USF Libraries
USF Digital Collections

Effects of a shallow-water hydrothermal vent gradient on benthic calcifiers, tutum bay, ambitle island, papua new guinea

MISSING IMAGE

Material Information

Title:
Effects of a shallow-water hydrothermal vent gradient on benthic calcifiers, tutum bay, ambitle island, papua new guinea
Physical Description:
Book
Language:
English
Creator:
Engel, Brienne
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Ocean acidification
Taphonomy
Dissolution
Foraminifera
Halimeda
Dissertations, Academic -- Cell, Micro, and Molecular Bio -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Ocean acidification is occurring in response to rapidly increasing concentrations of atmospheric CO2. Shallow-water hydrothermal vent systems have been proposed as natural laboratories for studying the effects of elevated pCO2 on benthic communities. Hydrothermal vents occur at depths of approximately 10m in Tutum Bay, Ambitle Island, Papua New Guinea; these vents are surrounded by a typical-appearing fringing coral-reef community. Groups of live specimens of seven species of reef-dwelling, larger benthic foraminifers, along with segments of calcareous green algae broken from live thalli, were collected from a reef location, placed in small mesh bags, and deployed for five days at six different sites along a gradient of temperature (29.6oC-59.3oC) and pH (5.9-8.1) with distance from a large hydrothermal vent in Tutum Bay. Foraminiferal taxa used in the experiment included Amphisorus hemprichii, a species with Mg-calcite porcelaneous shells, three species of Amphistegina that produce hyaline calcite shells, and three species with hyaline Mg-calcite shells (Heterostegina depressa and two Calcarina spp.). Several specimens of four of the seven foraminiferal species examined survived exposure to elevated temperatures of 59.3oC and low pH of 6.2 for five days, while at least one specimen of each of the seven species survived exposure to 39.9oC and pH 5.9. Examination of shells at 600-1000x magnification using scanning electron microscopy revealed fine-scale dissolution in specimens up to 30m from the vent. Results of this experiment, as well as previously reported observations from the study site, indicate that the calcifying reef-dwelling organisms examined can survive pH extremes that result in dissolution of their shells following death.
Thesis:
Thesis (MS)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Brienne Engel.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0004683
usfldc handle - e14.4683
System ID:
SFS0027995:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Effects of a Shallow W ater Hydrothermal Vent Gradient on Benthic Calcifiers, Tutum Bay, Ambitle Island, Papua New Guinea by Brienne E. Engel A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida M ajor Professor: Pamela Hallock Muller, Ph.D. Benjamin P. Flower, Ph.D. Edward S. VanVleet, Ph.D. Date of Approval: August 12, 2010 Keywords: ocean acidification, taphonomy, dissolution, Fora minifera, Halimeda Copyright 2010, Brienne E. Engel

PAGE 2

ACKNOWLEDGMENTS The unusual circumstances surrounding the initiation and completion of this thesis have left me indebted to a number of people whose schedules for Summer 2010 were modified in orde r to accommodate my needs. First, a thank you to my committee members, Drs. Ben Flower and Ted VanVleet for their input toward focusing the direction of this project and suggestions for data analyses, and to my major professor Dr. Pam Hallock Muller for c onceiving of this project and for continued help throughout the process. Thank you to Tony Greco for accommodating my crazy SEM schedule, for expert advice on the use of the SEM and EDAx, and for continually fixing the machines when they broke down as a r esult of my overuse. Thank you to Dr. Willem Renema and Dr. Johann Hohenegger for the Calcarina sp p. id entification. Thank you to the original Papua New Guinea collection team including Dr. Bryan McCloskey, Dr. Roy Price, Dr. Thomas Pichler and Dr. Pam H allock Muller. Financial support for the collection trip came from the National Science Foundation (grant BE: CBC# 0221834) awarded to Dr. Thomas Pichler. Personal financial support came from the USF Presidential Fellowship and the Anne and Werner Von Ro senstiel Endowed Fellowship. Finally, I'd like to thank all of my family and friends who have helped me deal with the stress of putting a thesis together in three months and for their forgiveness that I was not able to spend more time with them this summe r!

PAGE 3

i TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ............. iii LIST OF FIGURES ................................ ................................ ................................ ............. v LIST OF PLATES ................................ ................................ ................................ ............ vii ABSTRACT ................................ ................................ ................................ ..................... viii INTRODUCTION ................................ ................................ ................................ ............... 1 Tutum Bay Hydrothermal Vent System ................................ ................................ .. 2 Carbonate Minerals and Calcifying Organisms ................................ ....................... 4 Foraminifera ................................ ................................ ................................ ............. 6 Halimeda ................................ ................................ ................................ .................. 8 Taphonomy ................................ ................................ ................................ .............. 8 Energy Dis persive Spectroscopy ................................ ................................ ............. 9 Research Objectives ................................ ................................ ............................... 10 Hypotheses to be Tested ................................ ................................ ........................ 10 METHODS ................................ ................................ ................................ ........................ 11 Experimental Materials and Setup ................................ ................................ ......... 11 Imaging Techniques ................................ ................................ ............................... 12 Data Analyses ................................ ................................ ................................ ........ 14 RESULTS ................................ ................................ ................................ .......................... 16 Amphisorus hemprichii ................................ ................................ .......................... 1 6 Amphistegina spp ................................ ................................ ................................ .. 21 Calcarina spp ................................ ................................ ................................ ........ 34 Heteroste gina depressa ................................ ................................ .......................... 4 3 Halimeda tuna ................................ ................................ ................................ ........ 4 8 DISCUSSION ................................ ................................ ................................ .................... 53 CONCLUSIONS ................................ ................................ ................................ ................ 63 REFERENCES ................................ ................................ ................................ .................. 64 APPENDICES ................................ ................................ ................................ ................... 68

PAGE 4

ii ABOUT THE AUTHOR ................................ ................................ ................... END PAGE

PAGE 5

ii i LIST OF TABLES TABLE 1: Foraminiferal species, including their higher taxonomy, wall structure group and calcium carbonate mineral used to build their shells ................................ ................................ ................................ ........... 7 TABLE 2: Foraminifer al counts per site with species totals and site totals ............... 16 TABLE 3 : Descriptions of Amphisorus hemprichii images from each type of microscopy for each site (images described can be viewed in Plat e 1). ................................ ................................ ................................ ............... 19 TABLE 4 : Descriptions of Amphistegina lessonii images from each type of microscopy for each site (images described can be viewed in Plate 2).. ................................ ................................ ................................ .............. 24 TABLE 5 : Descriptions of Amphistegina lobifera images from each type of microscopy for each site (i mages described can be viewed in Plate 3). ................................ ................................ ................................ ............... 28 TABLE 6 : Descriptions of Amphistegina radiat a images from each type of microscopy for each site (images described can be viewed in Plate 4). ................................ ................................ ................................ ............... 32 TABLE 7 : Descriptions of Calcarina defrancii images from each ty pe of microscopy for each site (images described can be viewed in Plate 5). ................................ ................................ ................................ ............... 37 TABLE 8 : Descriptions of Calcarina gaudichaudii images from each type of microscopy for each site (images described can be viewed in Plate 6). ................................ ................................ ................................ ............... 41 TABLE 9 : Descriptions of Het erostegina depressa images from each type of microscopy for each site (images described can be viewed in Plate 7). ................................ ................................ ................................ ............... 46 TABLE 10 : Descriptions of Halimeda tuna images from each type of microscopy for each site (images described can be viewed in Plate 8). ................................ ................................ ................................ ............... 5 0

PAGE 6

iv TABLE 11 : Previously published Mg/Ca ratios for the species examined in this study and related species ................................ ................................ ............... 61 TABLE A 1 : Foraminiferal counts, Mg/Ca averages, and Mg/Ca standard deviations for foraminifers separated into three groups: dead, dying, or alive ................................ ................................ ............................... 69 TABLE A2: EDS data for all A. hemprichii specimens ................................ ..................... 70 TABLE A3: EDS data for all A. lessonii specimens ................................ .......................... 70 TABLE A4: EDS data for all A. lobifera specimens ................................ ......................... 71 TABLE A5: EDS data for all A. radiata specimens ................................ .......................... 71 TABLE A6: EDS data for all C. defra ncii specimens ................................ ........................ 71 TABLE A7: EDS data for all C. gaudichaudii specimens ................................ ................. 72 TABLE A8: EDS data for all H. depressa specimens ................................ ........................ 72 TABLE A9: EDS data for all H. tuna specimens ................................ ............................... 72

PAGE 7

v LIST OF FIGURES FIGURE 1: Location of Ambitle Island and the shallow wa ter hydrothermal vent studied (modified from Price and Pichler, 2005). ................................ 3 FIGURE 2: Physical, chemical, and biological trends along a transect leading away from active venting in Tutum Bay. ................................ ..................... 4 FIGURE 3: a) Location of Vent 4 (V 4) in relation to Ambitle Is land and b) the location of the transect line. .. ................................ ................................ ..... 12 FIGURE 4 : Mg/Ca r atio of A mphisorus hemprichii vs. distance from the v ent ........... 20 FIGURE 5 : Mg/Ca r atio of A mphisorus hemprichii vs. pH ................................ .......... 20 FIGURE 6 : Mg/Ca r atio of A mphisorus hemprichii vs. t emperature ............................ 21 FIGURE 7 : Mg/Ca ratio of A mphistegina lessonii vs. distance from the vent ............. 25 FIGURE 8 : Mg/Ca ratio of A mphistegina lessonii vs. pH ................................ ............ 25 FIGURE 9 : Mg/Ca ratio of A mphistegina lessonii vs. temperature .............................. 26 FIGURE 10 : Mg/Ca ratio of A mphistegina lobifera vs. d istance from the vent ............. 29 FIGURE 1 1 : Mg/Ca ratio of A mphistegina lobifera vs. pH ................................ ............ 29 FIGURE 1 2 : Mg/Ca ratio of A mphistegina lobifera vs. temperature ............................. 30 FIGURE 1 3 : Mg/Ca ratio of A mphistegina radiata vs. distance from the vent .............. 33 FIGURE 1 4 : Mg/Ca ratio of A mphistegina radiata vs. pH ................................ ............. 33 FIGURE 1 5 : Mg/Ca ratio of A mphistegina radiata vs. temperature .............................. 34 FIGURE 1 6 : Mg/Ca ratio of Calcarina defrancii vs. distance from the vent ................. 38 FIGURE 1 7 : Mg/Ca ratio of Calcarina defrancii vs. pH ................................ ................ 38 FIGURE 1 8 : Mg/Ca ratio of Calcarina defrancii vs. temperature ................................ 39

PAGE 8

vi FIGURE 1 9 : Mg/Ca ratio of Calcarina gaudichaudii vs. distance from the vent .......... 42 FIGURE 20 : Mg/Ca ratio of Calcarina gaudichaudii vs. pH ................................ ......... 42 FIGURE 2 1 : Mg/Ca ratio of Calcarina gaudichaudii vs. temperature ........................... 43 FIG URE 2 2 : Mg/Ca ratio of Heterostegina depressa vs. distance from the vent ........... 47 FIGURE 2 3 : Mg/Ca ratio of Heterostegina depressa vs. pH ................................ ......... 47 FIGURE 2 4 : Mg/Ca ratio of Heterostegina depressa vs. temperature ........................... 48 FIGURE 2 5 : Mg/Ca ratio of Halimeda tuna vs. distan ce from the vent ......................... 51 FIGURE 2 6 : Mg/Ca ratio of Halimeda tuna vs. pH ................................ ....................... 51 FIGURE 2 7 : Mg/Ca ratio of Halimeda tuna vs. temperature ................................ ......... 52 FIGURE 2 8 : Foraminiferal Mg/Ca ratio vs. state of foraminifer at sites 7.5 and 15m from the vent ................................ ................................ ...................... 58 FIGURE 2 9 : Species Mg/C a ratios vs. distance from the vent ................................ ....... 59

PAGE 9

vii LIST OF PLATE S PLATE 1: Amphisorus hemprichii Ehrenberg, 1839 ................................ .................. 18 PLATE 2 : Amphistegina lessonii d ................................ ..................... 23 PLATE 3 : Amphistegina lobifera Larsen, 1976 ................................ .......................... 27 PLATE 4 : Amphistegina radiata (Fichtel and Moll, 17 98) ................................ ........ 31 PLATE 5 : Calcarina defrancii ................................ ........................ 36 PLATE 6 : Calcarina gaudichaudii ................................ ................... 40 PLATE 7 : Heterostegina depressa ................................ .................. 45 PLATE 8 : Halimeda tuna (Ellis and Solander, 1786) Lamouroux, 1816 ................... 49

PAGE 10

viii Effects of a Shallow Water Hydrothermal Vent Gradient on Benthic Calcifiers, Tutum Bay, Ambitle Island, Papua New Guinea Brienne E. Engel ABSTRACT Ocean acidification is occurring in response to rapidly increasing concentrations of atmospheric CO 2 Shallow water hydrothermal ven t systems have been proposed as natural laboratories for studying the effects of elevated p CO 2 on benthic communities. Hydrothermal vents occur at depths of approximately 10m in Tutum Bay, Ambitle Island, Papua New Guinea; these vents are surrounded by a t ypical appearing fringing coral reef community. Groups of live specimens of seven species of reef dwelling, larger benthic foraminifers, along with segments of calcareous green alga e broken from live thalli, were collected from a reef location, placed in small mesh bags and deployed for five days at six different sites along a gradient of temperature ( 29.6 o C 59.3 o C ) and pH ( 5.9 8.1 ) with distance from a large hydrothermal vent in Tutum Bay. Foraminiferal taxa used in the experiment included Amphisorus he mprichii a species with Mg calcite porcelaneous shells, three species of Amphistegina that produce hyaline calcite shells, and three species with hyaline Mg calcite shells ( Heterostegina depressa and two Calcarina spp.). Several specimens of four of the seven foraminiferal species examined survived exposure to elevated temperatures of 59.3 o C and low pH of 6.2 for five days, while at least one specimen of each of the seven species survived exposure to 39.9 o C and pH 5.9

PAGE 11

ix Examination of shells at 600 1000 x magnification using scanning electron microscopy revealed fine scale dissolution in specimens up to 30m from the vent. Results of this experiment, as well as previously reported observations from the study site, indicate that the calcifying reef dwelling organisms examined can survive pH extremes that result in dissolution of their shells following death.

PAGE 12

1 INTRODUCTION Human activities have increased atmospheric carbon dioxide levels from 280ppmv (pre industrial) to 390ppmv (current); this trend is ex pected to double pre industrial le vels by the end of the century. The CO 2 that the ocean has absorbed since the industrial revolution directly corresponds to a 0.1 unit decrease in pH of the surface ocean in the coldest oceans and approximately 0.09 units in the warmest oc eans (Haugan and Drange, 1996). It is predicted there will be a drop of up to 0.5 pH units by the end of the 21st century using the projections of the IPCC Special Report on Emissions Scenarios (Caldeira and Wickett, 2005). Over the pas t decade scientists have become increasingly concerned about the effects of declining pH in surface waters on aquatic biotas, and especially on organisms that produce calcareous shells or skeletons (e.g., Kleypas et al., 2006) Hall Spencer et al. (2008) proposed that shallow water hydrothermal vent systems can be natural laboratories for studying the effects of elevated p CO 2 on communitie s in the vicinity of the vents. In 2003 and 2005, a research team from the University of South Florida and elsewhere v isited a hydrothermal system located in less than 10 m water depth just offshore Ambitle Island Papua New Guinea (Pichler and Dix, 1996; Pichler et al., 2006) Studies primarily focused on geochemistry (Price, 2008), microbiology, benthic invertebrates (K arlen et al., 2010 ), and benthic foraminiferal assemblages (McCloskey, 2009) However, an in situ experiment was conducted to test the responses of a suite of

PAGE 13

2 common benthic foraminifers, as well as segments of a common calcareous green alga, to the geoch emical conditions along the gradient away from one of the vents The foraminiferal and algal specimens from tha t experiment are the focus of this thesis. Tutum Bay Hydrothermal Vent System Hydrothermal vent systems occur over an extreme range of depths f rom intertidal to the abyss (Tarasov et al., 2005). Shallow water hydrothermal vent systems are defined as those occurring at depths less than 200m with a community composed of fewer "vent obligate" taxa and fewer symbiotrophic forms of life than their de eper counterparts (Tarasov et al., 2005). Deep sea vent communities survive by the process of chemosynthetic production while the presence of both light and geothermal fluids at shallow water vents allows the use of both photosynthetic and chemosynthetic primary production (Zeppilli and Danovaro, 2009). Hydrothermal vent systems alter local environments, providing a view into how different chemical, geological, biological and physical parameters respond to extreme conditions. In particular, shallow water systems are an easily accessible environment to study. Studies can be conducted on an ecosystem level to determine the effects of extreme conditions (e.g., low pH, high p CO 2 increased temperature, elevated levels of reduced compounds, or increased heavy metal concentration). The study site for this experiment was Tutum Bay, Ambitle Island, Papua New Guinea (Figure 1). It is a nearshore coral reef environment in the Tabar Feni island arc east of Papua New Guinea. The islands of this chain are composed of Pliocene to Holocene alkaline volcanoes (Pichler and Dix, 1996). Submarine, hydrothermal venting occurs in waters 5 10m deep. Two types of venting have been observed. (1) Focused

PAGE 14

3 discha rge of a clear, two phase fluid occurs at discrete ports 10 15cm in diameter with an estimated flow rate up to 300 400 L/min. Temperatures at these ports are between 94 o C and 98 o C and have associated hydrothermal precipitate accumulation of euhedral aragonite crystals, microcrystalline crusts of Fe oxyhydroxide, aragon ite, and ferroan calcite. (2) Dispersed or diffuse discharge of streaming gas bubbl es occurs directly through sandy or pebbly unconsolidated substrate, which undergoes shifts in location of tens of centimeters. These areas had no associated hydrothermal precipitates (Pichler and Dix, 1996). The gas from both types of discharge is 94 98% CO 2 making the waters slightly acidic (pH ~6) (Figure 2) The water is predominantly of meteoric origin (Pichler et al., 2006). FIGURE 1: Location of Ambitle Island an d the shallow water hydrothermal vent studied (modified from Price and Pichler, 2005).

PAGE 15

4 FIGURE 2: Physical, chemical, and biological trends along a transect leading away from active venting in Tutum Bay. From left to right, the y axes are number of for aminifer shells per gram of sediment, concentrations of As in mg/kg (sediment) and mg/L (pore water), pH, and temperature in o C. The values for As in pore waters were multipled by a factor of 104; the value at a distance of one meter is 0.9mg/L. The macr ofauna pie at a distance of 300m represents a sample taken at the reference site. In the legend on the right side, UC and UE indicate 'uncultured Crenarchaeota' and 'uncultured Euryarc haeota,' respectively. (F rom Pichler et al. 2006) Carbonate Mi nerals and Calcifying Organisms Ocean acidification has direct biological implications, especially for calcifying organisms. Such organisms build their shells mainly from one of two polymorphs of calcium carbonate (CaCO 3 ), specifically aragonite (e.g. calcare ou s algae, pteropods, corals) or calcite (e.g. coccolithophores most foraminifers ). These two minerals differ in solubility with aragonite ~ 50% more soluble than calcite. The 0.1 unit drop in pH of the world oceans corresponds to a reduction in the con centration of the carbonate ion (CO 3 2 ) of ~ 50% in surface seawater leading to arag onite and high magnesium (high Mg)

PAGE 16

5 calcite undersaturation at high latitudes (Ries et al. 2009) Gangst et al. (2008) predict a decrease in global aragonite production by 29%, calcite production by 13%, and total CaCO 3 production by 19% by the end of the 21st century compared to pre industrial values. If the entirety of the estimated glob al fossil fuel resources (~5000 Pg C) are released to the atmosphere in the upcoming c enturies the ocean will become undersaturated with respect to aragonite over nearly the entire surface. If, however, organic carbon rich shales and/or methane hydrates are found to be minable, emi ssions could reach up to 10,000 Pg C of CO 2 to the atmosphe re causing undersaturation with respect to calcite in the entire surface ocean (Caldeira and Wickett, 2005). Calcifying organisms build their shells through the following reaction: Ca 2+ + 2HCO 3 3 + CO 2 + H 2 O High Mg calcite results from lattice incorporation of magnesium into biogenic (or abiotic) calcite and results in a more soluble calcium carbonate phase. Partial dissolution of the shell can alter the chemical composition through pref erential leaching of Mg ( Savin and Douglas, 1973 ). to with respect to a specific mineral, the seawater is supersaturated with respect to that thermodynamic equilibrium with that mineral. However, supersaturation with respect to a calcium carbonate mineral d oes not always mean that there will be net precipitation of that mineral due to kinetic constraints and properties of seawater. Another parameter important to calcifying organisms is the magnesium/calcium ratio of seawater. Molar

PAGE 17

6 Mg/Ca ratios ( m Mg/Ca) le ss than 2 favor low Mg calcite, ratios between 2 and 5.3 favor high Mg calcite and aragonite, and ratios above ~5.3 favor aragonite precipitation ( Hardie, 1996 ). Foraminifera Members of the class Foraminifera (phylum Granuloreticulosa) are single celled m arine protozoans and are the most speciose members (with an estimated 10,000 modern species) of the Protozoa (Vickerman, 1992). They constitute the most diverse group of shelled microorganisms in modern oceans (Sen Gupta 1999) with distributions ranging from polar shelves to tropical coral reefs and from supralittoral sands and intertidal mudflats to the bottom of the deepest abyssal trenches (Gooday, 2002; Todo et al., 2005). They play an important role in biogeochemical cycling of inorganic and organic compounds with foraminiferal taxa varying in response to both the quantity of organic matter and its form (Loubere and Fariduddin, 1999). Second only to the coccolithophores, foraminifers are major oceanic producers of calcium carbonate. Langer (2008) es timated that together both larger and smaller benthic foraminifers along with planktonic foraminifers contribute 1.4 billion tons of calcium carbonate per year representing almost 25% of the present day calcium carbonate production in the world's oceans. They have been important sediment producers both in the modern age and in the fossil record as these single celled organisms have inhabited the ocean for over 500 million years (BouDagher Fadel, 2008) Simple forms appeared in the Cambrian and their evol ution has continued in a well recorded map of zonal stratigraphy, paleoenvironment, pal eobiology, and paleooceanography (BouDagher Fadel, 2008).

PAGE 18

7 The seven species of foraminifers used in this study are Amphisorus hemprichii Ehrenberg, 1839, Amphistegina l essonii d'Orbigny, 1826, A lobifera Larsen, 1976, A radiata (Fichtel and Moll, 1789), Calcarina defrancii d'Orbigny, 1826, C gaudichaudii d'Orbigny, 1840, and Heterostegina depressa d'Orbigny, 1826. They represent fou r different families (Amphisteginid ae, Calcarinidae, Nummulitidae, and Soritidae) and include porcelaneous members of the order Miliolida and hyaline Rotaliida ( Table 1). The porcelaneous group is defined by three layered calcitic, imperforate, non lamellar walls with a high percentage of rod like magnesium calcite crystals that have their axes randomly oriented in the embedding organic material. The hyaline calcareous representatives have a lamellar perforate wall structure composed of calcite crystals containing a variable percentage of magnesium and the mineralogical c axis oriented perpendicular to the shell surface (BouDagher Fadel, 2008). TABLE 1: Foraminiferal species including their higher taxonomy, wall structure group and calcium carbonate mineral used to build their shells Spe cies Order Family Group Shell Mineral Amphisorus hemprichii Miliolida Soritidae Porcelaneous High Mg Calcite Amphistegina lessonii Rotaliida Amphisteginidae Hyaline Low Mg Calcite Amphistegina lobifera Rotaliida Amphisteginidae Hyaline Low Mg Calcite Amphistegina radiata Rotaliida Amphisteginidae Hyaline Low Mg Calcite Calcarina defrancii Rotaliida Calcarinidae Hyaline High Mg Calcite Calcarina gaudichaudii Rotaliida Calcarinidae Hyaline High Mg Calcite Heterostegina depressa Rotaliida Nummulitidae Hyaline High Mg Calcite

PAGE 19

8 Halimeda Also examined in this study w ere segments of Halimeda a genus of macroscopic green algae that secrete aragonitic segments They are extremely important contributors to carbonat e sediments and are often underestimate d in the global calcium carbonate budget. For example, Halimeda bioherms in the Northern Great Barrier Reef contain one to four times more CaCO 3 sediment than adjacent ribbon reef facies (Rees et al., 2007). The Halimeda genus has 33 modern species that are widely distributed in tropical waters and are distinguishable by the morphological properties of the algal body, also called the th allus (Hillis 2001 and references therein ). The first recorded instance of Halimeda is from the Permian, however the ge nus did not begin to diversify until the late Cretaceous with its peak of success during the Cenozoic (Hillis, 2001). The species used in this study was Halimeda tuna (Ellis and Solander, 1786) Lamouroux, 1816. Taphonomy Taphonomy is the study of the pro cesses involved in an organism's incorporation into the fossil record. Important taphonomic processes include etching, abrasion, bioerosion, breakage, cement ation, degradation, dissolution, diagenesis, microbially mediated chemical reactions, predation, a nd transport (among others). Taphonomic processes are extremely important in determining how an assemblage of dead organisms has been modified since the organisms were alive. Many studies have been done to describe taphonomic signatures and patterns of s hell degradation of foraminifera l assemblages (e.g. Cottey and Hallock, 1988, Peebles and Lewis, 1991, Kotler et al., 1992, Berkeley et al., 2009) describ ing the differences that result from different processes. For example, in foraminifer s abrasion pr oduce s shell s that have small

PAGE 20

9 scratches and pitting of shell surfaces (Cottey and Hallock, 1988, Peebles and Lewis, 1991), while dissolu tion produces coarser textures that resemble karst topography (Peebles and Lewis, 1991). Shell architecture also plays an important role in dissolution. As discussed in Corliss and Honjo (1981), hyaline, trochospiral taxa are more resistant to dissolution than porcelaneous species. Among the hyaline perforate taxa, foraminifers with thin walls and a high density of pores undergo dissolution more readily than thicker walled, less porous taxa. Energy Dispersive Spectroscopy Energy dispersive spectroscopy, or EDS is a non destructive technique for determining el emental composition of a sample using a scanning electron micr oscope ( SEM ) The electron beam of the SEM hit s a sample caus ing the sample to release electromagnetic waves. This electromagnetic signal, in the form of x rays, is collected by a detector and converted into an electr on ic signal proportional to the energ y of the incoming x ray. Each electr on ic signal is added to a specific channel corresponding to its energy and each channel's count rates increase in an "add one" process Each channel corresponds to a specific element, as each element emits x rays of a charact eristic energy. As the channel s count rates increase, peaks are formed and the presence or absence of elements can be qualitatively determined. Quantitative analysis of samples can only be carried out under specific conditions as the peak intensi ties can be affected by a number of variables For example, heavier elements produce more x rays than lighter elements, there may be differences in x ray absorption and secondary fluorescence, and there is a natural variation in background radiation acros s the spectrum. Complex matrix effects prevent the direct comparison of

PAGE 21

10 x ray count rates of an unknown substance to the x ray count rates in a pure standard. In order to achieve accurate quantitative analyses, they should only be performed on homogeneou s samples with smooth, polished surfaces as rough topography may prevent x rays from reaching the detector. EDS is also not good for detecting trace elements and light elements as its minimum detect ion limit is approximately 1000 ppm, and elements lighter than Na are subject to errors ( A. Greco, 2010, personal communication) Research Objectives 1. To describe dissolution features of Halimeda tuna and an assortment of larger benthic foraminifers along a pH and temperature gradient produced by a shallow wate r hydrothermal vent 2. To determine Mg/Ca ratios of the Halimeda tuna and foraminifers using ED S and to analyze trends associated with t he pH and temperature gradient both within species and across evolutionary lineages. 3. To compare differences in disso lution among different evolutionary lineages of foraminifers both visually and chemically. Hypotheses to be T ested 1. P orcelaneous shells undergo dissolution more readily than hyaline shells. 2. Among the hyaline taxa, high Mg calcite shells undergo dissol ution more readily than calcite shells. 3. Partial dissolution will preferentially remove magnesium from shells resulting in lower Mg/Ca ratios. 4. Aragonitic Halimeda tuna which precipitates a more soluble polymorph of CaCO 3 than the calcitic foraminife rs, will undergo greater dissolution.

PAGE 22

11 METHODS Experimental Materials and Setup Several large (~100cm 2 ) pieces of reef rubble were collected at a depth of approximately 10m. Colle ction took place approximately 2 0 0m south of the area of hydrothermal ve nting in a coral dominated area deemed typical reef where the sediment was mostly carbonate coral reef substrate and had a visible foraminiferal assemblage Rubble was scrubbed with a small brush into a zipper sealed bag. All of the material collected sieve. Fixed volumes of approximately 5mL of homogenized rubble extract were added (approximate dimensions 6cm x 8cm). Halimeda tuna samples were also collected from the same site and added to the bags (approximately three segments per bag). Mesh bags were randomly assigned to six groups corresponding to distances away from Vent IV (7.5m, 15m, 30m, 90m, 180m, and a reference site close to the location of original rubble collection see Figure 3 ). These 12 bags were deployed on May 28th, 2005, by cable tying them to scuba weights and were left on site for five days.

PAGE 23

12 FIGURE 3 : a) Location of Vent 4 (V 4) in relation to Ambitle Island and b) the location of the transect line. F ive deployment locations are marked with red Xs. Distances from the vent at these locations are 7.5m, 15m, 30m, 90m, and 180m, respectively. The reference site was located at 10m d epth on reef substrate south of the vent field and therefore outside the area depicted in Figure 3a. One bag was not deployed, but rather, was immediately rinsed in distilled water and air dried as an example of the "starting material." Upon collection, the 12 bags that had been deployed were also rinsed in distilled water and air dried for later examination. Dried samples were left in their bags and stored in the Reef Indicators Lab under temperature controlled conditions at the College of Marine Scien ce, University of South Florida until May 2010. Imaging Techniques Individual foraminifers and Halimeda samples were mounted on aluminum SEM stubs coated with conductive carbon tape using paint brushes dipped in deionized water. They were then digitally photographed using a light microscope (Stemi 2000 C, Zeiss) with attached camera (AxioCam MRC5, Zeiss) and accompanying AxioVision software. Foraminifers were imaged at the highest magnification that would include the entire individual. Halimeda tuna spe cimens were imaged at three different magnifications to characterize the ir appearance and state of dissolution. Images were used for

PAGE 24

13 foraminiferal identification as well as verifying, based on symbiont color, whether individuals were dead or alive upon r etrieval of the sample bags. X ray spectra were generated for each of the individual foraminifers as well as three spectra for each sample of Halimeda tuna These were generated using an EDAx x ray detection system (Apollo 10, liquid nitrogen free, silic on drift detector, Ametek, Inc.) attached to a scanning electron mic roscope (Hitachi S 3500N). The spectra were collected under variable pressure mode with 30Pa of air in the SEM chamber. Conditions used included a working distance of 13mm, 15KV accelera ting voltage, and no aperture. Spectra were collected for 100 live H tuna samples. Spot size was adjusted to achieve a dead time of approximately 30% as per the manufacturer's recommendations and ranged from 83 92. The magnificat ion for each scan was determined by using the lowest magnification that yielded a field of view exclusively of a single organism with none of the stub visible. These magnifications ranged from 60x to 800x and were centered over the approximate middle of t he foraminifer shell The H tuna samples were each scanned three times at 60x. The three scans corresponded to three segments. EDAx Genesis software was used to view spectra and identify the elements present. Due to the limitations of quantitative EDS (homogeneous samples with smooth polished surfaces), true quantitative analysis could not be done. These factors affect overall count rates, but ratios between elements should remain the same. So to overcome the restrictions, Mg/Ca ratios were used. After digital imaging and EDS scans were complete, the SEM stubs with attached foraminifers and Halimeda tuna samples were sputter coated with AuPd alloy for eight

PAGE 25

14 minutes using a sputter coater (Hummer 6.2 Sputtering System, Anatech LTD, Alexandria, VA). Samples were then imaged with the Hitachi SEM under 5KV, size 4 aperture, 35 spot size and a 40mm working distance for maximal depth of field and surface texture detail. Foraminifers were imaged at the highest magnification that yielded an image of the e ntire shell Select ed individuals, one per species per deployment site, were also imaged at high magnification to view any alterations to the shell surface only visible under high magnification (600x for A. hemprichii 1000x for all other foraminifers ). Halimeda tuna samples were imaged three times, once each at 50x, 450x, and 1000x, each image from a different segment to obtain a comprehensive representation of the surface detail and overall dissolution of each specimen. Data Analyses Each of the 362 f oraminifers used in this experiment was assigned an identifying code and was classified to species level ( Table 2 ). Once positively identified, an image of a single foraminifer of each species was chosen for each site as a representative example for that site. Two individuals were chosen for the site closest to the vent (7.5m), one example of a relatively healthy specimen and one example of more extreme dissolution as determined by q ualitative descriptions based on specific shell characteristics and featu res of each taxon examined and how those features were influenced by exposure to vent fluids These images were assembled into plates with accompanying tables describing the state of the foraminifer based upon visual inspection of the different types of i mages: digital, full foraminifer SEM, close up SEM

PAGE 26

15 Ratios of the total counts of magnesium divided by the total counts of calcium (Mg/Ca) were determined directly in the EDAx Genesis software and were imported into Microsoft Excel. Averages, standard d eviations, one way ANOVAs and graphing were all done with Microsoft Excel. To determine which sets were significantly different, Tukey HSD tests were run by programming Microsoft Excel with the formulas from Zar (1998). A 95% significance level was used throughout.

PAGE 27

16 RESULTS A total of 362 foraminiferal specimens (Table 2) and 39 Halimeda segments were examined in this study. TABLE 2: Foraminifer al counts per site with species totals and site totals. 7.5m 15m 30m 90m 180m Ref. Site Starting Material Species Totals A. hemprichii 22 23 22 21 20 20 10 138 A. lessonii 5 5 5 5 10 4 2 36 A. lobifera 5 2 2 2 3 2 2 18 A. radiata 6 6 6 6 6 6 3 39 C. defrancii 3 13 13 6 5 5 3 48 C. gaudichaudii 10 11 10 10 10 10 8 69 H. depressa 2 2 2 2 2 2 2 14 Site T otals 53 62 60 52 56 49 30 362 Ref. Site is the Reference Site. Amphisorus hemprichii Amphisorus hemprichii was the only porcelaneous species used in this experiment. It has a brick like outer layer which also extends over the chamberlets. When the ou ter layer is dissolved away, chamberlets are exposed. Ten specimens were present in the sample that was killed and set aside before the start of the experiment (Table 2). These specimens appeared dark brown in their centers and lighter brown to white aro und the periphery; the brown color results from the presence of zooxanthellae within the shell (Plate 1). There were 20 to 23 specimens available from each of the experimental trials (Table 2). Specimens that were visually intact and exhibiting normal sy mbiont color were present in all trials. However, of the 22 experimental specimens

PAGE 28

17 from the 7.5m deployment site 9 were categorized as "dead" based on extensive ( >50% ) loss of surface layer 6 were categorized as "dying" based on a moderate ( 20 50% ) loss of surface layer and 7 were categorized as alive based on <20% loss of surface layer Percent loss was estimated visually from full foraminifer SEM images. The 9 dead specimens exhibited extensive dissolution visible at all magnifications. The highe r magnification (600x) SEM images revealed loss of fine detail and decreased surface topography in specimens from distances up to 30m from the vent (Plate 1; Table 3 ). Elemental ratios of Mg/Ca for A. hemprichii as determined by EDS were plotted against s ite data including: distance from the vent (Figure 4), average pH (Figure 5), and average temperature (Figure 6) (environmental data from McCloskey, 2009) Statistical analysis of the Mg/Ca ratio data revealed three significant groupings. Data for t he si te closest to the vent was significantly different from that for all other groups while the Mg/Ca ratios for starting material w ere significantly different from all other groups except the reference site and the site 30m from the vent.

PAGE 29

18 PLATE 1: Amphiso rus hemprichii Ehrenberg, 1839 T hree images of the same foraminifer for each site plus an additional foraminifer for the site 7.5m from the vent. The image on the left for each site is the digital photographic image, the middle is the full foraminifer S EM image and the image on the right is a close up SEM image taken at 600x magnification. Lengths listed under each of the SEM images correspond to the length of the yellow scale bars included in the pictures. 1mm 1mm Starting Material Reference Site 2mm 2mm 180m from V ent 90m from V ent 2mm 2mm 30m from V ent 15m from V ent 2mm 1mm 7.5m from V ent Specimen A 7.5m from V ent Specimen B

PAGE 30

19 TABLE 3: Descriptions of Amphisorus hemprichii images from each type of microscopy for each site (images described can be viewed in Plate 1). Site Light Micros copy Full Foram inifer SEM 600x SEM Starting Material Very dark brown in center, lighter around edges. Appears completely intact. Minimal loss of surface layer (<1%). Completely intact. Fine detail sharply visible on surface. Reference Site Light tan across entire foram inifer Appears completely intact. Minimal loss of surface layer (<1%). Completely intact. Fine detail sharply visible on surface. 180m from Vent Light tan across entire foram inifer Appears completely intact. Minimal loss of outer layer (<5%). Completely intact. Fine detail sharply visible on surface. 90m from Vent Very dark brown in center, lighter around edges. Appears completely intact. Minimal loss of surface layer (<1%). Completely intact. Fine detail sharply visible on su rface. 30m from Vent Darker brown in center, lighter around edges. Appears completely intact. Minimal loss of surface layer (<5%). Some holes in surface layer. Mostly intact. Some loss of fine detail smoother surface. 15m from Vent Darker brown in center, lighter around edges. Appears completely intact. Minimal loss of surface layer (<5%). Outer edges of foram inifer have some exposed underlying chamberlet structures. Some holes in outer layer. Mostly intact. Smooth surface almost complete los s of fine detail. 7.5m from Vent Specimen A Light colored loss of symbionts. Dissolution visible around center. Almost complete loss of surface layer. Exposed underlying chamberlet structures. No surface layer Can see through foram inifer Only u nderlying skeleton remains. 7.5m from Vent Specimen B Light colored loss of symbionts. Dissolution almost complete. On the way to total dissolution complete loss of surface layers and center of foram inifer No surface layer Can see through foram inifer Only underlying skeleton remains.

PAGE 31

20 FIGURE 4: Mg/Ca r atio of Amphisorus hemprichii vs. distance from the v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corre sponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test. Data p oints with two letters indicate significant similarity to more than one group. FIGURE 5: Mg/Ca r atio of Amphisorus hemprichii vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yell ow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey H SD test. Data points with two letters indicate significant similarity to more than one group.

PAGE 32

21 FIGURE 6: Mg/Ca r atio of Amphisorus hemprichii vs. t emperature. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the re ference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way AN OVA with post hoc Tukey HSD test. Data points with two letters indicate significant similarity to more than one group. Amphistegina spp. The three Amphistegina species build hyaline, low Mg calcite shells. Seven specimens were present in the sample that was killed and set aside before the start of the experiment (2 A. lessonii 2 A. lobifera 3 A. radiata ) (Table 2). These specimens appeared shades of golden brown, reflecting the presence of diatom symbionts within the dry cytoplasm within each shell ( Pl ates 2, 3, 4; Tables 4, 5, 6). Specimens available from each of the experimental trials ranged from 4 to 10 for A. lessonii and 2 to 5 for A. lobifera There were 6 for all experimental trials for A. radiata (Table 2). Specimens that were visually in tact and exhibiting normal symbiont color were present from all trials except those at the 7.5m site for A. lobifera and A. radiata All 5 of the A. lobifera at 7.5m were classified as "dead" based on a >50% loss of surface layer and 5 of the 6 A. radiat a at 7.5m were classified as dead while the other specimen was classified as

PAGE 33

22 "dying" based on a 20 50% loss of surface layer A mphistegina lessonii exhibited the greatest survival Of the 5 specimens at the 7.5m deployment site, 2 were classified as dead" based on a >50% loss of surface layer 1 as "dying" based on a 20 50% loss of surface layer and 2 as "alive based on <20% loss of surface layer Percent loss was estimated visually from full foraminifer SEM images. The A. lobifera and A. radiata showed similar trends of breakage and loss of the surface layer in individuals placed 15m from the vent with more extreme dissolution i n those from the 7.5m site ( Plates 3 4 ; Tables 5, 6) Elemental ratios of M g/Ca for the three Amphistegina spp as de termined by EDS were plotted against site data including: distance from the vent (Figures 7 10, 13 ), average pH (Figures 8, 11, 14 ), and average temperature (Figures 9, 12, 15 ) (environmental data from McCloskey, 2009). Statistical analysis of the Mg/Ca ratio data revealed two significant groupings in A. lessonii and A. lobifera Data for t he site closest to the vent was significantly different from that for the reference site in A. lessonii and the data for the site closest to the vent was significantly different from that for the site at 90m in A. lobifera Data for a ll sites were statistically similar for A. radiata

PAGE 34

23 P LATE 2 : Amphistegina lessonii d T hree images of the same foraminifer for each site plus an additional foraminifer fo r the site 7.5m from the vent. The image on the left for each site is the digital photographic image, the middle is the full foraminifer SEM image and the image on the right is a close up SEM image taken at 1000x magnification. Lengths listed under each of the SEM images correspond to the length of the yellow scale bars included in the pictures. Starting Material Reference Site 180m from Vent 90m from Vent 30m from Vent 15m from Vent 1mm 7.5m from Vent Specimen A 7.5m from Vent Specimen B

PAGE 35

24 TABLE 4: Descriptions of Amphistegina lesson ii images from each type of microscopy for each site (images described can be viewed in Plate 2). Site Light Microscopy Full Foram inifer SEM 1000x SEM Starting Material Deep golden brown symbiont color. Appears completely intact. Surface completely inta ct. Pustules have smooth edges. Rounded edges to individual pores. Smooth surface. Reference Site Deep golden brown symbiont color. Abnormally shaped. Appears completely intact. Surface intact. Pustules have smooth edges. Minimal breakage around pu stules. Some sharp edges to individual pores. Smooth surface. 180m from Vent Deep golden brown symbiont color. Appears completely intact. Minimal loss of surface layer (<1%). Pustules have smooth edges. Sharp edges to individual pores. Smooth surface 90m from Vent Lighter golden brown symbiont color. Appears completely intact. Surface completely intact. Pustules have sharp relief. Some sharp edges to individual pores. Smooth surface. 30m from Vent Very deep brown symbiont color. Appears comp letely intact. Surface completely intact. Pustules have sharp relief. Rounded edges to individual pores. Some crystals visible on surface. 15m from Vent Golden brown in center, loss of symbiont color around edges. Appears completely intact. Surface co mpletely intact. Dorsal view. Some sharp edges to individual pores. Some rough patches on surface. 7.5m from Vent Specimen A Light tan symbiont color. Appears completely intact. Individual pores visible. Shell breakage around pustules. Pustules ha ve sharp relief. Sharp edges to individual pores. Some connections between pores formed by dissolution. Smooth surface. 7.5m from Vent Specimen B White; complete loss of symbionts. Extensive breakage. Rough surface. Complete loss of outer layers. Extensive breakage around edges. Almo st complete loss of outer lamellae Very rough surface.

PAGE 36

25 FIGURE 7: Mg/Ca r atio of A mphistegina lessonii vs. d is tance from the v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond correspond s to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test. Data points with two letters indicate significant similarity to more than one group. FIGURE 8: Mg/Ca r atio of A mphistegina lessonii vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink d iamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference a s determined by a one way ANOVA with post hoc Tukey HSD test. Data points with two letters indicate significant similarity to more than one group.

PAGE 37

26 FIGURE 9: Mg/Ca r atio of A mphistegina lessonii vs. t emperature. Mg/Ca r atios are count rate ratios gener ated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test. Data points with two letters indicate significant similarity to more than one group.

PAGE 38

27 P LATE 3 : Amphistegina lobifera Larsen, 1976 T hree images of the same foraminife r for each site plus an additional foraminifer for the site 7.5m from the vent. The image on the left for each site is the digital photographic image, the middle is the full foraminifer SEM image and the image on the right is a close up SEM image taken at 1000x magnification. Lengths listed under each of the SEM images correspond to the length of the yellow scale bars included in the pictures. Starting Material Reference Site 1mm 180m from Vent 90m from Vent 1mm 30m from Vent 15m from Vent 1mm 7.5m from Vent Specimen A 7.5m from Vent Specimen B

PAGE 39

28 TABLE 5: Descriptions of Amphistegina lobifera images from each type of microscopy for each site (images described can be viewed in Plate 3). Site Light Microscopy F ull Foraminifer SEM 1000x SEM Starting Material Golden brown color with brown patterning. Small glassy center. Smooth surface. Crystals visible on surface. Rounded edges to individual pores. Smooth surface. Reference Site Light tan color with brown patterning. Small glassy center. Smooth surface. Crystals visible on surface. Sharp edges to individual pores. Smooth surface. 180m from Vent Golden brown color with brown patterning. Small glassy center. Smooth surface. Some sharp edges to individu al pores. Some connections between pores formed by surface loss. Smooth surface. 90m from Vent Light tan color with minimal patterning. Large glassy center. Smooth surface. Some loss of skeleton around edge. Sharp edges to individual pores. Smooth s urface. 30m from Vent Light tan color with brown patterning. Large glassy center. Smooth surface. Sharp edges to individual pores. Smooth surface. 15m from Vent White color with minimal patterning. Small brown center. Rough surface. Rough around ed ges. Minimal breakage. Almost complete surface loss. Rough under layer exposed. 7.5m from Vent Specimen A Light tan center with no patterning. Large glassy center. Extensive breakage. Extensive breakage. Visible pustules with sharp relief. Some su rface roughness. Complete surface loss. Rough under layer exposed. 7.5m from Vent Specimen B White color with no patterning or visible glassy center. Complete loss of outer layer. Extensive loss around edges. Extensive breakage. Complete surface loss Dissolution visible of rough under layer. Some breakage.

PAGE 40

29 FIGURE 10: Mg/Ca r atio of A mphistegina lobifera vs. distance from v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the y ellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tuke y HSD test. Data points with two letters indicate significant similarity to more than one group. FIGURE 11: Mg/Ca r atio of A mphistegina lobifera vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the referen ce site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA w ith post hoc Tukey HSD test. Data points with two letters indicate significant similarity to more than one group.

PAGE 41

30 FIGURE 12: Mg/Ca r atio of A mphistegina lobifera vs. t emperature. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond c orresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determ ined by a one way ANOVA with post hoc Tukey HSD test. Data points with two letters indicate significant similarity to more than one group.

PAGE 42

31 P LATE 4 : Amphistegina radiata (Fichtel and Moll, 1798) T hree images of the same foraminifer for each site plus an additional foraminifer for the site 7.5m from the vent. The image on the left for each site is the digital photographic image, the middle is the full foraminifer SEM image and the image on the right is a close up SEM image taken at 1000x magnification. Lengths listed under each of the SEM images correspond to the length of the yellow scale bars included in the pictures. 1mm Starting Material Reference Site 1mm 1mm 50 180m from Vent 90m from Vent 1mm 1mm 30m from Vent 15m from Vent 1mm 1mm 7.5m from Vent Specimen A 7.5m from Vent Specimen B

PAGE 43

32 TABLE 6: Des criptions of Amphistegina radiat a images from each type of microscopy for each site (images described can be viewed in Plate 4). Site Light Microscopy Full Foraminifer SEM 1000x SEM Starting Material Light golden brown color with brown patterning. Appea rs intact. Smooth surface. Crystals visible on surface. Some sharp edges of individual pores. Smooth surface. Reference Site Golden brown in center, lighter around edges. Minimal breakage along edge. Smooth surface. Minimal breakage along edge. Sharp edges of individual pores. Smooth surface. 180m from Vent Golden brown and light tan mottling with brown patterning. Appears intact. Smooth surface. Minimal breakage on surface. Sharp edges of individual pores. Smooth surface. 90m from Vent Golden brown color with brown patterning. Appears intact. Smooth surface. Extensive breakage of one chamber. Some sharp edges to individual pores. Some connections between pores formed by surface dissolution Smooth surface. 30m from Vent Golden brown and light tan mottling with brown patterning. Appears intact. Smooth surface. Minimal breakage on surface. Sharp edges of individual pores. Smooth surface. 15m from Vent Light tan color with brown patterning. Appears intact. Rough surface. Minimal break age on surface. Loss of surface layer. Rough under layer exposed. 7.5m from Vent Specimen A White with brown patterning. Breakage around edges and along one chamber. Rough surface. Rough edges. Extensive breakage of one chamber. Loss of surface laye r. Rough under layer exposed. 7.5m from Vent Specimen B White with no visible patterning. Extensive breakage across half the chambers. Complete surface loss. Rough edges. Extensive breakage of half the chambers. Loss of surface layer. Rough under l ayer exposed. Extensive breakage.

PAGE 44

33 FIGURE 13: Mg/Ca r atio of A mphistegina radiata vs. d istance fro m the v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way A NOVA with post hoc Tukey HSD test. FIGURE 14: Mg/Ca r atio of A mphistegina radiata vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one s tandard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way A NOVA with post hoc Tukey HSD test.

PAGE 45

34 FIGURE 15: Mg/Ca r atio of A mphistegina radiata vs. t emperature. Mg/Ca r atios a re count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letter s correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test. Calcarina spp. The two Calcarina species build hyaline, high Mg calcite shells. Eleven specimens were present in the sample that was killed an d set aside before the start of the experiment ( 3 C. defrancii 8 C. gaudichaudii ) (Table 2). These specimens appeared shades of golden brown, reflecting the presence of diatom symbionts within the dry cytoplasm within each shell (Plates 5, 6 ; Tables 7, 8 ). Specimens available from each of the experimental trials ranged from 3 to 1 3 for C. defrancii and 10 or 11 for C. gaudichaudii (Table 2). Specimens that were visually intact and exhibiting normal symbiont color were present in all trials However, of the 3 C. defrancii at 7.5m from the vent, 2 were classified as "dead" based on a >50% loss of surface layer and 1 was classified as "alive based on <20% loss of surface layer Of the 10 C. gaudichaudii at 7.5m from the vent, 6 were classified as "dea d" based on a >50% loss of surface layer 1

PAGE 46

35 as "dying" based on a 20 50% loss of surface layer and 3 as "alive based on <20% loss of surface layer Percent loss was estimated visually from full foraminifer SEM images. Dissolution is evident at 15m from the vent in C. defrancii and at 30m from the vent in C. gaudichaudii Eleme ntal ratios of Mg/Ca for the two Calcari na spp as determined by EDS were plotted against site data including: distance from the vent (Figures 16, 19 ), average pH (Figures 1 7, 20 ) and average temperature (Figures 18, 21 ) (environmental data from McCloskey, 2009). Statistical analysis of the Mg/Ca ratio data revealed no significantly different groupings in either species.

PAGE 47

36 P LATE 5 : Calcarina defrancii T hree ima ges of the same foraminifer for each site plus an additional foraminifer for the site 7.5m from the vent. The image on the left for each site is the digital photographic image, the middle is the full foraminifer SEM image and the image on the right is a c lose up SEM image taken at 1000x magnification. Lengths listed under each of the SEM images correspond to the length of the yellow scale bars included in the pictures. 2 2 Starting Material Reference Site 5 3 180m from Vent 90m from Vent 5 3 30m from Vent 15m from Vent 3 3 7.5m fr om Vent Specimen A 7.5m from Vent Specimen B

PAGE 48

37 TABLE 7: Descriptions of Calcarina defrancii images from each type of microscopy for each site (images described can be viewed in Plate 5). Site Light Microscopy Full Foraminifer SEM 1000x SEM Starting Mat erial Golden brown around center. White around edges and spines. Appears intact. Small needles across surface. Intact spines. Lots of topography on surface. Small needles visible. Lots of topography. Reference Site Golden brown around center. Whit e around edges and spines. Appears intact. Intact spines. Lots of topography on surface. Small needles visible. Lots of topography. 180m from Vent Completely white loss of symbionts. Appears intact. Intact spines. Lots of topography on surface. Fe w needles visible. Lots of topography. 90m from Vent Golden brown around center. White around edges and spines. Breakage of one chamber. Small needles across surface. Lots of topography. Breakage of one chamber. Small needles visible. Lots of topog raphy. 30m from Vent Light tan. Breakage of one chamber. Loss of surface topography/top layer. Breakage of one chamber and along some spines. Small needles visible. Lots of topography. 15m from Vent Golden brown around center. White around edges an d spines. Breakage of one chamber. Loss of surface topography. Breakage of one chamber. Few needles visible. Less topography smoother surface. 7.5m from Vent Specimen A Light tan. Appears intact. Increased fine detail due to loss of top surface la yers. Increased fine detail of needles and pores due to loss of top surface layers. 7.5m from Vent Specimen B Completely white loss of symbionts. Smooth surface. Smooth surface. Loss of topography/surface detail. Rough surface. No needles, no topog raphy.

PAGE 49

38 FIGURE 16: Mg/Ca r atio of Calcarina defrancii vs. distance from the v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the startin g material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way A NOVA with post hoc Tukey HSD test. FIGURE 17: Mg/Ca r atio of Calcarina def rancii vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and be low the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD t est.

PAGE 50

39 FIGURE 18: Mg/Ca r atio of Calcarina defrancii vs. t emperature. Mg/Ca r atios are count rate ratios generated b y EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of signif icant difference as determined by a one way ANOVA wi th post hoc Tukey HSD test.

PAGE 51

40 P LATE 6 : Calcarina gaudichaudii T hree images of the same foraminifer for each site plus an additional foraminifer for the site 7.5m from the vent. The imag e on the left for each site is the digital photographic image, the middle is the full foraminifer SEM image and the image on the right is a close up SEM image taken at 1000x magnification. Lengths listed under each of the SEM images correspond to the leng th of the yellow scale bars included in the pictures. 1mm 2mm Starting Material Reference Site 1mm 1mm 180m from Vent 90m from Vent 2mm 2mm 30m from Vent 15m from Vent 1mm 1mm 7.5m from Vent Specimen A 7.5m from Vent Specimen B

PAGE 52

41 TABLE 8: Descriptions of Calcarina gaudichaudii images from each type of microsco py for each site (images described can be viewed in Plate 6). Site Light Microscopy Full Foraminifer SEM 1000x SEM Starting Material Golden brown center. Lighter tan spines. Appears intact. Highly visible canal system across shell Spines are intact. Fine detail in pores visible and intact. Some topography on surface layer. Reference Site Light tan body and spines. Appears intact. Highly visible canal system across shell Spines are intact. Fine detail in pores visible and intact. Some topography on surface layer. 180m from Vent Light tan body and spines. Appears intact. Highly visible canal system across shell Minimal loss of surface layer (<1%). Fine detail in pores visible and intact. Minimal topography on surface layer. 90m from Vent L ight tan body and spines. Appears intact. Highly visible canal system across shell Increased visibility of fine detail. Fine detail in pores visible and intact. Some topography on surface layer. 30m from Vent Light tan body and spines. Appears intac t. Highly visible canal system across shell Minimal loss of surface layer (<5%). Fine detail in pores visible and mostly (>90%) intact. Some topography on surface layer. 15m from Vent Golden brown center. Lighter tan spines. Appears intact. Highly v isible canal system across shell Increased visibility of fine detail due to loss of surface layer. Fine detail in pores visible but mostly (>50%) missing. Minimal topography on surface layer. 7.5m from Vent Specimen A Completely white symbiont loss Some breakage of chambers. Increased visibility of fine detail due to almost complete loss of surface layer. Some breakage of chambers/spines. Fine detail in pores visible but mostly (>90%) missing. Surface layer missing exposing rough under layer. 7 .5m from Vent Specimen B Completely white symbiont loss. Extensive breakage visible on surface. Increased visibility of fine detail due to complete loss of surface layer. Extensive breakage of chambers/spines. No fine detail in pores. Surface layer missing exposing rough under layer.

PAGE 53

42 FIGURE 19: Mg/Ca r atio of Calcarina gaudichaudii vs. distance from the v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond correspon ds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA wi th post hoc Tukey HSD test. FIGURE 2 0: Mg/Ca r atio of Calcarina gaudichaudii vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA wi th post hoc Tukey HSD test.

PAGE 54

43 FIGURE 21: Mg/Ca r atio of Calcarina gaudichaudii vs. t emperature. Mg/Ca r a tios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test. Heterostegina depressa H eterostegina depressa build s hyaline, high Mg calcite shells. Two specimens were present in the sample that was killed and set aside before the start of the experiment (Table 2). These specimens appeared shades of golden brown, reflecting the presence of diatom symbionts within the dry cyto plasm within each shell (Plate 7; Table 9 ). Two s pecimens were available f rom each of the experimental trials (Table 2). Both specimens at 7.5m were classified as "dead" based on a >50% loss of surface layer while all others were classified as "alive" based on <20% loss of surface layer Percent loss was estimated visually fr om full foraminifer SEM images. Elemen tal ratios of Mg/Ca for H. depressa as determined by EDS were plotted against site data including: distance from the vent (Figure 22), average pH (Figures 23),

PAGE 55

44 and average temperature (Figures 24 ) (environmental data from McCloskey, 2009). Statistical analysis of the Mg/Ca ratio data revealed no significantly different groupings

PAGE 56

45 P LATE 7 : Heterostegina depressa T hree images of the same foraminifer for each site plus an additional foraminifer for t he site 7.5m from the vent. The image on the left for each site is the digital photographic image, the middle is the full foraminifer SEM image and the image on the right is a close up SEM image taken at 1000x magnification. Lengths listed under each of the SEM images correspond to the length of the yellow scale bars included in the pictures. 1mm 1mm Starting Material Reference Site 1mm 1mm 180m from Vent 90m from Vent 1mm 1mm 30m from Vent 15m from Vent 1mm 5 0 7.5m from Vent Specimen A 7.5m from Vent Specimen B

PAGE 57

46 TABLE 9 : Descriptions of Heterostegina depressa images from each type of microscopy for each site (images described can be viewed in Plate 7). Site Light Microscopy Ful l Foraminifer SEM 1000x SEM Starting Material Golden brown at thickest part of shell Light tan around edges. Minimal breakage. Smooth surface. Minimal breakage around edges and on surface. Sharp edges of individual pores. Smooth surface. Reference Site Dark brown at thickest part of shell Light tan around edges. Breakage likely due to handling. Smooth surface. Extensive breakage likely due to handling. Sharp edges of individual pores. Smooth surface. 180m from Vent Light tan at thickest part of shell White around edges. Smooth surface. Minimal breakage on surface. Sharp edges of individual pores. Smooth surface. 90m from Vent Golden brown at thickest part of shell White around edges. Smooth surface. Sharp edges of individual pores. S mooth surface. 30m from Vent Golden brown at thickest part of shell Light tan around edges. Minimal breakage. Smooth surface. Minimal breakage around edges and on surface. Sharp edges of individual pores. Smooth surface. Some loss of surface layer (<1%). 15m from Vent Light tan at thickest part of shell White around edges. Minimal breakage. Smooth surface. Minimal breakage around edges and on surface. Sharp edges of individual pores. Some roughness to surface. 7.5m from Vent Specimen A Com pletely white. Extreme breakage and loss of shell Rough surface loss of surface layer. Extreme breakage and loss of shell Complete loss of surface layer exposing rough under layer. Extreme breakage. 7.5m from Vent Specimen B Completely white. E xtreme breakage and loss of shell Rough surface loss of surface layer. Extreme breakage and loss of shell Complete loss of surface layer exposing rough under layer. Extreme breakage.

PAGE 58

47 FIGURE 22: Mg/Ca r atio of Heterostegina depressa vs. distance from the v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and be low the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test. FIGURE 23: Mg/Ca r atio of Heterostegina depressa vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test.

PAGE 59

48 FIGURE 24: Mg/Ca r atio of Heterostegina depressa vs. t emperature. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and th e yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc T ukey HSD test. Halimeda tuna : H alimeda tuna precipitates aragonit e One specimen, consisting of three segments, was killed and set aside before the start of the experiment (Table 2). Th is specimen appeared white, reflecting the calcium carbonate used to build the segments (Plate 8; Table 10 ). At all experimental sites, two specimens with three segments each were deployed. One specimen at 7.5m was completely decalcified. Elemen tal ratios of Mg /Ca for H. tuna as determined by EDS were plotted against si te data including: distance from the vent (Figure 25), average pH (Figure 26), and average temperature (Figure 27 ) (environmental data from McCloskey, 2009). Statistical analysis of the Mg/Ca ratio data revealed that the starting material was significantl y different from all experimental samples. The completely decalcified sample from 7.5m was excluded from plots and from the statistical analyses.

PAGE 60

49 P LATE 8 : Halimeda tuna (Ellis and Solander, 1786) Lamouroux, 1816 T hree images of the same sample of Halim eda for each site plus an additional sample for the site 7.5m from the vent. The image on the left for each site is the digital photographic image, the middle is a 50x magnification SEM image and the image on the right is a 1000x SEM image. Each picture is not necessarily from the same segment of the Halimeda but still from the same individual. Lengths listed under each of the SEM images correspond to the length of the yellow scale bars included in the pictures. 1mm 1mm Starting Material Reference Site 1mm 1mm 180m from Vent 90m from Vent 1mm 1mm 30m from Vent 15m from Vent 1mm 7.5m from Vent Specimen A 7.5m from Vent Specimen B

PAGE 61

50 TABLE 10 : Descriptions of Halimeda tuna images from each type of microscopy for each site (images described can be viewed in Plate 8). Site Lig ht Microscopy 50x SEM 1000x SEM Starting Material Opalescent white surface. Smooth surface. Individual hexagonal and pentagonal structures visible. Crystals on surface. Reference Site Opalescent white surface. Smooth surface. Extensive breakage from drying. Individual hexagonal and pentagonal structures visible. Crystals on surface. 180m from Vent Opalescent white surface. Some roughness. Extensive tissue loss from drying. Individual hexagonal and pentagonal structures visible. Crystals on surfac e. 90m from Vent Opalescent white surface. Smooth surface. Extensive breakage from drying. Individual hexagonal and pentagonal structures visible. Crystals on surface. 30m from Vent Opalescent white surface. Smooth surface. Extensive breakage from d rying. Individual hexagonal and pentagonal structures visible. Crystals on surface. 15m from Vent Opalescent white surface. Rough surface. Increased surface topography. Individual hexagonal and pentagonal structures visible. No crystals visible. 7.5m from Vent Specimen A Matte white surface with green showing through. Smooth surface. Individual hexagonal and pentagonal structures visible. Crystals on surface. 7.5m from Vent Specimen B Completely decalcified. Segment appears green. Extreme surfac e topography due to complete decalcification and loss of surface layer. Individual hexagonal and pentagonal structures visible. Crystals on surface. Image taken from segment with remaining surface layer.

PAGE 62

51 FIGURE 25: Mg/Ca r atio of Halimeda tuna vs. distance from the v ent. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation ab ove and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test A completely decalcified segment of H. tuna from 7.5m was excluded during averaging as the ratio was to o high to display on the graph. FIGURE 2 6: Mg/Ca r atio of Halimeda tuna vs. pH. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with post hoc Tukey HSD test A completely decalcified segment of H. tuna from 7.5m was excluded during averaging as the ratio was too high to display on the graph.

PAGE 63

52 FIGURE 27: Mg/Ca r atio of Halimeda tuna vs. t emperature. Mg/Ca r atios are count rate ratios generated by EDS. The pink diamond corresponds to the reference site and the yellow diamond corresponds to the value from the starting material. Error bars are plotted one standard deviation above and below the average value. Letters correspond to groups of significant difference as determined by a one way ANOVA with pos t hoc Tukey HSD test A completely decalcified segment of H. tuna from 7.5m was excluded during averaging as the ratio was too high to display on the graph.

PAGE 64

53 DISCUSSION Digital imaging was useful in visualizing whether individuals were alive or dead a t the end of the experiment based on symbiont color, or lack thereof. Full foraminifer SEM images were useful for describing the overall state of the organism and for judging how much dissolution had taken place. High magnification (600 1000x) SEM images proved to be the most useful for revealing trends in dissolution, exposing details not immediately apparent with the other imaging techniques SEM images that were close ups of foraminiferal pores and other shell surface features were a powerful tool to view the fine scale effects of dissolution. Even when the overall foraminifer appeared unaffected, loss of fine detail could often be seen under higher magnifications. Under magnifications as low as 1000x, lamellae were clearly visible in the perforate f oraminifers. A nalyses of elemental ratios of magnesium/calcium by EDS revealed trends not discernible through visual analysis alone. Evidence for dissolution indicated by a gradual decrease in the Mg/Ca ratio was seen in three of the species examined Temperature spikes of 10 15 o C occurred during the temperature logger deployment (McCloskey, 2009), so the temperatures listed for each site are averages and do not reflect the extremes that the samples endured. At t he site closest to the vent (7.5m) all taxa examined visibly exhibited the most extensive dissolution even though that site did not have the lowest pH recorded In three of the eight species tested, the

PAGE 65

54 average Mg/Ca ratio at the 7.5m site was significantly different from the ratio at the refe rence site or the starting material. In most of the other comparisons, data were too variable and sample sizes were too small to determine if differential dissolution of Mg was significant. Amphisorus hemprichii was the only porcelaneous species examined in the study. Shells of this species exhibited visually detectable dissolution (Plate 1, Table 3). The high magnesium calcite precipitated by Amphisorus hemprichii is also more susceptible to dissolution than low magnesium counterparts Dissolution pro gressed from the gradual loss of the surface layer to a subsequent loss of the entire center of the foraminifer. Some foraminifers that did not yet show any signs of dissolution were still noted to be withdrawing to inner chambers as visualized by the ret reat of symbiont color from the edges. The higher magnification (600x) SEM images revealed loss of fine detail and decreased surface topography at deployment distances 30m from the vent. Amphisorus hemprichii was noted to be the species most affected by dissolution using visual analysis. Elemental analysis agree d The A. hemprichii was the only species to exhibit three statistically significant groups of Mg/Ca ratios (Figure 4). The specimens from the site closest to the vent were significantly differe nt from specimens from all other sites while the starting material specimens were significantly different from specimens at all other sites except those at the reference site and the site 30m from the vent. The higher Mg/Ca ratio of the starting material may be due to higher magnesium concentrations in organic material and not necessarily a higher percentage of magnesium incorporated into its shell. When the average Mg/Ca ratio was plotted against temperature and pH, respectively, some interesting trends were illuminated. It appears

PAGE 66

55 that temperature had a greater effect on the state of the foraminifer than pH (Figures 5, 6). It is likely that the high temperatures close to the vent killed the foraminifers then subsequently the dead foraminifers began to dissolve in the low pH. This interpretation is based on the observation that the 15m site had the lowest measured pH (5.9) compared to the site 7.5m from the vent where the pH was 6.2. However, the temperature 7.5m from the vent was 59.3 o C compared to 39.9 o C at the site 15m from the vent. The difference of 20 o C seems to have played a larger role in the survival of foraminifers than the pH difference of 0.3 units. The three Amphistegina species build hyaline, low Mg calcite shell s. They were visually less affected than the porcelaneous and hyaline high Mg calcite foraminifers. Among the Amphistegina spp. A lessonii exhibited the highest tolerance for environmental extremes. There were individuals (e.g., 7.5m from Vent A, Plate 2) that seemed to r emain unaffected even at the site closest to the vent. A mphistegina lobifera and A. radiata showed similar trends of breakage and loss of the surface layer of individuals placed 15m from the vent, with more extreme dissolution in those from the 7.5m site (Plates 3, 4; Tables 5, 6). Amphistegina lessonii and A. lobifera exhibited similar elemental trends (Figures 7 9 and 10 12 ). Again, it appeared that temperature had the greater effect, killing foraminifers and leaving them vulnerable to dissolution. Mg /Ca ratios of A. lessonii from the site closest to the vent were significantly different from those deployed at the reference site. In A. lobifera the Mg/Ca ratios from the site closest to the vent were significantly different from those deployed 90m fro m the vent. For A. radiata specimens, there were no significant differences among sites Even though

PAGE 67

56 visually A. lessonii exhibited the least dissolution in terms of elemental analysis, A. radiata was the least affected. The two Calcarina species appea red to be the least affected by the hydrothermal venting by visual inspection of both the digital images and full foraminifer SEM images, yet upon examining the close up SEM images (1000x), the loss of fine detail in the form of needle like projects on the surface of the foraminifers and in their pores suggests that the hydrothermal venting had an effect across a much greater distance than is visible in the other images (Plates 5, 6; Tables 7, 8). The C. defrancii specimens lost most of their fine detail a t the site 15m from the vent while the C. gaudichaudii specimens were beginning to lose fine detail at the site 30m from the vent. If left on site for a longer time it is likely that the loss of fine detail would have proceeded to the point that it would be visible under lesser magnifications. The fact that the body morphology of these species has a much greater surface area than the other species tested would also contribute to increased dissolution. The two Calcarina species seemed relatively unaffecte d in terms of their elemental composition (Figures 16 21). Neither C. defrancii or C. gaudichaudii had any M g / C a r a t i o s f r o m sites that were significantly different than M g / C a r a t i o s f r o m any of the other sites. In general, the trends across sites, pH, and temperature are closer to a strai ght line than any sort of dependent trend. This is in contrast with the visual analysis which showed loss of fine detail up to 30m from the vent. Heterostegina depressa is a fragile foraminifer to work with. It is difficult to tell how much of the break age that occurred was due to dissolution and how much was due to human manipulation. It appears that dissolution was only beginning to take place at 15m from the vent while it had progressed to an extreme at the site 7.5m from the vent (Plate

PAGE 68

57 7; Table 9). T he extremely low numbers of specimens available (2 per site) prevented detection of any real trend s Halimeda tuna was included as a representative of calcifying organisms that use aragonite to build their skeletons. Again, it was difficult to tell ho w much of the altered appearance is evidence of dissolution versus normal cracking due to the air drying process. One point that was very evident, however, was the complete decalcification of one of the samples placed 7.5m from the vent (Plate 8, 7.5m fro m the vent B). All of the white calcium carbonate was gone leaving only green tissue remaining. In the second sample from 7.5m from the vent (Plate 8, 7.5m from the vent A), the same process had begun to occur and the green tissue could be viewed in areas of dissolution. Considering that the Halimeda samples were fragmented b e f o r e being placed in the bags, u n l i k e the living foraminifers, and were thus not able to actively maintain or repair their skeletons, they were fairly resistant to dissolu tion under all but the most extreme conditions. Halimeda tuna shows a different elemental trend than any of the foraminiferal species ( Figures 25 27). T he Mg/Ca ratio of the starting material is significantly different from all of the samples that were d eployed, possibly due to microbially mediated loss of organic material upon the organism's death. Loss of magnesium seemed to occur at a consistent rate across all of the sites, independent of pH and temperature. The large error bars associated with spec imens from the two sites closest to the vent are likely a consequence of three different states of foraminifers: dead, dying and alive. At more distant sites (30 180m) few, if any, individuals appeared to have been dying or dead at the end of the experim ent Foraminifers from the first two sites were distinguished visually using the full foraminifer SEM images. Individuals with more

PAGE 69

58 than 50% loss of their surface layers were determined to be "dead." Those with approximately 20 50% loss of surface layer were called "dying." Those with less than 20% loss of their surface layer were called "alive." The state of the foraminifer is also discernible by the amount of symbiont color visible on the shell. Healthy foraminifers of the species used in this study have a dark brown to golden brown color. As the foraminifers die, the color of the foraminifer shell fades to a light tan or white. After being separated into the three states and graphed versus their Mg/Ca ratio, t he trends within species seem to be ro bust, pointing to selective loss of magnesium as the foraminifer is dying or dead (Figure 28). This magnesium loss may be due to a decrease in organic material. F IGURE 28: Foraminiferal Mg/Ca ratio vs. state of f oraminifer at sites 7.5 and 15m from the vent Mg/Ca r atios are count rate ratios generated by EDS : >50% loss of surface layer was graded "dead," 20 50% loss of surface layer was graded "dying," <20% loss of surface layer was graded "alive." Percent loss was estimated visually from full foram inifer SEM images. Error bars are plotted one standard deviation above and below the average values. Missing error bars or missing columns indicate that 1 or 0 individuals fell under that category, respectively. For counts, averages, and standard deviat ions see Table A1

PAGE 70

59 When Mg/Ca ratios were plotted for all species the different shell mineralogies b e c a m e apparent (Figure 29 ). The three Amphistegina species clustered together as the only three representatives of foraminifers that use lo w Mg calcite to build their shell s. A mphisorus hemprichii C. defrancii C. gaudichaudii and H. depressa also clustered as they all precipitate hig h Mg calcite to build their shells The high Mg calcite group had a Mg/Ca ratio approximately double that of the low Mg c alcite group. The Halimeda tuna starting material had a Mg/Ca ratio approximately double that of the high Mg calcite group, but for all of the samples that were deployed, the H. tuna clustered with the low Mg calcite group. It would have been beneficial to have another organism that precipitates aragonite as a comparison for the aragonitic H. tuna FIGURE 29: Species Mg/Ca ratios vs. distance from the v ent. Mg/Ca r atios are count rate ratios generated by EDS. A completely decalcified segment of H. tu na from 7.5m was excluded during averaging as the ratio was too high to display on the graph. The Mg/Ca ratio values reported in this study were generated by a method (EDS) not commonly used. The method of EDS generates Mg/Ca ratios in count rate ratios as

PAGE 71

60 compared to the mmol/mol rat ios generated by more commonly used methods (e.g., ICP MS, AES MS electron microprobe) (Table 11 ). As such, the count rate data were converted to mCount/Count in order to directly compare the ratios. After this transform ation, the values from this study appeared to fall within the range of previously reported values for some of the species used in this study and related species Unfortunately, at this stage many of the larger benthic foraminifers do not have any report ed Mg/Ca values and of the seven studies done on these larger benthic foraminifers, including this study, six different techniques were used to determine Mg/Ca ratio s The values reported by Raja et al. (2005) and Raja and Saraswati (2007) are considerabl y higher than those reported by other studies for the high Mg calcite taxa. The fact that the data from this study appear to closely correspond to other previously reported values is reassuring, but a study needs to be conducted that compares several spec ies Mg/Ca ratios using a variety of techniques as a means of both calibrating previously reported data and recommending a methodology that will be the standard means of determining and reporting Mg/Ca ratios.

PAGE 72

61 TABLE 11 : Previously published Mg/Ca ratios for the foraminiferal species examined in this study and related species a Measured by EDS, average value from the Starting Material b Measured by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP AES) c Measured by Inductively Coupled Plasma Mass Spectrometry (ICP MS) d Measured by Titration after Acid Digestion e Measured by Electron Microprobe following 52 days of culturing f Measured by Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA ICP MS) The method of EDS is currently an underused exploratory technique which can be readily applied to calcifying organisms. Buster and Holmes (2006) used EDS in conjunction with chemical data from LA ICP MS in a study on brucite in the coral Montastraea faveolata Their study used EDS as a means of mapping out the surface and pinpointing areas of interest for further exploration with the LA ICP MS. More recently, Gooday et al. (2007) used EDS on new organic walled foraminifers from the Challenger Deep. Their study used EDS only as a mea ns of qualitatively determining the presence or absence of elements of interest and did not attempt to do any quantification.

PAGE 73

62 In general, specimens from the site 7.5m from the vent exhibited dissolution for all species tested. Specimens at the site 15m f rom the vent, however, were much less affected. This bodes well for Foraminifera as the 15m site had fairly extreme conditions (39.9 o C, 5.9pH). In the future, as the ocean temperatures increase and the pH decreases, Foraminifera appear to be well adapted to survive, although their shells may not be preserved in the sedimentary record. This is further supported by the fact that the sediment was predominantly siliciclastic to a distance of approximately 150 200m from the vent Beyond that, carbonate grain s occur red as an increasingly large percentage, until they made up the bulk of the sediment approximately 250 300m from the vent (McCloskey, 2009). The foraminifers used in this study were able to survive in areas where there was no visible carbonate in t he sediments. There is the potential for these to become "Lazarus taxa" in the future whereby they disappear from the fossil record, but are not extinct. They will then reappear in the record when the ocean's temperature and pH return to values that won' t dissolve foraminifer shells.

PAGE 74

63 CONCLUSIONS 1. F our out of seven larger benthic foraminiferal species examined exhibited some survival when exposed to elevated temperatures of 59.3 o C and low pH of 6.2 for five days, while all seven species exhibited som e survival when exposed to 39.9 o C and pH 5.9. 2. Amphisorus hemprichii underwent the most dissolution of all species tested. 3. Elevated temperature appeared to be a more important source of mortality than low pH over the range and time examined. 4. Live s pecimens seemed to be able to resist dissolution for a time. Following mortality, dissolution appeared to occur rapidly. 5. Dissolution occurred most readily in the porcelaneous species. In the hyaline species, the high Mg calcite species underwent disso lution more readily than the low Mg calcite species. 6. Many s pecimens exhibiting dissolution also exhibited preferential loss of magnesium relative to calcium as indicated by EDS. 7. More research should be done in to determine Mg/Ca ratios for larger ben thic foraminifers particularly the high Mg calcite taxa

PAGE 75

64 REFERENCES Berkeley, A., Perry, C.T., and Smithers, S.G. 2009. Taphonomic signatures and patterns of test degredation on tropical, intertidal benthic foraminifera. Marine Micropaleontology 73: 148 163. BouDagher Fadel, M. 2008 Evolution and Geological Significance of Larger Benthic Foraminifera Developments in Palaeontology & Stratigraphy, 21. Elsevier, Amsterdam, The Netherlands. Buster, N.A., and Holmes, C.W. 2006. Magnesium co ntent within the skeletal architecture of the coral Montastraea faveolata : locations of brucite precipitation and implications to fine scale data fluctuations. Coral Reefs 25: 243 253. Caldeira, K., and Wickett, M.E. 2005. Ocean model predictions of chemistry changes from carbon dioxide emissions to the atmosphere and ocean. Journal of Geophysical Research 110. C09S04, doi: 10.1029/2004JC002671. Corliss, B.H., and Honjo, S. 1981. Dissolution of deep sea benthonic Foraminifera. Micropaleontology 24 (4): 356 378. Cottey, T.L., and Hallock, P. 1988. Test surface degradation in Archaias angulatus Journal of Foraminiferal Research 18 (3): 187 202. Gang st R., Gehlen, M., Schneider, B., Bopp, L., Aumont, O., and Joos, F. 2008. Modeling th e marine aragonite cycle: changes under rising carbon dioxide and its role in shallow water CaCO 3 dissolution. Biogeosciences 5: 1057 1072. Gooday, A. 2002. Organic walled Allogromiids: aspects of their occurrence, diversity and ecology in marine hab itats. Journal of Foraminiferal Research 32: 384 399. Gooday, A.J., Todo, Y., Uematsu, K., and Kitazato, H. 2007. New organic walled Foraminifera (Protista) from the ocean's deepest point, the Challenger Deep (western Pacific Ocean). Zoological Jour nal of the Linnean Society 153: 399 423. Hall Spencer, J.M., Rodolfo Metalpa, R., Martin, S., Ransome, E., Fine, M., Turner, S.M., Rowley, S.J., Tedesco, D., and Buia, M. 2008. Volcanic carbon dioxide vents show ecosystem effects of ocean acidificatio n. Nature 454: 96 99.

PAGE 76

65 Hardie, L.A. 1996. Secular variation in seawater chemistry: an explanation for the coupled secular variation in the mineralogies of marine limestones and potash evaporites over the past 600 m.y. Geology 24 (3): 279 283. Haug an, P.M., and Drange, H. 1996. Effects of CO 2 on the ocean environment. Energy Conversion and Management 37 (6 8): 1019 1022. Hillis, L.W. 2001. The calcareous reef alga Halimeda (Chlorophyta, Byropsidales): a cretaceous genus that diversified in t he cenozoic. Palaeography, Palaeoclimatology, Palaeoecology 166: 89 100. Karlen, D.J., Price, R.E., Pichler, T., and Garey, J.R. 2010. Changes in benthic macrofauna associated with a shallow water hydrothermal vent gradient in Papua New Guinea. Paci fic Science 64 (3): 391 404. Kleypas, J.A, Feely R.A., Fabry, V.J., Langdon, C. Sabine C.L. and Robbins L.L. 2006. Impacts of Ocean Acidification on Coral Reefs and Other Marine Calcifiers. A Guide for Future Research, Report of a workshop spons ored by NSF, NOAA and the US Geological Survey. 88pp. Kotler, E., Martin, R.E., and Liddell, W. D. 1992. Experimental analysis of abrasion and dissolution resistance of modern reef dwelling foraminifera: implications for the preservation of biogenic car bonate. Palaios 7 (3): 244 276. Langer, M. 2008. Assessing the contribution of Foraminiferan protists to global ocean carbonate production. Journal of Eukaryotic Microbiology 55 (3): 163 169. Loubere, P. and Fariduddin, M. 1999. Benthic Foramin ifera and the flux of organic carbon to the seabed. In: Sen Gupta, B. (ed.), Modern Foraminifera Kluwer Press, Dordrecht. 181 199. McCloskey, B.J. 2009. Foraminiferal responses to arsenic in a shallow water hydrothermal vent system in Papua New Guine a and in the laboratory. Dissertation. University of South Florida. Peebles, M.W., and Lewis, R.D. 1991. Surface textures of benthic foraminifera from San Salvador, Bahamas. Journal of Foraminiferal Research 21 (4): 285 292. Pichler, T., and Dix, G.R. 1996. Hydrothermal venting within a coral reef ecosystem, Ambitle Island, Papua New Guinea. Geology 24 (5): 435 438. Pichler, T., Amend, J.P., Garey, J., Hallock, P., Hsia, H.P., Karlen, D.J., Meyer Dombard, D.R., McCloskey, B.J., and Price, R.E. 2006. A natural laboratory to study arsenic geobiocomplexity. Eos, Transactions American Geophysical Union 87 (23): 221 225.

PAGE 77

66 Price, R.E., and Pichler, T. 2005. Distribution, speciation and bioavailability of arsenic in a shallow water submarine hyd rothermal vent system, Tutum bay, Ambitle Island, PNG. Chemical Geology 224: 122 135. Price, R.E. 2008. Biogeochemical cycling of arsenic in the marine shallow water hydrothermal system of Tutum Bay, Ambitle Island, Papua New Guinea. Dissertation. University of South Florida. Raitzsch, M., D ue as J., de Nooijer, L.J., and Bickert, T. 2010. Incorporation of Mg and Sr in calcite of cultured benthic foraminifera: impact of calcium concentration and associated calcite satu ration state. Biogeosciences 7: 869 881. Raja, R., and Saraswati, P.K. 2007. A field based study on variation in Mg/Ca and Sr/Ca in larger benthic foraminifera. Geochemistry Geophysics Geosystems Q10012, doi:10.1029/2006GC001478 Raja, R., Saraswa ti, P.K., Rogers, K., and Iwao, K. 2005. Magnesium and strontium compositions of recent symbiont bearing foraminifera. Marine Micropaleontology 58: 31 44. Ries, J.B., Cohen A.L., and McCorkle, D.C. 2009. Marine calcifiers exhibit mixed responses to CO2 induced ocean acidification. Geology 37 (12): 1131 1134. Rees, S.A., Opdyke, B.N., Wilson, P.A., and Henstock, T.J. 2007. Significance of Halimeda bioherms to the global carbonate budget based on a geological sediment budget for the Northern Gre at Barrier Reef, Australia. Coral Reefs 26: 177 188. Savin, S.M., and Douglas, R.G. 1973. Stable isotope and magnesium geochemistry of recent planktonic Foraminifera from the South Pacific. Geological Society of America Bulletin 84: 2327 2342. Se gev, E., and Erez, J. 2006. Effect of Mg/Ca ratio in seawater on shell composition in shallow benthic foraminifera. Geochemistry Geophysics Geosystems Q10012, doi:10.1029/2006GC001478 Sen Gupta, B. 1999. Modern Foraminifera Kluwer Press Dordrech t The Netherlands. Tarasov, V.G., Gebruk, A.V., Mironov, A.N., and Moskalev, L.I. 2005. Deep sea and shallow water hydrothermal vent communities: two different phenomena? Chemical Geology 224: 5 39. Todo, Y., Kitazato, H., Hashimoto, J. and Gooday, A.J. 2005. Simple Foraminifera flourish at the ocean's deepest point. Science 307: 689.

PAGE 78

67 Toler, S.K., Hallock, P., and Schijf, J. 2001. Mg/Ca ratios in stressed foraminifera, Amphistegina gibbosa from the Florida Keys. Marine Micropaleontology 43: 199 206. Turner, J.V., Anderson, T.F., Sandberg, P.A., and Goldstein, S.J. 1986. Isotopic, chemical and textural relations during the experimental alteration of biogenic high magnesian calcite. Geochimica et Cosmochimica Acta 50: 495 506. Vicker man, K. 1992. The diversity and ecological significance of Protozoa. Biodiversity and Conservation 1: 334 341 Zar, J.H. 1998. Biostatistical Analysis Prentice Hall, New Jersey. 4th Ed. Zeppilli, D., and Danovaro, R. 2009. Meiofunal diversity and assemblage structure in a shallow water hydrothermal vent in the Pacific Ocean. Aquatic Biology 5: 75 84.

PAGE 79

68 APPENDICES

PAGE 80

69 TABLE A 1 : Foraminiferal counts, Mg/Ca averages, and Mg/Ca standard deviations for foraminifers separated into three groups: dea d, dying, or alive. DEAD DYING ALIVE Species Count Average St andard Deviation Count Average St andard Deviation Count Average St andard Deviation A. hemprichii 10 0.1043 0.00638 9 0.1227 0.01117 26 0.1279 0.00363 A. lessonii 2 0.0306 0.00154 2 0.0457 0.0 0411 6 0.0522 0.00858 A. lobifera 5 0.0388 0.01013 1 0.0494 N/A 1 0.0599 N/A A. radiata 5 0.0383 0.00657 2 0.0469 0.00453 5 0.0468 0.01028 C. defrancii 2 0.1316 0.01704 4 0.1121 0.01915 10 0.1186 0.02150 C. gaudichaudii 6 0.1097 0.01031 1 0.1152 N/A 14 0.1147 0.00829 H. depressa 2 0.1206 0.00692 0 N/A N/A 2 0.1208 0.01361 Those categorized as "dead" exhibited a >50% loss of surface layer "dying" exhibited a 20 50% loss of surface layer and "alive" exhibited <20% loss of surface layer Percent loss was estimated visually from full foraminifer SEM images. Includes only foraminifers from the two sites closest to the vent (7.5m and 15m).

PAGE 81

70 TABLE A2: EDS data for all A. hemprichii specimens. Includes the number of the bag of deployment, and the iden tifying code used for each foraminifer. Foraminifers were only separated into dead, dying, and alive at the first two sites from the vent. TABLE A3: EDS data for all A. lessonii specimens. Includes the number of the bag of deployment, and the identifyi ng code used for each foraminifer. Foraminifers were only separated into dead, dying, and alive at the first two sites from the vent.

PAGE 82

71 TABLE A4: EDS data for all A. lobifera specimens. Includes the number of the bag of deployment, and the identifying code used for each foraminifer. Foraminifers were only separated into dead, dying, and alive at the first two sites from the vent. TABLE A5: EDS data for all A. radiata specimens. Includes the number of the bag of deployment, and the identifying code u sed for each foraminifer. Foraminifers were only separated into dead, dying, and alive at the first two sites from the vent. TABLE A6: EDS data for all C. defrancii specimens. Includes the number of the bag of deployment, and the identifying code used for each foraminifer. Foraminifers were only separated into dead, dying, and alive at the first two sites from the vent.

PAGE 83

72 TABLE A7: EDS data for all C. gaudichaudii specimens. Includes the number of the bag of deployment, and the identifying code use d for each foraminifer. Foraminifers were only separated into dead, dying, and alive at the first two sites from the vent. TABLE A8: EDS data for all H. depressa specimens. Includes the number of the bag of deployment, and the identifying code used fo r each foraminifer. Foraminifers were only separated into dead, dying, and alive at the first two sites from the vent. TABLE A9: EDS data for all H. tuna specimens. Includes the number of the bag of deployment, and the identifying code used for each E DS scan.

PAGE 84

ABOUT THE AUTHOR Brienne Engel grew up outside Albany, New York in the town of East Greenbush. She was valedictorian of her graduating class from Columbia High School. For her undergraduate education, she attended Brandeis University on a ful l tuition Justice Louis Brandeis scholarship. While at Brandeis she double majored in Biochemistry (B.S.) and Biology (B.S.) and double minored in Chemistry and Art History. From her sophomore to senior years, she conducted research in the biochemistry d epartment in the Oprian Lab where her projects included the expression of Taxadiene Synthase in mammalian cells and the crystallization of rhodopsin. She spent the spring semester of her junior year abroad in the School for International Training's progra m, Australia: Natural and Cultural Ecology. Upon graduating, she moved to St. Petersburg, Florida to attend the University of South Florida, College of Marine Science. In August, 2010 Brienne will begin a Ph.D. program in Cancer Biology at the Moffitt Ca ncer Center on the University of South Florida's Tampa Campus.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004683
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Engel, Brienne.
0 245
Effects of a shallow-water hydrothermal vent gradient on benthic calcifiers, tutum bay, ambitle island, papua new guinea
h [electronic resource] /
by Brienne Engel.
260
[Tampa, Fla] :
b University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
502
Thesis (MS)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: Ocean acidification is occurring in response to rapidly increasing concentrations of atmospheric CO2. Shallow-water hydrothermal vent systems have been proposed as natural laboratories for studying the effects of elevated pCO2 on benthic communities. Hydrothermal vents occur at depths of approximately 10m in Tutum Bay, Ambitle Island, Papua New Guinea; these vents are surrounded by a typical-appearing fringing coral-reef community. Groups of live specimens of seven species of reef-dwelling, larger benthic foraminifers, along with segments of calcareous green algae broken from live thalli, were collected from a reef location, placed in small mesh bags, and deployed for five days at six different sites along a gradient of temperature (29.6oC-59.3oC) and pH (5.9-8.1) with distance from a large hydrothermal vent in Tutum Bay. Foraminiferal taxa used in the experiment included Amphisorus hemprichii, a species with Mg-calcite porcelaneous shells, three species of Amphistegina that produce hyaline calcite shells, and three species with hyaline Mg-calcite shells (Heterostegina depressa and two Calcarina spp.). Several specimens of four of the seven foraminiferal species examined survived exposure to elevated temperatures of 59.3oC and low pH of 6.2 for five days, while at least one specimen of each of the seven species survived exposure to 39.9oC and pH 5.9. Examination of shells at 600-1000x magnification using scanning electron microscopy revealed fine-scale dissolution in specimens up to 30m from the vent. Results of this experiment, as well as previously reported observations from the study site, indicate that the calcifying reef-dwelling organisms examined can survive pH extremes that result in dissolution of their shells following death.
590
Advisor: Pamela Hallock Muller, Ph.D.
653
Ocean acidification
Taphonomy
Dissolution
Foraminifera
Halimeda
690
Dissertations, Academic
z USF
x Cell, Micro, and Molecular Bio
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.4683