The simulated flux (mmol m -2 day -1 ) of A) dissolved inorganic nitrogen and B) dissolved inorganic phos phorus across the sediment water interface during 7 August, 1998. Negativ e numbers indicate a flux into the sediments. 85

PAGE 100

penetrating all the way to the shore near Tampa Bay and along the southern west coast of Florida (Fig. 30B), but di d not replicate the estuarine source of silicate suggested by the real world obs ervations. Simulated summer local stocks of silicate in the Big Bend region of Florida were larger than those observed. In contrast, the May silica te observations were greater than those simulated there. Simulated sediment NH 4 concentrations in August (Fig. 31B) were the same as simulated near-bottom NH 4 concentrations. Elevated stocks of ammonium near the mouth of the Mississi ppi River and southeast of Mobile Bay amounted to 6 mol NH 4 kg -1 Large fluxes of Nitrogen out of the sediments (Fig. 33A) occurred at these locations. Simu lated porewater nitr ate concentrations (Fig. 31A) also mirrored simulated near-bottom nitrate concentrations (Fig 29A) over most of the eastern Gulf of Mexico. Simulated porewater phosphate conc entrations (Fig. 32A) were also similar to their near-bottom counterparts (Fig 30A). On the whole, however, porewater phosphate concentrations were elevated nearer to the shore on the West Florida Shelf than the near-botto m phosphate concentrations. Flux of phosphate was predominantly into the sedim ents over the entire Eastern Gulf of Mexico (Fig. 33B), with the greatest influx of PO 4 near the mouth of the Mississippi River, like nitrogen. Again, t he influx of phosphate at the sediment surface is likely an artifact of th is models lack of deep sediments. The total simulated surface chlorophyll (Fig. 34A) was consistent with surface observations of chlorophyll (Fig. 34B) offshore, where concentrations 86

PAGE 101

reached 1.5 g l -1 The modeled pythoplankton stocks were not as high as those measured near the mouth of the Mississippi River and Tampa Bay. Simulated near-bottom chlorophyll (Fig. 35A) was elevated in the Southern West Florida Shelf, the Northeastern Gulf Coast and t he Northwestern Gulf Coast. Both simulated functional groups of near-bottom phytoplankton were limited primarily by phosphorus in their greatest concentra tions (Fig. 36). Light was the primary limiting factor for flagellates along the nor thern Gulf coast, while the near-bottom diatoms there managed to outcompete the flagellates under phosphorus limitation. There was no silica limitati on at any point in the simulation. Benthic chlorophyll was greatest al ong the 50 m isobath, where light limitation and nutrient reminer alization yielded chloroph yll stocks of 8.6 mg m -2 Maximum stocks of 12 mmol m -2 over the top 2 mm sedi ment were simulated near the Mississippi River (Fig. 37A). In their greatest concentrations, these benthic diatoms were limited mostly by phosphorus (Fig. 37B), like those simulated in the previous 1-D analysis (Da rrow et al 2003). Like the spring, CO 2 evaded the water column almost the entir e eastern Gulf of Mexico during the summer of 1998. 2. Case 2 After initiating an artificial bloom on 1 April 1998 to simulate the effects of recycled nutrients from a surface phyt oplankton bloom on the benthos, the simulated surface chlorophyll in case 2 reached maximum levels of 1.7 g l -1 on 87

PAGE 102

Figure 34. Cumulative surface ch lorophyll concentrations (g l -1 ) A) from diatoms and flagellates simulated in case 1 of the model and B) observed during the NEGOM ( ) and ECOHAB (+) cruises of August 1998. 88

PAGE 103

Figure 35. Cumulative near-bottom chlorophyll concentrations (g l -1 ) A) from diatoms and flagellates simulated in case 1 of the model and B) observed during the NEGOM ( ) and ECOHAB: Florida (+) cruises of August 1998. 89

PAGE 104

Figure 36. The simulated factors limiting the growth of A) diatoms and B) flagellates in the near-bottom waters of case 1 of the model during 7 August, 1998. White represents light, grey r epresents nitrogen and black represents phosphorus. 90

PAGE 105

Figure 37. Simulated A) Benthi c chlorophyll stocks (mg m -2 ) integrated over the top 2 mm of sediment dur ing 7 August 1998 and B) th e factors limiting their growth. White represents light, grey represents nitrogen and black represents phosphorus. 91

PAGE 106

15 April 1998 (Fig. 38A). By contrast, maximum simulated surface chlorophyll stocks on 15 April 1998 in case 1 of the model were only 0.765 g l -1 (Table 5). From the site of the artificial bloo m, surface phytoplankton drifted eastnortheastward during the 14 day period, resulting in enhanced chlorophyll concentrations near the shore relative to case 1 of the model (Fig 39). Some chlorophyll was present in the near-bottom waters through the sinking of live diatoms such that phy toplankton grew there off the recycled nutrients from sinking detritus and fecal pellets. Alt hough the maximum chlorophyll stocks in the near-bottom waters of the West Flori da Shelf were not different from case 1 (Table 5), simulated near-bottom chlorophyll stocks, gr eater than 0.5 g l -1 covered a larger area than in case 1 of the model (Fig. 40) by 15 April, 1998. Near-bottom phytoplankton in case 2 continued to thrive a week later and a small near-bottom chlorophyll plume extended northwestward, reaching the 20-m isobath by 23 April, 1998 (Fig. 41) in case 2. The simulated near-bottom phytopl ankton assemblage on the West Florida Shelf during 23 April, 1998 in bot h cases of the model was evenly mixed between diatoms and flagellates, with flagellates more prevalent near shore and diatoms more prevalent offshore. The near shore extent of diatom domination was greater, however, during case 2 of the model. Diatoms constituted more than 50% of the total phytoplankton assemblage nearly all the way to the coast in this case of the model, in contrast to case 1 where the diatom domination extended only to the 10-m isobath (Fig 42). In both cases of the model, the nearbottom diatoms were limited by nitrogen availability and the near-bottom 92

PAGE 107

Figure 38. Simulated total surface chlorophyll (g/l) from diat oms and flagellates during 1 April, 1998 in the A)case 2 and B) case 1 of the model. 93

PAGE 108

Figure 39. Simulated total surface chlorophyll (g/l) from diat oms and flagellates during 15 April, 1998 in the A) case 2 and B) case 1 of the model. 94

PAGE 109

Figure 40. Simulated total near-bottom chlorophyll (g/l) from diatoms and flagellates during 15 April 1998 in A) case 2 of the model and B) case 1 of the model. 95

PAGE 110

Figure 41. Simulated total near-bottom chlorophyll (g/l) from diatoms and flagellates during 23 April 1998 in A) case 2 of the model and B) case 1 of the model. 96

PAGE 111

Figure 42. Simulated phytoplankton dominance in the near-bottom waters of the eastern Gulf of Mexico during 23 April 1998 in A) case 2 of the model and B) case 1 of the model. Grey indicates t hat diatoms were the dominant functional group, while white indicates that flagellates were the dominant functional group. 97

PAGE 112

Figure 43. The simulated factors limiting the growth of A) diatoms and B) flagellates in the near-bottom waters of the eastern Gulf of Mexico during 23 April, 1998 in the case 2 of the model. White represents li ght, grey represents nitrogen and black represents phosphorus. 98

PAGE 113

Figure 44. Simulated benthic ch lorophyll stocks (mg m -2 ) integrated over the top 5 mm of sediment during 23 April, 1998 in the A) case 2 and B) case 1 of the model. 99

PAGE 114

Table 5. Comparison of the models simula ted chlorophyll stocks during the 2 model cases. Case 1 Case 2 WFS Max Surface Chl ( g l -1 ) on April 1 0.651 1.63 WFS Max Surface Chl ( g l -1 )on April 15 0.765 1.53 WFS Max Surface Chl ( g l -1 ) on April 23 0.765 0.798 WFS Max Near-Bottom Chl ( g l -1 ) on April 15 0.649 0.767 WFS Max Near-Bottom Chl ( g l -1 ) on April 23 0.735 0.757 WFS Max Benthic Chl (mg m -2 ) on April 15 6.99 6.99 WFS Max Benthic Chl (mg m -2 ) on April 23 8.42 8.42 100

PAGE 115

flagellates, which are more efficient nitrogen users were limited by phosphorous availability (Fig 43). Unlike previous studies (Darrow et al 2003), where a dump of diatoms led to increased benthic chlorophyll concentration s following a two week phytoplankton bloom, benthic chlorophyll concentrations did not differ from case 1 following the simulated surface bloom in case 2 (Fig 44). Near-bottom nutrient concentrations and porewater nutrient concentrations in case 2 of the model were also the same as case 1. 101

PAGE 116

Discussion The consistency of the simulat ed nutrient profiles produced by the calculated non-local exchange rates (Figs 14-16) with nutrient profiles observed on the SAB (Jahnke et al 2005) suggests that the present methodology for determining the non-local exchange rate is viable for future simulations. Indeed, calculated non-local excha nge rates ranged from 0-3 day -1 at the sediment surface compared to the exchange rate of 1.5 day -1 required to match observed sediment Si(OH) 4 profiles on the SAB (Jahnke et al, 2005). By using a non-local exchange parameterization to represent the exchange of particles and solutes across th e sediment/water interface, one is essentially turning a local 3 dimensional process into a 1 dimensional process. This is not a problem for the water-column because particles and solutes are rapidly mixed (relative to the 6 minut e time interval of the model) on the horizontal length scales of the porewater advective processes. Concentrations of any given chemical species can be assumed to be fairly homogeneous across any given sigma layer below the multi-kilometer scale grid squares of the model. Such an assumption is not valid in the sediments, where horizontal variability occurs on the scale of millimeters. By contrast, the 102

PAGE 117

underlying sediment environment is likely quite heterogeneous across the same grid squares as evidenced by variability in replicate samples of chlorophyll and nutrients (Fig 7., Table 1). Thus, the present parameterization of porewater advection as a non-local exchange process should not be used to make conclusions about the sediment environment itself. While it is likel y a good estimation of the exchange of dissolved chemical species across the sediment/water interface, it is not an accurate depiction of the process of porewater advection itself, particularly with the respect to the transport of particles to the deep sediments. A true coupling of the sediments to the water column would likely require a much more detailed and deeper, millimeter scale model of the sediment environment. Such inadequacies in the resolution of the sediment model may be responsible for this models failure to replicate high concentrations of benthic chlorophyll as previously simulated on the West Florida Shelf (Darrow et al 2003) and observed on the South Atlantic Bight (N elson et al 1999). On the continental shelf off the coast of North Carolina, benthic diatom stocks have been observed as high as 41 mg m -2 on the crests of sand ripples (Cahoon et al., 1990). The diatoms are typically not present, or present in much lower concentrations in the troughs of the sand ripples. The la ck of particulate transport to the deep sediments in the present model likely l ed to a significant underestimation of the overall nutrient flux from the water colu mn to the sediments, thus leading to nutrient limitation of the si mulated benthic diatoms. 103

PAGE 118

The flume experiments of Huettel et al (1996) show that water tends to enter the sediments at t he base of a sand ripple due to the pressure gradient generated by flow over the sand ripple. Pr essure is lowest at the crest of the ripple, resulting in an upwelling of deep nutrient rich pore waters out of the sediments at the crest of the ripple. This mechanism of pore water nutrient upwelling is consistent with the patterns of diatoms observed by Cahoon et al. (1990) off the coast of North Carolina. Be cause the current simulation is unable to resolve the horizontal scale of such ripples, the simulated benthic diatom populations are more representative of av erage concentrations over the a given grid space, rather than populations that might actually be observed at any given point in that particular grid space. In fact, average simulated concentrations of benthic chlorophyll in both cases of the model were similar to thos e observed on the West Florida shelf by G. Vargo in April and August 1993 (Fig. 11) and during July 2000 (Table 1). However, far higher average concentrations have been measured by G. Vargo in July and October 1992 (Fig 10) and August 2000 (Table 1) and also by J. Nelson in November 2001 (Table 1). Some sort of seasonal variabilit y in the overlying water column must lead to increased chlo rophyll in the benthos, but this model was unable to replicate any such conditions. 1998 and 1993 were, however, similar years in terms of freshwater nutrient input. Drought conditions existed over the southeastern United States (Lott, 1994)) and freshwater from the Mi ssissippi River did not reach the West Florida Shelf for another 3 months follo wing the coastal production cruises during 104

PAGE 119

which chlorophyll concentrations in the benthos were sampled (Weisberg, 1994). Thus, the model did at least match benthic algal growth observations on the West Florida Shelf in similar conditi ons lacking freshwater nutrient input. The highest simulated benthic chlo rophyll stocks occurred near the mouth of the Mississippi River (Fig. 18) w here the nutrient lim ited benthic diatoms benefited from fluxes of inorganic nitr ogen and phosphorus into the sediments (Fig 21). In contrast to t he chlorophyll stocks of > 30 mg m -2 simulated in the previous 1-D analysis (Darrow et al. 2001), the maximum chlorophyll stocks integrated over the top 5 mm of sediments in any case of this simulation were ~ 12 mg m -2 The benthic diatoms at the chlo rophyll maxima in both models were limited primarily by phos phorus availability (Fig. 25) and the near-bottom phytoplankton overlying the benthic chlorophyll maxima were light limited (Fig. 24). The benthic diatoms were also light limited to some extent in this area, perhaps explaining why they did not reach the levels simulated in the 1-D analysis, despite high porewater nutrient oncentrations. In the second case of the model, an artificial surface phytoplankton bloom was mimicked to represent the chlorophy ll plumes that grow in the riverine discharge of the Mississippi and Apalachi cola Rivers and are transported to the West Florida Shelf during the springtime when the prevailing winds favor eastward surface circulation (Gilbes et al., 1996). The present 3-D simulation was unable to match previous 1-D result s (Darrow et al, 2003) for benthic algal growth following the decline of such a surface phytoplankton bloom. There are a few possible reasons: 105

PAGE 120

The 1-D simulation provided no mec hanism for particles to enter the sediments. Any particle sinking through the water column was trapped in the deepest vertical layer of the model where it remained until it was consumed or dissolved. This artificial trapping mechanism in the model led to an equally artificial buildup of nutrients in the deepest vertical layer of the water column and a very large concentration gradient acro ss the sediment/water interface. Furthermore, only one sediment layer was represented in the 1-D simulation. So, rapid diffusion of nutrients into that la yer early in that simulation led to a very nutrient rich environment for the benthic algae. In the present simulation, particles were actively transported to multiple layers of sediment through pore water adv ection processes. There was neither a trapping effect in the nearbottom water column, nor in the surficial sediments. Also, in the present, 3-D simulation parti cles were mixed horizontally as they sank, diluting their concentrations as they were transported toward the seafloor. The biggest reason that the benthic diatoms in case 2 of the present model did not experience the rapid growth simulated during the 1-D analysis is that the near-bottom phytoplankton grew instead. Although there was an unrealistic buildup of nutrients in the near-b ottom waters of t he 1-D simulation (Darrow et al., 2003), the simulated nea r-bottom diatoms were unable to use those nutrients because they were light limit ed. The fallout of the surface bloom instead made it to the sedi ments where the shade adapt ed benthic diatoms grew on the recycled nutrients. In the present simulation, the near-bottom 106

PAGE 121

phytoplankton were not light limited, so t hey grew on the fallo ut of the surface bloom, using up the nutrients before they could be transpor ted to the sediments. The lack of near-bottom light limitation in the present simulation can chiefly be explained by the treatment of dissolved organic matter. In the 1-D analysis, 50% of all CDOM was consi dered to be colored and thus light absorbing (Darrow et al., 2003). Based on re search that was not available at the time of the 1-D model construction, the present simulation assumes 25% of all CDOM to be colored and thus light absor bing (Benner, 2002). Therefore, under similar surface irradiance conditions, more light made it to the near-bottom water in the present simulation than in the previous 1-D analysis. In the real world, diagenetic processes in the deep sediments result in enhanced nutrient concentrations there. Once these nutrients are upwelled as water flows over sediment ripples, t hey likely enhance benthic microalgal growth at the crests of the ripples The present simulation, testing the role of nutrients from the water column in enhancing benthic microalgal populations, assumes sediments are only 4 cm deep over the entire model grid. Such an assumption is unrealistic. In fact, some of the deepest sediments on the West Florida Shelf are located just northwest of Tampa Ba y (Hafen, 2001) where the November 2000 observations were made and also in the Florida Middle Grounds where the August 2000 observations were made. In any case, deep sediments and their interactions with surficial sediments we re not adequately modeled in the present simulation, so in that regard, a potentia lly important nutrient source for the benthic diatoms was ignored. 107

PAGE 122

Conclusions Although several observations show that benthic microalgae grow on the West Florida Shelf, it is still difficult to determine what causes variability in benthic microalgal populations over time. Clearly, nutrient concentrations in the shallow sediments of this model were not adequate to produce the high benthic chlorophyll concentrations observed in the field. Given the rapid flushing of most continental shelf sediments in the simulated Eastern Gulf of Mexico, pore water nutrient concent rations are often the same as nutrient concentrations in t he near-bottom waters. In nature, these gradients are typically steeper and nutri ents coming from the deeper sediments are more likely available to algae livin g in the surface sediments. In deeper waters, where sediments are not as rapidl y flushed, insufficient light exists to grow benthic microalgae. Nutrients are, however, efficiently transferred to the benthos on the continental shelf upon the decline of a surface phytoplankton bloom. Whether those nutrients are used up by near-bottom phytoplankton or make it all the way to the sediments is dependent on the amount of light reaching the bottom. There exists an irradiance window where near-bottom phytoplankton are light limited 108

PAGE 123

and the more shade adapted benthic algae are no t. It is within this window that benthic algae have the opportunity to us e nutrients from the overlying water column to grow whether they are new, upwelled nutrients or recycled nutrients from a dying surface bloom. 109

PAGE 124

References Admiraal, W. 1977. Experiments with mi xed populations of benthic estuarine diatoms in laborator y microecosystems. Botanica Marina 20, 479-485. Amon, R.M.W. and R. Benner. 1996. Bacterial utilizat ion of different size classes of dissolved organic matter. Limnology and Oceanography 41, 41-51. Azam, F., T.Fenchel, J.G. Field, J.S. Gr ay, L.A. Meyer-Reil and F. Thingstad. 1983. The ecological role of water-column microbes in the sea. Marine Ecology 10, 257-263. Banse, K. 1992. Grazing, temporal c hanges of phytoplankton, and the microbial loop in the open sea. Primary Productivity and Biogeochemical Cycles in the Sea ed: P.G. Falkowski and A.D. Woodhead. Plenum Press. Bates, J. D. 1963. Heavy mineral roconnaissance Florida west coast. Economic Geology 58, 1237-1245. Benner, R. 2002. Chemical composition and reactivity. In: Biogeochemistry of Marine Dissolved Organic Matter eds: D. Hansell and C. Carlson. Academic Press. Bidigare, R.R., M.E. Ondrusek, J.H. Morrow, and D.A. Kiefer. 1990. In vivo absorption properties of algal pigments. SPIE 1302, 290-302. Birdsall, B. C. 1977. Eastern Gulf of Mexico, Continental Shelf Phosphorite Deposits. Masters thesis. University of South Florida. Tampa, FL. 87 pages. Blumberg, A.F. and G.L. Me llor. A description of a three dimensional coastal ocean circulation model. In Threedimensional Coastal Ocean Models, ed N. Heaps, AGU, Wash ington D.C., pp 208-233, 1987. 110

PAGE 125

Bolden, S. 1990. Abundance, diet and foraging migrat ions of the tomtate (Haemulon aurolineatum) on an artificial and a natural reef in Onslow Bay, North Carolina. Masters thesis. Univer sity of North Carolina, Wilmington. pages. Boudreau, B.P. 1997. Diagenetic Models and their Implementation 417p. Springer-Verlag. Brooks, H. K. 1973. Geological oceanography. In: A Su mmary of Knowledge of the Eastern Gulf of Mexico, ed. Jones, J. I., R. E. Ring, M. O. Rinkel and R. E. Smith, State Univeristy Syst em of Florida Inst itute of Oceanography, St. Petersburg. IIE1 IIE49. Cahoon, L. B. and J. E. Cooke. 1992. Benthic microalgal production in Onslow Bay, North Carolina, USA. Marine Ecology Progress Series 84, 185-196. Cahoon, L. B. and C. R. Tronzo. 1992. Quantitative estimates of demersal zooplankton abundance in Onslow Bay, North Carolina, USA. Marine Ecology Progress Series 87, 197-200. Cahoon, L. B., R. S. Redman and C. R. Tronzo. 1990. Benthic microalgal biomass in sediments of On slow Bay, North Carolina. Estuarine and Coastal Marine Science 31, 805-816. Carder, K.L., R.G. Steward, G.R. Harvey and P.B. Ortner. 1989. Marine humic and fulvic acids: Their effects on remote sensing of ocean chlorophyll. Limnology and Oceanography 34, 68-81. Carlson C.A. 2002. Production a nd Consumption Processes. In: Biogeochemistry of Dissolved Organic Matter in the Ocean eds. D.A. Hansell and C.A. Carlson. Acade mic Press, San Diego. Pp. 91-151 Conmy, R. N. and P. G. Coble. 2002. Changes in the optical properties of colored dissolved organic matter in coas tal regions of the Gulf of Mexico between the Mississippi River and Florida Bay. AGU/ASLO conference abstract Darrow, B. P. 2002. Adventures in modeling: The West Florida benthic diatom massacre. XHAB 2002 Conferance Abstract Darrow, B. P., J. J. Walsh, G. A. Vargo, R. T. M. Jr., K. A. Fanning and J.-Z. Zhang. 2002. A simulation study of the growth of benthic microalgae following the decline of a su rface phytoplankton bloom. Continental Shelf Research 23, 1265-1283. 111

PAGE 126

Davis, C.C. 1948. Gym nodinium brevis pp. nov., A cause of discolored water and animal mortality in the Gulf of Mexico. Botanical Gazette. 109, 358360. De Jonge, V. N. and J. v. d. Bergs. 1987. Experiments on the resuspension of estuarine sediments cont aining benthic diatoms. Estuarine, Coastal and Shelf Science 24, 725-740. De Jonge, V. N. and J. E. E. V. Beusekom. 1995. Wind and tide induced resuspension of sediment and microphyt obenthos from tidal flats in the Ems estuary. Limnology and Oceanography 40, 766-778. Del Castillo, C. E., F. G ilbes, P. G. Coble and F. E. Muller-Karger. 2000. On the dispersal of riverine colored disso lved organic matter over the West Florida shelf. Limnology and Oceanography 45, 1425-1432. Delgado, M., V. N. D. Jonge and H. Peleti er. 1991. Effect of sand movement on the growth of benthci diatoms. Journal of Experimental Marine Biology and Ecology 145, 221-231. Del Giorgio, P.A. and J.J. Cole. 2000. Bacterial energetics and growth efficiency. In: Microbial Ecology of the Ocean Ed: D.L. Kirchman. WileyLiss 289-324. Dortch, Q. and T. E. Whit ledge. 1991. Does nitrogen or silicon limit phytoplankton production in the Mississippi Ri ver plume and nearby regions? Continental Shelf Research 12, 1296-1309. Ducklow, H. 2000. Bacterial producti on and biomass in the oceans. In: Microbial Ecology of the Oceans ed. D. Kirkman. Wiley Liss. El-Sayed, S. Z., W. M. Sackett, L. M. Jeff rey, A. D. Fredricks, R. P. Saunders, P. S. Conger, G. A. Fryxell, K. A. St eidinger and S. A. Earle. 1972. Chemistry, primary productivity, and benthic algae of the Gulf of Mexico. Serial Atlas of the Marine Environment, Folio 22. American Geographic Society. Eppley, R.W. 1972. Te mperature and phytoplankton growth in the sea. Fishery Bulletin 70, 1063-1085. Eriksson, P.G. and S.E.B. Weisner. 1999. An experimental study on effects of submersed macrophytes on nitrification and denitrification in ammoniumrich aquatic systems. Limnology and Oceanography 44, 1993-1999. 112

PAGE 127

Fanning, K. A., K. L. Carder and P. R. Betzer. 1982. Sediment resuspension by coastal waters: a potential mechanis m for nutrient re-cycling on the ocean's margins. Deep-Sea Research 29, 953-965. Fanning, K.A., R.H. Byrne, J.A. Breland, and P.R. Betzer. 1981. Geothermal springs of the West Florida continental shelf: evidence of dolomitization and radionuclide enrichment. Earth and Planetary Science Letters 52, 345-354. Gilbes, F., C. Tomas, J. J. Walsh and F. E. Muller-Karger. 1996. An episodic chlorophyll plume on the West Florida shelf. Continental Shelf Research 16, 1201-1224. Gordon, H.R., O.B. Brown, R.H. Evans, J.W. Brown, R.C. Smith, K.S. Baker, and D.K. Clark. 1988. A semianalytic radiance model of ocean color. Journal of Geophysical Research 93, 10909-10924. Gregg, W.W. and K.L. Carder. 1990. A simple spectral solar irradiance model for cloudless maritime atmospheres. Limnology and Oceanography. 35: 1657-1675. Hafen, M. 2001. West Florida Shelf Sand Waves Doctoral dissertation. University of South Florida. Hedges, J. 2002. Why dissolved organics matter. In: Biogeochemistry of Marine Dissolved Organic Matter eds: D. Hansell and C. Carlson. Academic Press. He, R. and R. H. Weisberg. 2003. A Loop Current intrusion case study on the West Florida shelf. Journal of Physical Oceanography 33, 465-477. Hu, C., F.E. Muller-Karger and P.W. Sawr zenski. 2006. Hurricanes, submarine groundwater discharge and Floridas red tides. Geophysical Research Letters 33, L11601. Huettel, M. and G. Gust. 1992. Impact of bioroughness on interfacial solute exchange in permeable sediments. Marine Ecology Progress Series 89, 253-267. Huettel, M., W. Ziebis and S. Forster. 1996. Flow-induced uptake of particulate matter in permeable sediments. Limnology and Oceanography 41, 309322. Humm, H. J. 1957. Rediscovery of A nadyomene menziesii, a deep-water green alga from the Gulf of Mexico. Bulletin of Marine Science Gulf Caribbean 6, 346-348. 113

PAGE 128

Hutchinson, G.E. 1961. The Paradox of the Plankton. American Naturalist. 95, 137-145. Iverson, R. L. and H. F. Bittaker. 1986. Seagrass distribution and abundance in Eastern Gulf of Mexico coastal waters. Estuarine, Coastal and Shelf Science 22, 577-602. Jahnke, R. A., J. R. Nelson, R. L. Marinel li and J. E. Eckman. 2000. Benthic flux of biogenic elements on the southeastern US continental shelf: influence of pore water advective tr ansport and benthic microalgae. Continental Shelf Research 20, 109-127. Jahnke, R, M Richards, J Nelson, C R obertson, A Rao and D Jahnke. 2005. Organic matter remineralization and porewater exchange rates in permeable South Atlant ic Bight continental shelf sediments. Continental Shelf Research 25:1433-1452. Joyce, E. A. J. and J. Williams. 1969. Rationale and pertinent data. In: Memoirs of the Hourglass Cruises, ed. Florida Department of Natural resources, St. Petersburg, FL. 1-50. Jumars, P.A., D.L. Penry, J.A. Baross, M.J. Perry, and B.W. Frost. 1989. Closing the microbial loop: dissolved carbon pathway to heterotrophic bacteria from incomplete ingestion, di gestion, and absorption in animals. Deep-Sea Research 20, 109-127. Kirk, J.T.O. 1995. Light and Photosynthesis in Aquatic Ecosystems. 511p. Cambridge University Press. Khromov, N. S. 1969. Distri bution of plankton in the Gulf of Mexico and some aspects of its seasonal dynamics. In : Soviet-Cuban Fishery Research, ed. Bogdanov, A. S., Washington, DC, USA. 36-56. Kuhl, M. and B.B. Jorgensen. 1994. The light field and microbenthic communities: radiance distributions and microscale optics of sandy coastal sediments. Limnology and Oceanography 39, 1368-1398. Lenes, J. M., B. P. Darrow, C. Cattrall, C. A. Heil, M. Callahan, G. A. Va rgo, R. H. Byrne, J. M. Prospero, D. E. Bate, K. A. Fanning and J. J. Walsh. 2001. Iron fertilization and the Trichodesmium response on the West Florida shelf. Limnology and Oceanography 46, 1261-1277. Li, M.Z. and C.L. Amos. 1995. SEDTRANS92: a sediment transport model for continental shelves. Computers and Geosciences. 21, 533-554. 114

PAGE 129

Lott, J.N. 1994. The U.S. summer of 1993. Weather 49, 370-383. Mallin, M. A., J. M. Burkholder and M. J. Sullivan. 1992. Contributions of benthic microalgae to coastal fishery yield. Transactions of the American Fisheries Society 121, 691-695. Manooch, C. S. I. 1977. Foods of t he Red Porgy, Pagrus Pagrus Linnaeus (Pisces: Sparidae), from Nort h Carolina and South Carolina. Bulletin of Marine Science 27, 776-787. Marinelli, R. L. 1992. Effects of polyc haetes on silicate dynamics and fluxes in sediments: Importance of species, animal activity and polychaete effects on benthic diatoms. Journal of Marine Research 50, 745-779. Marinelli, R. L., R.A. Jahnke, D.B. Craven, J.R. Nelson, and J.E. Eckman. 1998. Sediment nutrient dynamics on the Sout h Atlantic Bight continental shelf. Limnology and Oceanography 43, 1305-1320. Masserini, R. T. J. and K. A. Fa nning. 2000. A sensor package for the simultaneous determination of nanomolar c oncentrations of nitrite, nitrate, and ammonia in seawater by flourescence detection. Marine Chemistry 68, 323-333. Miller, D. C., R. J. Geider and H. L. MacIntyre. 1996. Microphytobenthos: The ecological role of the "secret gar den" of unvegetated, shallow-water marine habitats. II. Role in sedi ment stability and shallow-water food webs. Estuaries 19, 202-212. Millero, F. J., A. G. Dickson, G. Eischeid, C. Goyet, P. Guenther, K. M. Johnson, R. M. Key, K. Lee, D. Pu rkerson, C. L. Sabine, R. G. Schottle, D. R. W. Wallace, E. Lewis, and C. D. Winn. 1998. Assessment of the quality of the shipboard measurements of total alkalinity on the WOCE Hydrographic Program Indian Ocean CO2 survey cruises 1994. Marine Chemistry 63, 9. Millie, D. F., G. J. Kirkpatrick and B. T. Vinyard. 1995. Rela ting photosynthetic pigments and in vivo optical density s pectra to irradiance for the Florida red tide dinoflagellate, Gymnodinium breve. Marine Ecology Progress Series 120, 65-75. Moe, M. A. J. 1969. Biology of the Red Grouper Epinephelus Morio (Valenciennes) from the eastern Gulf of Mexico Florida Department of Natural Resources, St. Petersburg, FL. 115

PAGE 130

Mopper, K., X. Zhou, R.J. Ki eber, D.J. Kieber, R.J. Siko rski, and R. D. Jones. 1991. Photochemical degradation of dissolved organic carbon and its impact on the oceanic carbon cycle. Nature 353 60-61. Morel, A. and Y-H. Ahn. 1990. Optica l efficiency factors of free-living marine bacteria: Influence of bacterioplankt on upon the optical properties and particulate organic carbon in oceanic waters. Journal of Marine Research 48, 145-175. Morgan, M.G. and V. Dale. 2007. Hypoxia in the Northern Gulf of Mexico. An update by the EPA Science Advisory Board Environmental Protection Agency, United States of America. Muller-Karger, F. E., J. J. Walsh, R. H. Evans and M. B. Meyers. 1991. On the seasonal phytoplankton concentration and sea surface temperature cycles of the Gulf of Mexico as determined by satellites. Journal of Geophysical Research 96, 12645-12665. Mullin, M.M., E.F. Stewart, and F.J. Fug lister. 1975. Ingestion by planktonic grazers as a function of concentration of food. Limnology and Oceanography 20, 259-262. Nelson, D. M. and Q. Dort ch. 1996. Silicic acid depletion and solicon limitation in the plume of the Mississippi River: ev idence from kinetic studies in the spring and summer. Marine Ecology Progress Series 136, 163-178. Nelson, J. R., J. E. Eckman, C. Y. Rober tson, R. L. Marinelli and R. A. Jahnke. 1999. Benthic microalgal biomass and i rradiance at the sea floor on the continental shelf on the South Atl antic Bight: Spatial and temporal variability and storm effects. Continental Shelf Research 19, 477-505. Nelson, J.R. and C.Y. Roberston, 1993. De trital spectral absorption: Laboratory studies of visible light effects on phytodetritus absorption, bacterial spectral signal, and comparison to field measurements. Journal of Marine Researc 51, 181-207. Nordlie, F. G. 1990. Rivers and springs. In: Ecosystems of Florid a, ed. Myers, R. L. and J. J. Ewel, Unviersity of Ce ntral Florida Press, Orlando. 393-425. Ogawa, H., Y. Amagai, I. Koike, K. Kais er, and R. Benner. 2001. Production of refractory dissolved organi c matter by bacteria. Science 292, 917-920. Pakulski, J.D., R. Benner, T. Whitledge, R. Amon, B. Eadie, L. Cifuentes, J. Ammerman and D. Stockwell. 2000. Microbial metabolism and nutrient cycling in the Mississippi and Atchafalaya river plumes. Estuarine, Coastal and Shelf Science 50, 173-184. 116

PAGE 131

Paluszkiewicz, T., L. P. Atkinson, E. S. Posmentier and C. R. McClain. 1983. Observations of a loop current frontal eddy intrusion onto the West Florida Shelf. Journal of Geophysical Research 88, 9639-9651. Peng, T., T. Takahashi, W.S. Broeker and J. Olasfsson. 1987. Seasonal variability of carbon dioxide, nutrient s and oxygen in the northern North Atlantic surface water: observations and a model. Tellus 39B, 439-458. Pope, R.M. and E.S. Fry. 1997. Absorption spectrum (380-700 nm) of pure water. II. Integrati ng cavity measurements. Applied Optics 36, 87108723. Reid, J. A.; Jenkins, C. J.; Field, M. E.; Gardner, J. V.; Zimmermann, M.; Box, C. E.; Kneeshaw, T. A. 2001. usSEABED: Database Efforts in Marine Surficial Sediments of the US EEZ. American Geophysical Union, Fall Meeting 2001 abstract #OS11B-0367 Riedl, R.J., Machan, R., 1972. Hydrodynamic patterns in lotic intertidal sands and their bioclimatological implications. Marine Biology 13, 179-209. Riedl, R.J., Huang, N., Machan, R., 1972. The subtidal pump: a mechanism of interstitial water exchange by wave action. Marine Biology 13, 210-221. Roesler, C.S., M.J. Perry, and K.L. Carder. 1989. M odeling in situ phytoplankton absorption from total absorption spectra in productive inland marine waters. Limnology and Oceanography 34, 1510-1534. Ryder, P. D. 1985. Hydrology of the Floridan Aquifer System in West-Central Florida. United States Geological Survey Prof essional Paper 1403-F. Washington, DC. Schirripa, M. J., C. M. Legault and M. Ortiz. 1999. The Red Grouper fishery of the Gulf of Mexico: A ssessment 3.0. Southeast Fi sheries Science Center, Sustainable Fisheries Division. Miami, FL. Sinclair, W. C., J. W. Stewart, R. L. Knut illa, A. E. Gilboy and R. L. Miller. 1985. Types, features, and occurence of si nkholes. United States Geological Survey water-resources report 85-4126. Tallahassee, FL. Smith, R.C. and K.S. Baker. 1981. Optical properties of the clearest natural waters (200-800 nm). Applied Optics 20, 177-184. Smith, S. M. and G. L. Hitchcock. 1994. Nutrient enrichments and phytoplankton growth in the surface waters of the Louisiana Bight. Estuaries 17, 740753. 117

PAGE 132

Sundback, K. and W. Graneli. 1988. Influence of microphytobenthos on the nutrient flux between sediment and water: a laboratory study. Marine Ecology Progress Series 43, 63-69. Thomas, C. J. and L. B. Cahoon. 1993. Stable isotope analyses differentiate between different trophic pathways supporting rocky-reef fishes. Marine Ecology Progress Series 95, 19-24. Vargo, G. A., K. L. Carder W. Gregg, E. Shanley, C. He il, K. A. Steidinger and K. D. Haddad. 1987. The potential contribution of primary production by red tides to the West Florida shelf ecosystem. Limnology and Oceanography 32, 762-767. Walker, S.T. 1884. Fish morta lity in the Gulf of Mexico. Proc U.S. Natl Mus 6, 105-109. Walsh, J.J. and D.A. Dieterle. 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. Progress in Oceanography 34, 335-392. Walsh, J. J. and K. A. St eidinger. 2001. Saharan dust a nd Florida red tides: the cyanophyte connection. Journal of Geophysical Research 106, 1159711612. Walsh, J. J., R. H. Weisberg, D. A. Dieterl e, R. He, B. P. Darrow, J. K. Jolliff, K. M. Lester, G. A. Vargo, G. J. Kirkpatrick, K. A. Fanning, T. T. Sutton, A. E. Jochens, D. C. Biggs, B. Nababan, C. Hu and F. E. Muller-Karger. 2003. The phytoplankton response to intrus ions of slope water on the West Florida shelf: models and observations. Journal of Geophysical Research. 108, 3190-3213. Walsh, J.J., J.K. Jolliff, B. P. Darrow, J.M. Lenes, S.P. M ilroy, D.A. Dieterle, K.L. Carder, F.R. Chen, G.A. Vargo, R.H. Weisberg, K. A. Fanning, F.E. MullerKarger, K.A. Stedinger, C. A. Heil, C.R. Tomas, J.S. Prospero, T.N. Lee, G.J. Kirkpatrick, T.E. Whit ledge, D.A. Stockwell, T.A. Villareal, A.E. Jochens, and P.S. Bontempi. 2006. Red Tides in the Gulf of Mexico: Where, when, and why? Journal of Geophysical Research 111, 10031065. Weisberg, R.H. 1994. Transport of Mississi ppi River water to the west Florida shelf. In: Special NOAA Report-Coasta l Oceanographic Effects of Summer 1993 Mississippi River Floodin. EdL M.J. Dowgiallo. USDOC/NOAA Coastal Ocean Office/National Weather Service. 118

PAGE 133

Brian Darrow has been intere sted in computers since an early age. He wrote his first working computer program at age 9, beginning a lifelong j ourney as a self taught software engineer. Prior to hi s doctoral work in computer modeling marine ecosystems, Brian received a Bachelor of Science degree in biology from St. Louis University and a Master of Sc ience degree in Marine Science from the University of South Florida. Dubbed an honorary parrothead for his work studying the scum on the bottom of the ocean, Brian has thus far participated in 17 coastal research cruises and logged over 75 days at sea. He is the recipient of several awards and fellowships, incl uding the Gulf Oceanographic Charitable Trust Fellowship in Coastal Science.