USF Libraries
USF Digital Collections

Assessment of image analysis as a measure of Scleractinian coral growth

MISSING IMAGE

Material Information

Title:
Assessment of image analysis as a measure of Scleractinian coral growth
Physical Description:
Book
Language:
English
Creator:
Gustafson, Steven K
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Back reef
Belize
Coral growth
Coral reefs
Image analysis
Massive scleractinian
Modeling
Patch reefs
Dissertations, Academic -- Marine Science -- Masters -- USF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Image analysis was used to measure basal areas of selected colonies of Montastraea annularis and Porites astreoides, following the colonies over a three-year period from 2002 to 2004. Existing digital images of permanently-marked quadrats in the Caye Caulker Marine Reserve, Belize, were selected based on image quality and availability of images of selected quadrats for all three years. Annual growth rates were calculated from the basal-area measurements. Mean growth rates (radial skeletal extension) for M. annularis and P. astreoides were 0.02 cm yr-1 and -0.20 cm yr-1, respectively. Basal area measurements demonstrated a large degree of variability. Increases were approximately balanced by declines giving the impression of stasis. By removing negative values and correcting by 25% to allow for comparison with vertical growth rates, mean values increased to ~0.5 cm yr-1 for M. annularis and ~0.8 cm yr-1 for P. astreoides.Basal area as a growth measure was compared to methods used in earlier studies. A new growth index based on basal area and perimeter was proposed and modeled. This growth index can be useful for reporting growth measured from basal areas and comparable other methods. The index also measures negative growth, or mortality, which conventional methods cannot do.
Thesis:
Thesis (M.A.)--University of South Florida, 2006.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Steven K. Gustafson.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 53 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001788789
oclc - 133467481
usfldc doi - E14-SFE0001449
usfldc handle - e14.1449
System ID:
SFS0025768:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001788789
003 fts
005 20070523131500.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 070523s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001449
040
FHM
c FHM
035
(OCoLC)133467481
049
FHMM
090
GC11.2 (ONLINE)
1 100
Gustafson, Steven K.
0 245
Assessment of image analysis as a measure of Scleractinian coral growth
h [electronic resource] /
by Steven K. Gustafson.
260
[Tampa, Fla] :
b University of South Florida,
2006.
3 520
ABSTRACT: Image analysis was used to measure basal areas of selected colonies of Montastraea annularis and Porites astreoides, following the colonies over a three-year period from 2002 to 2004. Existing digital images of permanently-marked quadrats in the Caye Caulker Marine Reserve, Belize, were selected based on image quality and availability of images of selected quadrats for all three years. Annual growth rates were calculated from the basal-area measurements. Mean growth rates (radial skeletal extension) for M. annularis and P. astreoides were 0.02 cm yr-1 and -0.20 cm yr-1, respectively. Basal area measurements demonstrated a large degree of variability. Increases were approximately balanced by declines giving the impression of stasis. By removing negative values and correcting by 25% to allow for comparison with vertical growth rates, mean values increased to ~0.5 cm yr-1 for M. annularis and ~0.8 cm yr-1 for P. astreoides.Basal area as a growth measure was compared to methods used in earlier studies. A new growth index based on basal area and perimeter was proposed and modeled. This growth index can be useful for reporting growth measured from basal areas and comparable other methods. The index also measures negative growth, or mortality, which conventional methods cannot do.
502
Thesis (M.A.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 53 pages.
590
Adviser: Pamela Hallock Muller, Ph. D.
653
Back reef.
Belize.
Coral growth.
Coral reefs.
Image analysis.
Massive scleractinian.
Modeling.
Patch reefs.
690
Dissertations, Academic
z USF
x Marine Science
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1449



PAGE 1

Assessment of Image Analysis as a M easure of Scleractinian Coral Growth by Steven K. Gustafson A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: Pamela Hallock Muller, Ph.D. Gary T. Mitchum, Ph.D. Edward W. Burkett, Ph.D. Date of Approval March 29, 2006 Keywords: back reef, Belize, coral growth, coral reefs, image analysis, massive scleractinian, modeling, patch reefs Copyright 2006, Steven K. Gustafson

PAGE 2

ACKNOWLEDGEMENTS I would like to thank my major professo r, Dr. Pamela Hallock Muller, for her patience, her help, and for her high standards in science and writing. I would also like to acknowledge my committee members, Dr. Edward Burkett and Dr. Gary Mitchum for their thoughtful sugge stions and guidance. Special thanks also to Dr. Mary Balcer for getting me started dow n this path; to Dr. Edward Burkett, again, for introducing me to coral reefs, and for the opportunity to do this study; to Dr. Melanie McField for her friendship, her hospitality a nd shining example; and to Dr. Edward Van Vleet for his help in getting me restarted after a long hiatus. Last, but certainly not least, I would like to thank my family for putting up with and helping me through this process.

PAGE 3

i TABLE OF CONTENTS LIST OF TABLES.............................................................................................................iii LIST OF FIGURES...........................................................................................................iv ABSTRACT....................................................................................................................... vi 1. INTRODUCTION.........................................................................................................1 Introduction to tropical coral reefs...........................................................................1 Reefs of Belize.........................................................................................................2 Important factors affecting coral growth.................................................................4 Coral growth rate as environmental indicator..........................................................8 Coral growth rate measurement...............................................................................9 Objectives..............................................................................................................11 Data Source............................................................................................................12 2. METHODS..................................................................................................................15 Image Collection....................................................................................................15 Image Selection......................................................................................................16 Size Classification.....................................................................................20 Data Analysis.........................................................................................................20 Growth Index Model..............................................................................................22 Hypothesis testing..................................................................................................25 3. RESULTS..................................................................................................................... 27 Growth Rate...........................................................................................................27 Modeling................................................................................................................31

PAGE 4

ii 4. DISCUSSION...............................................................................................................33 Coral Growth Rate.................................................................................................33 Improvements to Methods and Recommendations for Further Research..............35 Growth index............................................................................................35 Image analysis...........................................................................................35 Image Quality............................................................................................37 Basal-area Variability...............................................................................39 5. CONCLUSIONS.........................................................................................................40 Coral Growth Rates................................................................................................40 Image analysis........................................................................................................41 LITERATURE CITED.....................................................................................................42 APPENDICES..................................................................................................................49 Appendix A. Combined mean measured basal area (cm2) for M. annularis and P. astreoides by site and year standard deviation..................50 Appendix B. Mean measured basal area (cm2) for M. annularis and P. astreoides by site and year standard deviation.............................50 Appendix C. Combined mean change in basal area (cm2 yr-1) for M. annularis and P. astreoides by site and time period standard deviation............................................................................51 Appendix D. Mean cha nge in basal area (cm2 yr-1) for M. annularis and P. astreoides by site and time period standard deviation.............51 Appendix E. Combined mean growth rate (cm yr-1 radial extension) for M. annularis and P. astreoides by site and time period standard deviation............................................................................52 Appendix F. Mean growth rate (cm yr-1 radial extension) rate for M. annularis and P. astreoides by site and time period standard deviation............................................................................52 Appendix G. Mean growth rate (cm yr-1 radial extension) for M. annularis and P. astreoides by size class and time period standard deviation..........................................................................................53

PAGE 5

iii LIST OF TABLES Table 1. Synopsis of published growth rates for Montastraea annularis and Porites astreoides Growth rates are averag e skeletal extension in cm yr-1.................................................................................................................10

PAGE 6

iv LIST OF FIGURES Figure 1. Map of Monitoring Si tes Referenced in the Study............................................13 Figure 2. CCRS diver mapping a quadrat.........................................................................16 Figure 3. CCRS quadrat map. Numbers a nd symbols indicate coverage type................16 Figure 4. Image of quadrat with reference frame and grid...............................................17 Figure 5. Partially processed image of a quadrat..............................................................18 Figure 6. Screen shot of area measurement application...................................................19 Figure 7. Basal-area shapes used in the growth-index models. “Shadow” lines added to show basic shapes used to construct the figures. One-unit growth was modeled by increasing the radii of circular shapes by one unit and calculating the area and perimeter accordingly...................................22 Figure 8. Construction of Figure 8B.................................................................................23 Figure 9. Construction of shape 8D. S1 and S2 are the parts of the overlapping circles. OL is the overlapping area Semicircle and Square are the basic shapes used the construct the figure.........................................................24 Figure 10. Growth rate distribution with normal curve superimposed.............................21 Figure 11. Mean growth rate (cm yr-1 radial skeletal increase) for M. annularis and P. astreoides colonies by time period SE. Data in appendix E............28 Figure 12. Mean growth rate (cm yr-1 radial skeletal increase) for M. annularis colonies by site and time sp an SE (there were no M. annularis colonies measured at site D). Data in appendix F..........................................29 Figure 13. Mean growth rate (cm yr-1 radial skeletal increase) for P. astreoides colonies by site and time sp an SE (there were no P. astreoides colonies measured at site G; only one P. astreoides colony was measured at site J so SE was not calculated). Data in appendix F.................29

PAGE 7

v Figure 14. Mean growth rate (cm yr-1 radial skeletal increase) for M. annularis colonies by size class and time peri od SE (Small=diameter < 5cm, Medium=diameter >= 5cm and < 10cm, Large=diameter >= 10cm). Data in appendix G..........................................................................................30 Figure 15. Mean growth rate (cm yr-1 radial skeletal increase) for P. astreoides colonies by size class and time peri od SE (Small=diameter < 5cm, Medium=diameter >= 5cm and < 10cm, Large=diameter >= 10cm). Data in appendix G..........................................................................................30 Figure 16. Ratio of basal area increase to perimeter for one-unit growth rate (Pearson r=1.00, p=0.00). A, B, C and D correspond to the shapes in Figure 22..........................................................................................................32 Figure 17. Ratio of growth-index-cacluate d basal area increase to actual basal area increase for one-unit growth rate (Pearson r=0.99, p=0.00). A, B, C and D correspond to the shapes in figure 22................................................32 Figure 18. Growth rate vs depth for growth axes. Multi-axis = average of minimum, intermediate and maximum axes. After Hubbard and Scaturo, 1985...................................................................................................36

PAGE 8

vi Assessment of Image Analysis as a M easure of Scleractinian Coral Growth Steven K. Gustafson ABSTRACT Image analysis was used to measure basal areas of selected colonies of Montastraea annularis and Porites astreoides following the colonies over a three-year period from 2002 to 2004. Existing digital im ages of permanently-marked quadrats in the Caye Caulker Marine Reserve, Belize, were selected based on image quality and availability of images of selected quadrats for all three years. A nnual growth rates were calculated from the basal-area measurements. Mean growth rates (radial skeletal extension) for M. annularis and P. astreoides were 0.02 cm yr-1 and -0.20 cm yr-1, respectively. Basal area measurements dem onstrated a large degree of variability. Increases were approximately balanced by decl ines giving the impression of stasis. By removing negative values and correcting by 25% to allow for comparison with vertical growth rates, mean values increased to ~0.5 cm yr-1 for M. annularis and ~0.8 cm yr-1 for P. astreoides. Basal area as a growth measure was compared to methods used in earlier studies. A new growth index based on basal area and perimeter was proposed and modeled. This growth index can be useful for reporting growth measured from basal areas and comparable other methods. The index also measures negative growth, or mortality, which conventional methods cannot do.

PAGE 9

1 1. INTRODUCTION Introduction to tropical coral reefs Tropical coral reefs are among the planet’s mo st biologically diverse ecosystems. The number of species living on coral reefs has b een estimated to be as high as 3 million, of which approximately ten percent have b een studied and described (Adey, 2000). High diversity makes coral reefs valuable as a biochemical resource. Tropical coral reefs are home to a diverse assemblage of sessile invertebrates such as corals, tunicates, bryozoans, and sponges. Being firmly attached to the substrate, these animals are unable to avoid environmental perturba tions, predators, or other stressors. Consequently, many engage in chemical wa rfare, using compounds synthesized by the host, by the endosymbionts, or sequestered from the host’s food. These compounds are used to deter predation, fight disease, pr event overgrowth by fouling and competing organisms, and to capture prey. Because of their unique structures and properties, these compounds are an important and, as yet, larg ely untapped source of natural products with enormous potential as pharmaceuticals, nut ritional supplements, enzymes, pesticides, cosmetics, and other novel commer cial products (Bruckner, 2002). Many coral reefs act as pr otective barriers to ocean waves, providing sheltered lagoons conducive to seagrass and mangrove communities, minimizing coastal erosion and providing nurseries for a multitude of organisms. In a recent World Resource Institute research report, th e value of shoreline protecti on provided by Caribbean reefs

PAGE 10

2 was estimated to be between $700 million and $2.2 billion yr-1 (Burke and Maidens, 2004). All coral reefs provide structure for the thousands of resident fish and invertebrate species, which, in turn, support local economie s through fisheries and tourism. A study of Hawaii’s coral reefs calculates their total economic value, combining the annual figures for tourism, amenities, fisheries, and biodiversity, to average $364 million yr-1 (Cesar et al., 2002). The average annual recreational value alone is $304 million (Cesar and van Beukering, 2004). The understanding of reef-building (hermatypic) coral growth is critical if we hope to pr otect and preserve the valuable resources that tropical coral reefs are. Reefs of Belize Charles Darwin called the Belize Barrier R eef the most remarkable reef in the West Indies (Darwin, 1846). Stretching some 250 kilomete rs along the Mesoamerican coast, it is the largest barrier reef in th e Western Hemisphere. Major studies of the geology and morphology of the Belizean reef s have been carried out (Stoddart, 1962, Stoddart, 1963; and others (cite d in McField, et al., 2001)) bu t studies of the community structure of this vast system are less comm on (McField et al., 2001). Even rarer are studies of reef communities on the numerous patch reefs in Belize’s shelf lagoon. The studies of these patch reefs that do exist are vi rtually all restricted to the southern lagoon (Muzik, 1982; Lasker and Coffroth, 1983; Ar onson et al., 1998; Aronson et al., 2002a). Indeed, an exhaustive search of the literat ure yielded only two st udies on northern shelf lagoon patch reefs in Belize (Mazzullo et al., 1992; Burkett et al., 2002). Community structures differ substantially fr om north to south. The southern patch

PAGE 11

3 reefs are in deeper, higher-energy water than their northern counterpa rts. The southern shelf lagoon reef communities we re historically dominated by Acropora palmat a Ellis and Solander 1786 in the hi gher energy zones, and by Acropora cervicornis Lamarck 1816 in more sheltered areas (Mazzullo et al ., 1992; Aronson et al., 1998). The northern shelf lagoon reefs were dominated by Montastraea annularis Ellis and Solander 1786 (Mazzullo et al., 1992; Burkett et al., 2002). There were, however, extensive stands of Acropora palmat a Lamarck 1816 and smaller stands of Acropora cervicornis Lamarck 1816 in the northern lagoon (Burkett et al., 2002; local residents, personal communication). The white-band epidemic of the late 1970s and 1980s killed most Acropora colonies throughout the Caribbean (Aronson a nd Precht, 2001b; Aronson et al., 2002a). Also, during 1983-84, a myster ious pathogen decimated Caribbean populations of Diadema antillarum Philippi 1845, a primary reef herb ivore (Carpenter, 1990; Lessios, 1995). With reduced grazing, blooms of brown algae dominated most of the shallow reefs. Local over-fishing and anthropoge nic nutrification intensified this trend (McClanahan and Muthiga, 1998). Aronson and Precht (2001a) reported that at Carrie Bow Caye, Belize, coral coverage on the fore reef declined dramatically since the 1980s while macroalgal cover increased from less than 5% to more than 60%. Aronson and Precht (2001a) proposed three causes for these dramatic shifts in Caribbean coral reef community structure. First, coral morality due to natural and anthropogenic phenomena has reduced live covera ge and increased available substrate for colonization by algae. Second, the mass mortality of Diadema antillarum in 1983 – 1984 and local over-fishing of parrotf ish and surgeonfish greatly re duced herbivory. Third, the

PAGE 12

4 abundance of available substrate and loss of herbivores has allowed filamentous algae and macroalgae to proliferate, preventing co ral recruitment (Arons on and Precht, 2001a). During the fall of 1995, an unprecedented mass bleaching event affected approximately 50% of Belizean scleractinian corals but with low mortality (McField, 1999). During the late summer and fall of 1998, the Belize Barrier Reef system suffered another mass bleaching. The latter event resulted in increased coral mortality in the fore reef community. On the back reef and on th e patch reefs of the shelf lagoon, some areas suffered 100 percent mortality (Aronson et al., 2000; Aronson et al., 2002b). In addition to the bleaching events, the reefs of Belize suffered further disturbance from three major hurricanes in a four-year period: Mitch (1998), Keith (2000) and Iris (2001). Mitch and Keith heavily damaged the shallow-water reefs in the northern shelf lagoon (Burkett et al., 2002; McField, M. D ., personal communication). Hurricane Iris had a much reduced effect as it battered the southern reefs. Important factors affecting coral growth Many factors affect coral growth. Arguabl y, the most important is the coral-algal symbiosis. Reef-building (hermatypic) corals have a symbiotic relationship with certain dinoflagellate algae that live w ithin the corals’ tissues. The coral-algal symbiosis is best adapted to clear, nutrient-poor water (Hallock et al., 1993; Wood, 1993). Under these conditions the unicellular symbionts, called zooxanthellae, are kept in a nitrogendeprived state. Without access to sufficient dissolved inorganic nitrogen (DIN), one of the key components of protein, the symbionts grow and reproduce very slowly. The coral host provides just enough DIN from its own wa ste products to maintain its symbionts’

PAGE 13

5 photosynthetic capabilities, and to allow the alga e to reproduce at a ra te that sustains a stable population size at a level that is most beneficial to the coral (Trench, 1987). As long as their photosynthetic systems remain intact, the symbionts continue producing photosynthate at rates dictated by the availabl e light. With limited DIN, the algae cannot use all of their photosynthate. The portion that would have gone to fuel growth and reproduction, beyond what the host allows, is se creted into the host’s cells where it used by the host coral for its energy needs. Most of the coral’s energy budget is made up of lipids from its symbionts (Falkowski et al., 1993 ). This is the reas on that zooxanthellate corals can do so well in highly oligotrophic wa ters. The occasional prey that come into contact with coral host’s tent acles supply sufficient protein for the corals to grow and reproduce. Factors that re duce the flow of lipids from the symbionts ultimately cause stress in the coral host. Stre ssed corals grow more slowly. Hermatypic corals are especially vulnerabl e to excess nutrients, particularly DIN (Koop et al., 2001). As DIN is added to the wa ters bathing the coral reef, several things occur which negatively affect the coral-algal symbiosis. The corals, being permeable to the seawater, cannot keep their algal symbiont s as deprived of nitrogen as they can in nutrient-poor water. The symbionts are able to take up nitrogen that has permeated the host cells from the now DIN-enriched environm ent (Muscatine et al., 1979; Domotor and D’Elia, 1984). With more DIN, the alg ae can make more protein for growth and reproduction. Energy required to make protei n comes from the photosynthate that would have been excreted by the algae if they were nutrient-deprived (Falkowski et al., 1993). Consequently, there is less photosynthate fo r the host. Also, with more nutrients available in the water column, free-livi ng phytoplankton populations increase. This

PAGE 14

6 decreases water clarity and hence available light for photosynthesis by the coral’s symbionts, further reducing the host’s access to energy supplies (Tomascik and Sander, 1987). Furthermore, as the coral’s symb iont population grows, the algae’s oxygen demands, when not photosynthesizing, reduce ox ygen available for the host. When the algae are producing photosynthate, they ar e also producing oxygen, which can reach toxic concentrations with elevated symbiont populati ons (Lesser and Shick, 1989). Symbiont population increases can cause st ress in the host from reduced photosynthate (energy) for respiration, reduced oxygen for respiration during darkness, and oxidative stress during the photo period. Over the last 100 years, human activity has resulted in environmental changes such as warming oceans, air and water pollution, excess dissolved nutrients, and increased ultraviolet radiation (Knowlton, 2001) These changes have been blamed for extensive disease and mortality in coral reef communities (Richardson et al., 1998). Coral reefs are uniquely vulnerable to these changes due to their close proximity to coastlines and the ocean surface. Warmi ng oceans are the result of global warming which has been attributed largel y to increasing carbon dioxide (CO2) levels in the atmosphere (the greenhouse eff ect). Increased atmospheric CO2 concentrations lead to increased concentrations in the oceans as well. This in turn acidifies the water slightly but sufficiently to dissolve scleractinian cora l skeletons at rates that may exceed coral calcification capacity, causing reefs to sh rink (Caldeira and Wickett, 2003; Hallock, 2005). Chronic stress weakens corals making them more susceptible to disease. In the past 40 years, many coral pathologies have b een identified. Black Band disease was one

PAGE 15

7 of the first to be identified (Rutzler a nd Santavy, 1983) and one of most widespread (Green and Bruckner, 2000). White Band dis ease was also one of the earliest to be identified (Gladfelter, 1982) and is curren tly the only coral disease known to cause major changes in the composition and structure of reefs (Green and Bruckner, 2000). A host of other diseases have been id entified since these initial few were described. White Pox, Yellow Blotch diseas e, Red-Band disease, Dark-Spots disease, Yellow Band disease; the list is long and growing (Bruckner, 2001; Gil-Agudelo and Garzn-Ferreira, 2001; Green and Bruckner, 2000). It is widely accepted that the effects of climate change are causing coral bleaching (U. S. State Department, 1999). Bleached co rals appear white, or “bleached,” because they have lost symbionts, the symbionts ha ve lost pigment, or both. Exposure to high light levels, increased ultraviolet radiati on, temperature or salinity extremes, high turbidity and sedimentation resulting in reduced light levels, and other factors have been shown to cause coral bleachi ng (Glynn, 1996; Kushmaro et al., 1996). If the bleaching is not too severe and the cond itions causing the bleaching do not persist, the bleached colonies can regain their resident symbi onts within several weeks to months (Glynn, 1996). Otherwise, the corals may eventual ly starve or succumb to the elevated temperatures. Seven major episodes of bleaching have o ccurred since 1979. These events have been primarily attributed to increased sea water temperatures associated with global climate change and El Nio/La Nia events, wi th a possible synergistic effect of elevated ultraviolet and visible light (Hoegh-Guldbe rg, 1999). In 1995, a mass bleaching event affected reefs, e.g., in Belize, that had no hi story of bleaching. In 1998, the most severe

PAGE 16

8 and extensive bleaching on record occurred, re sulting in mass mortality (Aronson, et al., 2000; McField, 1999). In a report presented to the U.S. Coral Reef Task Force in 1999, the U.S. State Department (2004) warned: “In 1998 coral reefs around the world appear to have suffered the most extensive and severe bleach ing and subsequent mortality in modern record. In the same year, tropical sea surface temperatures were the highest in modern record, topping off a 50-year trend for some tropical oceans. These events cannot be accounted for by localized stressors or natura l variability alone. The geographic extent, increasing frequency, and regional severity of mass bleaching events are likely a consequence of a steadily rising baselin e of marine temperatures, driven by anthropogenic global warming.” Scleractinian coral reefs have existed si nce the late Triassic period (Achituv and Dubinsky, 1990; Stanley and Fautin, 2001). Fo r some 200 million years coral reefs have survived the ravages of mass extinctions and climate change. Whether or not coral reefs will be able to survive the 21st century is an important and relevant question. In its 2000 report, the Global Coral Reef Monitoring Netw ork states that approximately 25 percent of coral reefs worldwide have been effectivel y lost and another 40 pe rcent could be lost by 2010 unless urgently needed action is taken (Wilkinson, 2001). Coral growth rate as environmental indicator Brown and Howard (1985) suggested that coral growth rate is a good individualbased parameter for measuring declining environmental quality on reefs (also see Buddemeier and Kinzie, 1976.). There is however, conflictin g evidence of how

PAGE 17

9 nutrification affects coral skeletal extens ion rates (Hudson, 1981; Co rtes and Risk, 1985; Brown et al., 1990; Rogers, 1990). A possibl e reconciliation is the "Janus effect" (Edinger, 1991, cited in Risk et al., 2001), whereby nutrient enhancement, up to a certain critical level, can increase coral growth rates. When this level is reached, nutrification becomes deleterious and growth rates dec line (Tomascik and Sander 1985, Risk et al., 1995). This increased growth in the presence of increased nutrients appears to be lowdensity skeletal extens ion (Risk et al., 2001). Coral growth rate measurement A majority of the publishe d scleractinian growth st udies used methods that required harvesting living cora l colonies or taking core samples from living coral colonies. These methods used density bands in X-radiographs of thin cross-sections of coral skeletons, alizarin-red dye markers, or both, to measure growth rates as annual skeletal extensions (Table 1). Using image analysis to compare basal areas offers a non-destructive method of calculating growth rates. Connell et al. ( 1997) used an “image-analysis” method to measure coral colony basal area, incorporat ing standard photography, tracing projected images and measuring the areas of the trac ings with an electronic planimeter. The method used in my study improves on the Connell etal. method by eliminating the processing of photographic film and the project ing of images for tr acing. Using digital photography also makes it possible to displa y the images immediately, saving time and resources by eliminating “wasted” shots, a nd since the images are already in digital format, it is not necessary to use a planimeter to measure areas. Furthermore, this method

PAGE 18

10 does not require harvesting, coring, dyeing or otherwise disturbing li ve coral colonies, nor does it require the use of X-ray equipment. Table 1. Synopsis of published growth rates for Montastraea annularis and Porites astreoides Growth rates are average skeletal extension in cm yr-1. Author Year Location Species Growth Technique Carricart-Gavinet & Merino 2001 Campeche Bank, Mexico M. annularis 0.87 X-ray Carricart-Gavinet et al. 1994 Campeche Bank, Mexico M. annularis 0.86 X-ray Dustan 1975 Dancing Lady Reef, Jamaica M. annularis 0.47 – 0.68 Aliz. Red Gladfelter et al. 1978 Buck I, V. I. M. annularis 0.76 Aliz. Red P. astreoides 0.31 Goreau & Macfarlane 1990 Discovery Bay, Jamaica M. annularis 0.62 Direct (nail) Highsmith et al. 1983 Carrie Bow Cay, Belize M. annularis 0.37 – 0.98 X-ray P. astreoides 0.29 – 0.69 Hubbard & Scaturo 1985 Cane Bay & Salt River, M. annularis 0.2 – 0.9 X-ray P. astreoides 0.19 – 0.31 Hudson et al. 1994 Biscayne Bay, Fl, US M. annularis 0.7 – 0.9 X-ray Logan & Tomascik 1991 Bermuda P. astreoides 0.2 X-ray Tomascik & Sander 1985 Barbados M. annularis 0.61 – 1.24 X-ray Van Veghel & Bosscher 1995 Leeward reef, Curacao, NA M. annularis 1.27 – 1.81 X-ray Reporting growth rates for scleractinian co lonies is somewhat problematic, given the large range in size. Area measurements by themselves do not give a growth rate. Change in basal area yields a ra te in areal units per time unit. However, this measure is biased toward the larger colonies A one percent change in a 1,000-cm2 colony will add ten cm2 in basal area while 100 percent change in a five-cm2 colony will add only five cm2 in basal area. Using percent change as a measure is biased toward the small

PAGE 19

11 colonies. A five-cm2 change in a five-cm2 colony is a 100-percent change while the same change in 1,000-cm2 colony is a 0.5-percent change. Ra dial skeletal extension is less affected by colony size but the assumption must be made that the colonies are more or less circular, which is not necessarily the case, especially for fragmented colonies. Proposed here is a growth index that woul d be useful for calculating growth rates that more accurately reflect the colony shap e and size, and are more comparable to those found in the literature. This index is calcu lated from the area and perimeter information obtained from image analysis and is based on the assumption that coral colonies grow by increasing their basal areas in all directions whereby a one-unit “radial” increase would add approximately one areal unit for each unit of its perimeter. Objectives The primary goals of this study are to: • Assess the use of image analysis to m easure coral growth using existing data. • Assess basal area as a measure of coral growth • Develop methods to compare basal area measurements to conventional radial measurements A secondary goal is to use the data from im ages analysis to address the following: • Did growth rates differ between years? • Did growth rates differ between species? • Did growth rates differ between sites? • Did growth rates differ with colony sizes?

PAGE 20

12 Data Source Data for my study were collected as pa rt of a collaborative effort between Caribbean Coral Reef Studies (CCRS) at th e University of Wisc onsin-Superior (UWS) and the Caye Caulker Marine Reserve (CCMR) and its supporting agencies in Belize, C.A. The CCMR was established 1999. This 9,670-acre reserve includes the Caribbean Sea surrounding the northern end of Caye Caulker and that portion of the Belize Barrier Reef system that lies to the ea st and southeast of the island between the Caye Chapel Channel and the North Channel (Fig. 1). CCRS is a long-term undergraduate resear ch program at UWS established in 1991 under the direction Dr. Edward Burkett. In January 2002, CCRS set up new monitoring sites in the CCMR. Ten sites were selected on back reef and lagoon patch reefs in the CCMR (Fig. 1) based on one or more of the following criteria: The reef community was represen tative of the general area. The site contained living coral, but dama ge from various sources (e.g., hurricanes, boat traffic, etc.) was evident. The site had the potential to be used by tourists. The site was located near sources of potential environmental impact.

PAGE 21

13 Figure 1. Map of Monitoring Sites Referenced in the Study Sites A, H and I, near the Caye Caulker and Caye Chapel channels, sites B, C, G and J, near the most developed areas on Caye Caulker, and sites F and E, at maximum distance from the developed areas on Caye Caulker, are typical of the M. annularis -

PAGE 22

14 dominated patch reefs in the CCMR. Site D, on the back reef, is typical of areas where large Acropora palmata stands were formerly common. Sites E and H are slightly outside the CCMR due to the lack of exact co ordinates for the rese rve boundaries at the time of site selection. On each site, a 50-meter transect was laid ou t with stations at two-meter intervals. These stations were permanently marked and labeled for year-t o-year location of sampling quadrats. Data collected by CCRS indicated post-d isturbance recruitmen t. These reefs appeared to be in an early successional stage (e.g., Grigg and Maragos, 1974; Grigg, 1983) as most of the scleractinia n colonies studied were 0 – 4 cm in radius (Burkett et al., 2002). The majority of M. annularis colonies measured by Burkett et al. (2002) were also in the 0 – 4 cm range. Gr owth rate studies indicate that M. annularis grows at a rate of approximately 0.4 – 1.2 cm yr-1, radially, depending on en vironmental conditions (Dustan, 1975; Gladfelter et al., 1978; Hudson et al., 1994; a nd others). Growing at 1.2 cm yr-1, these colonies would have been appr oximately 3 years old when measured, indicating that they recruited after Hurrica ne Mitch and the 1998 bleaching event, but before Hurricane Iris (after Edmunds, 2000). Porites astreoides Lamarck 1816 grows radially at a rate of approximately 0.2 – 0.7 cm yr-1, dependent on environmental conditions (Gladfelter et al., 1978; Highsmith et al., 1983; Hubbard and Scaturo, 1985; Logan and Tomascik, 1991). Most P. astreoides colonies measured by Burkett et al. (2002) ha d radii in the 0-4 cm range. The implied 0.2 – 0.7 cm yr-1 radial increase indicates that it is likely that these colonies also recruited between Hurricanes Mitch and Iris.

PAGE 23

15 2. METHODS Image Collection On each site (Fig. 1), a 50-meter transect was laid out with stations at two-meter intervals. These stations were permanently marked and labeled for year-to-year location of sampling quadrats. Each January in 2002, 2003 and 2004, CCRS divers drew maps of each quadrat (Fig. 2). A 0.5-m2 (70.7cm x 70.7cm) reference frame, made from inch PVC pipe and strung with a heavy monofilame nt nylon reference grid, was placed on the substrate at each tag on the transects, taking ca re to align the grid with the axis of the transect. All life forms were drawn to scale, identified and recorded on Mylar data forms which were pre-printed with a grid matc hing that of the reference frame (Fig. 3). A short video sequence of each quadrat was recorded using a Canon Elura 10 digital video camera mounted in a Quest DH-3P Delfin Pro underwater housing. Where depth allowed, the camera view angle was held perp endicular to the quadr at at the minimum distance that allowed the entire reference frame to be included in the image. In shallower locations where it was not possible to include the entire reference frame in the image, the quadrats were videographed in sections. The video sequences recorded by CCRS were examined and the one best frame for each qua drat was captured as a JPEG image using Adobe Premier. For quadrats where the depth was too shallow for a single image, several partial images of the quadrat were captured.

PAGE 24

16 Figure 2. CCRS diver mapping a quadrat. Figure 3. CCRS quadrat map. Numbers and symbols indicate coverage type. Image Selection The images vary in quality. Only th ose with proper focus, lighting and orientation were selected for analysis. Furt hermore, only quadrats with images from all years were included in the study so that the fate of individual coral colonies could be tracked. No suitable images from site A were available. Among the images that were suitable for analysis, only M. annularis and P astreoides colonies appeared in numbers sufficient to yield meaningful informati on. Therefore, my study included only these species. A total of 162 quadrats (54 from each year) were selected for image analysis. The images of the quadrats were processed using Adobe Photoshop. For the shallow quadrats represented by multiple imag es, a single complete image was assembled by scaling and edge-matching the partial imag es. For each quadrat image, contrast and color were optimized for edge definition, and a measurement scale was determined by measuring the distance in pixels between the sides of the reference frame along the grid line that best represented the sc ale of the quadrat. In some cases the tension of the grid caused the sides of the frame to distort. In images where this had happened, it was

PAGE 25

17 necessary to draw lines from corner to corner along the sides that are perpendicular to the measurement axis in order to ob tain an accurate scale (Fig. 4). Figure 4. Image of quadrat with reference frame and grid A transparent layer was then added to the image, on which each colony under study was labeled and outlined, resulting in a monochrome polygon representing the area of the colony (Fig. 5). The quadrat maps co rresponding to selected images were used to aid species identification wh ere necessary. Any areas no t covered by the colony, but completely bounded by the colony (e.g., dead spot s, cover by other organism, etc.) were also outlined so that they were not included in the basal area calcula tion. This layer was then exported as a bitmap image for use in ca lculating basal area. Only colonies that were completely visible in images from all years were analyzed. A total of 915 colonies (305 from each year) were analyzed. Coral colony Quadrat ID tag “Straightened” sides Measurement axis

PAGE 26

18 Figure 5. Partially processed image of a quadrat. Basal areas of the colonies in the images were measured using a software application designed and developed by Burke tt and Gustafson (1995) and modified for this project (Fig. 6). The pixels that we re part of each polygon representing a colony were identified using a seed -fill, four-nearest-neighbor algorithm (Heckbert, 1990). A record of the year, site, quadrat, species, id entification number, area (in pixels) and scale for each colony was written to a file for further analysis. Basal area, in cm2, and growth rate, in cm yr-1, were calculated using the output files from the area-measurement application and Microsoft Excel. Basal area was calculated as 2imageScale quadArea ls colonyPixe where colonyPixels represents the number of pixels contained in a colony polygon, quadArea is the area of the quadrat in cm2, and imageScale is the distance between the sides of th e quadrat in pixels. The quadrat size Processed colony Unprocessed colony Outlining a colon y Dead s p ot Colon y label

PAGE 27

19 for this study is 0.5m2 or 5000cm2. Annual growth rate was calculated as n basalArea basalArean year year where basalAreayear is the basal area of a colony in cm2 for a particular year and n is the number of years for which the rate is calculated. The growth rate is essentially the change in radius of a colony’s basal area, assuming that the colony is approximately circular. Figure 6. Screen shot of area measurement application. Pixel scale Taxa list Processed colon y Un p rocessed colon y Scale measured from here: To here:

PAGE 28

20 Size Classification Individual colonies were a ssigned to one of three cla sses based on size. Edmunds et al. (1998) considered colonies less than approximately 5cm in diameter to be recruits. Using this as the threshold for the small size class, again, making the assumption that individual colonies are more or less circular, colonies less than 20cm2 in basal area ( basalArea diameter 2 ) were classified as small. Size classes medium and large were arbitrarily chosen to represent colonies where diameter was greater than or equal to 5cm and less than 10cm, and colonies where di ameter was greater than or equal to 10cm, respectively. Data Analysis To determine whether to use parametric or nonparametric statistical tests in the data analysis, the data were tested for meeti ng the assumptions of the parametric tests. Data assumptions for the parametric test between-subjects ANOVA, are a normal distribution and homogeneity of variance. The growth-rate data for the 2002-03, 2003-04 and 2002-04 time periods were tested fo r normality of distribution using the Kolmogorov-Smirnov test, and for homogeneity of variance using Levene’s test. The tests indicated that none of th e distributions for the three time periods were normal, but that the variance was homogenous.

PAGE 29

21 To determine whether the di stribution had a si gnificant effect on the analysis, one-way ANOVA, and Kruskal-Wallis tests we re run on growth rate vs. time period. The ANOVA reported no significant differences while the Kruskal-Wallis test did, demonstrating a distribution effect. Figure 7. Growth rate distributio n with normal curve superimposed. The growth-rate data included many negative and zero values (Fig. 7). In order to natural-log transform these data, the absolute value of the data set’s minimum value was added to each value in the data set. Thes e data were then natural-log transformed and tested again with Kruskal-Wallis and ANO VA as described above. The results for ANOVA did not show significant difference, but Kruskal-Wallis did, indicating nonparametric tests were necessary. The pr ocess for determining whether to use parametric or nonparametric tests was applied to the data set for grow th rate vs. species, 3.00 2.00 1.00 0.00 -1.00 -2.00 -3.00 -4.00 Growth Rate cm-yr 200 150 100 50 0 Frequency

PAGE 30

22 growth rate vs. site, and grow th rate vs. size class. A majority of these cases also indicated nonparametric tests. All statistical tests we re performed using SPSS v13.0. Growth Index Model A series of one-unit growth rates were modeled for four different shapes, each having a basal area of approximately 40 cm2 (Fig. 8). The shapes were constructed from simple geometric figures so that basal area increases due to oneunit, omni-directional growth could be readily calcu lated. One-unit growth was m odeled by increasing the radii of circular portions of each shape by one unit and calculating its area and perimeter accordingly. Figure 8. Basal-area shapes used in the growth-index models. “Shadow” lines added to show basic shapes used to construct the figures. One-unit growth was modeled by increasing the radii of circular shapes by one unit and calculat ing the area and perimeter accordingly. A B C D

PAGE 31

23 For shape A (Fig. 8A), a perfect circle, the area and perimeter were calculated as 2 ir andir 2, respectively, where ri where is radius for the model iteration. Shape B (Fig. 8B) was constructe d from four circles of radius ri and one square with side lengths of 2 r0 where r0 is radius for the first model iteration (Fig. 9). The figure area was calculated as 24 chord circleArea where circleArea is the area of one of the circles, and chord is the distance betw een intersections of adjacent circumferences. The figure perimeter was calculated as torArc sec 4where sectorArc is the length of and arc circumscribing a sector, with a c hord length of chord of one of the circles. Shape C (Fig. 8C) was constructed from two semicircles of radius ri and three squares with side lengths of 2 ri. The figure area was calculated as ir r circleArea 6 20 where circleArea is the area of the two semicircular portions combined. The figure perimeter was calculated as012 r circumf where circumf calculated asir 2, is the circumference of the figure’s two semicircles combined. Figure 9. Construction of Figure 8B. Secto r chord2 sectorArc chord

PAGE 32

24 Shape D (Fig. 8D) was constructed from eight semicircles of radius r two squares with side lengths of 2 r0, and two rectangles with lengths of r0 + ri and widths of 2 r0 (Fig. 10). The figure area was calculated as gle tan rec a overlapAre circleArea 3 4 where circleArea is the combined area of two of the figure’s semicircles, overlapArea is the area where two circles, each constructe d from two semicircles, overlap, and rectangle is the rectangular area between the semicircles, and was calculated as i ir r r 2 2 60 The figure perimeter was calculated as 04 6 4 r overlapArc circumf where circumf is the combined circumference of two semicircles, and overlapArc is the segment of the circumference of this hypothetical circle that overlaps an adjacent hypothetical circle. Figure 10. Construction of Figure 8D. S1 and S2 are the parts of the overlapping circles. overlapArc overlapArea S1 S1 S2 S2

PAGE 33

25 Growth Index The growth index was computed as n adjRadius adjRadiusn p p where adjRadius is a radius adjusted to reflect the area-to-perimeter relationship, p is the time period for which the index was calculated for a particular colony, and n is the number of years between time periods. AdjRadius was calculated as basalArea circwhere circ is the circularity of basalArea, the basal area of a coral colony. Circularity is a measure of how close the shape being measured is to a perfect circle. A perfect circle has a circularity value of 1.00 while values for non-circular shapes are less than 1.00, with the least circular having the lowest value. Circularity is calculated as 24perimeter basalArea where basalArea and perimeter are the basal area of coral colony and its perimeter, respectively. The growth index is essentially a “radius” adju sted to more accurately give a growth rate appropriate for the shape of the basal area. The correlation between area increase a nd perimeter was analyzed using the Pearson two-tailed correlation test. The correlation between measured basal-areaincrease and modeled basal-area-increas e was analyzed with the same test. Hypothesis testing The hypotheses tested are as follows: Basal area increase and perimeter are not correlated, H0: p = 0.0 Basal area increase and perimeter are correlated, Ha: p 0

PAGE 34

26 Modeled basal-area-increase and measured basal-area-increase are not correlated, H0: p = 0.0 Modeled basal-area-increase and measured basal-area-increase are correlated, Ha: p 0 Mean growth rates did not differ between years, H0: year1 = year 2 Mean growth rates differed among years, Ha: year 1 year 2 Mean growth rates did not differ among species, H0: species 1 = species 2 Mean growth rates differed among species, Ha: species 1 species 2 Mean growth rates did not differ among sites, H0: site 1 = site 2 Mean growth rates differed among sites, Ha: site 1 site 2 Mean growth rates did not differ among size classes, H0: size 1 = size 2 Mean growth rates differed among size classes, Ha: size 1 size 2

PAGE 35

27 3. RESULTS Growth Rate Mean growth rate for M. annularis ranged from -0.05 cm yr-1 for 2002-03 to 0.09 cm yr-1 for 2003-04 (Fig. 11). The overall rate (2002-04) was 0.02 cm yr-1. KruskalWallis tests showed significant differences in growth rates between time periods for M. annularis. The growth rate 2002-03 was less th an the rate for 2003-04 (p=0.002). The rate for 2003-04 was greater than the rate for 2002-04 (p=0.031). The rate for 2002-03 was less than the rate for 2002-04 (p=0.046). For P. astreoides growth rates ranged from -0.22 cm yr-1 for 2002-03 to -0.19 cm yr-1 for 2002-03. The overall rate (2002-04) was -0.20 cm yr-1. Kruskal-Wallis tests showed no significant differences in gr owth rates between time periods for P. astreoides (Fig. 11). Note that 2002-04 values are not averag es of the 2002-03 and 2003-04 values. The 2002-04 values are derived from direct comp arison of the 2002 and 2004 basal area values.

PAGE 36

28 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 M. annularis (N=273) P. astreoides (N=32)Mean Growth Rate (cm yr-1) 2002-03 2003-04 2002-04 Figure 11. Mean growth rate (cm yr-1 radial skeletal increase) for M. annularis and P. astreoides colonies by time period SE‡. Data in appendix E. Mean growth rates for M. annularis plotted by site and time span showed a wide range of values. Values for M. annularis were generally positive and less extreme (Fig. 12) than the values for P. astreoides, which were generally ne gative (Fig. 13). KruskalWallis tests showed significant differences in the rates between years. The rate for 200203 was greater than the rate 2003-04 for site B (-0.37 cm yr-1 vs. 0.36 cm yr-1, p=0.000), site F (-0.16 cm yr-1 vs. 0.14 cm yr-1, p=0.009) and G (-0.39 cm yr-1 vs. 0.41 cm yr-1, p=0.000), and less than the rate 2003-04 for site H (0.29 cm yr-1 vs. -0.06 cm yr-1, p=0.029). The 2003-04 rate was greater than the 2002-04 rate for sites B (0.36 cm yr-1 vs. 0.00 cm yr-1, p=0.000) and G (0.36 cm yr-1 vs. 0.01 cm yr-1, p=0.005). The 2002-04 rate was less than the 2002-04 rate for sites B (-0.37 cm yr-1 vs. 0.00 cm yr-1, p=0.000) and G (-0.39 cm yr-1 vs. 0.01 cm yr-1, p=0.013). ‡ Error bars represent standard error. Data from image analysis was statis tically analyz ed with nonparametric tests. It is possible that error-bar over lap can occur where there is a statistically significant difference.

PAGE 37

29 Montastraea annularis-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5B (N=25)C (N=55)D (N=0)E (N=12)F (N=61)G (N=37)H (N=37)I (N=24)J (N=22)Mean Growth Rate (cm yr-1) 2002-03 2003-04 2002-04 Figure 12. Mean growth rate (cm yr-1 radial skeletal increase) for M. annularis colonies by site and time span SE (There were no M. annularis colonies measured at site D). Data in appendix F. Figure. 13 shows growth rates for P. astreoides plotted by site and time span. There were no significant di fferences in growth rates between time periods. Porites astreoides-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5B (N=2)C (N=2)D (N=8)E (N=3)F (N=7)G (N=0)H (N=2)I (N=7)J (N=1)Mean Growth Rate (cm yr-1) 2002-03 2003-04 2002-04 Figure 13. Mean growth rate (cm yr-1 radial skeletal increase) for P. astreoides colonies by site and time span SE (There were no P. astreoides colonies measured at site G; only one P. astreoides colony was measured at site J so SE wa s not calculated). Data in appendix F.

PAGE 38

30 Montastraea annularis-1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 1.25N=71 N=72 N=130N=78 N=65 N=130N=71 N=67 N=135Small Medium LargeMean Growth Rate (cm yr-1) 2002-03 2003-04 2002-04 Figure 14. Mean growth rate (cm yr-1 radial skeletal increase) for M. annularis colonies by size class and time period SE (Small=diameter < 5 cm, Medium=diameter >= 5cm and < 10cm, Large=diameter >= 10cm). Data in appendix G. Porites astreoides-1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 1.25N=5 N=7 N=20N=2 N=9 N=21N=4 N=9 N=19Small Medium LargeMean Growth Rate (cm yr-1) 2002-03 2003-04 2002-04 Figure 15. Mean growth rate (cm yr-1 radial skeletal increase) for P. astreoides colonies by size class and time period SE (Small=diameter < 5 cm, Medium=diameter >= 5cm and < 10cm, Large=diameter >= 10cm). Data in appendix G.

PAGE 39

31 Kruskal-Wallis tests showed signifi cant differences in growth rates by size classes between time periods (Fig. 14, 15). For Small M. annularis colonies the rate for 2002-03 was less than the rate for 2003-04 (-0.05 cm yr-1 vs. 0.19 cm yr-1, p=0.001) and the rate for 2003-04 was greater than the rate for 2002-04 (0.19 cm yr-1 vs. -0.13 cm yr-1, p=0.000). For Medium M. annularis colonies the rate for 2002-03 was greater than the rate for 2003-04 (-0.02 cm yr-1 vs. 0.07 cm yr-1, p=0.043), the rate for 2002-03 was greater than the rate for 2002-04 (-0.02 cm yr-1 vs. -0.03 cm yr-1, p=0.382) and the rate for 2003-04 was greater than the rate for 2002-04 (0.07 cm yr-1 vs. -0.03 cm yr-1, p=0.022). For Large M. annularis colonies the rate for 2002-03 was less than the rate for 2003-04 (-0.07 cm yr-1 vs. 0.04 cm yr-1, p=0.357), the rate for 2002-03 was less than the rate for 2002-04 (-0.07 cm yr-1 vs. 0.12 cm yr-1, p=0.052) and the rate for 2003-04 was less than the rate for 2002-04 (0.04 cm yr-1 vs. 0.12 cm yr-1, p=0.271). Kruskal-Wallis tests showed no significant differences in growth rates for P. astreoides size classes between time periods. Modeling Pearson correlation coefficient for basalarea increase vs. perimeter for shapes A (circle), B (elongate capsule), C (four-lobe ) and D (elongate eight-lobe) were 1.00 (p=0.000), indicating a strong positive rela tionship (Fig. 16). Pearson correlation coefficient for modeled basal-area increase vs. measured basal-area increase was 0.986 (p=0.000) for shape A, 1.000 (p=0.000) for shapes B and C, and 0.990 (p=0.000) for shape D, indicating a strong positive re lationship here as well (Fig. 17).

PAGE 40

32 0.98 0.99 1.00 1.01 1.02 020406080100 Model IterationsArea Increase/Perimete r A B C D Figure 16. Ratio of basal area increase to peri meter for one-unit growth rate. A, B, C and D correspond to the shapes in Figure 7. 0.8 0.9 1.0 1.1 1.2 1.3 1.4 020406080100 Model IterationsGrowth Index/Actual A B C D Figure 17. Ratio of growth-index-ca cluated basal area increase to ac tual basal area increase for oneunit growth rate. A, B, C and D co rrespond to the shapes in Figure 7.

PAGE 41

33 4. DISCUSSION Coral Growth Rate It was clear during the analys is that images from 2003 were the most variable in quality which made it difficult to obtain an accurate scale in many cases. This is likely a key reason that the 2002-03 and 2003-04 time pe riods showed more variability in the values calculated from basal area than did the 2002-04 time period. Also, my data were collected for basal-area measurement and due to the limited number of suitable images available, random image selecti on was not possible. Therefore, my data represent only those selected images and should not be inte rpreted to represent the conditions in the field. Data for M. annularis from sites E, H and I share at tributes. The coral colonies measured at these sites had the highest overa ll mean growth rates and the majority of these colonies were in the Large size class (g reater than 10cm in diameter). All three sites are close to channels in the barrier reef. Site E is on the southern margin of the North Channel. Sites H and I are located just north of the Caye Chapel Channel and just south of the Caye Caulker Channel. Nutrient uptake, gas exchange, and feeding depend on the flow of water over and around the coral (Goldshmid et al., 2004). Hist orically, growth may have been enhanced by the channels’ tidal currents, delivering nutri ent necessary for growth in the form of

PAGE 42

34 plankton, and maintaining better water quality and stabilizing surface temperatures with daily tidal flushing. It is widely accepted that scleractinian mortality is inversely related to size (Edmunds and Gates, 2004). Perhaps the larger colonies at these sites were able to survive disturbances better than their smaller counterparts. However, the larger colonies assessed at these sites also could be an artifact of image selection. My data from sites C, F and I had the highest mean basal areas for P. astreoides. Data from Sites C, F and I had neither the highest overall mean change-in-basal-area rates nor the highest overall m ean growth rates. Data from Site H had the highest rates for these measures with those from sites D and E tied for second place. Data from these three sites also were unique in that they showed positive change in basal area and growth while mean basal-area data from other the other sites declined. Porites astreoides is an opportunistic species capable of withstandi ng higher nutrient and sediment loads than M. annularis (Tomascik and Sander, 1987; Martin, 1998). This could account for the some of the difference in the distribution of P. astreoides and M. annularis. Site C is closest to Caye Caulker Village and coul d be affected by nutrient runoff and sediment resuspended by boat traffic, favoring P. astreoides. Another possible explanat ion of the distribution is that there is less competi tion for space at Site C. The higher basal area change and growth rates for P. astreoides at sites E and H could be due to the phenomena that affect M. annularis at these sites. Again, differences must be interpreted with caution, as they may simply be artifacts of image selection. The published growth rates for M. annularis are approximately 1 cm yr-1, radially (Table 1). Published rates for P. astreoides are approximately half of that (Table 1). The

PAGE 43

35 growth rates measured in this study for these species were 0.02 cm yr-1 and -0.20 cm yr-1, respectively. Improvements to Methods and Recommendations for Further Research Growth index A subset of the images used in my study was reanalyzed to obtain perimeter data for each colony so that circularity and growth index could be calculated. When growth indices were computed for these colonies, it became apparent that a significant change in circularity for a particular colony over a ti me period affected the results to the point where the growth rates were unusable (see Buddemeier and Kinzie, 1976.). Significant changes in basal circularity are not uncommon in coral colonies. Disease and trauma can change the shape of colony dramatically. Ther efore, only growth rates of colonies that maintain approximately the same basal shap e between measurements should be used for comparison with growth rates from “traditional” studies. Image analysis Image analysis measures horizontal skeletal extension while the harvesting and coring methods generally measure skeletal extension along the axis of maximum extension. This discrepancy makes it difficult to compare results of other growth studies to this study. Hubbard and Scaturo (1985) used a multi-axis method measuring skeletal extension along the maximum (vertical at depths less than approximately 20 m), intermediate and minimum (horizontal) growth axes. They show growth rates plotted against depth with the rates for the minimumgrowth axis and the mean of the maximum-

PAGE 44

36 and minimum-growth axes falling below those of the single-axis (maximum growth) method at depths less than approximately 30 m (Fig. 18). This demonstrates that the horizontal-axis growth rates are considerably less that those for the vertical axis in this depth range. The difference between maximumand minimu m-axis growth rates shown is approximately 25 percent. Increasing th e mean overall growth rates from this study by 25 percent would make the rate for M. annularis 0.03 cm yr-1 and the rate for P. astreoides 0.05 cm yr-1, a little closer to agreement with published growth rates for these species but still quite low. Growth Rate vs. Depth for Growth Axes0.0 0.2 0.4 0.6 0.8 1.0 010203040 Depth (m)Growth Rate (cm yr-1) Multi-axis Maximum Minimum Figure 18. Growth rate vs. depth for growth axes. Multi-axis = average of minimum, intermediate and m aximum axes. A fter Hubbard and Scaturo, 1985. Measuring skeletal extension from co res and cross-sections cannot detect decreases in colony size, while image analys is can, as my study demonstrates. The distribution of growth-ra te-area data shows that more than half the mean values were less than zero (Fig. 7). In the context of compar ing growth rates with those in the literature,

PAGE 45

37 negative growth rates are meani ngless. If growth rates ar e calculated only where change in basal area is non-ne gative, the rate for M. annularis is 0.41 cm yr-1 and the rate for P. astreoides is 0.61 cm yr-1. Increasing these rates by 25 percent to simulate vertical growth would make the rates 0.51 cm yr-1 and 0.77 cm yr-1, respectively. These values are comparable to published growth rates (Table1). Image Quality Connell, at al. (1997) reported that they were able to identify objects greater than 0.5 cm2 in area. The smallest objects identifi able in images used in this study were approximately 2 cm2 in area. The images of the CCRS quadrats were of varying quality in terms of lighting, focus, framing and collim ation (perpendicularity to the focal plane). This being a dataset of opportunity, ther e was no control of these image-quality parameters. These images were originally in tended to be a photographic record of the quadrats of which hand-drawn, in-situ maps were created for the CCRS study. While useful for this purpose, many proved unsuitabl e for the kind of detailed image analysis needed for my study. For species-specific gr owth, measurements of individual colonies for each year were required, further restricti ng the pool of suitable images. Furthermore, the cost of digital photography equipment at the time the images were acquired prohibited the use of a sufficiently large pixel ma trix to capture the fine detail needed for accurate and precise analysis. High-resolution equipment is now more affordable. CCRS returned to Caye Caulker in January 2005 to continue the long-term study. The photographic equipment used on this expedition, a Canon PowerShot A95 with an Ikelite #6140.80 housing, was

PAGE 46

38 far superior to that used prev iously and of lower cost. With its five-mega-pixel resolution and automatic focus, the camera was able to capture the 0.5m2 quadrats, in ambient light, at an image size of 2,272 X 1,704 pixels, with the detail required for the identification of most scleractinians. A skilled photographer could capture close-up images with detail sufficient to identify virtually a ll visually identifiable species. The major difficulty with image analysis is the lack of depth inherent in any twodimensional representation of three-dimensional objects. Th e difference between actual basal area and apparent basal area can be signi ficant in images of groups of objects with the amount of relief that can be encountered ov er short distances on a coral reef. Another problem encountered in image analysis is coll imation error. Again, the actual basal area and the apparent basal area may be significan tly different if the sight axis is not perpendicular to the plane of reference for the image. The use of dual cameras mounted on a framework could be employed to produce stereo images that could be analyzed using ray tracing and triangulat ion to measure depth of fiel d. If the proposed framework had a leveling system, collimation error could also be corrected. Spring-loaded, telescoping leg extensions at th e corners of the frame base with lock/unlock controls at the top of the frame, in combination with a spirit-bubble level indicator, would allow for quick and precise photography in less than optimal conditions. Working in shallow water may prevent cap turing the entire quadrat (or other subject) in one image. An accurate and rela tively fast method of assembling a mosaic of image “tiles” would also be of significant benefit. Adding a height adjustment to the proposed framework so that the cameras could be raised and lowered to suit depth would allow “tiled” quadrat images to be matched easily and accurately

PAGE 47

39 Basal-area Variability Basal area appears to be a very dynamic para meter. My data show that there is substantial growth and mortality of massive scleractinians but the mean growth rates suggest very little change. High temporal and spatial variabil ity in physiological responses, including growth or lesion healing, may be a ch aracteristic of corals under stress, as reported by Fi sher et al. (in press).

PAGE 48

40 5. CONCLUSIONS Coral Growth Rates Mean growth rates (radial skeletal extension) for M. annularis and P. astreoides were 0.02 cm yr-1 and -0.20 cm yr-1, respectively. By removing negative values and correcting by 25% to allow for comparison with vertical growth rates, mean values increased to ~0.5 cm yr-1 for M. annularis and ~0.8 cm yr-1 for P. astreoides. Did species-specific growth rates differ between years? There were statistically-significant di fferences in mean growth rate for M. annularis. The rate for 2002-03 was greater than the rate for 2003-04 while the 2003-04 was less than the rate for 2002-04. The limited sample size did not reveal stat istically significant differences in mean growth rate for P. astreoides between years. Did species-specific growth rates differ between species? The overall mean growth rates for M. annularis and P. astreoides were not significantly different.

PAGE 49

41 Did species-specific growth rates differ between sites? There were statistically-significant differences for M. annularis between site E and sites C and J, between site H and Site C, and between site I and sites B, C, F and G. The limited sample size did not reveal statis tically differences in growth rates for P. astreoides among sites. Did species-specific growth rates differ with colony sizes? There were no statistically significant differences between size classes for M. annularis or P. astreoides. Thus, the image analysis methods detected si gnificant differences. However, the process of selection of images was not random, ther eby limiting the applicability of these results to the images analyzed. Results should not be used to interpret c onditions at sites from which the images were collected. Image analysis Image analysis is useful as a coral grow th measure. Its utility, however, depend on image quality. Proper resolution, focus, li ghting, collimation and measurement scale are critical for precise measurements. The proposed growth index yields growth rates more comparable to growth rates from “conventional” studies. Particular a ttention must be paid to changes in colony basal shapes between measurements. Substa ntial change in circularity can render the measurements meaningless. Also, growth indices for elongate-s haped colonies are less comparable to published rates

PAGE 50

42 LITERATURE CITED Adey, W. H. 2000. Coral Reef Ecosystems and Human Health: Bi odiversity Counts! Ecosystem health 6(4): 227-236 Achituv, Y. and Z. Dubinsky. 1990. Evol ution and zoogeography of coral reefs. In: Dubinsky, Z., Ed. Ecosystems of the Wo rld: Coral Reefs. Elsevier, New York, NY Aronson, R. B. and W. F. Precht. 2001a Applied Paleoecology and the Crisis on Caribbean Coral Reefs. Palaios 16: 195 – 196 ___________ and __________. 2001b. White-band disease and the changing face of Caribbean coral reefs. Hydrobiologia 460: 25-38 ___________, _____________ and I. G. Macintyre. 1998. Extrinsic control of species replacement on a Holocene reef in Belize: The role of coral disease. Coral Reefs 17: 223-230. ___________, _____________, ________________ and T. J. T. Murdoch. 2000. Coral bleach-out in Belize. Nature 405: 36 W. F. Precht, M. A. Toscano and K. H. Koltes. 2002a. The 1998 bleaching event and its aftermath on a coral reef in Belize. Marine Biology 141: 435-447 ___________, I. G. Macintyre, W. F. Precht, T. J. T. Murdoch, and C. M.Wapnick. 2002b. The expanding scale of species tur nover events on coral reefs in Belize. Ecological Monographs 72: 233-249 Brown, B. E., and L. S. Howard. 1985. Asse ssing the effects of "str ess" on reef corals. Advances in Marine Biology 22: 1-63 __________, M. D. Le Tissier, T. P. Scoffin, A. W. Tudhope. 1990. Evaluation of the environmental impact of dredging on intertid al coral reefs at Ko Phuket, Thailand, using ecological and physiol ogical parameters. Marine Ecology Progress Series 65: 273-281 Bruckner, A.W. 2001. Coral health and mortal ity: Recognizing signs of coral diseases and predators. In: Humann and Deloach (eds.), Reef Coral Identification. Jacksonville, FL: Florida Caribbean Baha mas New World Publications, Inc. pp. 240-271.

PAGE 51

43 Bruckner, A. W. 2002. Life -saving products from coral re efs. Issues in Science & Technology. 18(3): 39-44 Buddemeier, R. W., and R. A. Kinzie III. 1976. Coral Growth. In: Oceanography and Marine Biology Annual Review, Barnes, H. (ed.), Aberdeen University Press. pp 183-225 Burke, L. and J. Maidens. 2004. Research re port: Reefs at Risk in the Caribbean. World Resources Institute, Washington, DC. http://marine.wri.org/pubs_description.cfm?PubID=3944 Burkett, E. W. and S. K. Gustafson. 1995. Us e of video image analysis for assessment of coral reef community structur e. In: Proceedings of St. Mary’s University research symposium, Winona, MN: 1 10 ___________, L. Digman, J. Gasele, J. Lind, J. Lind, M. Poore, S. Putz, E. Slattery, and K. Stephenson. 2002. Caye Cauker Resear ch Initiative: Phas e I Report. Annual project report to Caye Caulker Marine Reserve, Belize, C. A. Caldeira, K. and M. E. Wickett. 2003. Anthropogenic carbon and ocean pH. Nature 425: 365 Carricart-Gavinet, J. P. and M. Merino. 2001. Growth responses of the reef-building coral Montastraea annularis along a gradient of con tinental influe nce in the southern Gulf of Mexico. Bulleti n of Marine Science 68(1): 133-146 _____________, G. Horta-Puga, M. A Ruiz-Zrate a nd E. Ruiz-Zrate. 1994. Tasas restrospectivos de crecimiento del coral hermatpico Montastrea annularis (Scleractinia: Faviidae) en arrecifes al sur del Golfo de Mxico. Revista de Biologia Tropical 42(3): 515-521 Carpenter, R. C. 1990. Mass mortality of Diadema antillarum I. Long-term effects on sea urchin population-dynamics and co ral reef algal communities. Marine Biology 4: 67-77 Cesar, H. S. J., P. J. H. van Beukering, W. Pintz, and J. Dierking. 2002. Economic valuation of the coral reefs of Hawai‘i. Hawai‘i Coral Reef Initiative, University of Hawai‘i, Honolulu. http://www.hawaii.edu/ssri/hcri/f iles/cesar_noaa_fina l_report_01-02.pdf Cesar, H. S. J., and P. J. H. van Beukering. 2004. Economic valuation of the coral reefs of Hawai'i. Pacific Science 58(2) 231-242 Connell, J. H., T. P. Hughes and C. C. Wallace. 1997. A 30-year study of coral abundance, recruitment, and disturbance at several scales in space and time. Ecological-Monographs 67(4): 461-488

PAGE 52

44 Cortes, J., M. J. Risk. 1985. A reef under si ltation stress: Cahuita, Costa Rica. Bulletin of Marine Science 36: 339-356 Darwin, C. 1846. Geological Observations on Coral Reefs, Volcanic Islands, and on South America: being the Geology of the Voyage of th e Beagle, under the Command of Capt. FitzRoy, during the Years 1832-36. Ward Lock & Co., London, Melbourne & Toronto, 1910 [fi rst published Smith, Elder & Co., London, 1846]. Domotor, S. L. and C. F. D’Elia. 1984. Nutrient uptake kinetics and growth of zooxanthellae maintained in laboratory culture. Marine Biology 80: 93-101 Dustan, P. 1975. Growth and Form in the Reef-Building Coral Montastraea annularis. Marine Biology 33: 101-107 Edmunds, P. J. 2000. Recruitment of sclera ctinians onto the skelet ons of corals killed by black band disease. Coral Reefs 19: 69-74 ____________ and R. D. Gates. 2004. Si ze-Dependent Differences in the Phytophysiology of the Reef Coral Porites astreoides. Biological Bulletin 206: 61-64 ____________, R. B. Aronson, D.W. Swanson, D. R. Levitan, and W. F. Precht. 1998. Photographic Versus Visual Census Techniques for the Qu antification of Juvenile Corals. Bulletin of Marine Science 62: 937-946 Falkowski, P. G., Z. Dubinski, L. Muscat ine and L. McCloskey. 1993. Population Control in Symbiotic Corals. BioScience 43(9): 606-611 Fisher, E M., J. E. Fauth, P. Hallock, C. M. W oodley. In press. Lesion regeneration rates in reef-building corals (Montastraea spp.) as indicators of colony condition: strengths and caveats. Mari ne Ecology Progress Series Gil-Agudelo, D.L. and J. Garzn-Ferreira. 2001. Spatial and seasonal variation of dark spots disease in coral communities of the Santa Marta area (Columbian Caribbean). Bulletin of Marine Science 69: 619-630 Gladfelter, W.B. 1982. Whiteband disease in Acropora palmata: Implications for the structure and growth of sh allow reefs. Bulletin of Marine Science 32: 639-643 Gladfelter, E. H., R. K. Mona han, and W. B. Gladfelter. 1978. Growth Rates of Five Reef-Building Corals in the Northeastern Caribbean. Bulletin of Marine Science 28: 728-734 Glynn P. 1996. Coral reef bleaching: facts, hypotheses and implications. Global Change Biology 2:495– 510

PAGE 53

45 Goldshmid, R., R. Holzman, W. Daniel a nd A. Genin. 2004. Aeration of corals by sleep-swimming fish. Limnol ogy and Oceanography 49(5): 1832-1839 Goreau, T. J. and A. H. MacFarlane 1990. Reduced growth rate of Montastrea annularis following the 1987 to 1988 coral-bleac hing event. Coral-Reefs 8(4): 211-216 Green, E. P., A. W. Bruckner. 2000. The si gnificance of coral di sease epizootiology for coral reef conservation. Biol ogical Conservati on 96(3): 347-361 Grigg, R.W. 1983. Community Structure Suc cession and Development of Coral Reefs in Hawaii, USA. Marine Ecol ogy Progress Series 11: 1-14 _________ and J. E.Maragos. 1974. Recolonization of Hermatypic Corals on Submerged Lava Flows in Hawaii. Ecology (Washington, D. C.) 55: 387-395 Hallock, P. 2005. Global change and m odern coral reefs: New opportunities to understand shallow-water carbonate depos itional processes. Sedimentary Geology 175: 19 -33 _________, F. E. Muller-Karger, J. C. Halas. 1993. Coral Reef Decline. Research & Exploration 9 (3): 358-378 Heckbert, P. 1990. A Seed Fill Algorithm. In: Graphics Gems. Andrew Glassner, (ed.), Academic Press, Boston, pp. 275-277, 721-722 Highsmith, R. C., Lueptow, R. L. and S. C. Schonberg. 1983. Growth and bioerosion of massive corals. Marine Ecol ogy Progress Series 13: 261-271 Hoegh-Guldberg, O. 1999. Climate change, co ral bleaching and the fu ture of the world's coral reefs. Marine-and-Freshwater-Research 50(8): 839-866 Hubbard, D. K., and S. Scaturo. 1985. Grow th Rates of Seven Species of Scleractinian corals from Cane Bay and Salt River, St. Croix, USVI. Bulletin of Marine Science 36(2): 323-338 Hudson, J. H. 1981. Growth rates in Montastrea annularis: a record of environmental change in the Florida Keys. Proceedings 4th International Coral Reef Symposium 2: 233-240 ___________, K. J. Hanson, R. B. Halley, and J. L. Kindiner. 1994. Environmental Implications of Growth Rate Changes in Montastrea annularis: Biscayne National Park, Florida. Bulleti n of Marine Science 54: 647-669 Knowlton, N. 2001. The future of coral r eefs. Proceedings of th e National Academy of Sciences of the United States 98(10): 5419-5425

PAGE 54

46 Koop, K., D. Booth, A Broadbent, J. Brodi e, D. Bucher, D. Capone, J. Coll, W. Dennison, M. Erdmann, P. Harrison, O. Hoegh-Guldberg, P. Hutchings, J. Williamson and D. Yellowlees. 2001. ENCORE: The Effect of Nutrient Enrichment on Coral Reefs. Synthesi s of Result and Conclusions. Marine Pollution Bulletin 42(2): 91-120 Kushmaro, A., Y. Loya, M. Fine, and E. Rose nberg. 1996. Bacterial infection and coral bleaching. Nature 380: 396 Lasker, H. R. and M. A. Coffroth. 1983. Oc tocoral Distributions at Carrie Bow Cay, Belize. Marine Ecology Pr ogress Series 13: 21-28 Lesser, M. P., and J. M. Shick. 1989. Phototadaption and defenses against oxygen toxicity in zooxanthellae from natura l populations of symbiotic cnidarians. Journal of Experimental Marine Biology 134: 129-141 Lessios, H. A. 1995. Diadema antillarum 10 years after mass mortality: Still rare, despite help from a competitor. Pr oc. Royal Soc. London Ser. B 259: 331-337 Logan, A., and T. Tomascik. 1991. Extensi on growth rates in two coral species from high-latitude reefs of Berm uda. Coral Reefs 10: 155-150 Martin, R. E. 1998. Catastrophic Fluctuati ons in Nutrient Levels as an Agent of Mass Extinction: Upward Scaling of Ecological Processes? In: Biodiversity Dynamics: Turnover of Populations, Taxa, and Communities. McKinney, M. L. and J. A. Drake (eds). Columbia Univ. Press, New York, chap. 17. http://www.earthscape.org/r3/mckinney/mckinney17.html Mazzullo, S. J., K. E. Anderson-Underwood, C. D. Burke and W. D. Bischoff. 1992. Holocene Coral Patch Reef Ecology and Sedimentary Architecture, Northern Belize, Central America. Palaios 7: 591-601 McClanahan, T. R. and N. A. Muthiga. 1998. An ecological shift in a remote coral atoll of Belize over 25 years. Envi ronmental-Conservation 25: 122-130 McField, M. D. 1999. Coral Response During and After Mass Bleaching in Belize. Bulletin of Marine Science 64: 155-172 ____________, P. Hallock, and W. C. Jaap. 2001. Multivariate Analysis of Reef Community Structure in the Belize Barrier Reef Complex. Bulletin of Marine Science 69: 745-758 Muscatine, L., H. Musada and R. Burnap 1979. Ammonium uptake by symbiotic and aposymbiotic corals. Bulletin of Marine Scie nce 29(4): 572-575 Muzik, K. 1982. Octocoralia Cnidaria from Carrie-Bow Cay Belize. Smithsonian Contributions to the Mari ne Sciences 12: 303-310

PAGE 55

47 Richardson, L. L., W. M. Goldberg, and K. G. Kuta. 1998. Florida's mystery coral-killer identified. Nature 392: 557-558 Risk, M. J., P. W. Sammarco, E. N. Edinger. 1995. Bioerosion in Acropora across the continental shelf of the GBR. Coral Reefs 14: 79-86 ________, J. M. Heikoop, E. N. Edinger and M. V. Erdmann. 2001. The assessment 'toolbox’: community-based reef evaluati on methods coupled with geochemical techniques to identify sources of stre ss. Bulletin of Marine Science 69: 443-458 Rogers, C. S. 1990. Responses of coral reefs and reef organisms to sedimentation. Marine Ecology Progress Series 62: 185-202 Rutzler, K., and Santavy, D.L. 1983. The bl ack band disease of Atla ntic coral reefs. I. Description of the Cyanophyte pathogen: P.S.Z.N.I. Marine Ecology 4: 301-319 Stanley, G. D. Jr., D. G. Fautin. 2001. Th e Origins of Modern Corals. (Scleractinia). Science 291: 1913 Stoddart, D. R. 1962. Three Caribbean atolls: Turneffe Islands, Lighthouse Reef and Glover’s Reef, British Honduras. Atoll Research Bulletin 87: 31-49 ____________ 1963. Effects of Hurricane Hat tie on the Brittish Honduras reefs and cays, October 30-31, 1961. Ato ll Research Bulletin 95: 1-142 Tomascik, T. and F. Sander. 1985. Effects of eutrophication on the growth of the reefbuilding coral Montastrea annularis. Marine Biology 87: 143-155 __________ and ________. 1987. Effects of eutrophicat ion on reef-building corals II. Structure of coral communities on fringing reefs, Barbados, West Indies. Marine Biology 94: 53-75 Trench, R. K. 1987. Dinoflagellates in non-para sitic symbioses. In: Taylor, F. J. R (ed.), Biology of Dinoflagellates. Oxford Bost on: Blackwell Scientific Publications pp 530-570 U.S. State Department. 1999. Coral Bleach ing, Coral Mortality, and Global Climate Change. Report released by the Bu reau of Oceans and International Environmental and Scientific Affairs. http://www.state.gov/www/global/global_ issues/coral_reefs/990305_coralreef_rpt .html U. S. State Department. 2004. Variation in lin ear growth and skeletal density within the polymorphic reef building coral Montastrea annularis. http://usinfo.state.gov/produc ts/pubs/biodiv/coral.htm

PAGE 56

48 Van Veghel, M. L. J., and Bosscher, H. (1995). Variation in Linear Growth and Skeletal Density within the Polymorphic Reef Building Coral Montastrea annularis. Bulletin of Marine Science, 56(3): 902-908 Wilkinson, C. 2001. Status of Coral Reef s of the World: 2000. Global Coral Reef Monitoring Network Report. http://www.aims.gov.au/pages/research/coralbleaching/scr2000/scr-00gcrmn-report.html Wood, R. 1993. Nutrients, Predation and th e History of Reef Building. Palaios 8(6): 526-543

PAGE 57

49 APPENDICES

PAGE 58

50 Appendix A. Combined mean measured basal area (cm2) for M. annularis and P. astreoides by site and year standard deviation Year Site N 2002 2003 2004 Total B 2756.70 68.62 47.4162.17 56.7475.48 53.62 68.25 C 5762.58 60.60 61.5856.06 57.5858.53 60.58 58.12 D 855.38 17.93 63.0021.91 60.5025.41 59.63 21.24 E 15106.13 84.33 115.6778.12 123.6794.53 115.16 84.26 F 6883.01 142.81 80.71150.18 82.94165.95 82.22 152.53 G 3760.27 56.36 48.6542.29 62.0359.98 56.98 53.27 H 39244.46 378.32 258.97386.70 258.62399.95 254.02 385.12 I 31122.52 216.15 121.65234.77 135.84268.67 126.67 238.30 J 23108.17 105.38 106.57114.20 104.52114.66 106.42 109.85 Total 305101.08 181.83 100.46189.02 104.23200.66 101.92 190.46 Appendix B. Mean measured basal area (cm2) for M. annularis and P. astreoides by site and year standard deviation Year Species Site N 2002 2003 2004 Total M. annularis B 25 58.4870.99 49.6864.09 59.7277.73 55.9670.33 C 55 58.6455.86 60.2556.03 55.5856.54 58.1655.83 D 0 0.000.00 0.000.00 0.000.00 0.000.00 E 12 119.5086.03 124.0881.63 139.6795.00 127.7585.64 F 61 77.00146.46 74.23153.87 80.77174.29 77.33157.79 G 37 60.2756.36 48.6542.29 62.0359.98 56.9853.27 H 37 255.43385.57 269.92394.22 269.51407.95 264.95392.46 I 24 119.79237.78 128.58262.07 142.46298.41 130.28263.62 J 22 105.18106.85 108.36116.56 104.91117.34 106.15111.93 Total 273 101.42189.81 101.94198.06 106.77210.36 103.38199.36 P. astreoides B 2 34.5020.51 19.0012.73 19.5012.02 24.3314.40 C 2 171.00113.14 98.0060.81 112.50113.84 127.1784.18 D 8 55.3817.93 63.0021.91 60.5025.41 59.6321.24 E 3 52.6760.93 82.0062.75 59.6772.34 64.7858.28 F 7 135.4398.20 137.14104.66 101.8657.42 124.8186.29 G 0.000.00 0.000. 00 0.000.00 0.000.00 H 2 41.5013.44 56.5036.06 57.0039.60 51.6725.92 I 7 131.86129.38 97.86107.05 113.14137.07 114.29119.58 J 1 174.00— 67.00— 96.00— 112.3355.34 Total 32 98.1389.94 87.7877.40 82.5677.88 89.4981.34 Grand Total 305 101.08181.83 100.46189.02 104.23200.66 101.92190.46

PAGE 59

51 Appendix C. Combined mean change in basal area (cm2 yr-1) for M. annularis and P. astreoides by site and time period standard deviation Period Site N 2002-2003 2003-2004 2002-2004 Total B 27-9.30 11.06 9.3315. 30 0.026.93 0.0213.78 C 57-1.00 24.84 -4.0021. 46 -2.508.96 -2.5019.57 D 87.63 31.44 -2.5036. 23 2.5616.04 2.5628.22 E 159.53 30.55 8.0025. 06 8.7717.16 8.7724.31 F 68-2.31 22.06 2.2435. 74 -0.0419.79 -0.0426.74 G 37-11.62 28.06 13.3826. 07 0.8810.91 0.8824.98 H 3914.51 56.84 -0.3674. 62 7.0830.63 7.0856.80 I 31-0.87 38.67 14.1939. 62 6.6632.28 6.6637.11 J 23-1.61 33.21 -2.0416. 36 -1.8317.71 -1.8323.34 Total 305-0.62 32.86 3.7737. 94 1.5819.89 1.5831.19 Appendix D. Mean change in basal area (cm2 yr-1) for M. annularis and P. astreoides by site and time period standard deviation Period Species Site N 2002-2003 2003-2004 2002-2004 Total M. annularis B 25-8.809.11 10.0415.71 0.62 5.99 0.62 13.36 C 551.6219.75 -4.6720.31 -1.53 7.47 -1.53 17.01 D 00.000.00 0.000.00 0.00 0.00 0.00 0.00 E 124.5832.38 15.5821.02 10.08 18.91 10.08 24.53 F 61-2.7718.43 6.5429.83 1.89 18.23 1.89 23.01 G 37-11.6228.06 13.3826.07 0.88 10.91 0.88 24.98 H 3714.4958.27 -0 .4176.66 7.04 31.39 7.04 58.26 I 248.7927.24 13.8841.66 11.33 31.43 11.33 33.57 J 223.1824.55 -3.4515.24 -0.14 16.12 -0.14 19.00 Total 2730.5230.49 4.8237.46 2.67 19.20 2.67 30.02 P. astreoides B 2-15.5033.23 0.500.71 -7.50 16.26 -7.50 18.03 C 2-73.0052.33 14. 5053.03 -29.25 0.35 -29.25 51.39 D 87.6331.44 -2.5036.23 2.56 16.04 2.56 28.22 E 329.335.69 22.3315.50 3.50 6.50 3.50 24.07 F 71.7144.91 -35.2959.50 -16.79 26.15 -16.79 45.96 G 00.000.00 0.000.00 0.00 0.00 0.00 0.00 H 215.0022.63 0.503.54 7.75 13.08 7.75 13.46 I 7-34.0054.57 15.2934.57 -9.36 32.23 -9.36 44.59 J 1-107.00— 29.00— -39.00 — -39.00 68.00 Total 32-10.3448.22 -5 .2241.42 -7.78 23.32 -7.78 38.74 Grand Total 305-0.6232.86 3.7737.94 1.58 19.89 1.58 31.19

PAGE 60

52 Appendix E. Combined mean growth rate (cm yr-1 radial extension) for M. annularis and P. astreoides by site and time period standard deviation Period Site N 2002-2003 2003-2004 2002-2004 Total B 27 -0.41 0.48 0.340. 38 -0.030.27 -0.030.49 C 57 -0.02 0.68 -0.200. 78 -0.110.33 -0.110.63 D 8 0.28 1.12 -0.121. 25 0.080.61 0.081.00 E 15 0.35 0.82 0.030. 80 0.190.41 0.190.70 F 68 -0.15 0.59 0.060. 85 -0.040.40 -0.040.64 G 37 -0.39 0.76 0.410 .67 0.010.32 0.010.69 H 39 0.31 0.94 -0.060. 93 0.120.48 0.120.82 I 31 -0.07 0.86 0.230 .65 0.080.56 0.080.71 J 23 -0.10 0.81 -0.080. 47 -0.090.40 -0.090.58 Total 305 -0.07 0.77 0.060 .78 0.000.41 0.000.68 Appendix F. Mean growth rate (cm yr-1 radial extension) rate for M. annularis and P. astreoides by site and time period standard deviation Period Species Site N 2002-2003 2003-2004 2002-2004 Total M. annularis B 25-0.370.30 0.360.38 0.000.18 0.00 0.42 C 550.040.60 -0.210.77 -0.090.30 -0.09 0.59 D 00.000.00 0.000.00 0.000.00 0.00 0.00 E 120.130.75 0.300.51 0.220.45 0.22 0.57 F 61-0.160.54 0.140.75 -0 .010.37 -0.01 0.59 G 37-0.390.76 0.410.67 0.010.32 0.01 0.69 H 370.290.96 -0.060. 96 0.120.49 0.12 0.84 I 240.150.39 0.220.60 0.190.33 0.19 0.45 J 220.020.57 -0.120.43 -0.050.36 -0.05 0.46 Total 273-0.050.67 0.090.74 0.020.36 0.02 0.61 P. astreoides B 2-0.851.86 0.040.06 -0 .400.90 -0.40 1.01 C 2-1.720.74 0.071.51 -0 .820.39 -0.82 1.11 D 80.281.12 -0.121.25 0.080.61 0.08 1.00 E 31.220.45 -1.050.93 0.090.26 0.09 1.12 F 7-0.030.98 -0.691.28 -0.360.55 -0.36 0.97 G 00.000.00 0.000.00 0.000.00 0.00 0.00 H 20.510.80 -0.010.14 0.250.47 0.25 0.48 I 7-0.821.50 0.260.87 -0 .280.98 -0.28 1.18 J 1-2.82— 0.91— -0.96— -0.96 1.87 Total 32-0.221.36 -0.191.08 -0.200.68 -0.20 1.07 Grand Total 305-0.070.77 0.060.78 0.000.41 0.00 0.68

PAGE 61

53 Appendix G. Mean growth rate (cm yr-1 radial extension) for M. annularis and P. astreoides by size class and time period standard deviation Period 2002-2003 2003-2004 2002-2004 Total Species Size N MeanSDN M eanSDN MeanSDN MeanSD M. annularis S 71 -0.05 0.42780.190.4271-0. 130.312200.00 1.16 M 72 -0.02 0.58650.070.5867-0. 030.252040.03 1.41 L 130 -0.07 0.821300.040.931350. 120.403950.10 2.15 Total 273 -0.14 1.822730.311.93273-0 .030.978190.13 4.72 P. astreoides S 5 0.75 0.7320.370.404-0. 200.59110.92 1.72 M 7 0.23 1.3990.050.619-0.570.8425-0.28 2.84 L 20 -0.62 1.3421-0.341.2619-0.030.5860-1.00 3.17 Total 32 0.89 0.74320.620.40320. 350.26961.86 1.40 Grand Total 305 0.52 0.493050.490.503050. 260.279151.28 1.25