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 002001394
007 cr mnu|||uuuuu
008 090430s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002635
Anderson, Kelley L.
Can waters around Durney Key, Pasco County, Florida, support coral recruitment to artificial substrates?
h [electronic resource] /
by Kelley L. Anderson.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 105 pages.
Thesis (M.S.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: To determine whether an artificial reef installation is feasible, there must be a thorough characterization of the habitat. An understanding of both small-scale and large-scale environmental processes is needed to determine factors that potentially will influence the reef. Large-scale processes include coastal circulation, wave climate, and sediment dynamics that take place over spatial scales of tens to hundreds of kilometers in the region of the reef. Small-scale processes include the physical characteristics in the immediate vicinity of the reef the local current, wave and tide characteristics, temperature, salinity, and suspended and bottom sediments at a proposed reef site. The city of Port Richey, Florida, was considering installing an artificial reef of porcelain modules near Durney Key, a dredge spoil island just offshore.To assist in determining the feasibility of this proposal, I pursued three objectives: a) to characterize the oceanographic setting of Durney Key, including hydrodynamics, water quality and invertebrate biota; b) to investigate the potential for successful coral recruitment and growth in Durney Key waters; and c) to determine if porcelain is a suitable substrate for settlement of the larvae of coral species present in west central Florida. An array of Acoustic Doppler Current Profilers (ADCPs) were used to measure water velocity, water stage and temperature around Durney Key. ADCP data showed currents around Durney Key are tidally dominated with velocities increasing in winter due to frontal passages. Seasonal stage variation ranges from 0.29 m (11.4 in) to 0.64 m (2.1 ft) and seasonal temperature ranged from 10Â¨C and 35Â¨C for winter and summer, respectively.Atmospheric data from the Port Richey COMPS site showed average wind speeds were higher in winter (3.7 m/s or 12.4 ft/s) than summer (3.1 m/s or 10.2 ft/s), corresponding to increased average water velocities. Inorganic nutrients, salinity and pH were measured and compared to data from patch reefs in the Florida Keys to characterize the water quality and determine its suitability for coral recruitment and growth. Compared to Florida Keys patch reef waters, Durney Key water salinity averaged 12 parts per thousand (ppt) lower, pH was more variable with a lower minimum, and total phosphorus was much higher. Ceramic and porcelain recruitment tiles deployed to investigate larval recruitment were colonized by turf, coralline and macroalgae, with barnacles recruiting secondarily. Sediment cores revealed foraminiferal and molluscan assemblages characteristic of productive estuarine conditions.The Durney Key area was deemed not suitable for coral recruitment and growth on an inshore artificial substratum because of temperature extremes, potential for minimal water movement during summer, frequent occurrences of low salinity and pH, and high total organic phosphorus. Faunal studies demonstrated that the dominant recruitment reflects the common coastal/estuarine biota, which does not include reef-building corals.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Pamela Hallock Muller, Ph.D.
x Marine Science
t USF Electronic Theses and Dissertations.
Can Waters Around Durney Key, Pasco County, Florida, Support Coral Recruitment to Artificial Substrates? by Kelley L. Anderson A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: Pamela Hallock Muller, Ph.D. Sekeenia Haynes, Ph.D. Walter Jaap, B.S. Mark Luther, Ph.D. Date of Approval: November 10, 2008 Keywords: estuary, Pithlachascotee, recruitment, ADCP, artificial reef, benthic, sediments Copyright 2008, Kelley L. Anderson
Acknowledgments Caring and dedicated teachers are a rare treasure, which I why I am deeply thankful to have been advised by one. My major advisor, Dr. Pamela Hallock Muller has shown u nending patience, support and encouragement throughout my graduate career. My committee members, Sekeenia Haynes, Walter Jaap and Mark Luther have all been invaluable in their advice, support and assistance. An immense thank you goes out to the Reef Indic ators Lab for their assistance and general moral support. I am also indebted to Sherryl Gilbert for her patient guidance and programming expertise, as well as Vembu Subramanian for his COMPS data assistance. This work would not have been possible without the financial and vessel support of the City of Port Richey, Florida, especially the ingenuity and foresight of Jerry Calhoun. The willingness of the City to partner with graduate students for research needs is a great example of community cooperation. I sincerely thank all of the City staff for their support and assistance throughout this project. I am deeply indebted to Victor Levesque of the United States Geological Survey (USGS), without whom I could not have completed the ADCP current analysis. Hi s willingness to help both in the field and with software understanding is greatly appreciated. Clay, advice and facilities to make the recruitment tiles were donated by the St. Petersburg Clay Company, to which I am deeply grateful. Jon Fajans of SEAKey s graciously provided invertebrate taxonomic help. Dixie Hollins of Citrus Mining & Timber donated lime rock pieces for deployment studies, for which I am indebted. Water quality data were provided by the SERCFIU Water Quality Monitoring Network, which i s supported by SFWMD/SERC Cooperative Agreements #C 10244 and #C 13178 as well as EPA Agreement #X994621 94 0. I thank all my wildlife ecology professors at the University of Florida who encouraged and pushed me to pursue my dreams. My family and friend s have offered priceless support, assistance and inspiration through good times and bad I dedicate this thesis to anyone to cares enough to teach others why we should care
i Table of Contents List of Tables iii List of Figures vi Abstract viii 1. Introduction 1 Genesis of Project 2 Habitat Description 4 Objectives 7 Rational for Assessment Study 7 Phys ical Parameters 7 Chemical Parameters 9 Organic Pollutants 9 Metals 11 Biological Parameters 12 Recruitment Studies 13 Bioindicators 13 2. Methods 17 Physical Parameters 18 Chemi cal Parameters 19 Se diment Texture 22 Biological Parameters 24 Recruitment Tiles 24 Sediment Pu sh Cores 27 Lime Rock 28 FORAM Index 29 3. Results 31 Physi cal Parameters 31 Chemical Parameters 41 Water Samples 41 Pollutants in Sedimen t Sample 42 Se diment Texture 43
ii Biologi cal Parameters 45 Foraminifera 45 Moll uscan Assemblage 48 Recruitment 49 4. Discussion 55 Physi cal Parameters 55 Chemi cal Parameters 56 Water Samples 56 Pollutants in Sediment Samples 60 Sediment Texture 61 Biologi cal Parameters 61 Foramini feral Assemblage 62 Molluscan Ass emblage 64 Species Ass emblage on Tiles 67 Recruitment 68 Recommendations for Future Work 69 5. Conclusions 70 Ref erences Cited 71 App endices 85 Appendix I Physic al Data 86 Appendix II Chemi cal Data 96 Appendix III Sediment Texture 101 Appendix I V Biological Data 102
iii List of Tab les Table 1 1995 Land Use and Land Cover in the Springs Coast Basin 2 Table 2 Parameters evaluated with corresponding investigation 17 techniques Table 3 Scale size for particle sizes of sediment samples 23 Table 4 Funct ional groups with example genera as originally defined 30 by Hallock et al. (2003) and modified by Carnahan (2005) and Ramirez (2008) Table 5 Root mean squared tided and detided water velocities for 31 summer and winter currents at Durney Key Table 6 Water velocity variability between spring and neap tides in 32 summer, where V e is the eastward component and V n is the northward component Table 7 Water velocity variability bet ween spring and neap in winter; 32 abbreviations as in Table 6 Table 8 Percent velocity variation attributable to tides in summer 36 2007 Table 9 Percent of velocity variation attributable to tides in winter 36 2008 Table 10 A verage stage height in meters as rec orded by A DCPs in 38 summer 2007 and winter 2008 with stage height difference due to seasonality shown in last column Table 11 Water characteristics as measured using an Ultrameter 41 Table 12 Inorganic nutrient concentrations in w ater column at mean 42 low water where MDL = Method Detection Limit
iv Table 13 Summary of organic compounds in surficial sediment sample 43 from the south side of Durney Key, where MRL = Minimum Reporting Limit and I = the rep orted value is between the laboratory method detection limit (MDL) and the laboratory method reporting limit (MDL), adjusted for actual sample prepara tion data and moisture content Table 14 Concentrations of metals of concern as analyz ed in surfici al 43 sediment sample Table 15 Mass sediment samples remaining after dissolution with 50% 44 HCL to estimate % carbonate insoluble Table 16 Foraminiferal assemblage for sediment from scrubbed lime 45 rock Table 17 The av erage FORA M Index values for samples of sediment 46 from scrubbed lime rock with associated number of foraminiferal shells Table 18 Foraminiferal assemblage from sedim ent core samples 47 Table 19 Raw foraminiferal shell counts a nd FORA M Index values 47 with associated sediment sample characteristics for sediment core subsamples Table 20 Molluscan species as semblage pe r one gram subsample 48 Table 21 Summary stat i s tics for average cover and variability of 50 recruitment classes on tiles where C.A = Coralline algae, T.A .C. = Total A rea Covered and B.A .F. = Brown A lgal Film Table 22 Raw foraminiferal counts from ceramic recruitment tile 19 A 51 Table 23 Density of foraminifera l taxa found o n porcelain tile 19 A per 52 square centimeter Table 24 Raw foraminiferal counts from ceramic recruitment tile 19 B 52 Table 25 Density of foraminiferal taxa found on porcelain tile 19 B per 53 square centimeter Table 26 Specie s found on ceramic tile 19 A 53 Table 27 Species found on ceramic tile 19 B 54
v Table 28 Percent coral cover at patch reef sites where water quality 56 monitoring stations coincide with CREMP monitoring stations where WQMN = Water Quality Monitoring Network and CREMP ID = Coral Reef Ev aluating and Monitoring Project Table 29 Water characteristics in the Keys patch reef sites that 57 coincide with CREMP monitoring sites from quarterly sampling carried out d uring 2006 2007 Table 30 Summary of physical variables at Howard Frankland Reef, 58 Tampa Bay, Florida Table 31 Summary statistics for selected water quality variables in the 59 Keys patch reef sites that coincide with CREMP monitorin g sites from quarterly sampling carried out during 2007 Table 32 a Bivalve assemblage organized by feeding mode, where 66 DS = surface deposit feeder, DC = chemosymbiotic deposit feeder, and SU = suspension feeder Table 32 b Gastropod assembl age organized by feeding where mode 66 CP = predatory carnivores, CB = browsing carnivores, HR = herbivores on rock, rubble or coral substrates, HP = herbivores on plant or algal substrates, and HM = herbivores on fine grained substrates
vi List of Figures Figure 1 Figure 1 Map of Florida indicating Pasco County in red 3 with an image of Pasco County outlined in red Figure 2 Image of an EcoReef module being emplaced at Bunaken 4 National Park, Indonesia Figure 3 Durney Key with locations of various samples, where triangles 17 represent ADCP locations with blue being ADCP A, red ADCP B and orange ADCP C; circle representing location of surficial sediment co llection and numbers representing stakes of recruitment tiles Figure 4 Sieve set on shaker used in grain size analysis 23 Figure 5 Handmade porcelain recruitment plates with texturized 25 pattern Figure 6 PV C stake number one wit h recruitment tiles attached 25 Figure 7 Gridded tray with sediment sample, artist brush and water 28 Used to isolate foraminiferal and molluscan shells for identification Figure 8 Comparison of tided and detided summer north south water 33 velocities from the south side of Durney Key (ADCP A) Figure 9 Comparison of tided and detided summer north so uth water 33 velocities from the south side of Durney Key (A DCP A ) Figure 10 Comparison of tided and detided winter e ast west water 34 velocities from the so uth side of Durney Key (ADCP A) Figure 11 Comparison of tided and detided winter north south water 34 velocities from the so uth side of Durney Key (ADCP A) Figure 12 Correlation of detided east west summer 2007 water 36 velocity a nd wind for Durney A where the red trendline indicates a log regression and the black trendline indicates a linear regression
vii Figure 13 Correlation of detided north south summer 2007 water 36 vel oc ity and wind for Durney A where the red trendline indicates a log regression and the black trendline indicates a linear regression Figure 14 Correlation of detided east west winter 2008 water velocity 37 and wind for Durney A where the red tren dline indicates a log regression and the black trendline indicates a linear regression Figure 15 Correlation of detided north south winter 2008 water velocity 37 and wind for Durney A 2007 where the red trendline indicates a log regression and the black trendline indicates a linear regression Figure 16 Summer 2007 water height from the south side of Durney 39 Key (A DCP A ) where crescent moons represent neap tides and circular moons represent spring tides, the black circle r epresents the new moon and blue circle represents the full moon Figure 17 Winter 2008 water height from the south side of Durney Key 39 (A DCP A ) where crescent moons represent neap tides and circular moons represent spring tides, the black circle represents the new moon and blue circle represents the full moon Figure 18 Temperature as recorded by ADCP A in summer 2007 40 Figure 19 Temperature as recorded by ADCP A in winter 2008 40 Figure 20 Grain size distribution by weight percent of sample for 44 sediment cores ; core 1 wa s collected 10 m from shore, Core 2 at 20 m and Core 3 at 30 m from shore. Figure 21 Average coverage area per class for each time step grouped 50 by position on the stake, i.e top vs. bottom Figure 22 MDS plot displaying the degree of similarity between bottom 51 and top tile positions with all ti me steps included, with both brown algal film and total area covered classes removed Figure 23. Example of a deformed Miliolinella from ceramic tile 19A 53
viii Can Waters around Durney Key, Pasco County, Florida, Support Coral Recruitment to Artificial Substrates? Kelley L. Anderson ABSTRACT To determine whether an artificial reef installation is feasible, there must be a thorough characterization of the habitat. An understanding of both small scale and large scale environmental processes is needed to determine factors that potentially will influence the reef. L arge scale processes include coastal circula tion, wave climate, and sediment dynamics that take place over spatial scales of tens to hundreds of kilometers in the region of the reef. Small scale processes include the physical characteristics in the immediate vicinity of the reef the local current wave and tide characteristics, temperature, salinity, and suspended and bottom sediments at a proposed reef site. The city of Port Richey, Florida, was considering installing an artificial reef of porcelain modules near Durney Key, a dredge spoil island just offshore. To assist in determining the feasibility of this proposal, I pursued three objectives: a) to characterize the oceanographic setting of Durney Key, including hydrodynamics, water quality and invertebrate biota; b) to investigate the potenti al for successful coral recruitment and growth in Durney Key waters; and c) to determine if porcelain is a suitable substrate for settlement of the larvae of coral species present in west central Florida. A n array of A coustic Doppler Current Profilers (A DC Ps) were used to measure water velocity, water stage and temperature around Durney Key. A DCP data showed currents around Durney Key are tidally dominated with velocities increasing in winter due to frontal passages. Seasonal stage variation ranges from 0 .29 m (11.4 in) to 0.64 m (2.1 ft) and seasonal temperature ranged from 10C and 35C for winter and summer, respectively. A tmospheric data from the Port Richey COMPS site showed a verage wind speeds were higher in winte r (3.7 m/s or 12.4 ft/s) than summer (3.1 m/s or 10.2 ft/s ), corresponding to increased average water velocities. Inorganic nutrients, salinity and pH
ix were measured and compared to data from patch reefs in the Florida Keys to characterize the water quality and determine its suitability for coral recruitment and growth. Compared to Florida Keys patch reef waters, Durney Key water salinity averaged 12 parts per thousand (ppt) lower, pH was more variable with a lower minimum, and total phosphorus was much higher. Ceramic and porcelain recruitm ent tiles deployed to investigate larval recruitment were colonized by turf, coralline and macroalgae, with barnacles recruiting secondarily. Sediment cores revealed foraminiferal and molluscan assemblages characteristic of productive estuarine conditions The Durney Key area was deemed not suitable for coral recruitment and growth on an inshore artificial substratum because of temperature extremes, potential for minimal water movement during summer, frequent occurrences of low salinity and pH, and high total organic phosphorus. Faunal studies demonstrated that the dominant recruitment reflects the common coastal/ estuarine biota, which does not include reef building corals.
1 Introduction Increasing human coastal populations typically increase deleterious anthropogenic impacts to near sho re ecosystems. While Floridas population growth has slowed to its lowest rate in thirty years Pasco County s growth has not, ranking thirty first in the nation for c ounty growth rates in 2007. Pasc os population has grown 18.45% from 1990 to 2000 and 23.4% from 2000 to 2006 (U.S. Census Bureau, 2006) leading to concerns of negative impacts to its watershed Located directly north of the most dense ly populated county in Florida, Pinellas and south of the l ess densely populated Hernando County, Pasco County represents a transitional boundary between the urbanized Tampa Bay metropolitan area and the more rural nature coast. Located in the Springs Coast Watershed, the Pithlachascotee River is a blac kwater stream that starts in Hernando County as channeled flow through the Masaryktow n Canal, flows southwest to Crews Lake. From Crews Lake it flows approximately 40.2 km ( 25 miles ) into the Gulf of Mexico at the city of Port Richey. S ubstantial amounts of the river drain underground to the Floridan aquifer. In low flow conditions, most of i ts water originates from ground water seepage, while in high flow conditions surface water runoff from the surrounding watershed constitute s most of the flow (Single ton et al. 2006). With increasing amounts of impervious surfaces there will be increasingly rapid runoff directly into the river rather than being filtered through vegetation and sediment. Land uses of the Spring s Coast Basin are displayed in Table 1. T he Springs Coast Basin is divided into three physiographic regions based on underlying sediments and topographic relief: Coastal Swamp, Gulf Coastal Lowlands and Brooksville Ridge. The Coastal Swamp region parallels the coast, extending two to five miles inland and is characterized by tidal marshes and coastal swamps (Singleton et al. 2006). Low elevations of less than 3 meters (10 feet ) are typical with poorly drained, organic soils that overlie the Floridan aquifer system. The Brooksville Ridge runs northwest southeast through the central portion of t he basin and has elevations of 21.3
2 83.8 m ( 70 275 feet ) (Singleton et al. 2006). The Gulf Coastal Lowlands are a poorly drained triangular region in the southern portion of the basin lying between the Coastal Swamp, the cliffs of the Pamlico Scarp on the west, and the Brooksville Ridge on the east (Singleton et al. 2006). Table 1 1995 Land Use and Land Cover in the Springs Coast Basin ( Florida Department of Environmental Protection 2006 ). Land Use/Land Cover Acres Percent of Total Acres Urban and Built up 243,303 34.0 Agriculture 93,963 13.1 Rangeland 9,949 1.4 Upland Forests 186,573 26.0 Water 10,306 1.4 Wetlands 158,358 22.1 Barren Land 2,985 0.4 Transportatio n, Communication, and Utilities 11,055 1.5 Total 716,492 100 Genesis of Project An artificial reef project off the mouth of the Pithlachascotee River was proposed by the city of Port Richey as part of a snorkeling park for citizens and tourists. The proposed emplacement site is directly south of Durney Key, a spoil island located 2.6 km (1.6 miles) off the mouth of the Pithlachascotee River (Fig. 1 ). Artificial reefs have been constructed from a wide variety of materials, from old cars to prefabricat ed modules and are used for an equally wide variety of reasons. An artificial reef can be defined as one or more objects of natural or human origin deployed purposefully on the seafloor to influence physical, biological, or socioeconomic processes relate d to living marine resources (Seaman Jr., 2000). Uses of these artificial reefs include enhancement of fishery production, enhancement of recreation through snorkeling, SCUBA diving, and fishing opportunities, as well as conservation of biodiversity and r estoration of water quality and ecosystems. The goal of the proposed artificial reef near Durney Key was to provide tourists and citizens with an easily accessible recreation destination where they could observe
3 coral T he City of Port Richey specifical ly wanted to have Scleractinia (stony coral) and Octocorallia (soft coral such as sea whips) recruit to the proposed reef. The close proximity to both the coastal and inland waterways as well as public boat ramps and public parks with canoe/kayak ramps m akes the area a popul ar recreational site The Pithlachascotee Rive r Canoe Trail was officially designated as part of Floridas Statewide System of Greenways and Trails ( Florida Department of Environmental Protection, 2008 a ) and the island sees heavy freq uent and camping traffic on weekends. For the proposed artificial reef, Port Richey considered using EcoReefs modules. EcoReefs are prefabricated, pH neutral, fired unglazed bone china that can be built to the clients size specifications. Ea ch module consists of two flat circular pieces connected to arms on all sides of the module ( Fig. 2 ). These modules are designed to have a high surface area to volume ratio to allow water flow through and around the modules, and encourage recruitment. Figure 1 Map of Florida indicating Pasco County in red with an image of Pasco County outlined in red (F lorida Department of Environmental Protection).
4 To determine feasibility and probable success of meeting the goals set for the proposed artificial reef, a thorough site assessment is required However, there is limited oceanographic data available for the Pasco County coastal waters e.g., there is n o water atlas such as those maintained by the counties bordering Pasco to the north and south. With the passage of the Beach Act in October 2000 there is now moni toring for fecal coliforms and Enterococcus by the Department of Environmental Protection (F lorida Department of Health, 2008) Water is sampled on a weekly basis at Brasher Park which is located at the mouth of the Pithlachascotee at 28 17.108' N (28.285 N) and 82 43.948 W (82.732 W) The installation of the Coastal Ocean Monitoring and Pr ediction System (COMPS) meteorological station ( established via a cooperative effort between Pasco County Division of Emergency Management and the University of South Florida ) at the mouth of the Pithlachascotee River in 2002 provided meteorological data f or the area. Thus, this project serves to provide a much needed baseline study of the Durney Key area that can assist the City of Port Richey in their artificial reef site determination. Habitat Description The headwaters of the Pithlachascotee River ar e protected as part of the Starkey Wilderness Area, but increasing residential development from the boundary of the wilderness area downstream to Port Richey has resulted in increasingly polluted stormwater runoff. Pollutants include oil and other automob ile fluids, paint and Figure 2 Image of an EcoReef module being emplaced at Bunaken National Park, Indonesia. Photo credit: 2004 www.ecoreefs.com
5 construction debris, yard and pet wastes, pesticides and litter. Deleterious effects of urban runoff include contaminating streams, rivers and bays, harming aquatic life, and increasing the risk of flooding by clogging storm drains a nd catch basins. The Pithlachascotee River is listed as having two state impairments by the Environmental Protection Agency (EPA), namely pathogens and organic enrichment/oxygen depletion (United Stated Department o f Environmental Protection, 2006 ). Thes e impairments are based on data from 2002, although the same impairments are listed in 1998 (United Stated Department of Environmental Protection 2006 ). There are no Total Maximum Daily Loads (TMDLs) set for the waterway; the expected submittal date of T MDLs is the end of 2011. The impact of increased groundwater exploitation from development on coastal estuaries is of particular concern. A Stream Condition Index (SCI) assessment was conducted on the Pithlachascotee River in April 2005 to gather data on the biological health of the ri ver. These data will be used to determine the TMDLs for the river. The SCI is based on ten measurements that assess the ecologica l integrity of the aquatic macr oinvertebrate community. Sampling consists of twenty dipnet sw eeps of the most productive habitats found in a 100 m (328.1 ft) stretch of the stream (Fore et al., 2007). Each metric is calibrate d for the subecoregion in which it falls in this case the peninsula (Florida Department of Environmental Protection, 200 8a). Organisms are processed in the lab and data generated on the species assemblages and abundance are used to calculate ten biological metrics which have been shown to respond predictably to human disturbance (Florida Department of Environmental Protect ion, 2008a ). If the Index score falls between 73 and 100, it is considered good; if it falls between 46 and 72, it is fair; and so on until poor. At the time the SCI was conducted, the stream was approximately 0.4 m deep and water velocity was 0.2 m/sec. Dissolved oxygen was 4.94 mg/l, pH wa s 7.48 and temperature was 21.4 C (Florida Department of Environmental Protection, 2005). The river received an SCI score of 62, placing it in the fair category. This was interpreted to mean the macroinvertebrate community was healthy and thus the low dissolved oxygen levels were natural fluctuation. The suggestions from this SCI were to expand Best Management Practices (BMPs) to the private sector as well as public agencies by engineering retrofits, riparian zone stabilization, vegetative swales and
6 creating wetlands. Homeowners are encouraged to limit fertilization and remove yard waste to reduce stormwater runoff pollution. The upper portions of the river are still surrounded by relatively rural land uses whil e the lower portion is relatively urbanized, especially around Port Richey and New Port Richey. As a result of the stormwater runoff from the urbanized area, nitrogen and phosphorus levels, as well as bacteria and protozoans, have increased (Singleton et al. 2006). Spoil islands can play a critical role in coastal ecology, acting as a buffer to wave energy pounding the coastline, providing nutrients to surrounding waters via decomposing vegetation, and often containing mangroves which act as nurseries for various fisheries (Whitney et al., 2004). Durney Key the spoil island of interest, is located at 28 16' 59.12"N and 82 45' 7. 10"W off Pasco County, Florida (Fig. 1) and is composed of dredge fill material from a nearby boat channel. The island is approxi mately 65 m (213.1 ft) wide and 177 m (580.7 ft) long, located about 1.1 km (0.684 mi) from the closest coastline. Black mangroves ( Avicennia germinans Linnaeus) grow on Durney Key, among exotics such as Australian Pine ( Casuarina equisetifolia Linnaeus) and Brazilian Pepper ( Schinus terebinthifolius Raddi). The vegetation provides perches to many shore birds; brown pelicans, terns, egrets, herons and the occasional white pelican use this island for resting, hunting and nesting (personal observation). A vian biodiversity is high, with 140 species having been recorded at a nearby popular Florida birding trail site, Robert K. Rees Park, which is located on Green Key (Florida Fish and Wildlife Conservation Commission 2006 ). Durney Key is surrounded by seag rass beds composed of five different species of seagrasses, including Halodule wrightii (Ascherson) Thalassia testudinum (Koenig) Syringodium filiforme (Ktz) Halophila decipiens (Ostenfeld) and some sparse Ruppia megacarpa (Mason). These grass beds s upport a rich array of macrofauna, including invertebrates such as crustaceans, conchs and whelks, as well as fish (including sport fish such as snook), sea turtles and dolphins. Nearby offshore benthos are primarily hardbottom, i.e., lithified seafloor, h abitats. Common hardbottom species include calcareous algae such as Halimeda and red calcareous algae, boring mollusks ( Lithophaga sp.), boring sponges ( Cliona sp.), echinoderms and hydroids ( Eudendrium sp.) (Obrochta et al., 2003; Dix et al., 2005).
7 Lept ogorgia virgulata (sea whips) are the most common local octocoral while Cladocora arbuscula (tube corals) and Carijoa riisei (snowflake coral) are the common scleractinians (Dix et al., 2005). These hardbottom habitats could provide a recruitment source t o colonize other areas. The area is a popular recreational fishing site, with many fishers canvassing the surrounding flats and Anclote Power Plant outflow. The power plant, owned by Progress Energy, is a coal fired power plant located approximately 4.8 km (3 mi) south of Durney Key. The surrounding Anclote River estuary was studied as part of a multi disciplinary effort to characterize and document the environmental conditions before the power plant construction began in 1973 (Rolfes, 1974; Dietz, 1976; Weiss, 1978; and Szedlmayer, 1982) Objectives The installation of an artificial reef near Durney Key a dredge spoil island just offshore was under consideration by the city of Port Richey, Florida. T he city requested assistance in determining the p otential for coral growth on the artificial reef modules. T o assist in determining the feasibility of this proposal I pursued three objectives: to characterize the oceanographic setting of Durney Key off Port Richey, Florida, including hydrodynamics, w ater qu ality and invertebrate biota; to investigate the potential for successful coral recruitment and gro wth in Durney Key waters; and to determine if porcelain is a suitable substrate for settlement of the larvae of coral species present in west centra l Florida. Rational for A ssessment Strategy Physical P arameters Durney Key is typical of southwestern Florida with warm summers and generally mild winters with short cold snaps. Summer air temp eratures averaged from 2002 2004 was 26.7C (80.1F) and winter temperatur es average was 19.1C (66.6F), with relative
8 humidity averaging 70% (COMPS, 2008). Water temperatures averaged from the same time period were the same as the air temperatures (Florida Department of Health, 2008). A verage annual r ainfal l in Pasco County is 140 cm (55 in) the majority of which is received from June through September The predominant wind direction is east west and wind speeds average 3.1 m/ s in summer and winter wind speeds average 3.7 m/ s. A verage tidal range as measu red by the COMPS station is 1.2 m (3.9 ft). Summer salinity averaged 23.7 and 28.2 parts per thousand (ppt) around Durney Key during spring/ summer and fall/ winter, respectively. The flats around Durney Key are shallow, generally 1 3 m (3.3 9.8 ft) whi ch makes understanding the tidal variation critical if a structure is to remain submerged. To understand the full range of tidal variation, both neap and spring tides must be examined. Neap tides occur when the Moon and the Sun are separat ed by 90 when viewed from the E arth. This occurs at first and third quarter, when tidal forces due to the gravitational attraction of the Sun partially cancel those of the Moon yielding minimal tidal range Spring tides occur when the sun and moon lay in a strai ght li ne on either side of the Earth and tidal attraction cause an additive effect. Spring tides result in larger tidal range i.e. higher than average high tides, lower than aver age low tides, compressed slack water periods and stronger than average tidal cur rents. There are approximately seven days between spring and neap tides (Stewart, 2006). Understanding the primary driver for current velocities is an important step towards characterizing an areas flow regime. Many protected areas on Floridas west coa st have flow r egimes dominated by tidal flow (He and Weisberg, 2002 ). Wind can be another important factor contributing to curre nt velocities in shallow areas (Weisberg et al., 2001 ). Wind events occur more commonly in winter in Florida, with summer wind events resulting primarily from hurricanes and tropical storms. Winter winds are associated with frontal passages in which wind direction is uniform and persists long enough to alter the wat er level If the set up produced by the front is large enough t here will be a corresponding change in water level ; however if the area of interest is near shore there will be minimal water velocity change due t o water piling up at the coast (Stewart, 2006) Estuarine influence is another potential factor influencing water velocity due to temporal changes in river flow
9 Chemical Parameters Salinity typically ranges from 19.8 parts per thousand from November March to 16.9 ppt from April to October, at the COMPS site. There is no known record of other oceanographic p arameters near Durney Key, although there is USGS operated instrumentation installed on the U.S. Route 19 bridge the Pithlachascotee River which measures pH, dissolved oxygen, water color CO 2 and other parameters (United States Geological Survey, 2008). D ata from the USGS instrumentation are not comparable to samples taken at Durney Key since the U.S. Route 19 bridge is 2500 m (1.6 mi) upriver from the mouth of the Pithlachascotee. Understanding the pollutants in the environment surrounding Durney Key is important to determining if corals will be able to grow t here. This area is especially interesting because it is located just north of a power plant (latitude 28 11' 25.0080" and longitude 82 47' 14.9999" ; 1729 Baileys Bluff, Holiday, FL 34691). This p ower plant releases nickel compounds to the water, and many chemicals to the air, including lead, dioxin and mercury compounds ( United States Department of Environmental Protection, 2006). Organic Pollutants Dioxin is the common name for the family of halogenated organic compounds. The basic structure of two of the more common dioxins, polychlorinated dibenzodioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs), comprises two benzene rings joined by either a single (furan) or a double oxygen bridge (dioxin). PCDD/PCDFs (PCDD/Fs) have been shown to bioaccumulate in humans and wildlife due to their lipophilic properties, and are known teratogens and mutagens. Potential sources of these contaminants include agricultural run off, atmospheric deposition and particulates from municipal waste incineration. Chlorine atoms are attached to the basic structure at any of eight different places on the molecule, positions 1 4 and 6 9. The toxicity of PCDD/Fs depends on the number and position of the chlorine at oms; only congeners (i.e. related chemicals) that have chlorines in the 2, 3, 7, and 8 positions have been found to be
10 significantly toxic. Out of the 210 PCDD/F compounds in total, only 17 congeners (7 PCDDs and 10 PCDFs) have chlorine atoms in the rele vant positions to be considered toxic by the NATO Committee on the Challenges to Modern Society (NATO/CCMS) international toxic equivalent (I TEQ) scheme. Organochlorine compounds and organochlorine pesticides are among the most widespread and persistent e nvironmental contaminants. These substances can enter the sea via rivers, atmospheric deposition, spills and dumping o f dredged material. Most of them have short residence times in the water column, due to their hydrophobic character (low water solubilit y), which leads to bioaccumulation and strong sorption onto suspended particulate matter. The suspended matter is carried to the bottom water and finally trapped in marine sediments. For some organ o chlorine compounds, different chemical transformation an d microbial decomposition may occur, e.g. 1,1,1 trichloro 2,2 bis( p chlorophenyl)ethane ( DDT ) to 1,1 bis (4 ch lorophenyl) 2,2 dichloroethene) ( DDE ) However, complete mineralization of organochlorine compounds does usually not take place, or is an extre mely slow process (Dannenberger, 1996). Polychlorinated Biphenyls (PCBs) were shown to bioaccumulate on Tern Island, Hawaii, where PCBs 118, 138, and 153 were the predominant congeners found. These PCB congeners are commonly detected in other marine speci es (Bavel et al., 1996; Hope et al., 1997) and are potentially toxic (deSwart et al. 1996). Some of these congeners are known inducers of cytochrome P 450 dependent mixed function oxidases (Janz and Metcalfe, 1991) and potential xenoestrogens, i.e. synt hetic substances that differ from those produced by living organisms and imitate or enhance the effect of estrogens. Xenoestrogens are part of a heterogeneous group of chemicals that are hormone or endocrine disruptors (Safe and Gaido, 1998), and/or immun osuppressants (Miao et al., 2000). The bioaccumulation of PCBs is a complex phenomenon that involves many factors, such as chemical lipophilicity, species, gender, breeding condition, tissue composition, and metabolic capacity of the animal (Skaare, 1996). PCBs are taken up by marine organisms via two principal routes: direct absorption and feeding on contaminated organisms (Miao et al, 2000). Dissolved PCBs are taken up rapidly over the gills as a result of equilibrium partitioning between water and fish lipids (Barron, 1990). The low
11 water solubility and high lipophilicity of PCBs favor their bioconcentration, bioaccumulation, and low excretion from fish. Phytoplankton and algae, to which pollutants sorb from the water, constitute a primary link that i ntroduces contamination into the pelagic food web (Harding et al., 1997) Metals Potentially toxic metals are often referred to as heavy metals despite there being no widely accepted definition of the term at this time (Duffus, 2002). A recommendation to develop a new classification reflecting our understanding of the chemical basis of toxicity was made by the International Union of Pure and Applied Chemistry (IUPAC) but currently no such classification has been accepted. Throughout this document, the term heavy metal will refer to specific potentially toxic metals. Potentially toxic heavy metals can be released during natural weathering of rocks, ore, minerals and volcanoes. These natural sources may be augmented by anthropogenic inputs (Siegel, 2 002). These metals may be moved by advective transport and mixing. The metals often a ccumulate in the sediment when sedim ent sorbed metals settle out of suspension (Kennish, 1992). Estuarine sources of heavy metals are primarily freshwater influx, atmos pheric, and anthropogenic inputs. Riverine inputs are the especially important in estuaries. Anthropogenic inputs can add to the natural riverine load of heavy metals as well as contribute directly in many ways such as sewage sludge disposal, coal power plants, ash disposal, smelting and dredged spoil dumping. Uptake of metals by estuarine organisms can occur through diffusion or ingestion of food and/or particulate inorganic matter. These metals can be stored in the skeletal structure or intracellular matrices of an organism and can be excreted in feces, eggs, and molted material. In humans, metals are distributed to tissues and organs after incorporation. Excretion typically occurs through the kidneys and digestive tract, but metals may persist for years to decades in the liver, bones and/or kidneys. Even low level metal exposure is suspected to contribute to impaired functioning and chronic disease (Hu, 2002). There was no previous data on heavy metals around the Durney Key area to reference, th us the heavy metal levels found in this study will be useful for future comparisons.
12 Biological Parameters Conducting a holistic assessment of a proposed artificial reef site that includes investigating the parameters important to the end goals of the r eef is critical to the success of the reef. Often pre site emplacement studies are less than thorough, or lacking all together (Seaman Jr., 2000). A local example of the deleterious and expensive results of this lack of planning is the numerous loose tir es across south Florida that caused extensive damage to benthic growth and are now being recovered at substantial cost (Florida Department of Environmental Protection, 2008b ). Pre emplacement planning and site analysis may help to prevent costly mistakes and to help meet the goals set for the reef. Since the goal for the proposed Durney Key reef was to have coral and sponge growth on EcoReef artificial reef modules, part of the site analysis consisted of investigating what recruited to similar substrates and using bioindicators to determine if the habitat could support coral growth. Physiologically and ecologically (and roughly phylogenetically), the Scleractinia can be divided into two groups, those that contain zooxanthellae in their tissues (the zooxan thellate corals) and those that do not and instead rely on actively catching prey (the azooxanthellate corals). Worldwide both groups have approximately the same number of species. Zooxanthellate species (ZS) (the endodermic tissues contain zooxanthellae a dinoflagellate alga of the genus Symbiodinium ) are restricted to the photic zone and ZS are typically found in tropical subtropical regions in depths that rarely exceed 70 m. The azooxanthellate are ubiquitous, but most common in cooler, deep water ( down to 6300 m) or cryptic shallow water environments, such as caves. Both ZS and azooxanthellate species actively catch prey, but the ZS are mixotrophic meaning they are able to obtain energy from both predation and zooxanthellae. The shallow water, mi xotrophic corals require low nutrient conditions to thrive, thus large nutrient influxes create less than ideal habitat for coral growth (Hallock, 1981; Hallock and Schlager, 1986). With the limited nutrient data available for the Durney Key area, the on e proxy for heightened nutrient levels caused by wastewater is provided by bacterial sampling through the Beach Act. Under this act, water samples over 35 enterococci per 100 ml and
13 over 199 fecal coliform organisms per 100 ml of water are considered unsa fe for swimming When water samples are tested and shown to be above these levels the site is re tested immediately. If the results of the re test are still above these levels, a health advisory warning is issued (Florida Department of Health, 2008). S ince sampling began at Brasher Park in August of 2000 there have been 56 advisories issued, there was one in 2000, two in 2002, 15 in 2003, 14 in 2004, eight in 2005, nine in 2006 and seven in 2007 Recruitment studies Recruitment studies allow one to determine what could recruit onto an artificial reef prior to emplacement. While many factors can influence what organisms recruit onto the substrate, there appear to be several important factors: location and size of emplacement, time of emplacement, and succession (Carter et al., 1985; Svane and Petersen, 2001). Early successional organisms may follow the inhibition model of succession (sensu Connell and Slyater, 1977), where if early colonists persist, they exclude or suppress subsequent colonists o f all other species. This is corroborated by Birklands ( 1977 ) study showing barnacle growth to exclude coral growth, either by preemption of space or by more rapid growth. Subsequent succession may be closely tied to successful invasion of other organism s (Carter et al., 1985). Fish can also play an important role in determining species composition of substrates by selectively grazing certain species, potentially facilitating the persistence of early successional communities on the substrate (Carter et a l, 1985). With succession being exceedingly difficult to forecast, often the simplest option is to submerge substrate i n the proposed emplacement area. Comparisons to recruitment in similar habitat s can also be valuable. Recruitment data from artificia l reef sites in Tampa Bay, Florida were used for comparison in this study. Bioindicators Criteria that emphasize physical and chemical attributes of water are unsuccessful surrogates for measuring biotic integrity (Karr, 1981); a more holistic view i s needed. Biological communities reflect watershed conditions since they are sensitive to changes
14 in a wide array of environmental factors. Benthic invertebrates are used extensively as indicators of estuarine environmental status and trends because nu me rous studies have demonstrated that benthos respond s predictably to many kinds of natural and anthropogenic stresses ( Karr, 1981; Weisberg et al., 1997). Benthic invertebrates generally have limited mobility and cant avoid adverse conditions such as hypo xia and anthropogenic contaminant accumulation. This immobility is advantageous in environmental assessments because, unlike most pelagic fauna, benthic assemblages reflect environmental conditions (Weisberg et al., 1997). These advantages have led to be nthic invertebrates often being used as bioindicators. Bioindicators are organisms whose presen ce and quantity provide specific information of surrounding environmental conditions (Wilson, 1994). The better known a relationship between an organism and it s environment is, the more useful the bioindicator will be (Wilson, 1994). Foraminifers are shelled protists that can be excellent indicator organisms for reef and coastal environments (Hallock et al., 2003). Benthic macrofaunal communities are often uti lized as environmental health indicators because benthic animals are relatively sedentary, exhibit tolerances to stress, and have important roles in nutrient cycling (Dauer, 1993). Some advantages to using foraminif ers as bioindicators are that foraminife rs 1) are abundant, diverse, and widespread in marine environments, 2) have relatively short life spans as compared with long lived colonial corals facilitating differentiation between long term water quality and episodic stress events, 3) are relatively small (most < 2 mm) and abundant, permitting statistically significant sample sizes to be collected quickly and relatively inexpensively, 4) record environmental changes by changes in foraminiferal assemblages in the sediment and 5) exhibit species specif ic responses to ecological conditions (Yanko et al., 1999; Hallock et al., 2003). Benthic foraminifers inhabit a similar environment to what a newly settled coral polyp would, and thus make good bioindicators for potential coral recruitment habitat. B enthic foraminifers have been used extensively to indicate marine and brackish environmental conditions (Alv e, 1995). Foraminifers have been used as bioindicators for a variety of processes including water velocity, settling effects, tidal effects, sedime nt transport anthropogenic impacts such as heavy metals, pesticides, eutrophication, and
15 salinity changes as well as global effe cts such as UVB radiation (Alve, 1995 ; Yanko et al., 1999). On the Florida reef tract, s pecies assemblage s have shifted fro m long lived symbiont bearing taxa to small, fast growing, heterotrophic taxa over the past several decades, probably responding to altered nutrient flux (Cockey et al., 1996). This observed change agrees with models formulated by Hallock and Schlager (19 86), Birkeland (1988), and Hallock (1988) that predict community response to gradually increasing nutrient flux whether naturally or anthropogenically induced, should favor phytoplankton, benthic algae, and heterotrophic taxa lacking algal symbionts. Fo raminifers have been organized into functional groups that reflect the ir ecological roles ( Hallock et al. 2003). The larger benthic foraminifers (LBFs) are symbiont bearing and intolerant of eutrophication, thus they are seen to decline with increasing n utrient loads, while smaller rotaliids and miliolids increase in relative abundance (Cockey et al., 1996). The sm aller foraminifers of Miliolida and Rotaliida are typically 0.05 0.5 mm in diameter and have life spans of several days to weeks (Lo eblich and Tappa n, 1987). The majority do not have algal endosymbionts and thus are purely heterotrophic, feeding on bacteria and microalgae captured by the pseudopodia as they move along a substrate (Sen Gupta, 1999). Miliolids and rotaliids generally increase in abundance as their food supply increases, provided respiration is not limited by oxygen supply (Cockey et al., 1996). If organic carbon loading reduces nocturnal oxygen concentrations, the benthic foraminiferal a ssemblage will shift to hypoxia tolerant, o pportunistic taxa such as Ammonia ( Order Rotaliida ), which is a recognized eutrophication indicator (Alve, 1995). Assemblage shift has been described along gradients approaching a source of pollution by Alve (1995). A transitional assemblage is created when the natural assemblage, dominated by longer lived species, is gradually replaced by smaller, faster growing species as bacterial and microalga l food supplies increase. A stress tolerant assemblage may result if nutrient pollution is sufficient t o cause substantial organic carbon loading and subsequently, hypoxia. In my study, I examined foraminiferal assemblages in the sediments and recruiting onto artificial substratum as bioindicators of whether or not the Durney Key site could
16 support calci fying organisms that host algal symbionts. The assemblages can also indicate whether high stress conditions are limiting even the heterotrophic benthic community, such as bivalves, snails and sea urchins.
17 Methods A map of Du rney Key and sampling is presented in Figure 3, with locations of various procedures noted. The parameters evaluated are listed in Table 2. Figure 3. Durney Key with locations of various samples, where triangles represent ADCP lo cations with blue being ADCP A, red ADCP B and orange ADCP C; blue solid circle representing location of surficial sediment collection, open black circles representing sediment cores with C1 = core 1, C2 = core 2, C3 = core 3 and numbers representing stake s of recruitment tiles. Table 2 Parameters evaluated with corresponding investigation techniques. Parameter Category Method Employed Specific Parameter Physical Acoustic Doppler Current Profiler (ADCP) Water temperature, pressur e (indicating depth), current velocity, current direction Data from COMPS station Air temperature, wind speed, wind direction, relative humidity, rainfall Chemical Water samples Inorganic nutrients (NO3 + NO2 NO2 NH4+, Si(OH)4, TP) salinity, temper ature, pH, total dissolved solids Sediment samples O rganic pollutants suite and heavy metals suite Sedimentological Sediment push cores Grain size analysis 1 2 3 4 5 6 8 7 1 0 9 1 1 1 6 1 2 1 7 1 4 1 5 1 3 1 8 20 1 9 N 0 1 k m ( 0 6 m i ) C1 C2 C3
18 Table 2 (Continued) Percent carbonate Biological Recruitment tiles Recruitment rates of ben thic biota Sediment push cores I nvertebrate and foraminiferal assemblages Lime rock F oraminiferal density and assemblage Physical Parameters An array of three Argonaut XR Acoustic Doppler Current Profil ers (ADCPs) were deployed for full lunar cycle s during the summer of 2007 and winter spring of 2008. The ADCPs frequency was set at 3.0 MHz to yield a resolution of 0.2 m (0.66 ft) cell size; the blanking distance was set to 0 .2 m (0.66 ft) The ADCPs are c apable of measuring ten fixed size cells w ith an eleventh dynamic, automatically adjusting cell or bin which is the height of each portion of the water column where velocity is measured A velocity measurement was recorded every six minutes, in addition to temperature and pressure (depth). The temperature sensor has a resolution of 0.01 C and an accuracy of +/ 0.1 C Each ADCP has a three beam transducer for measuring water velocity in three dimensions (3D) and a compass/two axis tilt sensor to ensure uniformity in depth calculations. The c ompass/tilt sensor has a resolution o f 0.1 for heading, pitch and r oll and an accuracy of +/ 0.5 for the heading and +/ 1.0 for both pitch and roll. The built in piezoresistive type of pressure sensor, or strain gauge, allows readings accurate to 0. 1% of the full range of water levels around the island. These data are critical for ensuring that any proposed submerged structures are placed at an appropriate depth to avoid potential collisions with vessels that frequent the area. These water depth ti me series form an important baseline of accurate water levels to compare with future data sets. A time series of stage levels was produced from these pressure readings by utilizing Sontek/YSIs ViewArgonaut version 2.0 softwares algorithm. Two of the A DCPs were deployed on the southern side of Durney Key and one on the northern side of the island, between the shore and the channel, to understand the flow around the island (Fig. 3) Each ACP was readied for deployment by being painted with
19 anti fouling p aint, connected to an external battery pack and covered in duct tape for an added layer of fouling protection. Data were stored internally until the units were recovered The ADCPs were deployed for over a month to capture the full variability of a ti dal cycle in summer ( July 16, 2007 to August 23, 2007 ) and winter ( Febr uary 14, 2008 to April 10, 2008) to capture seasonal variability. The water velocity was correlated with wind velocity data from a nearby Coastal Ocean Monitoring and Prediction System (CO MPS) meteorological station located at the mouth of the Pithlachascotee River. The COMPS Port Richey Station is physically located at 28 17.108' N (28.285 N) and 82 43.948' W (82.732 W) inside the city of Port Richey' s Brasher Park and is capable of m easuring wind speed/direction, relative humidity/air temperature, barometric pressure, a nd precipitation. COMPS data are available in near real time online at http://comps1.marine.usf.edu/pas/index.shtml. ADCP water velocity data were quality checked usin g Sontek/YSIs ViewArgonaut v. 2.0 software. Data were visually analyzed and anomalous data (i.e. irregularly large values related to deployment and recovery) removed from beginnings and endings of time series. Temperature and stage data were visually checked and anomalous data removed. The remaining data were exported to Microsoft Excel and plotted to show velocity variation through time. These data were uploaded into a MATLAB routine to be detided using the Fo reman Tidal Analysis package (Po wlowic z et al., 2002). Detided theoretically implies the influence of tides has been removed from the data. The term tided is used here to refer to water velocity data that has not been altered and detided is used to refer to water velocity data that has h ad the influence of tides removed. Detided water velocity data were plotted against wind velocity in Microsoft Excel version 12.0.1 to determine correlation strength. Summer spring and neap tides were compared, as were winter spring and neap tides. Che mical Parameters Water samples were collected in small Nalgene bottles in the same manner for both the nutrient analysis and for use on site in an Ultrameter manufactured by Myron L
20 Company The water sample was decanted immediately into the Ultrameter to analyze for temperature, salinity, pH and total dissolved solids (TDS). Water samples were collected for inorganic analysis at comparable tide heights to maintain consistency. Water was collected in 30 ml acid washed Nalgene bottles opened under the surface of the water air interface and closed under the water to avoid contaminating the sample with the concentrated nutrients on the biofilm at the water air interface. Excess water was decanted from the sample to allow room for expansion during freezi ng. Samples were frozen at 12 o C until analysis. These samples we re analyzed using a continuous flow analyzer (CFA) according to the protocols put forth by Gordon et al. (1993) by Kent Fannings lab at the College of Marine Science, University of South Florida. The CFA utilizes a multichannel peristaltic pump to mix samples and chemical reagents in a continuously flowing stream to automate colorimetric analysis. These data were compared to data from patch reef s in Florida Keys tract. Sediment samples scooped from the sediment water interface in to acid washed Nalgene containers were analyzed for persistent organic pollutants and metals. Surface samples were used to approximate the environment to which coral polyps would be exposed Thi rteen of the pr iority pollutant metals as defined by the Environmental Protection Agency (EPA) are: Antimony (Sb), Arsenic (As), Beryllium (Be), Cadmium (Cd), Chromium (Cr), Copper (Cu), Lead (Pb), Mercury (Hg), Nickel (Ni), Selenium (Se), Silver (Ag), Thallium (Tl), and Zinc (Zn). These priority pollutants were part of the suite of elements analyzed by as described in the following pararaph A sediment sample was shipped to ActLabs in Ancaster, Ontario, Canada in a cooler with a liquid temperature blank for metal analy sis covering a suite of 60 elements, listed in Appendix II The sample was split into two subsamples that were prepared in two different manners: one portion was pulverized in totality, and the other portion was sieved at 230 m to anal yze only the finer sands and muds (of the 95 g sieved, 44.5 g were fines that were analyzed) Both subsamples were analyzed using Inductively Coupled Plasma Optical Emission Spectrometry (ICP OES) and Mass Spectroscopy (MS). ICP OES is able to measure elemental concentrati ons to very low detection limits (ppm to ppb) for multiple elements. Each sample was placed in solution using single and/or mixed acids, partial leaches, or fusion techniques using fluxes. Next, the sample
21 solution was introduced to a radio frequency exc ited plasma at approximately 8000 K. Atoms are excited until they emit wavelength specific photons or light that is characteristic of a specific element. The concentration of an element in the sample is directly related to the number of photons produced upon excitation (ActLabs, 2008). Mercury was analyzed in both subsamples differently from the above methods also by ActLabs, on a Perkin Elmer FIMS 100 cold vapor Hg analyzer. A 0.5 g sample was digested with aqua regia at 95 C to yield the stable divalen t form of mercury. Stannous chloride was used to reduce divalent mercury to the volatile free state and the concentration was determined via the absorption of light at 253.7 nm by mercury vapor. Argon was bubbled through the mixture to transport the mercu ry atoms into an absorption cell that was placed in the light path of an Atomic Absorption Spectrophotometer. The maximum amount absorbed (peak height) is directly proportional to the concentration of mercury atoms in the light path (ActLabs, 2008 ). A sur ficial sediment sample was collected with a n amber glass jar at 28.52 N, 82.98 W ; excess water was decanted. The sample was sent in a cooler with a temperature blank to ENCO Laboratories, Inc. in Orlando, Florida. The sample was analyzed fo r organochlorine pesticides and PCB s by EPA SW 846 methods 8081/8082 and polynuclear aromatic hydrocarbons by SW 846 EPA method 8270 via selected ion monitoring (SIM) ( United States Department of Environmental Protection, 2008) The results for analytes over the minimum detection limit (MDL) are included. The MDL or Minimum Detection Limit is a statistically determined value that represents the limit of what the particular instrumen t of measure can detect for each analyte There are specific rules on det ermining MDLs that are outlined by the regulatory agencies in this case the EPA T he rules governing how data are report ed requires that MDL be used However, since the MDL is a statistically derived value it may not necess arily be what can actually b e see n with a real reference standard. Thus the us e of the Method Reporting Limit ( MRL ), also referred to as a Practical Quantification Limit ( PQL ) by some, which is a value that can more realistically be measured with reference materials.
22 Sediment Te xture Three sediment cores (10 cm in length and 7 cm in diameter (3.9 x 2.8 in) were taken on July 15, 2007, at 0.5 (1.6 ft) 1 (3.3 ft) and 1.5 (4.9 ft) meters depth on the south side of Durney Key, 10, 20 and 30 meters south of Durney Key, respectively (Fig. 4) Sediment cores were numbered 1, 2, a nd 3, with 1 being the 10 m (32.8 ft) from shore at 0 .5 m (1.6 ft) depth, core 3 being 30 m (98.4 ft) from shore and 1.5 m (4.9 ft) depth and core 2 in the middle. Each core was divided at five centimeters d epth, into two subsamples representing top and bottom halves Each core half was washe d with deionized water over a 0.063 mm sized sieve to remove the mud and fine sand fraction, i.e., < 0.063 mm Both fractions were allowed to settle overnight and the e xcess water was siphoned off after the sediment fell out of suspension T hen fractions were dried in an oven between 45 and 55C. Subsamples between 30 and 100 grams of the s and fractions ( 0.063 mm ) were dry sieved according to methods described by Folk (1980). The sieve set was secured on top of a shaker with r ubber bands (Fig. 4) Each subsample (3 to 20 grams) was stirred and o rganic matter, e.g., algal clumps was removed if necessary and placed in the top, coarsest sieve. Each sample was shaken fo r ten minutes and visually checked for completeness, i.e., that no sediment was remaining in the top sieve. Each sieve was emptied onto tared beakers and weighed to the nearest milligram.
23 Weight percent of each size fraction was calculated by divid ing the weight of each subsample retained per sieve by the total weight of the sample, and multiplying by 100%. The Udden Wentworth scale (Table 3) was used to classify sediments by grain size. To determine the percent carbonate in the sediment, o ne gram dry sub samples from each core half were dissolved by adding 50% HCL to the beakers, letting sit for 12 hours, agitating, adding more HCL, letting sit for 12 hours, then checking for any further bubbling when agitated again (which there was none). The sa mples were then rinsed with fresh water, dried using the same methodology described for the grain size analysis, and remaining sediment weighed. Table 3 Scale size for particle sizes of sediment samples (Blatt et al., 1972). Gra in size class Range (mm) Phi size Gravel/Granule > 2 1 Very coarse sand < 2 1 0 Coarse sand < 1 0.5 1 Medium sand < 0.5 0.25 2 Fine sand < 0.25 0.125 3 Very fine sand < 0.125 0.063 4 Silt and clay < 0.063 > 4 Figure 4. Sieve set on shaker used in grain size analysis.
24 Biological Parameters Recru itment Tiles Benthic recruitment and succession were investigated by installing ceramic and porcelain tiles attached to 20 PVC rods on the south side of Durney Key. The PVC rods were placed in two roughly parallel rows of ten in 1.5 m (5 f t ) of water at mean low tide (Fig. 4) Recruitment tiles were made of porcelain clay to be similar to the artificial reef module of interest, ceramic tiles were also used to compare to the porcelain substrate (Fig. 5 ). Each rod had two tiles opposite each other at the seafloor, one made of porcelain and one ceramic, and a second pair again one porcelain and one ceramic, placed 30 cm (11.8 in) above (Fig. 6) These tiles were installed on June 22, 2007, and were monitored by recording images of each tile digitally on a monthly basis until September 11, 2007. Images were recorded using a Cannon PowerShot A630, 8.0 mega pixel camera in underwater housing. The images were taken on planar level with the tile to minimize image distortion. Each ceramic tile was approximate ly 10.2 cm 2 (4 in) in area and 1 cm (0.4 in) thick. These were hand made of Helios Porcelain clay from Highwaters Clay Company. Helios is a grolleg kaolin clay with a bright white color. This was chosen to most closely resemble the bone china used in Ec oReef modules, which the city government of Port Richey was interested in installing at the Durney Key site (Fig. 2) On the opposite side of the stake from the Helios Porcelain tile was a ceramic bathroom tile of equal size attached with the unglazed su rface facing away from the stake. Both the Helios Porcelain tiles and the ceramic bathroom tiles have texturized patterns to encourage settlement (W. Jaap, personal communication).
25 Figure 5. Handmade porcelain recruitment plates with texturized patter ns. Biological growth on these tiles was analyzed with Coral Point Count with Excel extensions (CPCe), a tool designed by the National Coral Reef Institute (NCRI) ( Kohler and Gill, 2006) The perimeter of each individual or colony on each tile was traced to calculate the total area of each recruitment category The areas of growth traced were grouped into five categories: macroalgae, turf algae, coralline algae, barnacle, and unknown. A sixth class was created by subtracting the summation of the a forementioned Figure 6. PVC stake number one with recruitment tiles attached. Tiles are paired with each upper and lower pair havi ng one porcelain and one ceramic tile.
26 five categories (i.e. total area covered) from the total surface area to account for the ubiquitous brown algal film. Each tile photo was given a unique identification tag that represented the location of the tile, substrate, and date the p hoto was taken Image analysis was utilized as a substitute to identifying the organisms on the tiles by hand as physical evidence of the recruited organisms is limited to two tiles. The other stakes and tiles disappeared from the field site sometime af ter September 11, 2007, which severely limited the ability to identify the organisms present. The recruitment coverage data were analyzed in the statistical package PRIMER e v.6 (Plymouth Routines in Multivariate Ecological Research). Nonparametric appr oaches to data analysis were used in this work to avoid problems arising from the non normal distribution of variables and from the low sample to variable ratio in the data sets. Using PRIMER, Bray Curtis similarity matrices were constructed for each time step using square root transformed data. This transformation is commonly used as it down weights the importance of the highly abundant species. Non metric multi dimensional scaling (MDS) plots and cluster analyses were derived from these similarity matr ices (Clarke and Warwick, 2001) SIMPER (similarity percentages) routines were run on the square root transformed Bray Curtis similarity matrices to understand which recruitment classes were responsible for the clustering patterns, or variation thereof ( Clarke and Warwick, 2001). SIMPER routines identify variables that primarily account for Bray Curtis dissimilarities by decomposing average Bray Curtis dissimilarities between all pairs of samples, one from each group, into percent contributions from each variable and lists the variables in decreasing order of contribution, for each variable contributing to greater than 90% similarity within each group or dissimilarity between groups. Outputs include average abundance, average similarity, a ratio of simil arity to standard deviation, percent contribution, and the cumulative percent contribution of each class. To understand the similarity between the recruitment patterns, a one way analysis of similarity (ANOSIM) was conducted with tile position as the fact or (Field et al., 1982 ; Clark e and Warwick 2001 ). ANOSIM is a nonparametric technique designed to allow comparisons for multivariate data sets in a manner similar to analysis of variance. This procedure uses the Sorenson similarity matrix to calculate R = (rB rW)/[0.25 n (n
27 1)]. Where rW is the average of all rank similarities for samples within the same group, rB is the average of all rank similarities for samples between different groups, and n is the total number of samples. Values of R near 1 indicate complete separation of sample groups, while values near 0 indicate no separation between groups (Field et al., 1982). Having determined R, ANOSIM then randomly assigns samples to different groups to generate a null distribution for R, i.e., Mo nte Carlo tests with 999 permutations (Hope, 1968), and to test whether within group samples were more closely related to each other than would be expected at random. The two recovered tiles were frozen until they could be anal yzed for species assemblages of invertebrates, primarily mollusks, and foraminifers. The assemblages on both tiles were analyzed under stereomicroscope and my identifications, excluding foraminifers, were verified by J. Fajans, an expert in invertebrates (Baker et al., 2004; Fajans e t al., 2007 ). To subsample the foraminiferal assemblage, a 100 cm 2 (39.4 in 2 ) grid was used to analyze subsamples of each 10x10 cm 2 (4x4 in 2 ) t ile. Ten random numbers were generated using the random number s generator in Microsoft Excel 2004 for each til e and these ten 1 cm (0.4 in) squares in the grid were outlined on each tile. Foraminiferal assemblage and density were analyzed within each square under a stereomicroscope. Foraminifers were manipulated using a fine artist' s brush moistened with dionize d water (tip size 3/0 to5/0) and identified to the species level when possible. Those not readily identifiable were removed and placed on a cardboard micropaleontological slide, which was coated thinly with dilute Elmers glue. Foraminifers were identifie d using characteristics defined by Loeblich and Tappan (1987). Sediment Push Cores Sand fr actions (>0.063 mm) from the grain size analysis were analyzed using a stereomicroscope as outlined in Hallock et al. (2003). Sediment was stirred thoroughly and a small scoop spatula was used to take a scoop (approximately 1 gram) from the center, bottom, of the mixed sample to represent the spectrum of grain sizes. Scooped sediment was weighed, to the nearest milligram, distributed on a gridded tray (Fig. 7) and
28 examined under a stereomicroscope I ntact molluscan skeletons, shells or bodies and foraminifer shells were manipulated using a fine artist' s brush moistened with d e ionized water (tip size 3/0 to5/0). All species were identified to the species level when possible and those not readily identifiable were removed. Any obviously worn or worked remains were not included. The removed specimens were placed on a cardboard micropaleontological slide, which was coated thinly with dilute water soluble glue. Foram inifers were identified using characteristics defined by Loeblich and Tappan (1987). Molluscs were identified using characteristics defined by Abbott (1974). Lime rock Six pieces of unattached lime rock were collected from one to two meters depth o ff Durney Keys south shore on February 14, 2008. The rocks were placed in sealable plastic bags with seawater. Each bag was emptied into a small bucket where the entire rock surface was scrubbed with a soft bristled toothbrush three times. The rock was removed from the bucket and the sediments were allowed to settle out for thirty minutes before excess water was decanted. Excess water was visually checked for any foraminifers present at the water air interface and if present these were held back with a toothbrush. This slurry was allowed to settle overnight and excess water was siphoned off. The remaining slurry was dried in an oven between 45 and 55C. Five subsamples of the sediment/algal/attached fauna matrix were r emoved by dis aggregating the dri ed slurry, stirring it to ensure it was mixed thoroughly and a small spatula scoop was Figure 7. Gridded tray with sediment sample, artist brush and water used to isolate foraminiferal and molluscan shells for identification.
29 scooped from the bottom up (approximately 0.2 g). These subsamples were scattered in a gridded tray and foraminiferal shells identified to species level when possible, again using a stereomicroscope to manipulate the foraminifers with a fine artist' s brush moistened with d e ionized water (tip size 3/0 to5/0) as outlined in Hallock et al. (2003) (Fig. 7) The sediment scrubbed from the lime rock was dried and five 0.2 g su bsamples were examined, foraminifers identified and counted. Five subsamples of 0.2 g were found to be sufficient to survey the majority of the species by creating rarefaction curves for each piece of lime rock. FORAM Index The FORAM Index (FI) is a single metric bioindicator that uses the relative abundances of foraminiferal taxa in functional groups as defined by Hallock et al (2003) and modified according to Ramirez (2008) ( Table 4 ). Both Ammonia and Nonion have been found in low oxygen foramini feral assemblages ( Sen Gupta, 1999). Ammonia and Haynesia were among the genera listed as having a known tolerance to pollution by Yanko et al (1999). Symbiotic foraminifers typically indicate lower nutrient, sufficient light conditi ons, as these are the conditions where the energy return from algal endosymbiont s outweigh s the costs of harboring them. The FI was developed in reef settings using surficial sediment samples to analyze the foraminiferal assemblage to determine the suitab ility of the surrounding benthos for other organisms with algal s ymbionts. The calculation steps of the FI are shown below: FI = (10 P s ) + (P o ) + (2 P h ) where: P s = N s /T P o = N o /T P h = N h /T where: T = total number of foraminifers counted N s = number of taxa of sy mbiont bearing specimens N o = number of stress tolerant taxa N h = number of specimens of other small heterotrophic taxa
30 Table 4 Functional groups with example genera as originally defined by Hallock et al. (2003) and modified by Carnahan (2005) and Ramirez (2008). Foraminifer Functional Group Ex ample Genera Symbiont bearing miliolids Borelis Laevipeneroplis Peneroplis Archaias Broeckina Cyclorbiculina Sorites Symbiont bearing rotaliids Amphistegina Aste rigerina Heterostegina Smaller miliolids Cornuspira Vertebralina Wiesnerella Hauerina Miliolinella Pyrgo Quinqueloculina Schlumbergerina Triloculina Spiroloculina Articulina Other, smaller taxa Reussella Discogypsina Discorbis Planorbulina Rosalina Te xtularia Stress tolerant taxa Bolivina Cribroelphidium Elphidium Haynesina Nonionoides Nonion Ammonia Ammobaculites Trochammina
31 Results Physical Parameters ADCP data show that currents in the Durney Key area are tidally dominated with velocit ies increasing in winter due to frontal passages. Average wind speeds were higher in winte r (3.7 m/s or 12.4 ft/s) than summer (3.1 m/s or 10.2 ft/s ), resulting in increased average water velocities ( Table 5 ). Water velocity var ied with tidal intensity, being faster during spring tides than neap tides for both north/south and east/west directio ns and in both seasons (Tables 6 and 7) Wind tends to be stronger and more consistent in winter than summer in Florida, except for durin g tropical storms and hurricanes of which there were none in the vicinity of the ADCPs during their deployment East west velocities were consistently stronger than the north south water velocities ( Table 5 ). This trend is not surprising because the Florida coastline is oriented north south in the Durney Key vicinity Table 5 Root mean squared tided and detided water velocities for summer and winter currents at Durney Key. ADCP Water Direction Sum mer Velocity (cm/s) Summer Detided Velocity (cm/s) Winter Velocity (cm/s) Winter Detided Velocity (cm/s) A East/West 5.20 2.27 7.99 5.53 North/South 2.72 2.04 5.49 4.58 B East/West 5.44 2.31 8.72 4.89 North/South 1.81 1.57 4.33 3.53 C East/West 10.8 3.61 9.98 5.94 North/South 5.55 2.38 6.0 4.36 Average 5.26 2.36 7.08 4.81
32 Table 6 Water velocity variability between spring and neap tides in summer, where Ve is the eastward component and Vn is the northward component. I nstrument Water direction Tide type Velocity range (cm/s) ADCP A 07 Ve Neap 4.4 2.3 Spring 10.3 10.3 Vn Neap 2.2 1.4 Spring 4.7 5.9 ADCP B 07 Ve Neap 4.1 3.3 Spring 10.9 10.6 Vn Neap 1.2 .1 Spring 1.6 2.7 ADCP C 07 Ve Neap 0.5 3.2 Spring 10.6 10.6 Vn Neap .6 .2 Spring 1.6 2.7 Table 7 Water velocity variability between spring and neap tides in winter; abbreviation s as in Table 6 Instrument Water direction Tide type Ve locity range (cm/s) ADCP A 08 Ve Neap 7 5 Spring 13 12 Vn Neap 2.5 3.5 Spring 6.2 6.7 ADCP B 08 Ve Neap 8.1 4.4 Spring 16 14.7 Vn Neap 2.4 2.6 Spring 3.9 4.5 ADCP C 08 Ve Neap 7.2 5.5 Spring 14.6 15.5 V n Neap 3 5 Spring 6.8 9.5 The dominance of tides in determining water velocities is clear from the tidal pattern observed in the water velocity plots (Figs. 8 11) The percentage of variation in water velocity explained by tides, as determine d by Foreman Tidal Analysis ( Pawlowicz et al., 2002) are listed in Tables 8 and 9 These percentages are generally greater in summer than winter, when there is more wind influencing the water movement. This reinforces the visual concept presented in the stage and water velocity plots
33 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7 W a t e r V e l o c i t y ( c m / s ) W a t e r V e l o c i t y ( c m / s ) 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7 Figure 8 Comparison of tided and detided summer east west water velocities from the south side of Durney Key (ADCP A). Figure 9. Comparison of tided and detided summer north south water velocities from the south side of Durney Key (ADCP A).
34 W a t e r V e l o c i t y ( c m / s ) 2 / 1 4 2 / 2 1 2 / 2 9 3 / 7 3 / 1 4 3 / 2 1 3 / 2 9 4 / 6 M o n t h / D a y o f 2 0 0 8 2 / 1 4 2 / 2 1 2 / 2 9 3 / 7 3 / 1 4 3 / 2 1 3 / 2 9 4 /6 M o n t h / D a y o f 2 0 0 8 W a t e r V e l o c i t y ( c m / s ) Figure 10 Comparison of tided and detided winter east west water velocities from the south side of Durney Key (ADCP A). Figure 11 Comparison of tided and detided winter north south v elocities from the south side of Durney Key (ADCP A).
35 Table 8 Percent velocity variation attributable to tides in summ er 2007. ADCP Water direction Var iation predicted/var iation original (%) A Ve 79.0 Vn 51.0 B Ve 81.5 Vn 26.2 C Ve 86.6 Vn 80.4 Table 9 Percent of velocity variation attributable to tides in winter 2008. ADCP Water direc tion Var iation predicted/var iation original (%) A Ve 51.6 Vn 27.4 B Ve 68.6 Vn 25.0 C Ve 65.1 Vn 43.0 Wind speed was plotted against the detided water velocity to determine how much non tidally influenced velocity is explained by the wind speed. Both summer and winter show the same general pattern with predominately low wind speeds and low water velocities as shown in Figures 12 15 The correlations are statistically significant (p < 0.05) because the number of observations is large (N ~ 10 4 ) However, the regressions explain very little of the variability, i.e., the R 2 values are less than 1% for the summer data and less than 5% for the winter data, when there is m ore wind (Figs. 12 15) While winds may contribute partially to increased w ater velocity during frontal passages during winter when wind blows in one direction, the contribution is reduced by the proximity to shore. This is supported by the low r values in the wind:water veloci ty correlations (Figs. 12 15 ). Winds may contrib ute more to water level, especially during frontal passages when water level may be dominated by wind set up. Frontal passages are also marked by sharp temperatur e decreases, as seen in Figure 19
36 W in d S p e e d ( m / s ) V e W a te r V e lo c ity ( c m / s ) Wind Speed (m/ s) V n Water V elocity (cm/ s) Figure 12 Correlation of detided east west summer 2007 water velocity and wind for Durney A where red trendline indicates a log regression and the black trendline indicates a linear regression. Fi gure 13 Correlation of detided north south summer 2007 water velocity and wind for Durney A where the red trendline indicates a log regression and the black trendline indicates a linear regression.
37 Stage measurements were recorded by the piezoresistive pressure sensor on each ADCP. Each ADCP was set less than 10 cm (3.9 in) above the sediment water interface, thus the stage is the water height from the sediment top to the air sea interface at each V e W a t e r V e l o c i t y ( c m / s ) W i n d S p e e d ( m / s ) V n W a t e r V e l o c i t y ( c m / s ) W i n d S p e e d ( m / s ) Figure 14 Correlation of detided east west winter 20 08 water velocity and wind for Durney A where the red trendline indicates a log regression and the black trendline indicates a linear regression. Figure 15 Correlation of detided north south winter 2008 water velocity and wind for Durney A 2007 where the red tide trendline indicates a log regression and the black trendline indicates a linear regression.
38 location. Each stage record was averaged over the sampling period to get the average stage for each ADCP duri ng each sampling period ( Table 10 ). The complete stage record s for ADCP A in summer 2007 and winter 2008 are shown in Figures 16 and 17. The moon phases are overlain on the stage record to show the influence of spring and neap tides. The strong diurnal inequality has a sharply increased range during s pring tides and a much narrower range during neap tides as would be expected (Figs. 16 and 19 ). Seasonal stage variation ranges from 0.29 m (11.4 in) to 0.64 m (2.1 ft) which is a considerable amount relative to the shallow water depth surrounding Durney Key ( Table 10 ). The larger seasonal difference seen at ADCP C is suspect, as the units were at the same location for both deployments and both A and B had equal average seasonal stage differences. A mechanical issue may have cau sed this, or strong wind events creating different conditions on the north side of the island, where ADCP C was located. Table 10 Average stage height in meters as recorded by ADCPs in summer 2007 and winter 2008 with stage heigh t difference due to seasonality shown in last column. Unit Summer 2007 Winter 2008 Difference ADCP A 1.7 m 1.41 m 0.29 m ADCP B 1.3 m 1.01 m 0.29 m ADCP C 1.63 m 0.99 m 0.64 m
39 2 / 1 4 2 / 2 1 2 / 2 9 3 / 7 3 / 1 4 3 / 2 1 3 / 2 9 4 / 6 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 / 2 0 8 / 2 3 Month/ Day of 2007 Figure 16 Summer 2007 water height from the south side of Durney Key (ADCP A) where crescent moons represent neap tides and circular moons represent spring tides, the black circle represents the new moon and blue circle represents the full moon. Fi gure 17 Winter 2008 water height from the south side of Durney Key (ADCP A) where crescent moons represent neap tides and circular moons represent spring tides, the black circle represents the new moon and blue circle represents the full moon. Month/ Day of 2008
40 Temp erature records from Durney Key as recorded by ADCP A for both summer a nd winter are shown in Figures 18 and 19 Summer temperature ranged from 28 to 35 C while winter temperatures ranged from 10 to 27 C during ADCP deployments. 2 / 1 4 2 / 2 0 2 / 2 7 3 / 4 3 / 1 0 3 / 1 7 3 / 2 3 3 / 2 9 4 / 4 4 / 1 0 T e m p e r a t u r e ( C ) T e m p e r a t u r e ( C ) 7 / 1 6 7 / 2 6 8 / 5 8 / 1 5 8 / 2 3 M o n t h / D a y o f 2 0 0 7 Figure 18 Temperature as recorded by ADCP A in summer 2007. Figure 19. Tempera ture as recorded by ADCP A in winter 2008. Month/ day of 2008
41 Chemical Parameters Water Samples Results of monthly water samples collected between July 2007 and April 2008 analyzed for temperature, salinity, pH and total dissolved solids (TDS) are shown in Table 11 Seasonal salinity averaged 16.9 ppt in the spring/ summer and 19.8 ppt in the fall/ winter at the COMPS station, which is considerably lower than the salinities measured at Durney Key. Table 11 Water characteristi cs as measured using an Ultrameter. Date Time (EST) Depth (m) Temperature ( C) Salinity ( ppt ) pH TDS 7/16/07 14:00 0. 152 28.7 25.0 8.24 41.2 8/22/07 14:50 0.152 33 22.3 8.39 41.6 9/11/07 14:00 0.152 30.2 23.7 8.21 44.3 11/8/07 10:00 0.152 18.5 32.8 7.8 8 46.1 12/14/07 13:30 0.152 24.7 27.7 7.93 44.6 1/18/08 14:20 0.152 17.4 30.0 7.06 41.9 2/4/08 13:30 0.152 20.4 27.2 7.94 39.9 2/17/08 15:30 0.152 21.6 27.9 8.06 42.5 4/10/08 9:30 0.152 22.7 23.5 Average summer 30.6 23.7 8.3 42.4 Average winter 20.8 28.2 7.8 35.8 Inorganic nutrient measurements are summar ized by season in Table 12. The method detection limit (MDL) is the minimum quan tity of a substance able to be distinguished from a blank value with 95% confidence. MDLs are unique to the me thod as many analytes require unique methods of analysis. Analyses that involve multiple steps, e.g., acid digestion, allow for introduction of additional error, thus the MDL is the amount of the analyte able to be detected after factoring in the sum of p otential error. The results of each sample are listed in Tables 1 3 of Appendix II
42 Table 12 Inorganic nutrient concentrations in water column at mean low water where MDL = Method Detection Limit. Summer 2007 Variable Dep th (m) Median (M) Minimum (M) Maximum (M) N MDL NO 3 + NO 2 0.25 0.004 0 0.184 9 0.25 NO 2 0.25 0.014 0.008 0.048 9 0.03 NH 4 + 0.25 1.06 0.8 5 1.97 9 0.45 Si(OH) 4 0.25 0.099 0.041 0.24 9 0.88 TP 0.25 9.70 7.40 10.5 9 0.06 Winter 2008 NO 3 + NO 2 0.25 0.05 3 0 7.7 14 0.25 NO 2 0.25 0.010 0 0.15 14 0.03 NH 4 + 0.25 1.36 0.11 3.77 14 0.45 Si(OH) 4 0.25 0.5 0 0 1.90 14 0.88 TP 0.25 4.18 0.06 41.8 14 0.06 Pollutants in Sediment Samples F ive persistent organic pollutants (POPs) were found in the sediment sampl e above the minimum detection level (MDL) these were: delta benzene hexachloride (d BHC also known as lindane), dieldrin, endosulfan sulfate, gamma benzene hexachloride (g BHC, an isome r of lindane) and heptachlor ( Table 13 ). In addi tion to these five, there were another five POPs denoted with an I in the Flag column that had values between the laboratory method detection limit (MDL) a statistically derived value, and the laboratory method reporting limit (MRL) what the lab can see with a real reference standard MDLs are unique to the method of analysis because many analytes use different extraction and detection processes to quantify or measure the analyte. None of the semivolatile organic compounds were detected above the m inimum detection limit (MDL) in the analysis of the surficial sediment sample from Durney Key using EPA 8270C method. None of the polychlorinate biphenyl compounds were detected above the MDL in the analysis of surficial sediment sample from Durney Key us ing EPA methods 8082, 8083, 8084, 8085, 8086, and 8087 methods (United States Environmental Protection Agency, 2008) The concentrations of eight metals of concern are listed in Table 14 two of which
43 were below the detection limit and are denoted with ND. The full suite of metals analyzed is reported in Table 4 of Appendix II. Table 13 Summary of organic compounds in surficial sediment sample from the south side of Durney Key, where MRL = Minimum Reporting Limit and I = the reported value is between the labor atory method detection limit (MD L) and the labor atory method reporting limit (MR L), adjusted for actual sample preparation data and moisture content. ENCO Summary sample date: July 26, 2007 Analyte Results Flag MR L Units 4,4' DDD 0.002 I 0.0023 ppm 4,4' DDT 0.0023 I 0.0023 ppm Chlordane alpha 0.0012 I 0.0024 ppm delta BHC 0.0035 0.0023 ppm Dieldrin 0.0033 0.0023 ppm Endosulfan sulfate 0.0061 0.0023 ppm Endrin 0.0016 I 0.0023 ppm Endrin al dehyde 0.0019 I 0.0023 ppm gamma BHC 0.0024 0.0023 ppm Heptachlor 0.0038 0.0023 ppm Table 14 Concentrations of metals of concern as analyzed in surficial sediment sample. Analyte Symbol Sb As Be Cd Cr Hg Pb Ni Se Ag Tl Zn Unit ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm MDL 0.1 0.5 0.1 0.1 1 1 0. 000 5 0.5 0.1 0.05 0.05 0.5 Seived Sample (fines) ND 6.8 0.5 0.3 43 16.7 0.0 52 21.5 1.9 ND 0.26 24.5 Pulverized Sample (bulk) ND 2.1 0.1 ND 23 5.4 0.0 14 5 1.1 ND 0.11 6.7 S ediment Texture Grain size analysis results from one gram subsample s of the upper and lower sections of each sediment core are shown by weight of sample in Figure 20 and A ppendix III The median grain size for the top of Core 1 is < 0.5 and > 0.25mm wit h a median phi of 2, which is medium sand. All other samples were fine sand, < 0.25 and > 0.125 with a
44 median phi of 3. One gram sediment sub samples from each core half were tested for percent carbonate the resul ts of which are shown in Table 15 Figure 20 Grain size distribution by weight percent of sample for sediment cores ; core 1 wa s collected 10 m from shore, Core 2 at 20 m and Core 3 at 30 m from shore. Table 15 Mass sediment samples remaining af ter dissolution with 50% HCL to estimate % carbonate. Sample Sample wt (g) insoluble fraction % carbonate Sediment Core 1 bottom 1 77 23 Sediment Core 1 top 1 80 20 Sediment Core 2 bottom 1 74 26 Sediment Core 2 top 1 74 26 Sediment Core 3 botto m 1 91 9 Sediment Core 3 top 1 85 15
45 Biological Parameters Foraminifera Twenty one species representing ten genera were identified from the sediment matrix removed from six pieces of lime rock that were collected from the south side of Dur ney Key (Table 16 ). Table 16 Foraminiferal assemblage for se diment from scrubbed lime rock. Lime rock piece 1 2 3 4 5 6 Species Number of shells counted A. parkensonia 1 3 0 41 18 14 C. subpoeyana 8 22 28 22 17 15 E. galvestonense 0 1 1 1 0 0 E. gun teri f. mexicanum 0 0 0 0 0 1 E. gunteri f. salsum 2 9 4 1 4 5 E. gunteri f. typicum 4 9 8 0 0 0 E.galvestonense 0 0 0 0 0 0 H. depressulum 14 20 33 64 26 14 L. bermudezi 189 70 89 64 70 105 L. oblonga 70 0 2 0 1 0 M. circularis 23 17 11 131 0 0 M. labiosa 14 11 12 0 68 191 M. suborbicularis 0 0 0 0 0 1 M. subrotunda 3 2 3 0 0 2 N. depressulum cf Bock 71 1 0 0 0 0 1 P. rotunda 0 4 3 0 0 0 P. sidebottomi 13 15 3 6 34 23 Q. bosciana 70 35 53 0 2 12 Q. lamarckiana 2 23 7 0 0 6 Q. oblonga 1 0 0 0 0 0 Q. poeyana 6 3 0 0 0 0 Q. seminula 0 7 0 0 0 1 Q. subrotunda 0 0 0 0 2 0 R. floridensis 6 0 0 0 0 29 R. subaraucana 93 17 18 19 22 0 Spirillina vivipara 0 1 0 2 0 0 T. linniana 0 0 0 0 2 1 T. linneiana var. comis cf Bock 71 0 0 0 0 1 0 T. oblonga cf Boltovskoy et al. 80 0 0 0 0 0 1 T. trigonula 7 85 12 58 67 32 Sum 527 354 287 409 334 454
46 All species identified were heterotrophic, i.e., non symbiont bearing. The two most numerous species identified w ere Lachlanella bermudezi and Miliolinella labiosa both are considered other small taxa in the FORAM Index (FI). Both Lachlanella and Miliolinella have cosmopolitan distributions ( Loeblich and Tappan, 1987). FORAM Index (FI) values (Table 17) were comp uted f r om the data shown in Table 16. FI values were between 1.7 and 2, indicating the foraminiferal assemblage is dominated by smaller, heterotrophic benthic foraminifers, with stress tolerant taxa making up less than a third of the assemblage Twenty t wo species representing 13 genera were identified from one gram subsamples of the top half of the three sediment cores taken on the south side of Durney Key Ammonia and Elphidium were the two dominant genera overall (Table 18), both have cosmopolitan dis tributions (Loeblich and Tappan, 1987). Of the 481 foraminiferal shells counted, only three were from symbiont bearing taxa. Sediment Core 2, which was the intermediate depth and distance from shore, had the greatest foraminiferal diversity and density. FORAM Index values were calculated for each of these subsamples The FI values, number of foraminifers per gram, number of genera and sediment sample chara cteristics are listed in Table 19. Table 17. The av erage FORAM Index values for samples of sedime nt from scrubbed lime rock with associated number of foraminiferal shells. Lime rock ID Average FI Value Number of shells counted 1 1.95 527 2 1.82 354 3 1.74 287 4 1.69 409 5 1.80 334 6 1.89 454
47 Table 18 Foraminiferal assemblage from sedim ent core samples. Sample Core 1 top Core 2 top Core 3 top Stress tolerant Ammonia parkensonia 7 38 62 A. tepida 8 12 1 Elphidium discoidal f. typicum 0 2 2 E. galvestonense 1 7 30 E. galvestonense f. mexicanum 14 29 11 E. gunteri f. salsum 30 46 11 E. poeyana 7 2 0 H aynesina depressulum 7 13 4 Nonion depressulum 3 6 6 Other small tax a Cycloforina subpoeyana 0 8 52 Lachlanella bermudezi 2 1 0 Miliolinella labiosa 0 1 0 M subrotunda 6 8 0 Pseudotriloculina sidebottomi 0 3 0 Pyrgo spp. 1 0 0 Quinqueloculina bosciana 4 10 0 Q lamarckiana 1 9 0 Q poeyana 0 1 0 Q seminula 1 3 0 Triloculina trigonula 0 1 0 Symbiont Bearing Amphistegina gibbosa 0 0 2 Cyclorbiculina compressus 0 0 1 Total Forams Picked 92 207 18 2 Table 19 Ra w foraminiferal shell counts and FORAM Index values with associated sediment sample characteristics for sediment core subsamples. Sample SC1 T SC2 T SC3 T Mass Assessed (g) 1 1 1 Forams/Gram 92 207 180 Number of Genera 8 11 5 % Mud 0.48 0.22 0.50 Phi, Median Grain Size 2 3 3 FORAM Index 1.16 1.18 1.39
48 Molluscan assemblages Seven species representing seven genera of bivalves and 29 species representing 22 genera of gastropods were identified from one gram subsamples for each half of the three sedi ment cores The species assembla ge results are listed in Table 20 Table 20 Molluscan species assemblage per one gram subsample. Species SC 1 B SC 1 T SC 2 B SC 2 T SC 3 B SC 3 T Bivalves Abra aequalis 1 0 0 2 0 0 Anodontia alba 1 0 0 0 0 0 Gouldia cerina 0 0 0 0 0 2 Laevicardium mortoni 0 1 1 1 0 0 Musculus lateralis 0 1 0 0 0 0 Nucula proxima 0 8 2 9 14 21 Periglypta listeri 0 0 0 2 1 2 Strigilla mirabilis 14 0 0 0 0 2 Raeta plicatella 0 1 0 0 0 1 Totals 16 11 3 14 15 28 Gastropods Aclis underwoodae 0 0 0 0 1 0 Astyris lunata 0 1 1 2 0 1 Boonea i mpressa 0 0 0 0 1 0 Caecum nitidum 0 2 0 0 0 0 C. pulchellum 0 3 9 3 1 Capulus ungaricus 1 2 2 0 0 1 Cerithium spp. 0 1 0 0 0 0 C. eburneum 3 7 10 7 4 0 C. lutosum 1 7 3 9 0 6 C. m ulticostatum 0 0 1 0 0 2 C. muscarum 0 1 0 0 0 0 Cyclostremiscus pentagonus 0 3 0 1 0 0 Cylindrobulla beauii 0 1 0 0 0 0 Granulina ovuliformes 0 0 0 0 1 0 G vitrea 0 0 2 0 0 3 Haliotinella patinaria 0 10 0 0 2 2 Lima scabra scabra 0 2 0 1 5 0 Litt orina spp. 0 0 0 0 0 3 L. mespillum 0 1 0 0 3 0 Marginella spp. 0 0 0 0 2 0 Meioceras nitidum 0 1 4 4 8 0 Oliva sayana 0 0 0 0 1 18
49 Table 20 (Continued) Olividae spp. 0 0 0 0 0 1 Rissoina catesbyana 1 7 1 0 0 0 Truncatella pulchella 0 0 1 4 0 0 Tur bonilla aequalis 0 1 0 0 0 0 Volutidae spp. 0 0 1 0 0 0 Gastropod Totals 6 59 37 40 43 59 Total 22 52 36 36 30 43 Recruitment Early recruits to the tiles consisted of coralline, turf algae and macroalgae with more turf and macroalgae pr esent on the upper tiles (Fig. 21). Coralline algae coverage decreased through time while macroalgae peaked in August and turf algae c overage remained about the same. Barnacles occupied about 6% of the tile space on the upper tiles by August, although their coverage de creased by September The coverage classes of biotic growth on recruitment tile images were analyzed in PRIMER v. 6 to determine differences in recruitment based on the tiles position in the water column, i.e ., whether the tile was at the top of the s take or the bottom. Both porcelain and ceramic tiles were analyzed together because there was no significant difference in recruitment between the tile types. All three time steps were grouped together and recruitment was analyzed based on tile p osition in an MDS plot (Fig. 22 ). Percentages for each recruitment class and total ar ea covered are listed in Table 21; because some organisms overlapped, the sum of classes may be > 100 SIMPER and ANOSIM results of all time steps combined a re reported in A ppend ix I V
50 Figure 2 1. Average coverage area per class for each time step grouped by position on stake, i.e., top vs. bottom. Table 21 Summary stat i s tics for average cover and variability of recruitment cla sses on tiles where C.A. = Coralline algae, T.A.C. = Total Area Covered and B.A.F. = Brown Algal Film Area reported in percentage of tile covered. 7/16/07 Barnacle C .A. Macroalgae Turf Unknown T .A.C. B.A.F. Avg Bottom 0.00 16.10 0.08 2.04 0.00 18.21 100 Std. Dev. Bottom 0.00 15.17 0.34 4.62 0.00 15.67 15.67 Avg Top 0.00 10.27 0.71 12.86 0.00 23.97 100 Std. Dev Top 0.00 16.72 6.71 20.08 0.00 25.99 28.16 8/22/07 Avg Bottom 0.40 5.35 1.63 12.71 0.02 20.08 80.14 Std. Dev. Bottom 1.64 8.92 16.45 17.32 0.00 20.26 20.37 Avg Top 5.63 3.46 6.87 11.54 0.01 26.96 76.27 Std. Dev Top 9.75 5.60 19.56 24.62 0.00 31.91 31.91 9/11/07 Avg Bottom 0.35 4.53 1.38 10.75 0.02 16.99 86.23 Std. Dev. Bottom 1.00 7.88 6.46 16.36 0.10 19.99 19.99 Avg Top 4.18 2.57 5.11 8.58 0.01 20.05 84.60 Std. Dev Top 8.29 4.70 13.37 19.83 0.00 29.88 23.79
51 Fi gure 22 MDS plot displaying the degree of similarity between bottom and top tile positions with all ti me steps included, with both brown algal film and total area covered classes removed. Only two tiles, 19 A and 19 B, were recovered and analyzed for faunal assemblages. Tile 19 A was dominated by Miliolinella the majority of which were deformed (Table 22 ), tile 19 B had far fewer foraminifers (Table 24) A n image of a representative deformed Miliolinella is shown in Figure 23 Foraminiferal species as semblages are listed in Table 23 and 25 Table 22 Raw foraminiferal counts from ceramic recruitment tile 19 A. Sample Miliolinella Deformed Miliolinella % Deformed 1 30 20 66.7 2 61 42 68.9 3 55 33 60 4 24 16 66.7 5 20 16 80 6 50 22 44 7 50 28 56 8 21 11 52.4 9 46 29 63
52 Table 22 (Continued) 10 35 22 62.9 Total 392 239 61 Table 23 Density of foraminifera l taxa found on porcelain tile 19 A per s quare centimeter. Sample Number 1 2 3 4 5 6 7 8 9 10 Species Ammonia spp. 0 0 1 0 0 0 3 0 1 0 Cymbal operetta spp. 0 0 1 0 0 0 0 0 0 0 Lachanella bermudezi 0 0 1 0 0 0 2 0 2 0 Miliolinella spp. 20 50 50 21 46 95 95 61 55 24 Planorbulina spp. 0 0 1 0 0 0 0 0 0 0 Pseudotriloculina sidebottomi 1 0 0 0 0 0 0 0 0 0 Quinqueloculina laevigata cf Bock, 1979 0 0 0 2 0 0 0 Q poeyana cf Bock, 1971 2 0 0 0 0 0 0 0 0 Q subpoeyana cf Bock, 1979 0 0 0 2 0 0 0 Rosalina concinna 0 0 0 0 1 3 10 0 0 0 R floridensis 0 0 0 0 0 0 5 4 3 1 R spp. 0 1 0 0 0 0 3 0 0 0 R subaraucana 0 0 1 0 0 0 1 5 0 0 Triloculina laveona 1 0 0 0 1 0 6 1 0 0 T trigonula 18 2 5 3 5 13 9 2 6 13 Total 40 55 60 24 53 111 138 68 67 38 Table 24 Raw foraminiferal counts fr om ceramic recruitment tile 19 B. Sample Miliolinella Deformed Miliolinella % Deformed 1 0 0 0 2 0 0 0 3 0 0 0 4 1 0 0 5 1 1 100 6 0 0 0 7 4 1 25 8 5 0 0 9 5 2 40 10 6 2 33.3 Total 22 6 27.27
53 Table 25 Density of foraminiferal taxa found on porcelain tile 19 B per square centimeter. Sample Number: 1 2 3 4 5 6 7 8 9 10 Species Miliolinella spp. 0 0 0 1 1 0 4 5 5 6 Pseudotriloculina sidebottomi 1 0 0 0 0 0 0 0 Quinqueloculina subpoeyana 0 0 1 0 0 0 0 0 Q tengos 0 1 0 0 0 0 0 0 0 Rosalina floridana 1 0 0 0 0 0 0 0 0 0 R subaraucana 1 0 0 0 0 0 0 0 0 0 Triloculina trigonula 7 8 4 1 11 3 7 2 0 18 Total 9 8 6 2 13 3 11 7 5 24 Non foraminiferal taxa identified on the tiles are listed in Table 26 and 27 Both sides of the tiles were examined for this step and the smooth side of the tile refers to the non texturized back while the texturized side refers to the front, i.e. the side facing away from the PVC stake. There were also limited coralline algae and minimal filament ous brown algae present. Table 26 Species found on ceramic tile 19 A. Smooth side (back) Tex turized side (front) Taxon Number counted Taxon Number counted Balanus eburneus 7 Balanus eburneus 14 Pictada imbricata 1 Diplosoma spp. Colony 1 Serpulid w orms mature 29 Serpulid worms mature 2 Diplosma spp. Colonies 2 Serpulid worms recently settled 15 Echiuran 2 Echiuran recently settled 4 Figure 23. Example of a deformed Miliolinella from ceramic tile 19A.
54 Table 27 Species found on ceramic tile 19 B. Smooth side (back) Tex turized side (front) Taxon Number c ounted Taxon Number counted Balanus eburneus 2 Balanus eburneus 11 Brachiodontes exustus 8 Cyclopoid copepod 1 Gammaridae amhipods 12 Stylea plicata 1 Serpulid worms 11 Spheroma quadridutata 4
55 Discussion The basic c onclusion that can be drawn from my study is that the waters around Durney Key presently cannot support growth of reef building corals, i.e., zooxanthellate Scleractinia, specifically the guild of Caribbean species that construct the reef framework, e.g. A cropora, Montastraea, Colpophyllia, Diploria or Siderastrea The physical, chemical and biological data presented above all support this assessment; each category of data will be discussed further below. Interpretations of data collected for this study are limited because of the dearth of data available for the Durney Key area. This does not diminish their importance; rather it implies their future worth. The limited number of sample s is offset by the wide range of variables examined. Each variable, e .g. FI, recru itment, geochemical analyse s, etc., provides the first known data point s for this area, forming a baseline for future studies. The loss of all but two recruitmen t tiles did not allow for an in depth species assemblage analysis. The low reso lution of recruitment classes from photographs rather than assessments of species severely limited detailed analyses of recruitment patterns. While the primary objective of determining site suitab ility for coral growth was achieved, the proposal that recr uitment plates could more fu lly elucidate seasonal patterns and temporal recruitment fluctuation s was not verified The two recovered tiles did show promise for sampling foraminiferal assemblages. Physical Parameters Two parameters from of the ADCP da ta set of importance to predictions of coral recruitment and growth are temperature and water velocity ; neither indicate that an artificial reef would support stony coral populations at this site. Western Atlantic reef building corals thrive in G oldilock s scenario conditions in which the water is warm, but
56 not too warm between 18 and 32 C with the optimal range being 26 28 C (Birkeland, 1997). Temperatures around Durney Key have a far greater range, comparable to the Persian Gulf but too great for m ost Western Atlantic and Caribbean corals. Corals require good water flow, but not too strong (Birkeland, 1997) Since corals are sessile organisms incapable of relocation, they depend on water flow to move potential prey within their reach. Reduced wa ter flow will reduce prey encounter rates as well as potentially limiting diffusion, which may impact physiological rates (Sebens and Johnson, 1991; Finelli et al., 2007). I suspect the average summer water velocities would fall below good flow for cora ls as these velocities fall within the range of concern discussed by Sebens and Johnson (1991). The strong tidal influence is the most striking feature of the ADCP data set and one with management implications The main driver of water velocities is tides and there is a strong semi diurnal inequality present. The seasonal stage difference, due to thermal expansion and contraction, is notable for two reasons. I f a structure were to be submerged, an extra 0.7 m (2.3 ft) would need to be added to the plann ed dep th from the summer stage height. S econdly there are many recreational users of the area who should be aware of this seasonal difference to avoid damage to their watercraft and the surrounding seagrass beds. Chemical Parameters The chemical data assembled during this study further indicate that recruitment of stony corals to an artificial reef at Durney Key would be unlikely. Salinity was consistently too low and pH was frequently too low. The presence of high total phosphorus and of potentiall y toxic elements in the sediments are also of concern. Water samples To interpret water quality characteristics, I compared Durney Key data to available water quality data for patch re efs in the Florida Keys ( SERC FIU 2008) P atch reef sites were chose n by determining which SERC sampling sites overlapped with
57 CREMP patch reef sites having total coral cover (both stony and octocoral) grea ter than 15%, as shown in Table 28 ( Callahan et al. 2007 ). Salinity is significantly lower than the Keys sites, 12.1 ppt lower on average in summer and 8 ppt lower on average in winter (Table 29 ) which is attributable to the outflow of the nearby Pithlachascotee River; the riverine input may also contribute to the elevated total phosphorus Temperatures around Durney Key show much greater variability than the Keys sites, the shallow depth may explain this. Similarly winter pH values were commonly less than 8, which would also be problematic (Feely et al., 2004) Table 28 Percent coral cover at patch reef sites wher e water quality monitoring stations coincide with CREMP monitoring stations where WQMN = Water Quality Monitoring Network and CREMP ID = Coral Reef Evaluating and Monitoring Project. WQMN ID CREMP ID % Total Stony Coral Cover % Total Octocoral Cover Admir al Patch (224) Molasses Reef Channel (9P4) 24 23.3 Grecian Rocks (400) Porter Patch (9P3) 3.53 11.9 Turtle Harbor (212) Turtle (9P1) 5.34 29.6 Coffins Patch Channel (248) W. Turtle Shoal (7P1) 12.7 22.2 Coffins Patch Channel (248) Dustan Rocks (7P2) 14 .7 27.5 Cliff Green (5P3) Boca Chica Mid (275) 15.3 27.5 Table 29 Water characteristics in the Keys patch reef sites that coincide with CREMP monitoring sites from quarterly sampling carried out during 2006 2007 (SERC FIU Water Quality Monitoring Netw ork). Site Date Time (EST) Depth (m) Temperature ( C) Salinity ( ppt ) Molasses Reef Channel 11/20/2006 12:10 6.8 23.3 36.4 Boca Chica Mid 12/14/2006 11:44 13.5 25.2 36.3 Molasses Reef Channel 12/19/2006 11:23 6.3 23.7 35.4 Coffins Patch Channel 1/24/200 7 12:33 6.8 25.4 36.2 Molasses Reef Channel 1/31/2007 11:06 6.3 21.5 37.2 Boca Chica Mid 2/21/2007 11:49 13.5 20.4 35.5 Coffins Patch Channel 4/27/2007 11:53 6.8 25.3 36.4 Molasses Reef Channel 6/7/2007 9:30 6.3 26.8 36.3 Boca Chica Mid 6/14/2007 13:4 9 13.5 29.6 35.6 Boca Chica Mid 8/23/2007 13:53 13.5 30.3 36.6
58 Table 29 (Continued) Coffins Patch Channel 9/6/2007 11:24 6.8 30.6 36.0 Molasses Reef Channel 9/26/2007 16:29 6.3 28.4 34.5 Boca Chica Mid 10/24/2007 15:52 13.5 28.7 35.9 Molasses Reef Cha nnel 11/26/2007 11:59 6.3 25.2 36.3 Coffins Patch Channel 11/29/2007 12:11 6.8 25.8 36.3 Average summer: 29.1 35.8 Average winter: 24.0 36.2 Water quality characteristics were also compared to local data from Tampa Bay, Florida (Table 30) (Dix et al., 2005). These data were more similar to the Durney Key waters with the largest distinction being a lower average pH of 7.8 in summer at Durney Key. Table 30 Summary of physical variables at Howard Franklan d Reef, Tampa Bay, Florida (Dix et al., 2005). Spring 2004 Sample Depth (m) Temp ( C) Salinity ( ppt ) Dissolved Oxygen (mg/l) pH N 9 10 10 10 10 Min 3.3 21.6 23.6 6.7 8.1 Max 4.8 21.8 23.9 7.2 8.1 Median 4.1 21.7 23.6 6.9 8.1 Mean 4.2 21.7 23.7 6.9 8.1 Std. Dev. 0.5 0.1 0.2 0.0 0.0 Fall 2004 N 10 10 10 10 10 Min 4 .3 30.0 20.3 4.1 8.0 Max 5.6 30.5 20.6 6.2 8.3 Median 5 .1 30.4 20.4 4.6 8.1 Mean 4.9 30.4 20.4 4.9 8.1 Std. Dev. 0.4 0.0 0.1 0.7 0.1
59 Because there is no water atlas or history of marine sampling at or around Durney Key, t h e nutrient data reported here are the first known Thus while t hese data cannot be comp ared to a historical data set, t hey can serve as a baseline for future studies. I nput from the Pithlachascotee as well as from coastal development, may be contributi ng to the high TP. My data do not indicate that other nutrient concentrations are problematic at Durney Key. Nutrients are generally the controlling factor for phytoplankton and other plant life if sunlight is abundant. However, inorganic nutrients are q uickly taken up by phytoplankton and thus may be underrepresented in samples taken from the wa ter column ( Laws and Redalje, 1979 ). The levels of T P in the waters around Durney Key argue for elevated phosphorus inputs to the waters which is not unexpected given the abundance of phosphate, especially in Miocene rocks and sediments (Hine et al. 2003) With TP levels an order of magnitude higher than levels found in the Florida Keys (Table 31) where eutrophication is already a concern, there is little doub t the eutrophic conditions would be an issue for any corals pres ent in the Durney Key area. Hallock ( 1988 ) argued the importance of oligotrophic conditions for coral survival, pointing out phosphate levels in oligotrophic waters seldom exceed 0.1 M, in eutrophic waters they often exceed 1 M. These guidelines place the Durney Key waters firmly in the eutrophic category. These nutrient concentrations help explain the rapid recruitment of algae and barnacles and support the conclusion that coral recruit s would be outcompeted. Table 31 Summary statistics for selected water quality variables in the Keys patch reef sites that coincide with CREMP monitoring sites from quarterly sampling carried out during 2007 (SERC FIU Water Quality Monitoring Network). Variable Depth (m) Median (M) Minimum (M) Maximum (M) N NO 3 + NO 2 0.25 0.228 0 .069 1.059 15 NO 2 0.25 0.043 0.007 0.112 15 NH 4 + 0.25 0.407 0.134 2.774 15 Si(OH) 4 0.25 0.106 0.012 3.234 15 TP 0.25 0.208 0.117 0.474 15
60 Pollutants in Sediment S amples The surficial sediment sample analyzed for persistent organic pollutants (POPs) was the first such analysis known for the Durney Key area. The five POPs above the minimum reporting level were delta BHC, gamma BHC, dieldrin, endosulfane sulfate (a breakdown product of endosulfane) and heptachlor, all of which are organochlorine pesticides. These POPs may have been introduced from runoff from the increasingly urbanized Pithlachascotee watershed. While none are present at alarmingly high levels, th ere is cause for concern if these compounds bioaccumulate in higher trophic level organisms that are harvested, such as tarpon and snook. The surficial sediment sample analyzed for metals was the first such analysis known for the Durney Key area. This sample was split and one subsample was sieved at 230 m and oversized sediment w as not included in the analysis; the other subsample was pulverized and analyzed in totality to avoid the confounding effects of increased sorption on smaller grain sizes. It i s difficult to determine the origin of measured metal concentrations, specifically whether they reflect natural concentrations or are anthropogenically enriched. Metals are naturally occurring in sediments and their concentrations vary widely with sedimen t type and grain size (Schropp et al., 1990). This variability makes comparisons to locations other than the one sampled difficult as well, and as there were no previous metal analyses in the area, there i s no comparative data included. However the resul ts fall within those seen in Biscayne Bay, Florida (Carnahan, 2005). Using past studies and standards set forth for sediment trace metal concentrations, there are several metals present at concentrations of concern. Two metals, Pb and Ni had concentrati ons over the effects range low, while Hg was over the effects range medium set forth by NOAAs National Status and Trends Program (National Oceanic and Atmospheric Association, 1999). Concentration of Cd would be of concern in fresh wa ter systems, b ut data points to marine organisms having higher tolerance for Cd (Long and Morgan, 1990). Concentration of Cu was higher than the concentration of concern for marine systems (Long and Morgan, 1990) and is more than three times higher than the 5 ppm conce ntration proposed by Klapow and Lewis (1979). Concentration of Pb
61 was ~ 15 times the lower range of the concentration of concern (Eisler, 1988). Concentration of Hg was ~ 3 times higher than the marine water quality standard purposed by Klapow and Lewis ( 1979). This analysis did not investigate speciation of the metals, which influences bioavailability If investigation into the potential relationship between heavy metals and foraminiferal test deformity is attempted, this would need to be determined. Sediment Texture Median grain sizes in sediment cores 1 and 2 were fine sands, with the exception of the top of Core 1, which was slightly coarser. There was little difference is grain size distribution between the top and bottom portions of the cores The sediment composition and grain size distribution is not representative of purely natural processes around Durney Key due to the dredged origin of the sediment. The lack of grain size difference between top and bottom portions is not surprising given the limited wave action with which to sort the sediment grains, and presence of baffling and binding agents (i.e. seagrasses and algae, respectively). Bioturbation, i.e., the biological turnover of sediment, can also alter the grain size distribution. The presence of molluscan shells and fragments in the sediment is reflected in the large percentage of carbonate present in the sediment, which ranged from 9 26 % as determined by acid dissolution. Biological Parameters Hardbottom habitats, i.e., lithi fied seafloor, are common along the west central Florida coast and provide important marine habitat ( Obrochta et al., 2003) Common species found on hardbottom outcrops include psammophytic green algae ( Caulerpales ) calcareous algae including Halimeda and red coralline algae, boring mollusks ( Lithophaga sp.), boring sponges ( Cliona sp.), and echinoderms ( Obrochta et al., 2003) Thirty two ahermatypic species (i.e., do not contain zooxanthellae) found in the Gulf of Mexico were described in the Hourglass m emoirs, six of which were found in shallow water (< 50m) (Cairns, 1977). Shallow areas with significant amounts of sediment
62 overlaying hardbottom typically support seagrasses, as is seen off of Pasco County, Florida. The proposed artificial reef emplace ment would require seagrass removal, which is undesirable due to the importance of seagrasses in the flats ecosystem. Moreover, seagrass removal would have secondary, u nintended deleterious effects on the desired coral gro wth by increasing sediment suspen sion and that might smother any coral s that recruited or were transplanted. Without the seagrass present to baffle water and bind the sediment together, there would be open, muddy expanses of sediment prone to being suspended in the water column. The lac k of seagrass would remove the vertical structure which allows baffling to occur, and provides a habitat for microbial and algal mats that act to bind the sediment surface, keeping the sediment in place Foraminiferal Assemblage The three different data sets for foraminiferal assemblages all demonstrate that the environment at Durney Key does not support calcifying organisms that host algal symbionts (i.e., larger benthic foraminifers and stony corals). The data set that best reflects what is living on h ard substratum at the site, the lime rock, indicates a diverse assemblage of smaller benthic foraminifers, with a significant number but not dominance by stress tolerant taxa. T he resulting FoRAM Index of ~1.7 1.9 supports this interpretation. The diag nosis based on this assemblage is that the environment is basically suitable for a typical west Florida coastal benthos dominated by algae, seagrass, mollusks, etc., but not for coral growth. The recruitment tiles indicate what recruited during the deploy ment time, which was predominantly Miliolinella This too supports the basic interpretation from the lime rock assemblage. However, the presence of so many deformed specimens is problematic; the best case scenario is that the malformations result from sa linity and pH variability, worst case scenario is they deformities are the result of toxic contaminants in the environment. The dominance of the stress tolerant Ammonia Elphidium assemblage in the sediments from the cores also indicates an environment not conducive to coral growth. This assemblage was predominate in restricted (sal i nities < 35 ppt ) in northern Biscayne
63 Bay (Ishman et al 1997) and common near urban Miami (Carnahan, 2005) and Tampa Bay, Florida (Dix, 2001) Both Ammonia and Elphidium are considered pollution tolerant (Sen Gupta, 1999) or, more broadly, stress tolerant (Hallock et al., 2003). Ammonia is particularly tolerant of hypoxia. Recent studies in mid to low latitude settings have reported Ammonia species to commonly be the most str ess tolerant taxa (Buzas Stephens and B uzas, 2005; Bergin et al., 2006; Burone et al., 2006; Unlu et al., 2006; Frontalini and Coccioni, 2008), often exhibiting shell deformities or even dissolution in extreme conditions. While both Ammonia and Elphidium both are stress tolerant genera, it is important to make the distinction that they do not only occur in marginal environments but that their predominance in foraminiferal assemblages typically indicates stressful conditions. Furthermore, their dominance as compared with the limerock and recruitment plate assemblage is likely taphonomic. The shells of Miliolinella in particular and the smaller miliolids in general are more brittle and more soluble than those of Ammonia and Elphidium and are much more like ly to dissolve or be broken down to finer pieces and therefore be under represented in the sediment assemblage. The loss of high magnesium calcite miliolids from the sediment assemblage that were present on the recruitment tiles and lime rock assemblage a lso likely reflects pH conditions unconductive to coral growth. By using the FI in an environment other than that which it was developed in, I tested the utility of the index in another setting as well as provided a baseline FI value for an area with no pr evious foraminifera l data. FI values for both lime rock and sediment samples will serve as a valuable baseline for future studies near Durney Key T here are no known previous studies of foraminiferal assemblages nearshore Pasco County and thus nothing di rectly comparable to my results Foraminiferal assemblage data from nearby Tampa Bay were similar, being dominated by Ammonia with Elphidium and Mi liolinella being subdominant in the open bay (Poag, 1981; Dix, 2001). All FI values fall below those associ ated with habitats conducive to living coral reefs (i.e., >4) indicating the city would not meet their goal of having coral growth on an artificial reef. One interesting result from the recovered tiles was the very high prevalence of severely deformed M ilioloinella present on tile 19 A. Foraminifera l shell deformities are
64 commonly accepted as resulting from environmental stressors both natural and anthropogenic While it is unfortunate there were only two tiles recovered, the utility of recruitment ti les to sample live foraminifera is tantalizing because it allows very fragile, deformed specimens to be observed When coral rubble or limestone is scrubbed, some fragile shells will be broken because morphological deformities appear to increase shell fra gility (Souder, unpublished data). Additionally, the potentially confounding issue of mechanical damage is removed. The problem of taphonomic loss is of course further compounded in sediment samples. Polluted areas can have dramatically higher inciden ces of deformed foraminifers (e.g., Yanko et al., 1998 ). Research on how foraminifers incorporate compounds into their shell and their utility as a bioindicator is growing (e.g. Bresler and Yanko, 2000; Samir and El Din, 2001; Erez, 2003). Foraminifers have been used as bioindicators in a variety of environments (e.g. Resig, 1960; Seiglie, 1968; Yanko et al., 1994; Sen Gupta et al. 1996; Alve, 2000; Elberling et al., 2003; Hayward et al., 2004; Tsujimoto et al., 2006). For those such as Murray and Alv e (1999) who dismiss including dead forami nifers in assemblage analyses, the use of recruitment tiles for foraminiferal recruitment may provide a useful solution to the problem of quantitatively sampling hard substratum I believe adding recruitment tile s to rubble, rock or sediment samples can add a useful dimension to the total assemblage picture. Molluscan Assemblage The predominance of mollusks and worm tubes in the sediment core subsamples indicate s moderate to relatively high nutrient flux (Daniel s, 2005). Prevalence of molluscan fragments in sediment indicates calcification by het erotrophs, predation and grazing as the dominant methods of obtaining energy. Despite the sediment being largely dredge spoil material, this conclusion appears to hold true. Molluscan fragments dominate the carbonate sediment around Durney Key representing grazers that feed on the plentiful algae and carnivores that prey on the grazers The predominance of mollusks was also found to exist in Tampa Bay where Mollusca w as the dominant phylum by weight at artificial reefs, although much of this was due to the predominance
65 of Asian Green Mussels ( Perna viridis) (Dix et al., 2005). Salinity has been correlated with changes in bivalve diversity, which has decreased in low s alinity and hypersaline environments ( Garrison et al., 2007). Species common to both the Durney Key molluscan assemblages and the Tampa Bay assemblages include Abra aequalis Astyris lunata Boonea impressa Caecum pulchellum Cerithium muscarum Laevicar dium mortoni Musculus lateralis and Nucula proxima (Dix et al., 2005). These data are the first known molluscan analysis of the Durney Key area, and as such will serve as a baseline for future comparisons. Molluscan assemblages identified in the sediment core subsamples were classified by functional signifi cance through researching their feeding method. Three of the bivalve species identified were deposit feeders (DS), one was a chemosymbiotic (i.e., able to derive energy from sulfur oxidizing bacteria ho used within the organism) deposit feeder (DC) and six were suspension feeders (SU) (Table 32 a ) (Todd, 2008). Deposit feeder species were represented by 91 specimens (of which 72 were N. proxima ), chemosymbiotic deposit feeder s were represented by one spe cimen and suspension feeder species were represented by 21 specimens. Fifteen of the gastropod species identified in the sediment core subsamples were herbivorous a nd 12 were carnivorous (Table 32 b ). Herbivorous species were represented by 128 specimens (87%) and the carnivorous species were represented by 19 specimens (13%). This is a typical trophic level distribution that could supports observed number of predators according to the 10% energy conversion rule of thumb (Bolen and Robi nson, 2003). Ener gy input to the system comes primarily from alg a e, seagrasses and the few mangroves present, the latter of which shed their leaves frequently thus providing ample detritus for bacteria, fungi and other detritus consumers.
66 Table 32 a Bivalve assem blage organized by feeding mode, where DS = surface deposit feeder, DC = chemosymbiotic deposit feeder, and SU = suspension feeder. Species Mode of feeding Bivalves DS DC SU Abra aequalis X Anodontia alba X Gouldia cerina X Laevicardium mortoni X Lima scabra scabra X Musculus lateralis X Nucula proxima X Periglypta listeri X Raeta plicatella X Strigilla mirabilis X Totals 3 1 6 Table 32 b. Gastropod assemblage organized by feeding where mode CP = predatory carnivores, CB = browsing carnivores, HR = herbivores on rock, rubble or coral substrates, HP = herbivores on plant or algal substrates, and HM = herbivores on fine grained substrates. Species Mode of feeding Gastropods CP CB HR/HP HM/HR HM HP Graphis underwoodae X Astyris lunata X Boonea impressa X Caecum nitidum X C. pulchellum X Capulus ungaricus X Cerithium spp. X C. eburneum X C. lutosum X C. multicostatum X C. muscarum X Cyclostremiscus pentagonus X Cylindrobulla beauii X Granulina ovuliformes X
67 Table 32 b (Continued) G. vitrea X Haliotinella patinaria X Littorina spp. X L. mespillum X Marginella spp. X Meioceras nitidum X Oliva sayana X Ol ividae spp. X Rissoina catesbyana X Truncatella pulchella X Turbonilla aequalis X Volutidae spp. X Total 5 5 5 8 2 1 Species Assemblages on Tiles The two recovered tiles co rroborate the recruitment image analysis conclusio ns, both having Balanus eburneus present. Both tiles had more B. eburneus present on the front face of the tile, i.e. the side facing away from the PVC stake, than the backside. This is intuitive as the water flow is restricted by the stake and opposit e tile on the backside. The front would presumably be the favored side for coral growth as well due to the more abundant light and unrestricted wate r flow, but again it appears that corals transplanted or larvae were able to recruit, would be pre empted by barnacles. This conclusion is consistent with findings of Birkeland (1977), including that coral survival increased on the shaded side of artificial substrata and in cryptic locations to avoid overgrowth Unfortunately, the tiles were lost after only three months, which was insufficient time to confirm that coral larvae would not recruit. However, as noted previously, physical and chemical conditions at Durney Key are not suitable for coral recruitment and growth.
68 Recruitment While having only im ages to analyze for recruitment data was limiting, two important results were obvious. The rapid recruitment of barnacles and macroalgae was a clear indication that the nutrient flux is too high to be conducive to co ral growth. Since most shallow water c orals are slow growing mixotrophic organisms, they will be outcompeted by faster growing barnacles and algae where conditions allow these to thrive. In broader terms, the K selected corals will be outcompeted by opportunistic r selected algae an d barnacle s. This occurs by three main pathways: space pre emption shading and direct overgrowth This iss ue was addressed by Birkeland ( 1977 ) when he stated Under conditions of high nutrient input from outside resources, r selected species can be the perpetually superior competitors for space, preventing K selected species from establishing themselves and gaining a refuge in size. (p. 15). The other o bvious result, also addressed by Birkland ( 1977 ) paper, is the more rapid recruitment on the top tiles than the bottom tiles. The average total area covered for all time steps was greater for the top tiles than the bottom tiles although the top tiles tended to have greater variability in recruitment than the bottom tiles. More rapid recruitment of r selected spec ies closer to the air sea interface makes intuitive sense, as there is more incoming energy in the form of sunlight. This explains the results discussed by Birkeland about coral recruits that while they grow faster on the upper surface of artificial sub strata on coral reefs, survival is greater in the shade on vertical and under surfaces. Growth of recruits is faster in shallow waters but survival increases with depth, at least to 20 m, as the light decreases. (p.15). The higher survival rate of recru its in less ideal locations may be ascribed to reduced competition. Both of these results argue against the emplacement of an artificial reef at Durney Key for the purposes of coral recruitment and growth. The lack of any coral recruits observed suppor ts this conclusion. While it would be possible to attempt to grow coral transplants, I suspect even sizable colonies would be rapidly smothered by algae. The ease of use and low cost makes recruitment plates a useful first step to determining the suitabi lity of a site for an artificial reef.
69 Recommendations for Future Work The development of a Pasco County Water Atlas, similar to the Pinellas and Hillsborough Water Atlases (http://www.wateratlas.usf.edu/) would be beneficial. These atlases act as repos itories of data freely accessible to anyone. Any such data should also be input to the National Oceanographic Data Center, which currently lists no data for the Durney Key area. An important finding that should be disseminated to local users is the seaso nal stage variation due to thermal expansion and contraction. These data should be incorporated into future nautical charts. Continued sampling of nutrient levels, geochemical analyses and FI would form a valuable monitoring program to assess ecosystem h ealth of the popular recreation area. Investigating the speciation of the heavy metals and therefore bioavailability of the metals would yield a more thorough assessment. Toxicity tests would also be a valuable component, but were not included in this st udy due to time and financial con straints. After the FI analysis is completed, the foraminiferal shells could be analyzed for presence of heavy metals, which may help to constrain the causes of shell deformities. As for the recreational possibilities of D urney Key, it undoubtedly will continue to be a popular site for boaters, including kayakers. The Key itself could benefit from removal of exotics, as well as addition of picnic facilities and interpretive signs regarding the island origin, native vegetat ion, and things to see while snorkeling or wading. Shallow areas could be restricted to swimmers, kayaks and canoes, to allow individuals to enjoy the dynamic seagrass communities while protecting both the persons and the seagrass communities from powerbo ats and powered personal watercraft. The emplacement of artificial reef structures could be problematic because that would require further dredging to increase water depth sufficient for the structures. Dredging would disrupt seagrass and increase turbid ity, and storms would likely bury the structures. And while organisms certainly would recruit to the structures, the same community of organisms would likely recruit that presently live on hard substrates in the surrounding area, i.e., barnacles, serpulid worms, assorted bivalves and algae. Thus, I anticipate that the costs of an artificial reef at Durney Key would far outweigh the benefits.
70 Conclusions The Durney Key area does not appear suitable for coral growth due to : o high temperature variabili ty o limited water flow o low salinity, o low winter pH, o relatively high total phosphorus levels o p resence of heavy metals and persistent organic pollutants, o rapid recruitment of opportunistic ( r selected ) species, o low Foram Index values and o high inc idence of deformed foraminifers. Tides appear to be the dominant force in both water velocities and stage levels. Seasonal stage differences are significant in the shallow flats environment around Durney Key. Recruitment plates offer a useful novel technique in sampling foraminiferal assemblages, especially when shell fragility is a concern. Cooperative projects between local municipalities and universities can be mutually beneficial.
71 References Cited Abbott, R. 1974. American Seashells Second Editio n Van Nostrand Reinhold Company. New York, USA. ActLabs. 2008. Actlabs Group of Companies: Methods. from http://www.actlabs.com/methods_usa.htm Alve, E. 1995. Benthic foraminiferal responses to estuarine pollution: A review. Journal of Foraminif eral Research 25:190 203. Alve, E. 2000 Environmental stratigraphy: A case study reconstructing bottom water oxygen conditions in Frierfjord, Norway, over the past five centuries, in Martin, R. E. (ed.). Environmental Micropaleontology Kluwer Academ ic/Plenum Publishers, Boston, p. 324 350. Alve, E. and J. Nagy. 1986. Estuarine foraminiferal distribution in Sandebut kta a branch of the Oslo Fjord. Journal of Foraminiferal Research 16: 261 284 Baker, P., S. Baker, and J. Fajans. 2004. Nonindigen ous marine species in the greater Tampa Bay ecosystem. Tampa Bay Es tuary Program Tech. Pub. Barron, M. 1990. Bioconcentration will water borne organic chemicals accumulate in aquatic animals? Environmental Science and Technology 24:1612 1618.
72 Ba vel B. V., P A Naf C Bergqvist, D. Broman, K. Lundgren, O. Papakosta, C. Rolff, B. Strandberg, Y. Zebuhr, D. Zook and C. Rappe. 1996. Levels of PCBs in the aquatic environment of the golf of Bothnia: benthic species and sediments. Marine Pollution Bull etin 32:210 218 Bergin, F., F. Kucuksezgin, E. Uluturhan, I. Barut, E. Meric, N. Avsar, and A. Nazik. 2006. The response of benthic foraminifera and ostracoda to heavy metal pollution in Gul f of Izmir (Eastern Aegean Sea). Estuarine Coastal and Shelf S cience 66(3 4): 368 386. Birkeland, C. 1977. The importance of rate of biomass accumulation in early successional stages of benthic communities to the survival of coral recruits. Proceedings, Third International Coral Reef Symposium p 15 21. Birkeland, C. 1997. Life and Death of Coral Reefs. Chapman and Hall, Ney York, NY. Blatt, H., G. Middleton and R. Murray. 1972. Origin of Sedimentary Rocks. PrenticeHall, Englewood Cliffs, NJ. Bolen, E. and W. Robi nson. 2003 Wildlife Ecology and Managemen t Prentice Hall, Upper Saddle River, NJ. Bresler, V. and V. Yanko. 1995 Chemical ecology: A new approach to the study of living benthic epiphytic Foraminifera Journal of Foraminiferal Research 25: 267 279. Burone, L., N. Venturini, P. Sprechmann, P Valente, and P. Muiz. 2006. Foraminiferal responses to polluted sediments in the M ontevideo coastal zone, Uruguay. Marine Pollution Bulletin 52(1): 61 73. Buzas Stephens, P., and M.A. Buzas. 2005. Population dynamics and dissolution of fo raminifer a in Nueces Bay, Texas. Journal of Foraminiferal Research 35: 248 258.
73 Cairns, S. 1977 Stony corals. I. Caryophylliina and Dendrophylliina (Anthozoa: Scleractinia) Mem. Hourglass Cruises 3(4): 1 29. Callahan, M., J. Wheaton, C. Beaver, S. Brooke, D. Johnson, J. Kidney, S. Kupfner, J. Porter, M. Meyers, S. Wade, M. Colella, and M. Bertin. 2007. Coral Reef Evaluation and Monitoring Project: 2006 Executive Summary EPA Steering Committee Meeting July 2007. Florida Fish and Wildlife Conservation Commiss ion. Carnahan, E. 2005. Foraminiferal assemblages as bioindicators of potentially toxic elements in Biscayne Bay, Florida. Thesis: College of Marine Science, University of South Florida, Tampa, FL. Car ter, J., A. Carpenter, M. Foster and W. Jessee. 1 985. Benthic Succession on an Artificial Reef Designed to Support a Kelp Reef Community. Bulletin of marine S cience 37 (1): 27. Clark, K. and R. Warwick. 2001. Changes in Marine Communities: An Approach to Statistical Analysis and Interpretations. PR IMER E Ltd, Plymouth, UK. Cockey, E., P. Hallock and B. Lidz. 1996. Decadal changes in benthic foraminiferal assemblages off Key Largo Florida. Coral Reefs 15:237 248. COMPS. 2008. COMPS Station Port Richey. http://comps1.marine.usf.edu/pas/index.sht ml Connell, J., R. Slatyer. 1977. Mechanisms of Succession in Natural Communities and Their Role in Commun ity Stability and Organization. The American Naturalist 111(982) :1119 Daniels, C. 2005. Coral Reef Assessment: An Index Utilizing Sediment Con stituents. Thesis: College of Marine Science, University of South Florida, Tampa, FL.
74 Dannenb erger, D 1996. Chlorinated Microcontaminants in Surface Sediments of the Baltic Sea Investigations in the Belt Sea, the Arkona Sea and the Pomeranian Bight Marine Pollution Bulletin 32(11):772 781. Dauer, D. M. 1993 Biological criteria, environmental health and estuarine ma crobenthic community structure. Marine Pollution Bulletin 26(5) : 8. de Swart, R., P. Ross, J. Vos and A. Osterhaus. 1996. Impai red i mmunity in harbor seals ( Phoca vitulina ) exposed to bioaccumulated environmental contaminants: review of a long term feeding study. Environmental Health Perspectives 104:823 828 Dietz, R. 1976. Food and feeding habits of Sciaenidae in the Anclote Anchorage. Thesis: College of Marine Science, University of South Florida, Tampa, FL. Dix, T. 2001. The distribution and ecology of benthic foraminifera of Tampa Bay, Florida. Thesis: College of Marine Science, University of South Florida, Tampa, FL. Dix, T., T. A sh, D. Karlen, B. Go etting, C. Holden and S. Estes. 2005 The Epifaunal Community Structure on A rtificial Reefs in Tampa Bay, Environmental Protection Commission of Hillsborough County Technical Report. Duffus, J.H. 2002. Heavy Metals -A meaningless term? IUPAC, Pure and Applied Chemistry 74:793 807. Eisler, R. 1988. Lead hazards to fish, wildlife, and invertebrates: a synoptic review. Biological Report 14. Laurel, MD: United States Fish and Wildlife Service, U.S. Department of the I nterior.
75 Elberling, B., K. Knudsen, P. Kristenses and G. Asmund. 2003. Applying foraminiferal stratigraphy as a biomarker for heavy metal contamination and mining impa ct in a fjord in west Greenland. Marine Environmental Research 55: 235 256. Erez J. 2003 The source of ions for biomineralization in Foraminifera and their implication for paleoceanographic proxies Biomineralization 54 : 115 149. Fajans, J. L. Sturmer, S. Baker, K. Hulen, and E. Cassiano. 2007. Whats in the Clam Bag? A Guide to Marine Organisms In, On and Around a Clam Culture Bag. http://shellfish.ifas.ufl.edu/clambag/index.html. Feely, R., C. Sabine, K Lee, W. Berelson, J. Kleypas, V. Fabry and F. Millero. 2004. Impact of anthropogenic CO 2 on the CaCO 3 system in the oce ans. Science 305(5682): 362 366. Field, J., K. Clarke, and R. Warwick. 1982. A practical strategy for analysing multispecies distribution patterns. Marine Ecology Progress Series 8:37 52. Finelli, C., B. Helmuth, N. Pentcheff, and D. Wethey. 2007. Intr acolony variability in photosynthesis by corals is affected by water flow: role of oxygen flux Marine Ecology Progress Series 349:103 110. Florida Department of Environmental Protection. 2005. EcoSummary: Pithlachascotee Ri ver at Starkey Blvd, upstrea m of bridge. from http://water.dep.state.fl.us/eswizard/ecosummary.asp? id=735 Florida Department of Environmental Protection. 2008a "West Central Region: Pithlachascotee River Canoe Trail." from http://www.dep.state.fl.us/gwt/guide/regions/westcentral /trails/pithlachoscotee.htm
76 Florida Department of Environmental Protection. 2008b Osborne Reef Waste Tire Removal Pilot Project. from http://www.dep.state.fl.us/waste/categories/tires/pages/osbornepilot.htm Florida Department of Health. 2008. "Paso County: Brasher Park Beach." from http://esetappsdoh.doh.state.fl.us/irm00beachwater/reshistory.aspx? SPID=221. Florida Fish and Wildlife Conservation Commission. 2006. West Florida Birding Trail. Wildlife Foundation of Florida, Tallahassee, Fl. http: //www.floridabirdingtrail.com Fore, L., R. Frydenborg, D. Miller, T. Frick, D. Whiting, J. Espy and L. W olfe. 2007 Development and testing of biomonitoring tools for macroinvertebrates in Florida streams (strea m condition index and biorecon). Prepare d for the Florida Department of Environmental Protection. Folk, R.L. 1980. Petrology of Sedimentary Rocks Hemphill. Austin, TX. Frontalini, F. and R. Coccioni. 2008. Benthic foraminifera for heavy metal pollution monitoring: A case study from the cent ral Adriatic Sea coast of Italy. Estuarine Coastal and Shelf Science 74: 404 417. Garrison, J. Jr., B. Henk, and R. Creel. 2007. Neoichnology of the micro tidal Gulf Coast of Texas: implications for paleoenvironmental and paleoecological interpreta tions of clastic rocks of the Cretaceous Western Interior Basin, U.S.A. Extended Abstracts, SEPM Research Conference on Ichnological applications in sedimentological and sequence stratigraphic problems, Price, Utah Gordon, L. J. Jennings Jr. A. Ross and J. Krest. 1993. A suggested Protocol For Continuous Flow Automated Analysis of Seawater Nutrients. In: WOCE Operations Manual WHP office Report 90 1, WOCE report 77 No. 68/91. 1 52.
77 H allock, P. 1981. Production of carbonate sediments by selected large benthic foramini fera on two Pacific coral reefs. Journal of Sedimentary Petrology 51(2): 467 474. H allock, P. 1988. The role of nutrient availability in bioerosion: consequences to carbonate buildups. Palaeogeography, Palaeoclimatology, Palaeoec ology 63:275 291. Hallock, P. and W. Schlager. 1986. Nutrient Excess and the Demise of Coral Reefs and Carbonate Platforms. Palios 1 (4): 389 398. Hallock, P., A. Hine, G. Vargo, J. Elrod, and W. Jaap. 1988. Platforms of the Nicaraguan Rise: Examples of sensitivity of carbonate sedimentation to excess trophic resources. Geology 16:1104 1107. Hallock, P., B. Lidz, E. Cockey Burhard, and K. Donnelly. 2003. Foraminifera as bioindicators in coral reef assessment and monitoring: The FORAM Index. Environ mental Monitoring and Assessment 81:221 238. Harding G., R. LeBlanc, P. Vass, R. Addison, B. Hargrave, S. Pearre Jr, A. Dupuis, P. Brodie. 1997. Bioaccumulation of polychlorinated biphenyls (PCBs) in the marine pelagic food web, based on a seasonal stud y in the southern Gulf of St. Lawrence, 1976 1977. Marine Chemistry 56:145 179 Hayward, B. W., H.R. Grenfell, K. Nicholson, R. Parker, J. Wilmhurst, M. Horrocks, A. Swales, and A.T. Sabaa. 2004 Foraminiferal record of human impact on intertidal estuar ine environment s in New Zealands largest city. Marine Micropaleontology 53: 37 66. He, R. and R. Weisberg. 2002. Tides on the West Florida Shelf. Journal of Plysical Oceanography 32:3455 3473.
78 Hine, A., G. Brooks, R. Davis Jr., D. Duncan, S. Locker, D. Twichell and G. Gelfenbaumf 2003. The west central Florida inner shelf and coastal system: a geologic conceptual overview and introduction to the special issue Marine Geology 200:1 18. Hope, A. C. A. 1968. A simplified Monte Carlo significance tes t procedure. Journal of the Royal Statistical Society Series B 30:582 598. Hope B, S. Scatolini, E. Titus and J. Cotter. 1997. Distribution patterns of polychlorinated biphenyl congeners in water, sediment and biota from MidwayAtoll (North Pacific Ocea n). Marine Pollution Bulletin 34:548 563 Hu, H. 2002. Human health and heavy metals exposure. In: McCally, M. (ed.), Life Support: The Environment and Human Health MIT Press, Cambridge, MA, pp. 65 82. Ishman, S.E., Graham, S., D' Ambrosio, J. 1997. Mode rn benthic foraminifera distributions in Biscayne Bay: Analogs for historical reconstructions. U.S. Geological Survey Open File Report, 97 34. Janz D. M. and D. C. Metcalfe. 1991. Nonadditive interactions of mixtures of 2,3,7,8 TCDD and 3,3",4,4" tetra chlorobiphenyl on aryl hydrocarbon hydroxylase inducation on rainbow trout ( Oncorhynchus mykiss ). Chemosphere 23:467 472 Karr, J. 1981. Assessment of Biotic In tegrity Using Fish Communities. Fisheries 6(6): 7. Kennish, M.J. 1992. Ecology of Estuaries : Anthropogenic Effects CRC Press Inc, Boca Raton, FL. Klapow, L.A. and R.H. Lewis. 1979. Analysis of toxicity data for California marine water quality standards. Journal Water Pollution Control Federation 51(8):2061 2070.
79 Kohler, K.E. and S.M. Gill. 2006. Coral Point Count with Excel extensions (CPCe): A Visual Basic program for the determination of coral and substrate coverage using random point count methodology. Computers and Geosciences 32( 9 ):1259 1269. Laws E. and D. Redalje. 1979 A new met hod for estimating phytoplankton growth rates and carbon biomass Marine Biology 62(1):73 79. Loeblich, A. and H. T appan. 1987 Foraminiferal Genera and Their Classification. New York, Van Nostrand Reinhold. Long, E. and L. Morgan 1990. The potenti al for biological effects of sediment sorbed contaminents tested in the National Status and Trends Program. Technical Memorandum NOS OMA 52 Miao, X. C. Swenson, K. Yanga gihara, and Q. Li. 2000. Polychlorinated Biphenyls and Metals in Marine Species fr om French Frigate Shoals, North Pacific Ocean. Archives of Ebvironmental Contamination and Toxicology. 38:464 471. Murray, J. and A lve, E. 1999 Taphonomic experiments on marginal marine foraminiferal assemblages: how much ecological information is pr eserved? Paleogeography Paleoclimatology Paleoecology 149(1 4): 183 197. National Oceanic and Atmospheric Association. 1999. Sediment Quality Guidelines developed for the National Status and Trends Program http://response.restoration.noaa.gov/book_shelf /121_sedi_qual_guide.pdf Obrochta, S., D. Duncan and G. Brooks 2003. Hardbottom development and significance to the sediment starved west central Florida inner continental shelf. Marine Geology 200: 291 306
80 Pawlowicz, R., B. Beardsley, and S. Lentz. 2002. Classical tidal harmonic analysis including error estimates in MATLAB using T_TIDE. Computers and Geosciences 28: 929 937. Poag, C. 1981. Ecological Atlas of Benthic Foraminifera of the Gulf of Mexico Marine Science In ternational, Woods Hol e, Massachusetts. Ramirez, A. 2008. Patch Reefs in Biscayne National Park, FL: Sediments, Foraminiferal Distributions, and a Comparison of Three Biotic Indicators of Reef Health. Thesis: College of Marine Science, University of South Florida, Tampa, FL. Resig, J. M. 1960. Foraminiferal ecology around ocean outf alls off Southern California in Pearson, E.A. (ed.). Waste Disposal in the Marine Environment Pergamon Press, New Yourk, p.104 121. Rolfes, J. 1974. Pattern of diversity and density in th e ichthyofauna of the Anclote Anchorage, Florida. Thesis: College of Marine Science, University of South Florida, Tampa, FL. Safe, S. and K. Graido. 1998. Phytoestrogens and anthropogenic estrogenic compounds. Environmental Toxicology and Chemistry 17: 119 126. Samir, A. and A. El Din. 2001 Benthic foraminiferal assemblages and morphological abnormalities as pollution proxies in two Egyptian bays Marine Micropaleontology 41(3 4): 193 227. Schropp, S., F. Lewis, H. Windom, J. Ryan, F. Calder and L. Bu rney. 1990. Interpretation of metal concentrations in estuarins sediments of Florida using aluminum as a reference element. Estuaries 13(3):227:235.
81 Seaman Jr., William. 2000. Artificial Reef Evaluation With Applications to Natural Marine Habitats CRC Press, Boca Raton Sebens, K. and A. Johnson. 1991. Effects of water movement on prey capture and distribution of reef corals. Hydrobiologia 216/217:247 248. Seiglie, G. A. 1968 Foraminiferal assemblages as indicators of high organic carbon con tent in sediments and polluted waters The American Association of Petroleum Geologists Bulletin 52: 2231 2241. Sen Gupta, B. R. Turner, and N. Rabalais. 1996. Seasonal oxygen depletion on continental shelf waters of Louisiana: Historical records on be nthic foraminifera: Geology 24: 227 230. Sen Gupta, B. 1999. Modern Foraminifera Kluwer Academic Publishers, Boston. SERC FIU W ater Quality Monitoring Network. 2008. Water quality monitoring project. http://serc.fiu.edu/wqmnetwork/FKNMS CD/index.htm Siegel, F.R. 2002. Environmental Geochemistry of Potentially Toxic Metals Springer, New York. Singleton, T., K. Petrus, G. Maddox, J. Paulsen, Z. Shelley, A. Lassiter, C. Kovach, D. Williams, J. Aguilo, and D. Worley. 2006. Water Quality Status Report Springs Coast Water Quality Status Report Florida Department of Environmental Protection Skaare, J. U. 1996. Environmental pollutants in marine mammals from the Norwegian coast and Arctic. The Science of the total environment 186:25 27 Stewart, R. 2 006. Introduction to Physical Oceanography Texas A & M University.
82 Svane, I. and K. Petersen 2001 On the Problem of Epibiosis Fouling and Artificial Reefs. Marine Ecology 22(3): 19. Szedlmayer, S. 1982. Distribution and abundance of Juvenile fishes along a salinity gradient in the Anclote River estuary, Tarpon Springs, Florida. Thesis: College of Marine Science, University of South Florida, Tampa, FL. Todd, J. 2008. Introduction to molluscan life habits databases http://eusmilia.geology.ui owa.edu/database/mollusc/mollusclifestyles.htm Tsujimoto, A., R. Nomura, M. Yasuha ra, H. Yamazai and S. Yoshikawa. 2006 Impact of eutrophication on shallow marine benthic foraminifers over the last 150 years in Osaka Bay, Japan Marine Micropaleontolo gy 60: 258 268. U. S. Census Bureau. 2006. People Quick Facts Pasco County http://quickfacts.census.gov/qfd/states/12/12101.html United States Department of Environmental Protection. 2006 Envirofacts Data Warehouse. from http://oaspub.epa.gov/e nviro/multisys2.get_list_tri? tri_fac_id=34691FLRDP1729B United States Department of Environmental Protection. 2008. New Test Methods Online. from http://www.epa.gov/osw/hazard/testmethods/sw846/new_meth.htm United States Geological Survey. 2008. N ational Water Information System: Web Interface. http://nwis.waterdata.usgs.gov/nwis/qwdata? site_no=02310300&agency_cd= USGS&begin_date=&end_date=&format=html_table&inventory_output=0&rdb_invento ry_ output=file&date_format=YYYY MM DD&rdb_compression=file& qw_sample _wide=0&submitted _form=brief_list
83 Unlu, S., B. Alpar S. Aydin, C. Akbulak, N. Balkis, I. Barut, E. Meric, A. Aksu, and C. Kirbasoglu. 2006. Anthropogenic pollution in sediments from the Gulf of Gemlik (Marmara Sea, Turkey): Cause result relationship. Fresenius Environmental Bulletin 15 ( 12A ): 1521 1530. Weisberg, R., Z. Li and F. Muller Karger. 2001. West Florida shelf response to local wind forcing: April 1998. Journal of Geophysical Research. 106:31239 31262. Weisberg, S. J. Ranasi nghe, D. Daniel, L. Schaffner, R. Diaz, and J. Frithsen. 1997. An Estuarine Benthic Index of Biotic Integr ity (B IBI) for Chesapeake Bay. Estuaries 20 (1): 149 158. Weiss, W. R. 1978. The Zooplankton of the Anclote Estuary, Florida. Thesis: College of Marine Science, University of South Florida, Tampa, FL. Whitney, E., B. Means and A. Rudloe. Priceless Florida: Natural Ecosystems and Native Species. Pineapple Press, Sarasota, Fl. 2004. Wilson, J. G. 1994. The Role of Bioindi cators in Estuarine Management. Estuaries 17(1): 94 101. Yanko, V., J. Kronfeld, and A. Flexer. 1994. Response of benthic Foraminiera to various pollution sources Implic ations for pollution monitoring. Journal of Foraminiferal Research 24( 1 ): 1 17. Y anko, V M. A hmad a nd M. K aminsky 1998. Morphological deformities of benthic foraminiferal tests in response to pollution by heavy metals: implications for pollution monitoring. Journal of Foraminiferal Research 28: 177 200.
84 Yanko, V., A. Arnold, and W. Parker. 1999 E ffects of marine pollution on benthic foraminifera In Modern Foraminifera B. Sen Gupta ed Amsterdam, Kluwer Press : 217 235.
86 Appe ndix I Physical Data Figure I 1. Comparison of tided and detided summer east west water velocities from the southern most side of Durney Key (A DCP B). Figure I 2. Comparison of tided and detided summer north south water velocities from the south ern most side of Durney Key (A DCP B). W a t e r V e l o c i t y ( c m / s ) 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7 W a t e r V e l o c i t y ( c m / s ) 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7
87 Figure I 3. Correlation of detided east west summer 2007 water velocity and wind for A DCP B where the red trendline indicates a log regression and the black trendline indicates a linear regression. Figure I 4. Correlation of detided north south water summer 2007 velocity and wind for A DCP B where the red trendline indicates a log regression and the black trendline indicates a linear regression. Wind Speed (m/ s) V n Water V elocity (cm/ s) V e Water V elocity (cm/ s) Wind Speed (m/ s)
88 Figure I 5. Summer water height from the southern most side of Durney Key (A DCP B) where crescent moons represent neap tides and circular moons represent spring tides with the black circles being new moons and blue circles being full m oons. Figure I 6. Comparison of tided and detided summer east west water velocities from the north side of Durney Key (A DCP C). 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 / 2 0 8 / 2 3 Month/ Day of 2007 W a t e r V e l o c i t y ( c m / s ) 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7
89 Figure I 7. Comparison of tided and detided summer north south water velocities from the north side of Durney Key (A DCP C). Figure I 8. Correlation of detided east west water velocity and wind for A DCP C 2007 where the red trendline indicates a log regression and the black trendline indi cates a linear regression. Wind Speed (m/ s) V e Water V elocity (cm/ s) 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7 W a t e r V e l o c i t y ( c m / s )
90 Figure I 9. Correlation of detided north south water velocity and wind for A DCP C 2007 where the red trendline indicates a log regression and the black trendline indicates a linear regression. Figure I 10. Summer water height from the north side of Durney Key (A DCP C) where crescent moons represent neap tides and circular moons represent spring tides with the black circles being new moons and blue circles being full moons. 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 / 2 0 8 / 2 3 Month/ Day of 2007 Wind Speed (m/ s) V n Water V elocity (cm/ s)
91 Figure I 11. Comparison of tided and detided winter east west water velocities from the southern most side of Durney Key (A DCP B). Figure I 12. Comparison of tided and detided winter n orth south water velocities from the southern most side of Durney Key (A DCP B). W a t e r V e l o c i t y ( c m / s ) 2 / 1 4 2 / 2 1 2 / 2 9 3 / 7 3 / 1 4 3 / 2 1 3 / 2 9 4 /6 M o n t h / D a y o f 2 0 0 8 W a t e r V e l o c i t y ( c m / s ) 2 / 1 4 2 / 2 1 2 / 2 9 3 / 7 3 / 1 4 3 / 2 1 3 / 2 9 4 / 6 M o n t h / D a y o f 2 0 0 8
92 Figure I 13. Correlation of detided east west water velocity and wind for A DCP B 2008 where the red trendline indicates a log regression and the black trendline indicates a linear regression. Figure I 14. Correlation of detided north south water velocity and wind for A DCP B 2008 where the red trendline indicates a log regression and the black trendline indicates a l inear regression. V e W a t e r V e l o c i t y ( c m / s ) W i n d S p e e d ( m / s ) V n W a t e r V e l o c i t y ( c m / s ) W i n d S p e e d ( m / s )
93 Figure I 15. Winter water height from the southern most side of Durney Key (A DCP B) where crescent moons represent neap tides and circular moons represent spring tides with the black circles being new moons and blue circles being full moons. Figure I 16. Comparison of tided and detided winter east west water velocities from the southern most side of Durney Key (A DCP C). 2 / 1 4 2 / 2 1 2 / 2 9 3 / 7 3 / 1 4 3 / 2 1 3 / 2 9 4 / 6 Month/ Day of 2008 W a t e r V e l o c i t y ( c m / s ) 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7
94 Figure I 17. Comparison o f tided and detided winter north south water velocities from the southern most side of Durney Key (A DCP C). Figure I 18. Correlation of detided east west water velocity and wind for A DCP C 2008 where the red trendline indic ates a log regression and the black trendline indicates a linear regression. V e W a t e r V e l o c i t y ( c m / s ) W i n d S p e e d ( m / s ) 7 / 1 6 7 / 2 2 7 / 2 9 8 / 5 8 / 1 2 8 /2 3 M o n t h / D a y o f 2 0 0 7 W a t e r V e l o c i t y ( c m / s )
95 Figure I 19. Correlation of detided north south water velocity and wind for A DCP C 2008 where the red trendline indicates a log regression and the black trendline indicates a linear regression. Figure I 20. Winter water height from the north side of Durney Key (A DCP C) where crescent moons represent neap tides and circular moons represent spring tides with the black c ircles being new moons and blue circles being full moons. 2 / 1 4 2 / 2 1 2 / 2 9 3 / 7 3 / 1 4 3 / 2 1 3 / 2 9 4 / 6 Month/ Day of 2008 W i n d S p e e d ( m / s ) V n W a t e r V e l o c i t y ( c m / s )
96 Appendix II Chemical Data Table II 1. Inorganic nutrient concentrations at patch reefs in the Florida Keys as sampled by SERC in winter 2006 through 2007. Date Site NO 3 2 (M) NO 2 (M) NH 4 + (M ) Si(OH) 4 (M) TP (M) 11/20/2006 Coffins Patch Channel 0.027 0.072 0.968 0.978 0.262 12/14/2006 Boca Chica Mid 0.136 0.047 1.05 2.33 0.384 12/19/2006 Molasses Reef Channel 0.044 0.087 0.806 0.700 0.449 12/19/2006 Grecian Rocks 0.138 0.073 2.94 0.114 0 .435 12/20/2006 Turtle Harbor 0.235 0.107 1.59 0.234 0.307 1/24/2007 Coffins Patch Channel 0.668 0.057 0.651 0.061 0.157 1/31/2007 Molasses Reef Channel 0.119 0.026 3.17 0.033 0.312 2/21/2007 Boca Chica Mid 0.195 0.042 0.572 2.31 0.327 3/1/2007 Turtle Harbor 0.241 0.022 0.190 2.07 0.260 3/1/2007 Grecian Rocks 0.233 0.056 0.835 0.153 0.292 4/27/2007 Coffins Patch Channel 0.112 0.014 2.905 0.096 0.343 5/3/2007 Turtle Harbor 0.097 0.052 3.79 0.007 0.307 5/4/2007 Grecian Rocks 0.241 0.043 0.185 0.042 0 .247 6/7/2007 Molasses Reef Channel 0.257 0.066 2.76 0.068 0.294 6/14/2007 Boca Chica Mid 0.187 0.020 4.057 0.021 0.232 8/23/2007 Boca Chica Mid 0.255 0.069 0.572 1.08 0.109 9/6/2007 Coffins Patch Channel 0.115 0.013 0.341 10.28 0.236 9/25/2007 Grecia n Rocks 0.372 0.048 0.169 0.105 0.170 9/25/2007 Turtle Harbor 0.301 0.033 0.213 0.313 0.141 9/26/2007 Molasses Reef Channel 0.405 0.112 0.559 0.118 0.150 10/24/2007 Boca Chica Mid 0.158 0.021 0.463 1.200 0.184 11/26/2007 Molasses Reef Channel 0.000 0.0 51 0.121 0.270 0.192 11/27/2007 Grecian Rocks 0.120 0.024 0.177 0.434 0.214 11/27/2007 Turtle Harbor 0.051 0.018 0.468 0.569 0.210 11/29/2007 Coffins Patch Channel 0.035 0.017 0.146 0.406 0.174 Mean 0.327 0.048 0.624 0.546 0.226 Std. Dev. 0.269 0.030 0.567 0.791 0.070
97 Table II 2. Inorganic nutrient concentrations for summer 2007 at Durney Key. Date NO 3 2 (M) NO 2 (M) NH 4 + (M) Si(OH) 4 (M) TP (M) 5/15/07 0.01 0.01 1.46 9.07 0.10 5/15/07 0.07 0.04 1.97 8.25 0.09 5/15/07 0.18 0.05 1.73 7.40 0.07 5/15/07 0.01 0.01 1.75 10.5 0.04 8/22/07 0 0.01 0.24 10.0 0.24 8/22/07 0 0.02 0.10 10.10 0.1 8/22/07 0 0 .01 0.11 9.36 0.11 8/22/07 0 0.02 0.10 10.10 0.095 8/22/07 0.01 0.01 0.11 9.36 0.113 8/22/07 0 0.01 0.24 10.04 0.235 Mean 0.03 0.02 0.78 9.42 0.12 Std Dev 0.058 0.013 0.82 0.96 0.066 Table II 3. Inorganic nutrient concentrations for fall and winter 2007 at Durney Key. Date NO 3 2 (M) NO 2 (M) NH 4 + (M) Si(OH) 4 (M) TP (M) 10/12/07 7.70 0.15 2.01 0.50 41.79 10/12/07 0.32 0.08 1.36 0.80 13.68 10/12/07 0.08 0.03 2.15 1.88 7.79 10/12/07 0.07 0.06 2.66 0.53 11.03 11/8/07 0.00 0.00 0.11 0.00 2.53 11/8/07 0.00 0.00 0.51 0.05 2.94 11/8/07 0.20 0.01 1.13 0.12 2.94 11/8/07 0.17 0.01 0.75 0.26 3.21 11/23/07 0.00 0.00 2.13 0.68 4.64 11/23/07 0.00 0.01 1.13 0.28 4.88 11/23/07 0.00 0.01 0.98 0.17 4.98 11/23/07 0.00 0.01 0.46 0.00 4.18 1/18/08 0.18 0.03 3.77 1.65 0.06 1/18/08 0.05 0 .02 3.24 1.63 0.11 1/18/08 0.04 0.02 3.16 1.90 0.06 Mean 0.59 0.03 1.70 0.70 6.99 Std Dev 1.97 0.042 1.13 0.71 0.06
98 Table II 4. Results of metals in surficial sediment sample as analyzed by Actlabs where ND denotes a concentration below th e detection limit. Analyte Symbol Unit Symbol DURNEY KEY (seived) DURNEY KEY (pulverized) MDL Analysis Method Ag ppm ND ND 0.05 MULT INAA/TD ICP/TD MS Al % 1.72 0.6 0.01 TD ICP As ppm 6.8 2.1 0.5 INAA Au ppb 7 ND 2 INAA Ba ppm 118 52 1 MULT INAA/TD IC P MS Be ppm 0.5 0.1 0.1 MULT INAA/TD ICP MS Bi ppm ND ND 0.1 MULT INAA/TD ICP MS Br ppm 306 63.4 0.5 INAA Ca % 18 10.3 0.01 TD ICP Cd ppm 0.3 ND 0.1 MULT INAA/TD ICP MS Ce ppm 36.9 9.3 0.1 TD ICP Ce ppm 47 13 3 INAA Co ppm 2.8 1.1 0.1 MULT INAA/TD ICP MS Cr ppm 43 23 1 MULT INAA/TD ICP MS Cs ppm 1.17 0.3 0.05 MULT INAA/TD ICP MS Cu ppm 16.7 5.4 0.2 MULT INAA/TD ICP MS Dy ppm 2 0.6 0.1 TD ICP Er ppm 1.1 0.3 0.1 TD ICP Eu ppm 0.61 0.2 0.05 TD ICP Eu ppm 0.6 0.3 0.2 INAA Fe % 1.08 0.32 0.01 INA A Ga ppm 4.2 1.6 0.1 TD ICP Gd ppm 3.5 1 0.1 TD ICP Ge ppm ND ND 0.1 TD ICP
99 Hf ppm 0.8 0.1 0.1 TD ICP Hf ppm 13 5 1 INAA Hg ppb 52 14 5 Hg FIMS Hg ppm ND ND 1 INAA Ho ppm 0.4 0.1 0.1 TD ICP In ppm ND ND 0.1 TD ICP Ir ppb ND ND 5 INAA K % 0.49 0. 17 0.01 TD ICP La ppm 23.9 6.4 0.1 TD ICP La ppm 20.4 5.2 0.5 INAA Li ppm 9.8 2.7 0.5 TD ICP Lu ppm 0.1 ND 0.1 TD ICP Lu ppm 0.24 0.08 0.05 INAA Mass g 1.06 1.75 INAA Mg % 1.04 0.27 0.01 TD ICP Mn ppm 134 67 1 TD ICP Mo ppm 6 4 1 TD ICP Na % 3.6 2 0.81 0.01 INAA Nb ppm 4.1 1.4 0.1 TD ICP Nd ppm 22.6 6.4 0.1 TD ICP Nd ppm 18 6 5 INAA Ni ppm 21.5 5 0.5 MULT INAA/TD ICP MS P % 0.054 0.015 0.001 TD ICP Pb ppm 16.6 3.5 0.5 MULT INAA/TD ICP MS Pr ppm 5.7 1.6 0.1 TD ICP Rb ppm 24.6 7.4 0.2 MULT I NAA/TD ICP MS Re ppm 0.008 0.001 0.001 TD ICP S % 1.31 0.29 0.01 TD ICP Sb ppm ND ND 0.1 INAA Sc ppm 2.8 1.2 0.1 INAA Se ppm 1.9 1.1 0.1 MULT INAA/TD ICP MS Sm ppm 3.6 1.1 0.1 TD ICP Sm ppm 4.1 1.2 0.1 INAA Sn ppm ND ND 1 TD ICP Sr ppm > 1000 494 0.2 TD ICP Ta ppm ND ND 0.1 MULT INAA/TD ICP
100 MS Tb ppm 0.4 0.1 0.1 TD ICP Tb ppm ND ND 0.5 INAA Te ppm 3.4 2 0.1 TD ICP Th ppm 8.8 1.8 0.1 MULT INAA/TD ICP MS Ti % 0.13 0.07 0.01 TD ICP Tl ppm 0.26 0.11 0.05 TD ICP Tm ppm 0.1 ND 0.1 TD ICP U ppm 2 .8 0.9 0.1 MULT INAA/TD ICP MS V ppm 46 18 2 TD ICP W ppm ND ND 1 INAA Y ppm 11.3 3.2 0.1 TD ICP Yb ppm 0.9 0.3 0.1 TD ICP Yb ppm 1.4 0.4 0.2 INAA Zn ppm 24.5 6.7 0.5 MULT INAA/TD ICP MS Zr ppm 28 6 1 TD ICP
101 Appendix I II Sediment T exture Table III 1. Grain size data (weight % ) for sediment cores for near to offshore gradation, where sediment core 1 is 10 m from shore, core 2 is 20 m from shore and core 3 is 30 m from shore. Median grain size indicated in bold. Sample Sample wt ( g) >2 mm >1 mm >0.5 mm >0.25 mm >0.125 mm >0.063 mm <0.063 mm Sediment Core 1 bottom 7.18 4.9 21.0 13.4 14.8 33.1 12.4 0.42 Sediment Core 1 top 9.63 6.6 16.4 8.7 9.8 38.1 19.9 0.48 Sediment Core 2 bottom 6.12 9.2 21.9 7.6 7.0 30.7 23.6 0.03 Sediment Co re 2 top 3.66 2.2 28.6 3.9 8.8 31.0 25.3 0.22 Sediment Core 3 bottom 10.0 1.3 2.2 3.9 4.5 48.5 39.1 0.58 Sediment Core 3 top 7.85 2.2 2.3 7.3 6.8 45.2 35.7 0.50
102 Appendix IV Biological Data Figure IV 1 MDS plot displaying the deg ree of similarity between tile positions at time one. Figure IV 2 Cluster diagram showing similarity of ceramic tiles by position (i.e. top vs. bottom) at time one.
103 Figure IV 3 Cluster diagram showing similarity of ceramic tiles by position (i.e. top vs. bottom) at time two. Figure IV 4 MDS plot displaying the degree of similarity between tile positions at time two.
104 Figure IV 5 MDS plot displaying the degree of similarity between tile positions at time three. Figure IV 6 Cluster diagra m showing similarity of ceramic tiles by position (i.e. top vs. bottom) at time three.
105 Table IV 1 A NOSIM results for 999 permutations of recruitment class similarity by position at each time step and all time steps combined. Date Global Statistic R Value Significance level # of permuted statistics >/= R value 7/16/07 0.172 0.1% 0 8/22/07 0.165 0.1 % 0 9/11/07 0.102 0.4 % 3 All time steps 0.079 0.1% 0 TableIV 2 SIMPER results of dissimilarity among recruitment classes for bottom tiles across all time steps Average similarity: 74.07 Recruitment Class Av.Abund Av.Sim Sim/SD Contrib% Cum.% Brown Algal Film 8.84 45.70 3.35 61.70 61.70 Total Area Covered 4.31 15.31 2.15 20.67 82.37 Coralline algae 2.84 7.94 1.05 10.72 93.09 Table IV 3 SIMPER results of dissimilarity among recruitment classes for top tiles across all time steps Average similarity: 67.89 Recruitment Class Av.Abund Av.Sim Sim/SD Contrib% Cum.% Brown Algal Film 8.48 39.93 2.58 58.82 5 8.82 Total Area Covered 4.83 14.86 1.80 21.89 80.71 Turf 2.54 5.39 0.87 7.95 88.66 Coralline algae 2.04 3.83 0.71 5.64 94.30 Table IV 4 SIMPER results of dissimilarity among recruitment classes for bottom and top ti les across all timesteps. Average dissimilarity = 30.63 Recruitment Class Av. Abund Av. Abund Av. Diss Diss/SD Contrib % Cum. % Total Area Covered 4.31 4.83 7.43 1.41 24.27 24.27 Coralline algae 2.84 2.04 6.46 1.33 21.08 45.35 Turf 2.10 2.54 6.10 1.27 19.91 65.26 Barnacle 0.35 1.74 4.30 0.94 14.06 79.32 Brown Algal Film 8.84 3.78 1.28 12.34 12.34 91.65