USF Libraries
USF Digital Collections

An evaluation of the Along Track Reef Imaging System (ATRIS) for efficient reef monitoring and rapid groundtruthing of E...

MISSING IMAGE

Material Information

Title:
An evaluation of the Along Track Reef Imaging System (ATRIS) for efficient reef monitoring and rapid groundtruthing of EAARL Lidar
Physical Description:
Book
Language:
English
Creator:
Caesar, Nicole O
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Patch reef
Biscayne National Park
Rugosity
Digital camera
Florida Keys
Dissertations, Academic -- Marine Science -- Masters -- USF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The Along-Track Reef-Imaging System (ATRIS) is a vessel-mounted, digital camera, depth sounder and Global Positioning System (GPS) package that facilitates the rapid capture of underwater images in shallow-water benthic environments. This technology has the potential to collect ecologically significant data, particularly in benthic habitats less than 10 m in depth, with better location referencing and in less time than is required for surveys carried out by Scuba divers. In October 2004, ATRIS was tested coincidently with SCUBA-assisted video along transects on five patch reefs in Biscayne National Park. Images from both data sets were subsampled, viewed, and benthic cover under random points were identified and counted. Digital-still images of reef benthos collected by ATRIS were of higher quality than SCUBA-acquired video imagery, allowing more reliable classification of benthos. "Substrate", which included areas of hard-ground, sand or rubble, was the most frequently identified benthic category (43%), followed by octocoral (21%), unidentifiable (19%), and macroalgae (12%). Total stony coral cover averaged less than 5%. ATRIS-acquired benthic-cover data were compared with rugosity data derived from the Experimental Advanced Airborne Research Lidar (EAARL), revealing no strong correlations, probably because much of the hard substrate patch reef topography was created by corals that have died in the past few decades. ATRIS, diver-acquired data, and EAARL provide different scales of information, all of which can be valuable tools for assessing and managing coral reefs.
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 Nicole O. Caesar.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 77 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 - 001790068
oclc - 144334104
usfldc doi - E14-SFE0001483
usfldc handle - e14.1483
System ID:
SFS0025802:00001


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 001790068
003 fts
005 20070613094330.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 070613s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001483
040
FHM
c FHM
035
(OCoLC)144334104
049
FHMM
090
GC11.2 (ONLINE)
1 100
Caesar, Nicole O.
3 245
An evaluation of the Along Track Reef Imaging System (ATRIS) for efficient reef monitoring and rapid groundtruthing of EAARL Lidar
h [electronic resource] /
by Nicole O. Caesar.
260
[Tampa, Fla] :
b University of South Florida,
2006.
520
ABSTRACT: The Along-Track Reef-Imaging System (ATRIS) is a vessel-mounted, digital camera, depth sounder and Global Positioning System (GPS) package that facilitates the rapid capture of underwater images in shallow-water benthic environments. This technology has the potential to collect ecologically significant data, particularly in benthic habitats less than 10 m in depth, with better location referencing and in less time than is required for surveys carried out by Scuba divers. In October 2004, ATRIS was tested coincidently with SCUBA-assisted video along transects on five patch reefs in Biscayne National Park. Images from both data sets were subsampled, viewed, and benthic cover under random points were identified and counted. Digital-still images of reef benthos collected by ATRIS were of higher quality than SCUBA-acquired video imagery, allowing more reliable classification of benthos. "Substrate", which included areas of hard-ground, sand or rubble, was the most frequently identified benthic category (43%), followed by octocoral (21%), unidentifiable (19%), and macroalgae (12%). Total stony coral cover averaged less than 5%. ATRIS-acquired benthic-cover data were compared with rugosity data derived from the Experimental Advanced Airborne Research Lidar (EAARL), revealing no strong correlations, probably because much of the hard substrate patch reef topography was created by corals that have died in the past few decades. ATRIS, diver-acquired data, and EAARL provide different scales of information, all of which can be valuable tools for assessing and managing coral reefs.
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 77 pages.
590
Adviser: Pamela Hallock Muller, Ph.D.
653
Patch reef.
Biscayne National Park.
Rugosity.
Digital camera.
Florida Keys.
690
Dissertations, Academic
z USF
x Marine Science
Masters.
773
t USF Electronic Theses and Dissertations.
4 0 856
u http://digital.lib.usf.edu/?e14.1483



PAGE 1

An Evaluation of the Along Track Reef Imag ing System (ATRIS) for Efficient Reef Monitoring and Rapid Groundt ruthing of EAARL Lidar by Nicole O. Caesar A thesis submitted in partial fulfillment of the requirements for the degree of Marine Science College of Marine Science University of South Florida Major Professor: Pamela Hallock Muller, Ph. D. John Brock, Ph. D. Deby Cassill, Ph. D. David Mann, Ph. D. Date of Approval: April 7, 2006 Keywords: patch reef, Biscayne National Pa rk, rugosity, digital camera, Florida Keys Copyright 2006, Nicole O. Caesar

PAGE 2

ACKNOWLEDGEMENTS I would like to acknowledge my major professor, Dr. Pamela Hallock Muller, for her commitment to her students and to her role as a research scientist. I thank Dr. Muller for the guidance and support provided to me during the course of my degree. I also thank my committee members: J ohn Brock, Deby Cassill and David Mann. Many thanks to W. Wright, the Princi pal Investigator fo r the NASA EAARL, for his very significant contributions to the succes s of this research project. V. Rabine is thanked for his dedication and skill as the Chief Pilot of the Cessna 310 aircraft used during the NASA EAARL lidar overflights of th e northern Florida Keys reef tract. The substantial contributions from R. Curry to field investigations within Biscayne National Park are greatly appreciated. Special thanks to P. Thompson for the development of the ATRIS, and D. Hickey for his professionali sm in handling the boat operations undertaken within this study. I would also like to tha nk my family, N. Piehl and T. Wood for their support and encouragement. The U.S. Geological Survey Coastal and Marine Geology Program funded this investigation as a component of the Geol ogic Studies of Coral Reefs Project. My stipends were provided by the University of South Florida/U.S. Geological Survey 2003 – 2006 cooperative agreement.

PAGE 3

i TABLE OF CONTENTS LIST OF TABLES ............................................................................................................iii LIST OF FIGURES ..........................................................................................................iv ABSTRACT ...................................................................................................................... vi 1. INTRODUCTION ..........................................................................................................1 The Along-Track Reef Imaging System (ATRIS) ..................................................2 Experimental Advanced Airbor ne Research Lidar (EAARL) ................................5 Topographic Complexity ........................................................................................7 Biscayne National Park ...........................................................................................7 Objectives ...............................................................................................................9 2. METHODS ...................................................................................................................1 0 Study Sites ............................................................................................................10 Transects ...............................................................................................................10 ATRIS Data Acquisition........................................................................................12 Diver-acquired Data...............................................................................................13 Image Post-processing...........................................................................................13 Benthic Classification and Calc ulation of Percent Cover......................................15 EAARL Data Acquisition......................................................................................16 Statistical Analyses................................................................................................17 3. RESULTS .................................................................................................................... .18 Percent Cover of Benthic Categories.....................................................................18 Comparison of ATRIS and Di ver-acquisition Methods .......................................22 Relationship between Benthic Cla ss, Depth and Rugosity Index .........................24 4. DISCUSSION ...............................................................................................................29

PAGE 4

ii Comparison of ATRIS and Di ver-acquisition Methods .......................................30 Benthic Class and Rugosity Index ........................................................................31 5. CONCLUSIONS ...........................................................................................................34 REFERENCES .................................................................................................................35 APPENDICES ..................................................................................................................39 Appendix I Diver-acquired benthi c category frequencies for each transect on each of the five reef sites ............................................40 Appendix II ARIS-acquired benthi c category frequencies for each transect on each of the five reef sites ............................................41 Appendix III Data utilized to crea te lower, mean and upper quartile benthic category distribu tions for ATRIS and diveracquired data sets combined; Figure 14 ........................................42 Appendix IV Raw data used for regr ession analysis of benthic category and depth versus rugosity indices .................................................43

PAGE 5

iii LIST OF TABLES Table 1. Summary of data collected with ATRIS and by a Scuba diver.................13 Table 2. Original and reclassified benthic categories identified for ATRIS and diver-acquired images.........................................................................16 Table 3. Summarized result s of the Kolmogorov-Smirnov Goodness-of-fit test compari ng the maximum differences between the ATRIS and diveracquired datasets; significant differences differences noted by asterisk (*)................................................................23 Table 4. Results of linear regressi ons of point-count data for each benthic category and each image, compared with the rugosity index for the index of the GPS coordinates of eac h image; significant correlations (p ) (p 0.05) are noted by asterisk (*)...........................................................24 Table 5. Results of linear regressi ons of water depths recorded during the ATRIS surveys over each of f our reefs, against the rugosity indices for indices for those GPS coordina tes; significant correlations (p 0.05) noted (p 0.05) noted by asterisk (*)..................................................................25 Table 6. Comparisons of relative advantages and disadvantages of image-data acquisition using AT RIS and diver-acquired video................30

PAGE 6

iv LIST OF FIGURES Figure 1. Components of the Along-Tr ack Reef Imaging System (ATRIS)...............3 Figure 2. Underwater digital image of ATRIS components mounted on a 3 m aluminium pole; shown here is the digital camera in its underwater housing, attached to the base of the pole.....................................................5 Figure 3. Orientation of th e Cessna aircraft during surveys using the Experi mental Advanced Airborne Research Lidar (EAARL) US Geological Survey, St. Petersburg, FL............................................. 6 Figure 4. Landsat image annotat ed to show the location of the study area within within Biscayne National Park, south Florida, USA...................................8 Figure 5. Composite image showing th e abundance of small patch reefs near the near the study area in Biscayne National Park, created using data acquired acquired using the EAARL system; Patch-reefs included in this study are noted by arrows..........................................................................11 Figure 6. Planned transects (tan lines) and actual vessel tracks (black lines) plotted plotted on a rugosity base map of an individual patch-reef; the map was was created from EAARL-acquired data...................................................12 Figure 7. Screen capture of the PointCount99 user in terface illustrating the analysis of an image acquired using ATRIS..............................................14 Figure 8. Comparison of per cent-cover data for all benthic categories, collected collected using ATRIS and diveracquired images, for Reef 1.................18 Figure 9. Comparison of per cent-cover data for all benthic categories, collected collected using ATRIS and diveracquired images, for Reef 2.................19 Figure 10. Comparison of percen t-cover data for all benthic categories, collected collected using ATRIS and diveracquired images, for Reef 3.................19

PAGE 7

v Figure 11. Comparison of percen t-cover data for all benthic categories, collected collected using ATRIS and diveracquired images, for Reef 4.................20 Figure 12. Comparison of percentcover data for all benthic categories, collected collected using ATRIS and diveracquired images, for Reef 5.................20 Figure 13. Comparison of mean percent-c over data for all benthic categories, collected using ATRIS and diveracquired images, for all reefs; standard standard error bars are shown………………………………………........21 Figure 14. Lower, mean and upp er quartile benthic cat egory distributions for ATRIS and diver-acquired data sets combined....................................22 Figure 15. Ratios of diver-acquired data to ATRIS data for each benthic class, averaged over the five reefs.......................................................................23 Figure 16. The frequency (maximum frequency = 50/image) of points classified as “substrate” identified for each image, plotted against the rugosity indices for the G PS coordinates of each image.................... 25 Figure 17. The frequency (maximum frequency = 50/image) of points classified as “unidentifiable” id entified for each image, plotted against the rugosity in dices for the GPS coordinates of each image.........26 Figure 18. Water depths recorded during the ATRIS surv eys of two transects over over Reef 1, plotted against th e rugosity indices for the GPS coordinates coordinates of those depths: ther e is no significa nt correlation between the between the data sets..................................................................................26 Figure 19. Water depths recorded during the ATRIS surv eys of two transects over over Reef 2, plotted against th e rugosity indices for the GPS coordinates coordinates of those depths (r2 = 0.118)....................................................27 Figure 20. Water depths recorded duri ng the ATRIS surveys of two transects over over Reef 3, plotted against th e rugosity indices for the GPS coordinates coordinates of those depths (r2 = 0.051)....................................................27 Figure 21. Water depths recorded duri ng the ATRIS surveys of two transects over over Reef 4, plotted against th e rugosity indices for the GPS coordinates coordinates of those depths (r2 = 0.063)....................................................28

PAGE 8

vi An Evaluation of the Along Track Reef Imag ing System (ATRIS) for Efficient Reef Monitoring and Rapid Groundt ruthing of EAARL Lidar Nicole O. Caesar ABSTRACT The Along-Track Reef-Imaging System (A TRIS) is a vessel-mounted, digital camera, depth sounder and Global Positioning Sy stem (GPS) package that facilitates the rapid capture of underwater images in sh allow-water benthic environments. This technology has the potential to collect ecologically signifi cant data, particularly in benthic habitats less than 10 m in depth, with better location referencing and in less time than is required for surveys carried out by Scuba divers. In Oc tober 2004, ATRIS was tested coincidently with SCUBA-assisted video along transects on five patch reefs in Biscayne National Park. Images from both data sets were subsampled, viewed, and benthic cover under random point s were identified and counte d. Digital-still images of reef benthos collected by ATRIS were of higher quality than SCUBA-acquired video imagery, allowing more reliable classificati on of benthos. “Substrate”, which included areas of hard-ground, sand or rubble, was the mo st frequently identi fied benthic category (43%), followed by octocoral ( 21%), unidentifiable (19%), an d macroalgae (12%). Total stony coral cover averaged le ss than 5%. ATRIS-acquired benthic-cover data were compared with rugosity data derived from the Experimental Advanced Airborne Research Lidar (EAARL), revealing no str ong correlations, probably because much of the hard substrate patch reef topography was crea ted by corals that have died in the past few decades. ATRIS, diver-acquired data, and EAARL provide different scales of information, all of which can be valuable to ols for assessing and managing coral reefs.

PAGE 9

1 1. INTRODUCTION Coral reefs offer a wealth of economic and social goods to large numbers of people (Moberg and Folke 1999). These produc tive, biologically diverse ecosystems not only provide coastal protection and recreation, but constitute approximately 10% of the world’s fisheries and contribu te 25% of the fish catch in developing nations (HoeghGuldberg 1999, Moberg and Folke 1999). Reef -system inhabitants have played a crucial role as sources of new biochemicals and dr ugs as certain corals sponges, mollusks and seaweeds have been found to possess antican cer, antimicrobial, anti-inflammatory and anti-coagulating properties (Hoegh-Guldbe rg 1999, Moberg and Folke 1999). Reef ecosystems are facing increased pr essure from pollution and human-induced disturbances (Call et al. 2003). A decline in stony coral species was documented by the Coral Reef Evaluation and Monitoring Project (CREMP) at 79% of their monitoring sites within the Florida Keys National Marine Sanctuary between 1996 and 2004 (CREMP Executive Summary 2004). The Florida Keys are affected by a suite of global, regional and local factors (Hallock 2005) and the decline of large areas of the Florida reef tract is thought to be due, in part, to elevated nitr ogen and phosphorous levels that are delivered via submarine groundwater discharge (Lapoint e et al. 2002, Finkl and Charlier 2003). Urbanization along the coast of Florida intensified during the mid 1900s when population densities approached 2500 persons per km2 (Finkl and Charlier 2003). This population growth led to intensive agriculture, pesticide use and shallow-injecti on wells contribute to increased nutrient loading and the resulting nutrification of reef waters (Dustan 1999, Griffin et al. 1999, Finkl and Charlier 2003). Other local f actors that contribute to reef degradation include ship groundings, anchor damage and intense usage by visitors; Dustan (1999) noted that seasonal visitors nearly double the populat ion of the Keys.

PAGE 10

2 The extensive coral bleaching even t of 1997 and 1998 prompted President Clinton’s Executive Order 13089 of June 1998, mandating “a comprehensive program to map and monitor US coral reefs” and for “r esearch aimed at identifying the major causes and consequences of degradation of coral r eef ecosystems” (Rogers et al. 2002). Thus, concerns regarding rapid coastline developm ent near the Florida Keys, Caribbean and Indo-Pacific reef systems reinforce the need for rapid and effective monitoring techniques (Brown 1988, Dustan 1999). Reliable maps and baseline data on globa l and regional ecosystems are essential for effective reef-monitoring pr ograms (Rogers et al. 2002). Furthermore, evidence of worldwide coral-reef degradation (Hochberg et al. 2003) reinforces the need for effective monitoring programs to provide baseline in formation to evaluate the effects of hurricanes, coral bleaching, coral disease and human activities on reef systems, and to provide quantitative data on reef communitie s and community structure (Ohlhorst 1988, Aronson 1994). Accurate habitat maps are essential tools for resource managers and scientists in understa nding the spatial dynamics and ge neral distribution of ecosystems (Riegl 2001). Coastal habitat maps are also a fundamental requirement for establishing coastal management plans (Mumby et al. 1997 ) and accurate maps of submerged reefs provide assessment tools that facilitate the in ventory of degraded and natural reef systems and reef structure (L indeman et al. 2001). The Along-Track Reef Imaging System (ATRIS) Coral reef monitoring efforts require significant amounts of funding, time and experience (Rogers and Miller 2001), and ad hoc approaches tend to be favored to reduce cost (Mumby and Harborne 1999). An arguable need exists fo r a method that facilitates the capture of ecologically sign ificant underwater data in a less labor-intensive and timeconsuming fashion than currently utilized methods. Scuba surveys are more labor intensive and time-consuming than shipboard surveys (Green et al. 1996) and vessel-ac quired imagery can provide a fast, accurate method of mapping corals (Riegl et al. 2001). Data collection of video imagery over

PAGE 11

3 5 km of transect would require approximat ely 17 diver hours, at a rate of 5 m per minute (Jaap and McField 2001), versus 2 hours for vessel-acquire d imagery along the same transect. Improved efficiency will be particularly useful in groundtruthing large quantities of shallow-water reef data, collected by remote sensing techniques such as the Experimental Advanced Airbor ne Research Lidar (EAARL). An image-acquisition method, the Along-Tr ack Reef Imaging System (ATRIS), was designed at the US Geological Survey Ce nter for Coastal and Watershed studies in St. Petersburg, Florida, by the remote sensi ng team lead by Dr. John Brock, to efficiently acquire digital images of shallow-water reef transects, by the incorporation of a camera system onto a research vessel. System co mponents include: a Nikon D1x digital camera, CSI Minimax GPS receiver, navigation co mputer, storage computer, three meter aluminum pole and a bathymetric survey acoustic sounder (Fig. 1). Figure 1. Components of the Along-Track Reef Imaging System (ATRIS).

PAGE 12

4 The camera is housed in a cylindrical PVC housing 27.9 cm high and 22.9 cm in diameter, and deployed on a 3 m aluminum pole a ttached to a survey vessel (Fig. 2). A transparent Lexan window provides a nadir view of reef substrat e. A flange plate at the top of the housing permits three waterproof connectors: 1) a supplier of twelve VDC camera power, 2) input of the GPS GGA string into the image header files and 3) an IEEE 1394 Firewire cable to permit image stor age directly onto onboard swappable 250 gigabyte (GB) hard drives. A Wide Area Augmentation System (WAAS) and Differential capable CSI-Minimax GPS supply vessel coordinates to the Nikon camera. A GPS antenna, mounted to the top of the pole, enables the GPS receiver to acquire image position and location. These GPS readings are sent to the camera to be incorporated into the header files of the digital images. HYPACK navigation software is utilized to follow pre-pla nned transect lines and to cont inuously log ve ssel position at approximately 1 reading per second. The ac oustic sounder is mounted to the bottom of the aluminum camera pole, providing camera /image range from the reef surface. These bathymetric readings are used to determine image scale and bottom coverage during data post-processing. A toggle switch connected to a small motor is used to raise and lower the aluminum pole to follow reef topography. Th e acoustic sounder disp lay is monitored to provide guidance in adju sting the pole height. The digital images are instantaneously tr ansferred to an on-board computer; this allows the duration of each transect to be limited only by the capacity of the storage media in the computer. A forward-looking video camera is also mounted on the base of the pole to provide an oblique view of appro aching substrate. This video feed, recorded on VHS tape, provides further aid in evaluati ng bottom conditions during each transect.

PAGE 13

5 Post-processing serves to correlate imag es with geographical position and depth. Geographic coordinates (Universal Transver se Mercator) and time (Universal Time Coordinated) are recorded in the camera image headers and the HYPACK navigation logs. During post-processing, vessel coordina te information from the navigation files is used to georectify the images as the camera headers do not contain information which relates to the orientation of the camera images. Experimental Advanced Airbor ne Research Lidar (EAARL) Topographic and hydrographic airborne la ser-mapping systems can perform low cost geomorphic surveys with approximatel y 10 cm vertical accuracy and spatial densities that exceed one elevation measurem ent per square meter (Brock and Sallenger 2001). These mapping systems fall into a category of remote sensing referred to as Light Detection and Ranging (Lidar) (Brock and Sallenger 2001). NASA’s EAARL is a 532 nanometer (nm) green wavelength raster-scanning Lidar, designed to survey shallow, su bmerged topography, subaerial topography and vegetation-covered topogra phy in a single flight (Wri ght and Brock 2002). EAARL measures bottom topography relative to the ai rcraft’s GPS position as opposed to mean Figure 2. Underwater digital image of AT RIS components mounted on a 3 m aluminium pole; shown here is the digital camer a in its underwater hou sing, attached to the base of the pole. 3 m aluminium pole Digital camera in underwater housing

PAGE 14

6 low water (Wright and Brock 2002). This sens or features a maximum pulse rate of 5000 hertz, with a vertical accuracy of approximate ly 15-20 cm, horizontal accuracy of less than 1 m, and a spot size of 15 cm with 1x1 m sample spacing (Brock et al. 2002). The 250-pound system is flown on a twin e ngine Cessna 310 (Fig. 3) and operates by recording the time-resolved intensity waveform of the pulsed return signal to the detector each nanosecond after a pulse is emitted. Topographic maps are created by processing the laser pulses recorded during a f light to create a digita l elevation model of the flown surface. EAARL’s high resolution makes it well su ited to rapidly capture detailed, structurally complex, rugose transects of shallow-water reefs (Brock et al. 2004). Scanning Lidars, such as EAARL, acquire spat ially dense data within a swath that may be hundreds of meters wide. EAARL produces a swath of 240 m at a flying altitude of 300 m. Acquiring this type of data by tr aditional means, such as ground-based range finders, is prohibitively time consumi ng and expensive (Sa llenger et al. 2003). Figure 3. Orientation of the Cessna aircraft during surveys using the Experimental Advanced Airborne Research Lidar (EAARL) US Geological Survey, St. Petersburg, FL.

PAGE 15

7 Topographic Complexity Habitat structural complexity encompasses va riations in vertical relief and is an important governing factor in the abundan ce patterns of many marine organisms (McCormick 1994). Habitat structural complexi ty, also termed rugos ity, contributes to ecological community structure (McCormick 1994, Friedlander and Parrish 1998), fish distribution (McCormick 1994, Chapman and Kramer 1999) and provides habitat and shelter for consumers. Szmant (1997) noted that structural complexity is essential to a reef system’s ability to acqui re nutrients, maintain high le vels of gross production and consumption, and maintain coral dominance over algae. Surveys that measure reef rugosity contribute to the long-term mon itoring of cumulative di sturbance and sudden catastrophic change due to hurricanes and sh ip groundings (Brock et al. 2004). These surveys also provide useful data in the st udy of fish assemblages and reef ecological structure and function (Fri edlander and Parrish 1998). Biscayne National Park Biscayne National Park is located at the southeast tip of the Florida peninsula (24 25’ N, 80 15’ W), and received National Pa rk status in 1980. The park’s southern boundary extends from the mainland to a point 9 km south of Pacific Reef and the northern boundary extends from the mainland to a point 6 km north of Fowey Rocks (Fig. 4). The park’s eastern boundary follow s the 18 m depth contour along the seaward margin of the reef tract (Hudson et al. 1994). Approximately 4000 patch reefs are located between Fowey Rocks and Broad Creek (Marszalek et al. 1977). “Patch reef” re fers to localized, elevated accumulations of corals, octocorals and associated benthos (G insburg et al. 2001). Prominent patch reef morphology includes linear and dome-shaped reef s (Ginsburg et al. 2001). Linear-type patch reefs, which are usually found seaward of dome-type reefs, are linear in plain view and are oriented end to end in single or multip le rows, exhibiting an orientation similar to the outer bank reefs (Marszalek et al. 1977). Dome-type patc h reefs range in size from a

PAGE 16

8 few meters to several hundred meters in diam eter, are roughly circular or elliptical in plain view and are generally le ss than five meters in height, though some exhibit as much as nine meters in vertical relief. This reef type usually occurs in clusters and supports a varied assemblage of stony corals, octocorals sponges and algae (Mars zalek et al. 1977). Dome-type reefs vary in both height above the sea-floor a nd substrate type, and are found at varying distances from shore (Marszalek et al. 1977). The density of patch reefs within Biscayne National Park and their locati on in relatively clear water make this area well suited to this study. Figure 4. Landsat image annotated to show the location of the study area within Biscayne Biscayne National Park, south Florida, USA. Anniversary reef

PAGE 17

9 Objectives The specific objectives of this study are 1) to evaluate the capability of ATRIS to capture ecologically significant underwate r data over shallow-water patch reefs by comparing ATRIS image transects with diver-a cquired video and 2) to utilize the data acquired by ATRIS to groundtru th topographic rugosity data acquired with the EAARL airborne laser mapping system.

PAGE 18

10 2. METHODS Study Sites This study utilized data collected over five patch reefs in Biscayne National Park. The patch reefs included in this study were dome-type reefs th at ranged in diameter from 45 m to 60 m, demonstrated comparable bent hic distributions, dept h and distance from shore. Reef benthos visually comprised of > 40% octocorals, 55% bare or sandcovered hard ground and < 5% stony corals a nd sponges. These sites were located just seaward of Anniversary reef (Fig. 5) and we re characterized by reef slopes visually estimated as < 80. Transects Predefined northeast and southwest orient ed transect lines we re overlaid on basemaps for survey purposes (Fig. 6). These base-maps were created from EAARL data to display benthic rugosity at each site. These maps were georectified and stored as TIFF images. Transect lines were oriented over va riations in reef bent hos as indicated by the base-map for each site to ensure that digi tal image transects represented the benthic variation present at each reef site.

PAGE 19

11 Figure 5. Composite image showing the abundan ce of small patch reef s near the study area in Biscayne Nati onal Park, created using data acquired using the EAARL system; Patch reef s included in this st udy are noted by arrows. Anniversary reef 1km

PAGE 20

12 ATRIS Data Acquisition ATRIS data were collected at five reef sites within Biscayne National Park on October 21st and 22nd, 2004 (Fig. 5). Digital underwate r images were collected with a Nikon D1x 5.3 megapixel camera at the rate of one image every two seconds with an AFS 18-70 mm f: 3.5 4.5 Nikkor lens. The fo cal length was set to 46 mm, providing a vertical Field of View (FOV) of 19.2 a nd horizontal FOV of 28.9. Image resolution was 3008 x 1960, with JPEG “normal compression”, resulting in an image-file size of 1.5 megabytes. Camera settings were selected to optimize image quality. The acoustic-sounder display was monitored to provide guidance in adjusting the camera-pole height. Image range was mainta ined at 1.5 m above the reef surface where possible, generating images with consistent spatial resolution. A 7.9 m Coastal Runner 2690 vessel was used and vessel speed was main tained at 0.7 m/s. This vessel speed, combined with the digital image acquisition rate and Nikon lens’ field of view provided minimal image overlap. Figure 6. Planned transects (tan lines) and ac tual vessel tracks (bl ack lines) plotted on a rugosity base map of an individual patch reef; th e map was created from EAARL-acquired data. Pre-planned transect lines in tan Executed vessel lines in black 2 0m

PAGE 21

13 Depth soundings from the acoustic sounder were wired directly into HYPACK navigation software, and supplie d image-range data at a ra te of approximately 10 per second. Diver-acquired Data During ATRIS data collection, a Scuba diver was towed by the survey vessel to allow Hi-8 video data to be collected coinci dentally over the pre-pl anned transects (Table 1). A conventional Sony DCR-VX1000 video ca mera recorder was utilized and housed in a commercial Ikelite underwater housing. The survey vessel was oriented to each transect prior to diver descent and initiation of data collection. Th e diver utilized a premeasured, nadir-oriented PVC pipe to maintain the desired range of 40 cm from the reef surface, capturing an image with an area of a pproximately 40 cm. No artificial lights were used. Reef ID Coordinates of reef center (latitude, longitude) No. of vessel-acquired digital images captured Seconds of diver video captured No. of still frames pulled from diver video Reef 1 e584724_n2809202 224 458 222 Reef 2 e584403_n2808689 101 323 154 Reef 3 e584252_n2808118 397 161 223 Reef 4 e584480_n2809116 131 105 248 Reef 5 e584368_n2808892 124 310 285 Image Post-processing ATRIS-acquired image sets were subsampled to represent vessel transit over each patch reef. Images containing seagrass beds that were photographe d during vessel transit to various sites were not utilized. Fift y random points were overlaid on each ATRIS image for classification purposes using Po intCount99 for Coral Reefs (Fig. 7). Table 1. Summary of data collected with ATRIS and by a Scuba diver.

PAGE 22

14 Coral-reef benthos were classified to species level where possible. Broader categories, including boulder, plate and branching coral, were utilized when species identification was not possible. Other categories included octocoral, s ponge, macroalgae, substrate and seagrass. Ambiguous or blurry substrate points that coul d not be identified were categorized as “unidentifiable”. Due to vari ations in camera to reef-surf ace range, image overlap varied from 0 to approximately 5%. The spr eadsheets produced by the PointCount99 software contained frequency counts of the bent hic categories that were classified in each image. Diver-acquired video transects were converted to still fr ames with an 8% overlap using RavenView software located at the Fl orida Wildlife Research Institute (FWRI). Image transects represented vessel transit ac ross each patch reef. Fifteen random points were overlaid on each image for image cl assification using PointCount99 . Benthic Categoryification and Ca lculation of Percent Cover Figure 7. Screen capture of the PointCount99 setup illustrating the analysis of an image acquired using ATRIS Figure 7. Screen capture of the PointC ount99 user interface illustrating the analysis of an image acquired using ATRIS.

PAGE 23

15 Benthic Classification and Ca lculation of Percent Cover The vessel speed and chosen image size allowed data points from ATRISacquired images to be classified to species le vel in some instances. This was not the case with the diver-acquired images. The smaller range of the diver-acquired images to the reef surface resulted in blurring, as the trans it speed was too great for the camera system being utilized. To make the comparison of th e two data sets possible, the finer ATRIS categories were incorporated into the broader categories identifiable in the diver-acquired data set (Table 2). The “substrate” category in cluded in the table indi cates areas of bare hard-ground, sand or rubble present on the reef. Reef-benthos category percentages were calculated for each data-acquisition method.

PAGE 24

16 Original ATRIS Categories Reclassified ATRIS Categories Original Diver Categories Acropora cervicornis Branching coral Branching coral M illepora alcicornis Porites porites Boulder coral Boulder coral Boulder coral Diploria strigosa M ontastraea cavernosa Siderastrea radians Plate coral Plate coral Plate coral Porites astreoides Octocorallia Octocoral Octocoral Porifera Sponge Sponge Substrate Substrate Substrate Macroalgae Macroalgae Macroalgae Seagrass Seagrass Seagrass Zoanthids Zoanthids Zoanthids Fish Other Other Image margin Unidentifiable Unidentifiable Unidentifiable EAARL Data Acquisition An EAARL mission was flown from August 2nd to 9th 2002 to collect a broad swath of high resolution data of the northern Florida reef tract. Approximately 70 GB of data were collected during th e survey which encompassed our study site within Biscayne National Park. The data were processed by the US Geological Survey (USGS) remotesensing team to produce a digital elevati on model of subsurface topography. Rugosity index values were generated for the survey coordinates by team leader Dr. John Brock, by dividing the actual surface area by the theore tical surface area of a horizontal flat plane covering the same spatial extent [s urface area of topogra phy within kernel) / Table 2. Original and reclas sified benthic categories identified for ATRIS and diver-acquired images.

PAGE 25

17 (surface area of the kernel kernel) for a 2 m kernel]. A kernel refers to the size of the area box that slides over each point of the E AARL digital elevation map to execute the calculation. Statistical Analyses Image data sets from the two survey methods did not meet the assumptions for parametric tests. A Kolmogorov-Smirnov Go odness-of-fit test was performed using Microsoft Excel software, to determine if benthic category representation differed significantly in the image tr ansects captured by the ATRIS and Scuba diver methods (Sokal and Rohlf 1973). Linear regression analysis (Sokal and Rohlf 1973) using Microsoft Excel software was used to determ ine if benthic cover and reef depth were correlated with rugosity. A probability of < 0.05 was considered reason to reject the null hypothesis of no significant di fference between test cases.

PAGE 26

18 3. RESULTS Percent Cover of Benthic Categories All five sites demonstrated similar distri butions in benthic cover (Fig. 8 to 12). The most common benthic types were s ubstrate, macroalgae, octocoral, and unidentifiable. All other categor ies individually averaged less than 5% cover in both data sets (Fig. 13). Overall, the diver-acquire d data set displayed a higher proportion of “unidentifiable” points than the ATRIS data se t due to the poor visual resolution of the diver-acquired data set. Reef 10 10 20 30 40 50 60Bra n chi n g c o r a l B o uld e r c o ral P l a te c o r a l O c t o c o r a l S p o n ge Su b s t rate Macr o alg a e Se a gra s s Zo a n t h i ds Other UnidentifiablePercent cover ATRIS Diver Figure 8. Comparison of percent-cover data for all benthic categories, collected usin g ATRIS and dive r -ac q uired ima g es for Reef 1.

PAGE 27

19 Reef 20 10 20 30 40 50 60Bra n c h ing cora l Boulder cora l P l ate coral Octocoral Sponge Sub s trate Macro a lgae Sea g rass Zoa n thid s Oth e r Un id e nti fi a b lePercent cover ATRIS Diver Reef 30 10 20 30 40 50 60Bra n c h ing cora l Boulder cora l P l ate coral Octocoral Spo n ge Sub s trate Macro a lgae Sea g rass Z oa n thid s Oth e r Un id e nti fi a b lePercent cover ATRIS Diver Figure 9. Comparison of percent-cover data for all benthic categories, collected using ATRIS and diver-acquired images, for Reef 2. Figure 10. Comparison of percent-cover data for all benthic categories, collected using AT RIS and diver-acquired images, for Reef 3.

PAGE 28

20 Reef 40 10 20 30 40 50 60Bra n chi n g c o r a l B o uld e r c o ral P l a te c o r a l O c t o c o r a l Sp o n g e Su b s t rate Macr o alg a e Se a gra s s Zoan t hids Other UnidentifiablePercent cover ATRIS Diver Mean percent cover for reefs 1 to 50 10 20 30 40 50 60Bra n chi n g c o ral B o uld e r coral P l a te c o r a l O c t o c o r a l Sp o n g e Su b s t rate Ma c r o a l g a e Seag r a s s Zoan t hi d s Other UnidentifiablePercent cover ATRIS Diver Figure 11. Comparison of percent-cover data for all benthic categories, collected using AT RIS and diver-acquired images, for Reef 4. Figure 12. Comparison of percent-cover data for all benthic categories, collected using AT RIS and diver-acquired images, for Reef 5.

PAGE 29

21 Mean percent cover for reefs 1 to 50 10 20 30 40 50 60Bra n chi n g c o ral B o uld e r coral P l a t e c o r a l Oct o c o r a l S p o n ge S u b s t r a te Ma c r o a l g a e Se a g r a s s Zo a n t h id s Othe r Un i den t ifi a blePercent cover ATRIS Diver Mean percent-cover values for the two camera methods (Fig. 13) were combined and new averages were calculated for each benthic category. Figure 14 displays these values, as they correspond to the lower (25%), mean (50%) and upper (75%) quartiles for each category. The benthic cate gories responsible for the ma jority of the variability displayed in Table 5 are macroalgae, unide ntifiable, octocoral and substrate. Figure 13. Comparison of mean percent-cove r data for all benthic categories, collected using ATRIS and diver-acquired images, for all reefs; standard error bars are shown.

PAGE 30

22 Percentile distribution of lower, mean and upper quartiles0 10 20 30 40 50 60 70Pla te c ora l Zoa n th i ds Sp o n g e Bra n c h in g c o ral Sea g ra s s Boulder coral O th e r M acroalgae Uni d en ti fi a b l e Oc to c o ral Substra t ePercent cover Comparison of ATRIS and Diver-acquisition Methods Figure 15 displays the ratio of diver-ac quired to ATRIS benthic cover data. A ratio of 1 indicates complete agreement in benthic cover between the two methods. Boulder coral, macroalgae and seagrass categories were under-represented in the diveracquired data set with means of 0.65, 0.39 and 0.42 respectively. The higher resolution ATRIS data set allowed the classification of mo re points into their applicable categories. The “other” and “unidentifiable” categories were substantially ove rrepresented in the diver-acquired data set, with m eans of 2.82 and 3.34 respectively. Figure 14. Lower, mean and upper quartile benthi c category distributions for ATRIS and diver-acquired data sets combined.

PAGE 31

23 Diver-acquired/ATRIS percent cover ratios0 0.5 1 1.5 2 2.5 3 3.5 4Bra n chi n g c o r a l B o uld e r c o ral P l a te c o r a l O c t o c o r a l Sp o nge Su b strate Macr o alg a e Se a gra s s Zoanthids Other UnidentifiableRatio Table 3 displays the results of the Kolmogorov-Smirnov Goodness-of-fit test. D (the maximum unsigned difference between the ATRIS and diver-acquired data sets), is significant at all reef sites. This indicates that the image data captured by the two survey methods was significantly diffe rent at all five reef sites. Reef site D 0.05 D Reef 1 0.037 0.17* Reef 2 0.035 0.21* Reef 3 0.047 0.13* Reef 4 0.040 0.19* Reef 5 0.036 0.09* Figure 15. Ratios of diver-acquired data to ATRIS data for each benthic category, averaged over the five reefs. Table 3. Summarized results of the Kolmogorov-Smirnov Goodness-of-fit test comparing the maximum differences between the ATRIS and diveracquired data sets; significant differences noted by asterisk (*).

PAGE 32

24 Relationship between Benthic Cate gory, Depth and Rugosity Index Navigation data were only collected for four of the five reefs as a consequence of a field error. Regression analyses were used to investigate the relationship between benthic category (Table 4), depth (Table 5) and the corresponding rugosity indices for those coordinates at four reef sites. The substrate and uni dentifiable bent hic categories correlated significantly to their corresponding rugosity values. Substrate frequencies were negatively correlated to rugosity (Fi g. 16) whereas “unident ifiable” frequencies were positively correlated to rugosity (Fi g. 17). Depth was signi ficantly correlated to rugosity for 3 of the 4 reef sites (Fig. 18 to 21). Benthic category n Slope p-value Intercept p-value F Value r 2 Branching coral 41 1.092 0.897 0.853 0.923 0.0170 0.0004 Boulder coral 94 4.109 0.479 -2.123 0.734 0.506 0.005 Plate coral 27 4.007 0.417 -2.679 0.610 0.680 0.026 Octocoral 280 -8.979 0.222 20.995 0.008 1.498 0.005 Sponge 37 2.420 0.170 -1.296 0.489 1.960 0.053 Substrate 310 -24.865 0.017* 51.463 5.31E-06 5.779 0.018 Macroalgae 277 7.645 0.227 1.543 0.820 1.469 0.005 Seagrass 22 -25.198 0.396 35 0.263 0.752 0.036 Other 138 1.454 0.623 -0.0233 0.994 0.242 0.002 Unidentifiable 247 86.815 1.97E-08* -83.724 5.15E-07 33.713 0.121 Zoanthids 20 2.005 0.605 -0.783 0.849 0.277 0.0152 Table 4. Results of linear regres sions of point-count data fo r each benthic category and each image, compared with the rugosity index for the GPS coordinates of each image; significant correlations (p 0.05) noted by asterisk (*).

PAGE 33

25 Reef n Slope p-value Intercept p-value r 2 Reef 1 472 0.803 0.381 5.094 5.14E-07 0.00164 Reef 2 463 8.897 3.22E-14* -4.598 0.000148 0.118 Reef 3 229 3.501 0.000582* 1.467 0.175 0.0510 Reef 4 329 7.001 4.43E-06* -2.518 0.121 0.0625 Substrate0 5 10 15 20 25 30 35 40 45 50 11.11.2 Rugosity indexFrequency Table 5. Results of linear re gressions of water depths recorded during the ATRIS surveys over each of four reefs, against the rugosity indices for those GPS coordina tes; significant correlations (p 0.05) noted by asterisk (*). Figure 16. The frequency (maximum frequenc y = 50/image) of points classified as “substrate” identifie d for each image, plotted against the rugosity indices for the GPS coordinates of each image.

PAGE 34

26 Unidentifiable0 5 10 15 20 25 30 35 40 45 50 11.11.2 Rugosity indexFrequency Figure 17. The frequency (maximum frequency = 50/image) of points classified as “unidentifiable” iden tified for each image, plot ted against the rugosity indices for the GPS coordinates of each ima g e. Figure 18. Water depths recorded during th e ATRIS surveys of two transects over Reef 1, plotted agai nst the rugosity indices for the GPS coordinates of those depths; there is no significant correlati on between the data sets. Reef 10 1 2 3 4 5 6 7 8 9 10 11.11.2Rugosity indexDepth (m)

PAGE 35

27 Reef 22 3 4 5 6 7 8 9 11.11.2 Rugosity indexDepth (m) Reef 32 3 4 5 6 7 8 9 11.11.2 Rugosity indexDepth (m) Figure 19. Water depths recorded during th e ATRIS surveys of two transects over Reef 2, plotted agains t the rugosity indices for th e GPS coordinates of those depths (r2 = 0.118). Figure 20. Water depths recorded during th e ATRIS surveys of two transects over Reef 3, plotted agai nst the rugosity indices for the GPS coordinates of those depths (r2 = 0.051).

PAGE 36

28 Reef 42 3 4 5 6 7 8 9 10 11.11.2 Rugosity indexDepth (m) Figure 21. Water depths recorded during th e ATRIS surveys of two transects over Reef 4, plotted agains t the rugosity indices for th e GPS coordinates of those depths (r2 = 0.063).

PAGE 37

29 3. DISCUSSION As coral cover on reef environments conti nues to decline, there is a strong need for large scale, periodic assessment of benthic composition (Aronson et al. 1994). Collection of photographic data provides reef researchers with a permanent, archivable record with which to estimate coral diversity and percent c over of corals, algae, sponges, and other sessile plants and animals (R ogers et al. 2002, Japp and McField 2001). Photographic surveys can be executed with di ver-acquired digital vi deo (traditional) or with vessel-oriented methods using either digital or video camera equipment. Surveys conducted by Scuba divers requi re time and well trained manpower, creating a need for an alternative method (G reen et al. 1996, Riegl et al. 2001). Diver surveys are limited by air supply, diver bottom time, and the skill of the divervideographer to maintain a constant distance from the reef surface. On the plus side, diver surveys have less potential for habita t damage and are not as limited by depth as vessel-mounted camera systems. Diver-collect ed video is labeled with a time tag, but other relevant field data must be hand recorded. The digi tal video data must be run through expensive frame-pulling software be fore the image analysis can begin. However, once this is complete, the small image sizes can be easily handled by most computer systems. Data collection with the ATRIS shipboard system has many advantages. System incorporation of an onboard G PS allows the images to be tagged with GPS coordinates and accurately recalled by location. The integr ation of this coordinate data in a header embedded in each image makes it possible to accurately and efficiently select a subset of images for use in a GIS platform. The GPS refe rencing of image files is also useful when seeking locations for more detailed study and in the development of a timeline series. This capability becomes especially useful when conducting small scale ecosystem

PAGE 38

30 studies, as it is possible to relate each image to a GPS position and reef depth. Depth readings can be used to produce a depth profile of the surveyed area and to facilitate the calculation of the benthic c over for each image. The ATRIS deployment methods and larger image footprint allo w large data sets to be gathered relatively quickly, with minima l effort, in clear, relatively shallow-water conditions. The capture of di gital stills (as opposed to vide o) decreases the processing required before images can be analyzed. Un fortunately, vessel outfitting must currently be done near an engineering shed prior to de parture for the field, as vessel measurements are required to secure the camera mount to the vessel. Rapid re sponse to sudden reef change may only occur if a previously outfitted vessel is in the area. If certain system malfunctions occur, such as damage to the motor responsible for the pole motion and fittings or to the underwater housing, field work must be postponed until repairs are completed. The reliance on a motor to control pole depth also increase s the possibility of damaging reef habitat and the camera housing. A summary of system comparisons can be found in Table 6. Characteristic ATRIS Diver System Labor Intensive Less More Time Efficient More Less Accurate image referencing via GPS Integral Not integral Preprocessing required before image analysis Less More Acquisition of depth profile of surveyed area Integral Not integral Image bottom coverage More Less Use in large-scale monitoring More Less Use in fine-scale monitoring High potential Current use Operational depth limitations < 10 m < 30 m, diving at greater depths require further technical support and certification Likelihood of substrate damage during survey Some Minimal Requirements for rapid response Limited to presence of outfitted vessel Trained videographer Image resolution 3008 x 1960 680 x 420 Table 6. Comparisons of relative advantag es and disadvantages of image-data acquisition using ATRIS and diver-acquired video.

PAGE 39

31 Comparison of ATRIS and Diver-acquisition Methods Benthic category representation was found to be significantly different between the two survey methods at all 5 reef sites. The two camera systems varied in image resolution, image range and spatial coverage ATRIS had a higher image resolution, larger spatial size per image and greater dist ance from the reef. The survey speed of 0.7 m/s resulted in good ATRIS image quality and poor diver-acquired image quality. Diveracquired image quality and the ability to cate gorize benthos to speci es level would have been improved had the diver transit speed b een reduced to ~ 0.033 m/s (Rogers et al. 2002). Benthic Category and Rugosity Index This section of the study sought to determine the degree to which benthic categories were correlated to rugosity valu es. The “substrate” and “unidentifiable” categories weakly correlated with rugosity. The negative correlation found between “substrate” frequency and rugosity, may be an indicator of reef decline. High “substr ate” frequencies present on the reef crests at my study sites were paralleled by very low stony coral cover < 5%. These areas have experienced a loss in three-dimensional reef habitat and demonstrate correspondingly low rugosity values. Palandro et al. (2003) utilized Landsat imag ery in their analysis of decadal-scale changes in benthic categories at Carysfort reef, and reported that the decline in coral cover through time was directly parallele d by an increase in “substrate”. My observations of coral cover are similar to th e value of 4% reported by Palandro et al. (2003) for Carysfort reef at the end of th eir study period and that of 6.6%, recorded within the Florida Keys National Marine Sanctuar y by CREMP in 2004 (CREMP Executive Summary 2004). As the Palandro et al. (2003) study show s, coral cover on Floridia reefs have declined dramatically over the past several decades. Thus, the relationship between

PAGE 40

32 “substrate” frequencies and rugosity values de monstrated in my st udy may indicate areas that are experiencing, or ha ve experienced, a loss in structural complexity. The positive correlation found between depth an d rugosity at three of the four reef sites was unexpected, as I assumed that reef -structural complexity would diminish with increasing depth. However, the following factors may explain why my original assumption was in error. The first factor is that the surveyed de pth range was relatively shallow, with a maximum depth of ~ 9 m. Thus, all depths within the survey area were well within the range suitable for reef growth (Hallock and Schlager 1986). For the reefs surveyed, the potential for reef growth would not be expected to decline with depth. Secondly, the surveyed patch reefs we re surrounded by narrow sand halos and seagrass beds. The seagrass habitats exhibite d low rugosity values and occurred at depths of ~ 7.5 m. However, images of the seagrass beds were removed from analysis because the target survey habitats were the patch reef s. Typically, the reef margins rose abruptly from the surrounding seagrass, so some of the highest rugosit y values were found at the reef margins, where the depth can shoal abrup tly from ~ 9 m to as little as 2 m over a linear distance of a few meters. My reef site s exhibited little topogr aphic variation on the tops of the patch reefs, while the most ab rupt changes in topography, i.e., the highest rugosity, occurred at the reef margins, which were al so adjacent to the deepes t water. Despite the lack of relationship between benthic categories, most notably coral cover, and rugosity, EAARL data can provide an invaluable resource for mapping patch reef location and distribution, su ch as that within Biscayne National Park (see Fig. 5). Although coral cover currently constitutes a sm all percentage of patch reef benthos, the three-dimensional structure produced by form er coral colonies continues to provide essential substrate for the attachment of octocorals, sponges and macroalgae. The topography provided by the limestone and octocora ls provide essential habitat and shelter for fish and invertebrates, while the macr oalgae provide a food source. Moreover, the topography of the patch reefs, and associated invertebrates and fish, still serve as appealing snorkeling and diving sites for vi sitors to Biscayne National Park.

PAGE 41

33 Thus, while my study did not demonstrat e that EAARL-acquire d rugosity data can be used to predict coral cover, EAARL da ta are extremely useful in detailed mapping and quantification of the occurr ence of patch reef habitat. The ATRIS system provides an efficient way to survey large areas of shallowwater benthic habitats (< 3 m) like seagrass beds, and record phys ical disturbance to habitats from hurricanes and ship groundings. The system also provides a cost-effective method to survey extensive ride tide and bl ack-water events, and to groundtruth data captured with remote-sensing techniques, as an area of ~ 3.5 km2 can be surveyed during a 5 hour field trip. Also, the differentiation of small-scale patterns in coral communities is possible with vessel-acquired imagery (Riegl et al. 2001). Thus, as coral communities continue to decline, systems such as ATRIS will become more useful in the documentation of be nthic resources at broa der scales such as coral cover versus algal dominance, coral disease detection, occu rrence and extent of bleaching events, and reef habitat change due to physical damage. Diver surveys will continue to be needed to execute fine-scale monitoring and identification to species level at depths of 30 m or greater, as ATRIS is limited to depths of < 10 m or less. Remotely operated or au tonomous vehicles equipped with video or digital cameras may also prove useful for deeper-water surveys. EAARL and ATRIS systems can provide e ffective tools for the assessment and subsequent management of shallow-water co ral reef resources and in determining the condition of benthic habitats. Each system provides a different scale and resolution of information and data acquired using these systems will provide an invaluable baseline against which to compare future surveys, to as sess the ability of reef systems to continue to provide essential habitat, or documen t their decline by phys ical and biological processes.

PAGE 42

34 5. CONCLUSIONS 1. Comparison of benthic data fr om ATRIS and diver-ac quired video yielded significant differences in be nthic categories, especial ly the “unidentifiable” category. This was likely an artifact of towing the diver too fast, preventing acquisition of an optimal data set. 2. Stony coral cover did not correlate with rugosity, whereas the “substrate” and “unidentifiable” categories w eakly but significantly corre lated with reef rugosity. 3. Depth positively correlated with rugosity on 3 of the 4 reef sites. 4. Lack of significant correlation between stony coral cover and rugosity was likely the result of very low coral cover (< 5 %) as a consequence of decline in coral cover in recent decades. 5. ATRIS, diver-acquired data, and EAARL provi de different scales of information; all can be valuable tools for a ssessing and managing coral reefs.

PAGE 43

35 REFERENCES Aronson RB, Edmunds PJ, Precht WE, Swans on DW, Levitan DR (1994) Large-scale long-term monitoring of Caribbean coral reef s: simple, quick, ine xpensive techniques. Atoll Research Bulletin 421: 1-19 Beaver CR, Jaap WC, Porter JW, Wheaton J, Callahan M, Kidney J, Kupfner S, Torres C, Wade S, Johnson D (2004) Coral Reef Ev aluation and Monitoring Project (CREMP), 2004 Executive Summary Brock J, Sallenger A (2001) Airborne topogr aphic Lidar mapping for coastal science and resource management. US Geologi cal Survery Open-File Report 01-46 Brock JC, Wright W, Clayton TD (2004) Lidar optical rugosity of coral reefs in Biscayne National Park, Florida. Coral Reefs 23: 48-59 Brock JC, Wright W, Sallenger AH (2002) Basis and methods of NASA airborne topographic mapper Lidar surveys for coastal stud ies. Journal of Coastal Research 18(1): 1-13 Brown BE (1988) Assessing environmental impa cts on coral reefs. In: Proceedings of the 9th International Coral Reef Sy mposium, Australia 1: 71-79 Call KA, Hardy JT, Wallin DO (2003) Coral reef habitat discrimination using multivariate spectral analysis and satellite remote sensing. International Journal of Remote Sensing 24(13): 2627-2639 Chapman MR, Kramer DL (1999) Gradients in coral reef fish dens ity and size across the Barbados marine reserve boundary: effect s of reserve protection and habitat characteristics. Marine Ec ology Progress Series 181: 81-96 Dustan P (1999) Coral reefs under stress: sour ces of mortality in the Florida Keys. Natural Resources Forum 23: 147-155 Finkl CW, Charlier RH (2003) Sustainability of subtropical coastal zones in Southeastern Florida: Challenges for urbanized coastal environments threatened by development, pollution, water supply and storm hazards. Journal of Coastal research 19(4): 934-943

PAGE 44

36 Friedlander AM, Parrish JD (1998) Habitat ch aracteristics affecting fish assemblages on a Hawaiian coral reef. Journal of Experime ntal Marine Biology and Ecology 224:1-30 Ginsburg RN, Gischler E, Kien e WE (2001) Partial mortalit y of massive reef-building corals: an index of patch r eef condition, Florida reef trac t. Bulletin Marine Science 69(3): 1149-1173 Griffin DW, Gibson III CJ, Lipp K, Riley K, Pa ul III JH, Rose JB (1999) Detection of viral pathogens by reverse tran scriptase of PCR and of micr obial indicators by standard methods in the canals of the Florida Keys Applied and Environmental Microbiology 65(9): 4118-4125 Green EP, Mumby PJ, Edwards AJ, Clark CD ( 1996) A review of remote sensing for the assessment and management of tropical coas tal resources. Coastal Managament 24:1-40 Hallock P (2005) Global change and modern coral reefs: new oppor tunities to understand shallow-water carbonate depositional processes. Sedimentary Geology 175:19-23 Hallock P, Schlager W (1986) Nutrient excess and the demise of coral reefs and carbonate platforms. Palaios 1(4): 389-398 Hochberg EJ, Atkinson MJ, Andrefouet S ( 2003) Spectral reflectan ce of coral reef bottom-types worldwide and implications for coral reef remote sensing. Remote Sensing of the Environment 85:159-173 Hoegh-Guldberg O (1999) Climate change, co ral bleaching and the future of the world’s reefs. Marine Freshw ater Research 50:839-866 Hudson JH, Hanson KJ, Halley RB, Kindinger JL (1994) Environmental implications of growth rate changes in Montastrea annularis : Biscayne National Park, Florida. Bulletin of Marine Science 54(3): 647-669 Jaap WC, McField MD (2001) Video sampli ng for monitoring coral reef benthos. Bulletin of the Biological Soci ety of Washington 10: 269-273 Lapointe BE, Matzie WR, Barile PJ (2002) Biotic phase shifts in the Florida Bay and fore reef communities of the Florida Keys: linkages with historical flows and nitrogen loading from Everglades runoff, in the Everglades, Florida Bay, and coral reefs of the Florida Keys: an Ecosystem Sourcebook. Porter JW and Porter KG, Eds, CRC Press, Boca Raton 629-648 Lindeman KC, Kramer PA, Jerald SA (2001) Co mparative approaches to reef monitoring and assessment: an overview. Bulletin of Marine Science 69(2): 335-337

PAGE 45

37 Marszalek DS, Babashoff G, Noel MR, Worl ey DR (1977) Reef distribution in South Florida. In: Proceedings of the 3rd International Coral Reef Symposium, Miami 2: 223229 McCormick MI (1994) Comparison of field methods for measuring surface topography and their associations with a tropical reef fish assembla ge. Marine Ecology Progress Series 112: 87-96 McField MD, Hallock P, Jaap WC (2001) Mu ltivariate analysis of reef community structure in the Belize barrier reef complex. Bulletin of Marine Science 69(2): 745-758 Moberg F, Folke C (1999) Ecological goods a nd services and coral reef ecosystems. Ecological Economics 29:215-233 Mumby PJ, Green EP, Edwards AJ, et al. ( 1997) Coral reef habita t-mapping: how much detail can remote sensing pr ovide? Marine Biology 130: 193-202 Mumby PJ, Harborne AR (1999) Development of a systematic classification scheme of marine habitats to facilitate regional manageme nt and mapping of Caribbean coral reefs. Biological Conservation 88: 155-163 Ohlhorst SL, Liddell WD, Taylor RJ, Taylor JM (1988) Evaluation of reef census techniques. In: Proceedings of the 6th International Coral Reef Symposium, Australia 2: 319-324 Palandro D, Andrefouet S, Muller-Karger FE, Dustan P, Hu C, Hallock P (2003) Detection of changes in coral reef comm unities using Landsat-5 TM and Landsat-7 ETM+ data. Canadian Journal of Remote Sensing 29(2): 201-209 Riegl BR, Korrubel JL, Martin C (2001) Mapping and monitoring of coral communities and their spatial patterns usi ng a surface-based video method from a vessel. Bulletin of Marine Science 69(2): 869-880 Rogers CS, Miller J (2001) Coral bleachi ng, hurricane damage, and benthic cover on coral reefs in St. John, U.S. Virgin Islands : a comparison of surveys with the chain transect method and videography. Bulleti n of Marine Science 69(2):459-470. Rogers CS, Miller J, Waara RJ (2002) Track ing changes on a reef in the US Virgin Islands with videography a nd sonar: a new approach. In: Proceedings of the 9th International Coral Reef Sym posium, Bali, Indonesia 2:1065-1069 Sallenger AH, Krabill WB, Swift RN, Brock J, List J, Hansen M, Holman RA, Manizade S, Sontag J, Meredith A, Morgan K, Yunkel JK, Frederick EB, Stockdon H (2003) Evaluation of airborne topogr aphic Lidar for quantifying beach changes. Journal of Coastal Research 19(1): 125-133

PAGE 46

38 Sokal RS, Rolph FJ (1995) Biometry 3rd ed. Freeman WH and Co mpany, New York Szmant AM (1997) Nutrient effects on co ral reefs: a hypothesis on the importance of topographic and trophic complexity to reef nutrient dynamics. In: Proceedings of the 8th International Coral Reef Sy mposium, Panama 2: 1527-1532 Wright W, Brock JC, Wright (2002) EAARL: A Lidar for mapping shallow coral reefs and other coastal environemtns. Presented at the 7th International Conference on Remote Sensing for Marine and Coastal Environments, Miami, Florida

PAGE 47

39 APPENDICES

PAGE 48

40 Diver data Reef 1 Reef 2 Reef 3 Reef 4 Reef 5 Benthic category Transect 1 Transect 2 Transect 1 Transect 2 Transect 1 Transect 2 Trans ect 1 Transect 2 Tran sect 1 Transect 2 Acropora cervicornis 0 0 0 0 2 0 0 0 0 0 Boulder 7 10 24 3 6 1 2 18 7 4 Branching 4 0 0 0 0 0 0 6 3 2 Unidentifiable 311 366 231 323 213 48 90 515 207 68 Water column 0 0 0 0 3 0 0 0 0 0 Macroalgae 104 32 27 62 43 34 12 17 56 118 Millepora alcicornis 0 0 0 0 0 0 0 0 2 0 Millepora complanata 0 0 1 0 0 0 0 0 0 0 Octocorallia 140 154 127 137 118 25 181 419 260 209 Plate 1 0 0 0 0 0 0 1 11 9 Porifera 4 2 1 0 0 1 0 0 13 1 Porites porites 0 0 0 0 4 0 0 0 0 0 Housing portal 17 5 21 12 10 7 4 16 16 4 Seagrass 0 2 7 2 0 10 0 9 7 0 Pvc pipe 1 0 0 1 0 1 0 2 3 0 Substrate 251 253 776 374 636 68 101 467 495 245 Zoanthidea 0 0 0 0 0 0 0 0 0 0 Total 840 824 1215 914 1035 195 390 1470 1080 660 Appendix I. Diver-acquired benthic ca tegory frequencies for each transect on each of the five reef sites.

PAGE 49

41 ATRIS data Reef 1 Reef 2 Reef 3 Reef 4 Reef 5 Benthic category Transect 1 Transect 2 Transect 1 Transect 2 Transect 1 Transect 2 Trans ect 1 Transect 2 Tran sect 1 Transect 2 Acropora cervicornis 0 0 6 11 5 0 0 0 0 0 Boulder 12 57 6 4 2 1 4 21 23 21 Image margin 32 16 46 13 12 8 19 40 49 31 Diploria strigosa 1 0 0 0 1 2 0 5 8 4 Unidentifiable 800 751 106 167 97 152 196 241 128 219 Fish 2 2 1 1 0 1 0 0 2 2 Macroalgae 599 259 520 346 353 264 197 170 662 334 Millepora alcicornis 4 5 16 9 0 2 3 10 32 8 Montastraea cavernosa 12 6 13 6 5 12 30 0 49 1 Octocorallia 594 499 582 426 173 179 390 416 1149 880 Plate 2 2 0 1 0 0 1 12 6 1 Porifera 12 2 9 0 3 10 3 9 15 9 Porites astreoides 4 1 13 0 1 1 1 5 28 6 Porites porites 0 0 4 0 0 0 2 5 2 0 Seagrass 0 58 41 14 22 23 0 33 43 0 Siderastrea radians 4 0 8 3 0 0 0 1 1 0 Substrate 1121 1462 1521 937 675 595 538 882 2062 1141 Zoanthidea 1 0 8 12 1 0 6 0 40 36 Total 3200 3120 2900 1950 1350 1250 1390 1850 4299 2693 Appendix II. ATRIS-acquired benthic category frequencies for each transect on each of the five reef sites.

PAGE 50

42 Benthic class N mean stdev cv P25 P50 P97_5 Plate coral 20 0.18 0.38 207.20 0.00 0.04 1.36 Zoanthids 20 0.19 0.37 200.76 0.00 0.00 1.34 Sponge 20 0.29 0.31 104.95 0.03 0.23 1.20 Branching coral 20 0.37 0.32 87.37 0.10 0.33 1.03 Seagrass 20 0.88 1.22 137.60 0.00 0.59 5.13 Boulder coral 20 1.17 0.67 57.37 0.63 1.11 2.54 Other 20 1.37 0.80 58.48 0.81 1.22 4.10 Macroalgae 20 11.78 7.24 61.49 4.67 12.39 26.25 Unidentifiable 20 19.38 11.80 60.88 9.46 19.09 44.42 Octocoral 20 21.47 8.93 41.60 14.65 19.40 46.41 Substrate 20 42.98 10.17 23.67 34.95 44.10 63.87 Appendix III. Data utilized to create lower, mean and upper quartile benthi c category distributi ons for ATRIS and diver-acquired data sets combined; figure 14.

PAGE 51

43 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 1 1.05 Substrate 50 8.22 1 1 1.03 Substrate 48 8.48 1 1 1.04 Substrate 41 8.33 1 1 1.05 Substrate 33 8.26 1 1 1.08 Substrate 38 7.98 1 1 1.08 Substrate 38 7.98 1 1 1.20 Substrate 22 7.12 1 1 1.14 Substrate 18 6.94 1 1 1.09 Substrate 18 7.04 1 1 1.11 Substrate 14 6.5 1 1 1.09 Substrate 16 6.6 1 1 1.16 Substrate 19 6.34 1 1 1.16 Substrate 16 6.41 1 1 1.16 Substrate 21 6.41 1 1 1.14 Substrate 20 6.08 1 1 1.20 Substrate 26 5.76 1 1 1.08 Substrate 18 5.99 1 1 1.09 Substrate 21 5.56 1 1 1.12 Substrate 19 5.7 1 1 1.27 Substrate 26 6.3 1 1 1.27 Substrate 14 6.3 1 1 1.22 Substrate 21 5.31 1 1 1.22 Substrate 12 5.31 1 1 1.07 Substrate 17 5.44 1 1 1.12 Substrate 23 5.28 1 1 1.12 Substrate 16 5.28 1 1 1.07 Substrate 21 4.82 1 1 1.07 Substrate 9 4.89 1 1 1.04 Substrate 6 4.53 1 1 1.04 Substrate 15 4.34 1 1 1.05 Substrate 22 4.54 1 1 1.18 Substrate 13 4.6 1 1 1.10 Substrate 24 3.96 1 1 1.02 Substrate 18 4.09 1 1 1.02 Substrate 18 4.09 1 1 1.04 Substrate 21 4.22 1 1 1.06 Substrate 15 3.88 1 1 1.07 Substrate 14 3.82 1 1 1.03 Substrate 15 4.09 1 1 1.06 Substrate 17 4.23 1 1 1.06 Substrate 20 4.18 1 1 1.18 Substrate 14 5.36 Appendix IV. Raw data used for regression anal ysis of benthic category and depth versus rugosity indices.

PAGE 52

44 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 1 1.09 Substrate 10 5.58 1 1 1.10 Substrate 14 5.92 1 1 1.16 Substrate 23 6.01 1 1 1.17 Substrate 17 6.63 1 1 1.11 Substrate 20 7.22 1 1 1.11 Substrate 17 6.84 1 1 1.02 Substrate 34 8.73 1 1 1.06 Substrate 50 8.89 1 1 1.03 Substrate 50 8.72 1 1 1.03 Macroalgae 2 8.48 1 1 1.04 Macroalgae 8 8.33 1 1 1.05 Macroalgae 1 8.26 1 1 1.08 Macroalgae 1 7.98 1 1 1.08 Macroalgae 5 7.98 1 1 1.20 Macroalgae 18 7.12 1 1 1.14 Macroalgae 28 6.94 1 1 1.09 Macroalgae 17 7.04 1 1 1.11 Macroalgae 17 6.5 1 1 1.09 Macroalgae 13 6.6 1 1 1.16 Macroalgae 12 6.34 1 1 1.16 Macroalgae 21 6.41 1 1 1.16 Macroalgae 8 6.41 1 1 1.14 Macroalgae 15 6.08 1 1 1.20 Macroalgae 7 5.76 1 1 1.08 Macroalgae 16 5.99 1 1 1.09 Macroalgae 12 5.56 1 1 1.12 Macroalgae 8 5.7 1 1 1.27 Macroalgae 5 6.3 1 1 1.27 Macroalgae 9 6.3 1 1 1.22 Macroalgae 8 5.31 1 1 1.22 Macroalgae 17 5.31 1 1 1.07 Macroalgae 10 5.44 1 1 1.12 Macroalgae 9 5.28 1 1 1.12 Macroalgae 16 5.28 1 1 1.07 Macroalgae 15 4.82 1 1 1.07 Macroalgae 24 4.89 1 1 1.04 Macroalgae 23 4.53 1 1 1.04 Macroalgae 17 4.34 1 1 1.05 Macroalgae 16 4.54 1 1 1.18 Macroalgae 15 4.6 1 1 1.10 Macroalgae 8 3.96 1 1 1.02 Macroalgae 9 4.09 Appendix IV. Raw data used for regression analysis (continued)

PAGE 53

45 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 1 1.02 Macroalgae 16 4.09 1 1 1.04 Macroalgae 21 4.22 1 1 1.06 Macroalgae 10 3.88 1 1 1.07 Macroalgae 7 3.82 1 1 1.03 Macroalgae 12 4.09 1 1 1.06 Macroalgae 16 4.23 1 1 1.06 Macroalgae 12 4.18 1 1 1.18 Macroalgae 12 5.36 1 1 1.09 Macroalgae 11 5.58 1 1 1.10 Macroalgae 12 5.92 1 1 1.16 Macroalgae 8 6.01 1 1 1.17 Macroalgae 18 6.63 1 1 1.11 Macroalgae 10 7.22 1 1 1.11 Macroalgae 19 6.84 1 1 1.04 Octocoral 1 8.33 1 1 1.05 Octocoral 2 8.26 1 1 1.08 Octocoral 1 7.98 1 1 1.14 Octocoral 2 6.94 1 1 1.09 Octocoral 8 7.04 1 1 1.11 Octocoral 6 6.5 1 1 1.09 Octocoral 9 6.6 1 1 1.16 Octocoral 7 6.34 1 1 1.16 Octocoral 8 6.41 1 1 1.16 Octocoral 19 6.41 1 1 1.14 Octocoral 8 6.08 1 1 1.20 Octocoral 11 5.76 1 1 1.08 Octocoral 12 5.99 1 1 1.09 Octocoral 13 5.56 1 1 1.12 Octocoral 17 5.7 1 1 1.27 Octocoral 13 6.3 1 1 1.27 Octocoral 23 6.3 1 1 1.22 Octocoral 14 5.31 1 1 1.22 Octocoral 15 5.31 1 1 1.07 Octocoral 21 5.44 1 1 1.12 Octocoral 11 5.28 1 1 1.12 Octocoral 16 5.28 1 1 1.07 Octocoral 10 4.82 1 1 1.07 Octocoral 17 4.89 1 1 1.04 Octocoral 21 4.53 1 1 1.04 Octocoral 16 4.34 1 1 1.05 Octocoral 12 4.54 1 1 1.18 Octocoral 14 4.6 Appendix IV. Raw data used for regression analysis (continued)

PAGE 54

46 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 1 1.10 Octocoral 15 3.96 1 1 1.02 Octocoral 18 4.09 1 1 1.02 Octocoral 15 4.09 1 1 1.04 Octocoral 8 4.22 1 1 1.06 Octocoral 22 3.88 1 1 1.07 Octocoral 23 3.82 1 1 1.03 Octocoral 16 4.09 1 1 1.06 Octocoral 17 4.23 1 1 1.06 Octocoral 14 4.18 1 1 1.18 Octocoral 21 5.36 1 1 1.09 Octocoral 26 5.58 1 1 1.10 Octocoral 21 5.92 1 1 1.16 Octocoral 14 6.01 1 1 1.17 Octocoral 13 6.63 1 1 1.11 Octocoral 13 7.22 1 1 1.11 Octocoral 12 6.84 1 1 1.05 Unidentifiable 14 8.26 1 1 1.08 Unidentifiable 10 7.98 1 1 1.08 Unidentifiable 6 7.98 1 1 1.12 Unidentifiable 50 8.01 1 1 1.12 Unidentifiable 50 8.01 1 1 1.12 Unidentifiable 50 8.01 1 1 1.20 Unidentifiable 50 7.46 1 1 1.20 Unidentifiable 50 7.46 1 1 1.20 Unidentifiable 50 7.46 1 1 1.28 Unidentifiable 50 8.06 1 1 1.20 Unidentifiable 8 7.12 1 1 1.14 Unidentifiable 2 6.94 1 1 1.14 Unidentifiable 50 6.94 1 1 1.09 Unidentifiable 6 7.04 1 1 1.11 Unidentifiable 8 6.5 1 1 1.09 Unidentifiable 9 6.6 1 1 1.16 Unidentifiable 8 6.34 1 1 1.16 Unidentifiable 3 6.41 1 1 1.16 Unidentifiable 2 6.41 1 1 1.14 Unidentifiable 7 6.08 1 1 1.20 Unidentifiable 3 5.76 1 1 1.08 Unidentifiable 2 5.99 1 1 1.09 Unidentifiable 4 5.56 1 1 1.27 Unidentifiable 4 6.3 1 1 1.27 Unidentifiable 1 6.3 1 1 1.22 Unidentifiable 2 5.31 Appendix IV. Raw data used for regression analysis (continued)

PAGE 55

47 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 1 1.22 Unidentifiable 2 5.31 1 1 1.07 Unidentifiable 2 5.44 1 1 1.12 Unidentifiable 2 5.28 1 1 1.12 Unidentifiable 1 5.28 1 1 1.07 Unidentifiable 2 4.82 1 1 1.04 Unidentifiable 2 4.34 1 1 1.02 Unidentifiable 4 4.09 1 1 1.06 Unidentifiable 2 3.88 1 1 1.07 Unidentifiable 5 3.82 1 1 1.03 Unidentifiable 4 4.09 1 1 1.18 Unidentifiable 3 5.36 1 1 1.10 Unidentifiable 3 5.92 1 1 1.16 Unidentifiable 1 6.01 1 1 1.11 Unidentifiable 4 7.22 1 1 1.11 Unidentifiable 1 6.84 1 1 1.13 Unidentifiable 50 7.34 1 1 1.17 Unidentifiable 50 7.8 1 1 1.11 Unidentifiable 50 8.3 1 1 1.11 Unidentifiable 50 8.3 1 1 1.05 Unidentifiable 50 8.72 1 1 1.02 Unidentifiable 16 8.73 1 1 1.08 Other 1 7.98 1 1 1.20 Other 2 7.12 1 1 1.09 Other 2 7.04 1 1 1.09 Other 1 6.6 1 1 1.16 Other 2 6.34 1 1 1.16 Other 1 6.41 1 1 1.20 Other 2 5.76 1 1 1.12 Other 2 5.7 1 1 1.27 Other 2 6.3 1 1 1.27 Other 1 6.3 1 1 1.22 Other 4 5.31 1 1 1.22 Other 3 5.31 1 1 1.12 Other 2 5.28 1 1 1.07 Other 1 4.82 1 1 1.18 Other 1 4.6 1 1 1.10 Other 1 3.96 1 1 1.02 Other 1 4.09 1 1 1.02 Other 1 4.09 1 1 1.07 Other 1 3.82 1 1 1.06 Other 1 4.18 1 1 1.09 Other 1 5.58 Appendix IV. Raw data used for regression analysis (continued)

PAGE 56

48 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 1 1.16 Other 1 6.01 1 1 1.11 Other 1 6.84 1 1 1.09 Boulder coral 1 7.04 1 1 1.11 Boulder coral 4 6.5 1 1 1.09 Boulder coral 2 6.6 1 1 1.20 Boulder coral 1 5.76 1 1 1.08 Boulder coral 2 5.99 1 1 1.12 Boulder coral 3 5.7 1 1 1.22 Boulder coral 1 5.31 1 1 1.12 Boulder coral 3 5.28 1 1 1.07 Boulder coral 1 4.82 1 1 1.18 Boulder coral 4 4.6 1 1 1.06 Boulder coral 3 4.18 1 1 1.09 Boulder coral 2 5.58 1 1 1.11 Boulder coral 3 7.22 1 1 1.10 Branching coral 2 3.96 1 1 1.03 Branching coral 2 4.09 1 1 1.12 Plate coral 1 5.7 1 1 1.27 Plate coral 2 6.3 1 1 1.17 Plate coral 2 6.63 1 1 1.11 Sponge 1 6.5 1 1 1.16 Sponge 2 6.34 1 1 1.16 Sponge 1 6.41 1 1 1.22 Sponge 1 5.31 1 1 1.12 Sponge 1 5.28 1 1 1.18 Sponge 2 4.6 1 1 1.06 Sponge 1 3.88 1 1 1.16 Sponge 3 6.01 1 1 1.18 Zoanthid 1 4.6 1 2 1.06 Macroalgae 2 6.89 1 2 1.06 Macroalgae 2 6.89 1 2 1.01 Macroalgae 1 6.76 1 2 1.06 Macroalgae 1 6.48 1 2 1.04 Macroalgae 15 6.19 1 2 1.02 Macroalgae 7 6.2 1 2 1.02 Macroalgae 14 6.2 1 2 1.08 Macroalgae 11 5.89 1 2 1.06 Macroalgae 12 5.62 1 2 1.06 Macroalgae 13 5.24 1 2 1.06 Macroalgae 9 5.24 1 2 1.03 Macroalgae 8 5.37 1 2 1.03 Macroalgae 14 5.2 Appendix IV. Raw data used for regression analysis (continued)

PAGE 57

49 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 2 1.05 Macroalgae 5 5.4 1 2 1.03 Macroalgae 7 5.26 1 2 1.03 Macroalgae 18 5.41 1 2 1.04 Macroalgae 9 5.85 1 2 1.06 Macroalgae 3 6.14 1 2 1.09 Macroalgae 11 6.16 1 2 1.16 Macroalgae 4 6.14 1 2 1.16 Macroalgae 1 6.93 1 2 1.02 Macroalgae 2 7.85 1 2 1.02 Macroalgae 6 7.74 1 2 1.02 Macroalgae 1 7.83 1 2 1.04 Macroalgae 5 7.81 1 2 1.06 Macroalgae 8 7.34 1 2 1.12 Macroalgae 5 6.67 1 2 1.13 Macroalgae 1 6.55 1 2 1.05 Macroalgae 4 5.7 1 2 1.11 Macroalgae 3 5.48 1 2 1.09 Macroalgae 7 4.94 1 2 1.08 Macroalgae 3 4.49 1 2 1.07 Macroalgae 10 4.36 1 2 1.14 Macroalgae 16 4.76 1 2 1.08 Macroalgae 1 4.77 1 2 1.06 Macroalgae 3 4.41 1 2 1.08 Macroalgae 1 4.43 1 2 1.10 Macroalgae 1 5.59 1 2 1.05 Macroalgae 5 5.28 1 2 1.06 Macroalgae 1 5.69 1 2 1.10 Macroalgae 7 5.97 1 2 1.18 Macroalgae 2 6.64 1 2 1.06 Substrate 28 6.89 1 2 1.06 Substrate 29 6.89 1 2 1.01 Substrate 33 6.76 1 2 1.01 Substrate 28 6.76 1 2 1.06 Substrate 31 6.48 1 2 1.03 Substrate 30 6.4 1 2 1.04 Substrate 29 6.19 1 2 1.02 Substrate 34 6.2 1 2 1.02 Substrate 23 6.2 1 2 1.08 Substrate 21 5.89 1 2 1.06 Substrate 15 5.62 1 2 1.06 Substrate 20 5.24 1 2 1.06 Substrate 23 5.24 Appendix IV. Raw data used for regression analysis (continued)

PAGE 58

50 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 2 1.03 Substrate 25 5.37 1 2 1.03 Substrate 24 5.2 1 2 1.05 Substrate 27 5.4 1 2 1.03 Substrate 22 5.26 1 2 1.03 Substrate 19 5.41 1 2 1.04 Substrate 21 5.85 1 2 1.06 Substrate 20 6.14 1 2 1.09 Substrate 27 6.16 1 2 1.16 Substrate 31 6.14 1 2 1.16 Substrate 34 6.93 1 2 1.01 Substrate 50 6.88 1 2 1.01 Substrate 50 6.81 1 2 1.05 Substrate 50 8.14 1 2 1.02 Substrate 48 7.85 1 2 1.02 Substrate 41 7.74 1 2 1.02 Substrate 39 7.83 1 2 1.04 Substrate 28 7.81 1 2 1.06 Substrate 15 7.34 1 2 1.12 Substrate 37 6.67 1 2 1.13 Substrate 16 6.55 1 2 1.05 Substrate 17 5.7 1 2 1.11 Substrate 22 5.48 1 2 1.09 Substrate 1 4.94 1 2 1.08 Substrate 16 4.49 1 2 1.07 Substrate 19 4.36 1 2 1.14 Substrate 20 4.76 1 2 1.08 Substrate 42 4.77 1 2 1.06 Substrate 16 4.41 1 2 1.08 Substrate 19 4.43 1 2 1.10 Substrate 20 5.59 1 2 1.05 Substrate 11 5.28 1 2 1.06 Substrate 14 5.69 1 2 1.10 Substrate 17 5.97 1 2 1.18 Substrate 14 6.64 1 2 1.05 Substrate 40 8.48 1 2 1.07 Substrate 49 8.42 1 2 1.06 Substrate 46 8.29 1 2 1.02 Substrate 50 8.35 1 2 1.01 Substrate 50 8.41 1 2 1.06 Octocoral 1 6.89 1 2 1.06 Octocoral 1 6.89 1 2 1.01 Octocoral 2 6.76 Appendix IV. Raw data used for regression analysis (continued)

PAGE 59

51 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 2 1.01 Octocoral 8 6.76 1 2 1.06 Octocoral 6 6.48 1 2 1.04 Octocoral 6 6.19 1 2 1.02 Octocoral 8 6.2 1 2 1.02 Octocoral 7 6.2 1 2 1.08 Octocoral 15 5.89 1 2 1.06 Octocoral 18 5.62 1 2 1.06 Octocoral 12 5.24 1 2 1.06 Octocoral 18 5.24 1 2 1.03 Octocoral 12 5.37 1 2 1.03 Octocoral 10 5.2 1 2 1.05 Octocoral 15 5.4 1 2 1.03 Octocoral 15 5.26 1 2 1.03 Octocoral 10 5.41 1 2 1.04 Octocoral 15 5.85 1 2 1.06 Octocoral 14 6.14 1 2 1.09 Octocoral 5 6.16 1 2 1.16 Octocoral 5 6.14 1 2 1.16 Octocoral 5 6.93 1 2 1.02 Octocoral 2 7.74 1 2 1.02 Octocoral 3 7.83 1 2 1.04 Octocoral 3 7.81 1 2 1.06 Octocoral 9 7.34 1 2 1.12 Octocoral 7 6.67 1 2 1.13 Octocoral 13 6.55 1 2 1.05 Octocoral 15 5.7 1 2 1.11 Octocoral 12 5.48 1 2 1.09 Octocoral 15 4.94 1 2 1.08 Octocoral 30 4.49 1 2 1.07 Octocoral 6 4.36 1 2 1.14 Octocoral 12 4.76 1 2 1.08 Octocoral 4 4.77 1 2 1.06 Octocoral 13 4.41 1 2 1.08 Octocoral 9 4.43 1 2 1.10 Octocoral 19 5.59 1 2 1.05 Octocoral 5 5.28 1 2 1.06 Octocoral 11 5.69 1 2 1.10 Octocoral 18 5.97 1 2 1.18 Octocoral 23 6.64 1 2 1.07 Octocoral 1 8.42 1 2 1.06 Octocoral 1 8.29 1 2 1.06 Unidentifiable 1 6.89 Appendix IV. Raw data used for regression analysis (continued)

PAGE 60

52 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 2 1.06 Unidentifiable 4 6.89 1 2 1.01 Unidentifiable 2 6.76 1 2 1.01 Unidentifiable 10 6.76 1 2 1.06 Unidentifiable 12 6.48 1 2 1.03 Unidentifiable 17 6.4 1 2 1.02 Unidentifiable 1 6.2 1 2 1.02 Unidentifiable 4 6.2 1 2 1.08 Unidentifiable 3 5.89 1 2 1.06 Unidentifiable 3 5.62 1 2 1.06 Unidentifiable 3 5.24 1 2 1.03 Unidentifiable 2 5.37 1 2 1.03 Unidentifiable 1 5.2 1 2 1.05 Unidentifiable 1 5.4 1 2 1.03 Unidentifiable 1 5.26 1 2 1.04 Unidentifiable 5 5.85 1 2 1.06 Unidentifiable 12 6.14 1 2 1.09 Unidentifiable 4 6.16 1 2 1.16 Unidentifiable 8 6.14 1 2 1.16 Unidentifiable 1 6.93 1 2 1.02 Unidentifiable 1 7.74 1 2 1.02 Unidentifiable 6 7.83 1 2 1.04 Unidentifiable 14 7.81 1 2 1.06 Unidentifiable 18 7.34 1 2 1.13 Unidentifiable 6 6.55 1 2 1.16 Unidentifiable 50 5.96 1 2 1.05 Unidentifiable 10 5.7 1 2 1.11 Unidentifiable 8 5.48 1 2 1.09 Unidentifiable 4 4.94 1 2 1.07 Unidentifiable 15 4.36 1 2 1.14 Unidentifiable 2 4.76 1 2 1.06 Unidentifiable 14 4.41 1 2 1.08 Unidentifiable 19 4.43 1 2 1.07 Unidentifiable 50 4.48 1 2 1.08 Unidentifiable 50 4.43 1 2 1.05 Unidentifiable 50 4.91 1 2 1.10 Unidentifiable 8 5.59 1 2 1.05 Unidentifiable 29 5.28 1 2 1.06 Unidentifiable 21 5.69 1 2 1.10 Unidentifiable 8 5.97 1 2 1.18 Unidentifiable 10 6.64 1 2 1.09 Unidentifiable 50 6.8 1 2 1.14 Unidentifiable 50 7.09 Appendix IV. Raw data used for regression analysis (continued)

PAGE 61

53 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 2 1.10 Unidentifiable 50 7.4 1 2 1.13 Unidentifiable 50 8 1 2 1.05 Unidentifiable 50 8.04 1 2 1.05 Unidentifiable 10 8.48 1 2 1.06 Unidentifiable 3 8.29 1 2 1.06 Other 1 6.89 1 2 1.01 Other 1 6.76 1 2 1.03 Other 1 6.4 1 2 1.02 Other 2 6.2 1 2 1.06 Other 1 5.62 1 2 1.06 Other 1 5.24 1 2 1.03 Other 1 5.37 1 2 1.03 Other 1 5.26 1 2 1.03 Other 1 5.41 1 2 1.09 Other 1 6.16 1 2 1.05 Other 2 5.7 1 2 1.11 Other 3 5.48 1 2 1.06 Other 1 5.69 1 2 1.18 Other 1 6.64 1 2 1.06 Boulder coral 1 5.24 1 2 1.05 Boulder coral 2 5.4 1 2 1.03 Boulder coral 3 5.26 1 2 1.03 Boulder coral 1 5.41 1 2 1.06 Boulder coral 1 6.14 1 2 1.09 Boulder coral 2 6.16 1 2 1.02 Boulder coral 1 7.83 1 2 1.12 Boulder coral 1 6.67 1 2 1.13 Boulder coral 14 6.55 1 2 1.05 Boulder coral 2 5.7 1 2 1.11 Boulder coral 2 5.48 1 2 1.09 Boulder coral 23 4.94 1 2 1.08 Boulder coral 3 4.77 1 2 1.06 Boulder coral 2 4.41 1 2 1.08 Boulder coral 1 4.43 1 2 1.10 Boulder coral 2 5.59 1 2 1.06 Boulder coral 2 5.69 1 2 1.03 Branching coral 1 5.37 1 2 1.08 Branching coral 1 4.49 1 2 1.06 Branching coral 2 4.41 1 2 1.08 Branching coral 1 4.43 1 2 1.06 Plate coral 1 5.62 1 2 1.03 Plate coral 1 5.37 Appendix IV. Raw data used for regression analysis (continued)

PAGE 62

54 Reef Transect Rugosity index Benthic category Frequency Depth (m) 1 2 1.03 Plate coral 1 5.2 1 2 1.03 Sponge 1 5.26 1 2 1.03 Sponge 1 5.41 1 2 1.06 Seagrass 18 6.89 1 2 1.06 Seagrass 13 6.89 1 2 1.01 Seagrass 12 6.76 1 2 1.01 Seagrass 3 6.76 1 2 1.03 Seagrass 1 6.4 1 2 1.16 Seagrass 2 6.14 1 2 1.16 Seagrass 9 6.93 2 1 1.01 Substrate 28 7.48 2 1 1.00 Substrate 35 7.4 2 1 1.01 Substrate 34 7.34 2 1 1.01 Substrate 25 7.15 2 1 1.03 Substrate 38 7.36 2 1 1.05 Substrate 27 6.72 2 1 1.09 Substrate 16 6.6 2 1 1.06 Substrate 20 6.15 2 1 1.08 Substrate 14 5.57 2 1 1.03 Substrate 20 5.5 2 1 1.07 Substrate 18 5.36 2 1 1.06 Substrate 22 5.08 2 1 1.08 Substrate 23 5.06 2 1 1.12 Substrate 19 5.13 2 1 1.11 Substrate 33 4.92 2 1 1.04 Substrate 35 4.5 2 1 1.04 Substrate 24 4.3 2 1 1.04 Substrate 13 4.44 2 1 1.04 Substrate 28 4.28 2 1 1.02 Substrate 26 4.12 2 1 1.01 Substrate 26 3.82 2 1 1.02 Substrate 30 3.73 2 1 1.02 Substrate 22 3.83 2 1 1.03 Substrate 16 3.9 2 1 1.01 Substrate 32 3.72 2 1 1.01 Substrate 26 3.65 2 1 1.01 Substrate 22 3.74 2 1 1.02 Substrate 29 3.76 2 1 1.03 Substrate 19 3.83 2 1 1.02 Substrate 24 3.66 2 1 1.04 Substrate 29 3.52 2 1 1.06 Substrate 12 3.91 Appendix IV. Raw data used for regression analysis (continued)

PAGE 63

55 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 1 1.05 Substrate 28 4.01 2 1 1.01 Substrate 25 3.58 2 1 1.02 Substrate 22 3.4 2 1 1.05 Substrate 28 3.54 2 1 1.04 Substrate 22 3.96 2 1 1.03 Substrate 31 3.9 2 1 1.04 Substrate 29 4.21 2 1 1.06 Substrate 20 4.09 2 1 1.07 Substrate 21 4.43 2 1 1.10 Substrate 31 4.37 2 1 1.09 Substrate 30 4.09 2 1 1.05 Substrate 22 4.32 2 1 1.03 Substrate 27 4.46 2 1 1.07 Substrate 26 4.68 2 1 1.10 Substrate 18 4.67 2 1 1.14 Substrate 21 4.97 2 1 1.12 Substrate 33 4.71 2 1 1.07 Substrate 27 4.94 2 1 1.06 Substrate 25 5.18 2 1 1.05 Substrate 29 5.28 2 1 1.21 Substrate 33 6.14 2 1 1.15 Substrate 27 6.46 2 1 1.12 Substrate 28 6.97 2 1 1.04 Substrate 38 6.86 2 1 1.08 Substrate 50 7.83 2 1 1.04 Substrate 50 7.62 2 1 1.00 Macroalgae 3 7.4 2 1 1.01 Macroalgae 1 7.34 2 1 1.01 Macroalgae 1 7.15 2 1 1.03 Macroalgae 4 7.36 2 1 1.05 Macroalgae 7 6.72 2 1 1.09 Macroalgae 11 6.6 2 1 1.06 Macroalgae 16 6.15 2 1 1.08 Macroalgae 22 5.57 2 1 1.03 Macroalgae 19 5.5 2 1 1.07 Macroalgae 15 5.36 2 1 1.06 Macroalgae 21 5.08 2 1 1.08 Macroalgae 13 5.06 2 1 1.12 Macroalgae 21 5.13 2 1 1.11 Macroalgae 5 4.92 2 1 1.04 Macroalgae 3 4.5 2 1 1.04 Macroalgae 11 4.3 Appendix IV. Raw data used for regression analysis (continued)

PAGE 64

56 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 1 1.04 Macroalgae 7 4.44 2 1 1.04 Macroalgae 9 4.28 2 1 1.02 Macroalgae 8 4.12 2 1 1.01 Macroalgae 10 3.82 2 1 1.02 Macroalgae 7 3.73 2 1 1.02 Macroalgae 12 3.83 2 1 1.03 Macroalgae 8 3.9 2 1 1.01 Macroalgae 3 3.72 2 1 1.01 Macroalgae 4 3.65 2 1 1.01 Macroalgae 10 3.74 2 1 1.02 Macroalgae 10 3.76 2 1 1.03 Macroalgae 9 3.83 2 1 1.02 Macroalgae 4 3.66 2 1 1.04 Macroalgae 5 3.52 2 1 1.06 Macroalgae 11 3.91 2 1 1.05 Macroalgae 7 4.01 2 1 1.01 Macroalgae 4 3.58 2 1 1.02 Macroalgae 16 3.4 2 1 1.05 Macroalgae 4 3.54 2 1 1.04 Macroalgae 16 3.96 2 1 1.03 Macroalgae 6 3.9 2 1 1.04 Macroalgae 2 4.21 2 1 1.06 Macroalgae 10 4.09 2 1 1.07 Macroalgae 16 4.43 2 1 1.10 Macroalgae 9 4.37 2 1 1.09 Macroalgae 6 4.09 2 1 1.05 Macroalgae 18 4.32 2 1 1.03 Macroalgae 16 4.46 2 1 1.07 Macroalgae 15 4.68 2 1 1.10 Macroalgae 17 4.67 2 1 1.14 Macroalgae 18 4.97 2 1 1.12 Macroalgae 12 4.71 2 1 1.07 Macroalgae 9 4.94 2 1 1.06 Macroalgae 8 5.18 2 1 1.05 Macroalgae 6 5.28 2 1 1.21 Macroalgae 2 6.14 2 1 1.15 Macroalgae 5 6.46 2 1 1.12 Macroalgae 3 6.97 2 1 1.04 Macroalgae 4 6.86 2 1 1.00 Octocoral 3 7.4 2 1 1.01 Octocoral 8 7.34 2 1 1.01 Octocoral 1 7.15 Appendix IV. Raw data used for regression analysis (continued)

PAGE 65

57 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 1 1.03 Octocoral 3 7.36 2 1 1.05 Octocoral 5 6.72 2 1 1.09 Octocoral 15 6.6 2 1 1.06 Octocoral 8 6.15 2 1 1.08 Octocoral 11 5.57 2 1 1.03 Octocoral 9 5.5 2 1 1.07 Octocoral 13 5.36 2 1 1.06 Octocoral 7 5.08 2 1 1.08 Octocoral 8 5.06 2 1 1.12 Octocoral 10 5.13 2 1 1.11 Octocoral 8 4.92 2 1 1.04 Octocoral 9 4.5 2 1 1.04 Octocoral 13 4.3 2 1 1.04 Octocoral 19 4.44 2 1 1.04 Octocoral 12 4.28 2 1 1.02 Octocoral 8 4.12 2 1 1.01 Octocoral 9 3.82 2 1 1.02 Octocoral 12 3.73 2 1 1.02 Octocoral 16 3.83 2 1 1.03 Octocoral 18 3.9 2 1 1.01 Octocoral 6 3.72 2 1 1.01 Octocoral 14 3.65 2 1 1.01 Octocoral 11 3.74 2 1 1.02 Octocoral 11 3.76 2 1 1.03 Octocoral 16 3.83 2 1 1.02 Octocoral 16 3.66 2 1 1.04 Octocoral 11 3.52 2 1 1.06 Octocoral 14 3.91 2 1 1.05 Octocoral 13 4.01 2 1 1.01 Octocoral 18 3.58 2 1 1.02 Octocoral 12 3.4 2 1 1.05 Octocoral 17 3.54 2 1 1.04 Octocoral 7 3.96 2 1 1.03 Octocoral 10 3.9 2 1 1.04 Octocoral 16 4.21 2 1 1.06 Octocoral 18 4.09 2 1 1.10 Octocoral 6 4.37 2 1 1.09 Octocoral 12 4.09 2 1 1.05 Octocoral 4 4.32 2 1 1.03 Octocoral 4 4.46 2 1 1.07 Octocoral 5 4.68 2 1 1.10 Octocoral 11 4.67 Appendix IV. Raw data used for regression analysis (continued)

PAGE 66

58 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 1 1.14 Octocoral 9 4.97 2 1 1.12 Octocoral 4 4.71 2 1 1.07 Octocoral 13 4.94 2 1 1.06 Octocoral 13 5.18 2 1 1.05 Octocoral 11 5.28 2 1 1.21 Octocoral 15 6.14 2 1 1.15 Octocoral 10 6.46 2 1 1.12 Octocoral 11 6.97 2 1 1.04 Octocoral 7 6.86 2 1 1.00 Unidentifiable 2 7.4 2 1 1.01 Unidentifiable 15 7.15 2 1 1.03 Unidentifiable 3 7.36 2 1 1.05 Unidentifiable 8 6.72 2 1 1.09 Unidentifiable 3 6.6 2 1 1.06 Unidentifiable 6 6.15 2 1 1.08 Unidentifiable 2 5.57 2 1 1.03 Unidentifiable 1 5.5 2 1 1.07 Unidentifiable 3 5.36 2 1 1.08 Unidentifiable 3 5.06 2 1 1.11 Unidentifiable 2 4.92 2 1 1.04 Unidentifiable 1 4.3 2 1 1.04 Unidentifiable 8 4.44 2 1 1.02 Unidentifiable 1 4.12 2 1 1.03 Unidentifiable 3 3.9 2 1 1.01 Unidentifiable 2 3.72 2 1 1.01 Unidentifiable 1 3.65 2 1 1.01 Unidentifiable 3 3.74 2 1 1.03 Unidentifiable 1 3.83 2 1 1.02 Unidentifiable 1 3.66 2 1 1.04 Unidentifiable 4 3.52 2 1 1.06 Unidentifiable 5 3.91 2 1 1.05 Unidentifiable 1 4.01 2 1 1.04 Unidentifiable 3 3.96 2 1 1.03 Unidentifiable 1 3.9 2 1 1.04 Unidentifiable 1 4.21 2 1 1.07 Unidentifiable 1 4.43 2 1 1.10 Unidentifiable 1 4.37 2 1 1.03 Unidentifiable 2 4.46 2 1 1.07 Unidentifiable 1 4.68 2 1 1.10 Unidentifiable 1 4.67 2 1 1.06 Unidentifiable 3 5.18 2 1 1.05 Unidentifiable 3 5.28 Appendix IV. Raw data used for regression analysis (continued)

PAGE 67

59 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 1 1.15 Unidentifiable 5 6.46 2 1 1.12 Unidentifiable 5 6.97 2 1 1.01 Other 1 7.48 2 1 1.01 Other 2 7.34 2 1 1.01 Other 2 7.15 2 1 1.05 Other 3 6.72 2 1 1.09 Other 1 6.6 2 1 1.08 Other 1 5.57 2 1 1.03 Other 1 5.5 2 1 1.08 Other 3 5.06 2 1 1.11 Other 2 4.92 2 1 1.04 Other 2 4.5 2 1 1.04 Other 1 4.3 2 1 1.04 Other 2 4.44 2 1 1.04 Other 1 4.28 2 1 1.02 Other 3 4.12 2 1 1.01 Other 1 3.82 2 1 1.03 Other 1 3.9 2 1 1.01 Other 1 3.74 2 1 1.03 Other 1 3.83 2 1 1.02 Other 1 3.66 2 1 1.05 Other 1 3.54 2 1 1.04 Other 1 3.96 2 1 1.03 Other 1 3.9 2 1 1.10 Other 1 4.37 2 1 1.03 Other 1 4.46 2 1 1.07 Other 1 4.68 2 1 1.10 Other 1 4.67 2 1 1.14 Other 1 4.97 2 1 1.12 Other 1 4.71 2 1 1.07 Other 1 4.94 2 1 1.05 Other 1 5.28 2 1 1.15 Other 2 6.46 2 1 1.12 Other 1 6.97 2 1 1.04 Other 1 6.86 2 1 1.09 Boulder coral 4 6.6 2 1 1.07 Boulder coral 1 5.36 2 1 1.04 Boulder coral 1 4.44 2 1 1.01 Boulder coral 2 3.82 2 1 1.01 Boulder coral 7 3.72 2 1 1.03 Boulder coral 2 3.83 2 1 1.02 Boulder coral 2 3.66 Appendix IV. Raw data used for regression analysis (continued)

PAGE 68

60 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 1 1.10 Boulder coral 1 4.37 2 1 1.07 Boulder coral 1 4.68 2 1 1.10 Boulder coral 2 4.67 2 1 1.06 Boulder coral 1 5.18 2 1 1.15 Boulder coral 1 6.46 2 1 1.12 Boulder coral 2 6.97 2 1 1.02 Branching coral 4 4.12 2 1 1.01 Branching coral 4 3.65 2 1 1.01 Branching coral 2 3.74 2 1 1.06 Branching coral 2 3.91 2 1 1.01 Branching coral 2 3.58 2 1 1.04 Branching coral 1 3.96 2 1 1.03 Branching coral 1 3.9 2 1 1.04 Branching coral 1 4.21 2 1 1.06 Branching coral 2 4.09 2 1 1.07 Branching coral 1 4.43 2 1 1.05 Branching coral 5 4.32 2 1 1.07 Branching coral 1 4.68 2 1 1.04 Plate coral 1 4.5 2 1 1.01 Plate coral 1 3.82 2 1 1.02 Plate coral 1 3.73 2 1 1.03 Plate coral 1 3.9 2 1 1.01 Plate coral 1 3.74 2 1 1.06 Plate coral 6 3.91 2 1 1.04 Plate coral 1 4.21 2 1 1.09 Plate coral 1 4.09 2 1 1.01 Sponge 1 3.82 2 1 1.03 Sponge 1 3.9 2 1 1.03 Sponge 2 3.83 2 1 1.02 Sponge 2 3.66 2 1 1.04 Sponge 1 3.52 2 1 1.05 Sponge 1 4.01 2 1 1.14 Sponge 1 4.97 2 1 1.01 Seagrass 21 7.48 2 1 1.00 Seagrass 7 7.4 2 1 1.01 Seagrass 5 7.34 2 1 1.01 Seagrass 6 7.15 2 1 1.03 Seagrass 2 7.36 2 1 1.05 Zoanthid 1 4.32 2 2 1.03 Substrate 50 7.37 2 2 1.03 Substrate 34 7.37 2 2 1.14 Substrate 24 6.88 Appendix IV. Raw data used for regression analysis (continued)

PAGE 69

61 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 2 1.06 Substrate 24 6.5 2 2 1.06 Substrate 21 6.26 2 2 1.06 Substrate 18 5.8 2 2 1.07 Substrate 18 5.8 2 2 1.05 Substrate 25 5.1 2 2 1.06 Substrate 15 4.69 2 2 1.04 Substrate 28 4.89 2 2 1.05 Substrate 27 4.6 2 2 1.04 Substrate 24 4.48 2 2 1.08 Substrate 25 4.6 2 2 1.06 Substrate 16 4.1 2 2 1.04 Substrate 21 3.94 2 2 1.02 Substrate 19 3.75 2 2 1.03 Substrate 17 3.88 2 2 1.02 Substrate 19 3.62 2 2 1.02 Substrate 26 3.55 2 2 1.04 Substrate 33 3.8 2 2 1.05 Substrate 36 3.82 2 2 1.02 Substrate 18 3.64 2 2 1.02 Substrate 9 3.55 2 2 1.03 Substrate 23 3.63 2 2 1.03 Substrate 14 3.8 2 2 1.01 Substrate 26 3.59 2 2 1.07 Substrate 12 3.53 2 2 1.08 Substrate 15 3.48 2 2 1.07 Substrate 14 3.84 2 2 1.09 Substrate 22 4.31 2 2 1.03 Substrate 22 4.55 2 2 1.05 Substrate 21 4.6 2 2 1.08 Substrate 32 5.14 2 2 1.25 Substrate 32 5.3 2 2 1.28 Substrate 22 6.66 2 2 1.16 Substrate 37 6.58 2 2 1.07 Substrate 48 6.92 2 2 1.02 Substrate 50 6.89 2 2 1.03 Macroalgae 1 7.37 2 2 1.14 Macroalgae 7 6.88 2 2 1.06 Macroalgae 8 6.5 2 2 1.06 Macroalgae 17 6.26 2 2 1.06 Macroalgae 12 5.8 2 2 1.07 Macroalgae 11 5.8 2 2 1.06 Macroalgae 13 4.69 Appendix IV. Raw data used for regression analysis (continued)

PAGE 70

62 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 2 1.04 Macroalgae 12 4.89 2 2 1.05 Macroalgae 4 4.6 2 2 1.04 Macroalgae 6 4.48 2 2 1.08 Macroalgae 11 4.6 2 2 1.06 Macroalgae 10 4.1 2 2 1.04 Macroalgae 12 3.94 2 2 1.02 Macroalgae 10 3.75 2 2 1.03 Macroalgae 8 3.88 2 2 1.02 Macroalgae 6 3.62 2 2 1.02 Macroalgae 9 3.55 2 2 1.04 Macroalgae 8 3.8 2 2 1.05 Macroalgae 5 3.82 2 2 1.02 Macroalgae 7 3.64 2 2 1.02 Macroalgae 17 3.55 2 2 1.03 Macroalgae 6 3.63 2 2 1.04 Macroalgae 12 4.89 2 2 1.03 Macroalgae 6 3.8 2 2 1.01 Macroalgae 6 3.59 2 2 1.07 Macroalgae 20 3.53 2 2 1.08 Macroalgae 9 3.48 2 2 1.07 Macroalgae 12 3.84 2 2 1.09 Macroalgae 9 4.31 2 2 1.03 Macroalgae 16 4.55 2 2 1.05 Macroalgae 15 4.6 2 2 1.08 Macroalgae 14 5.14 2 2 1.25 Macroalgae 10 5.3 2 2 1.28 Macroalgae 18 6.66 2 2 1.16 Macroalgae 3 6.58 2 2 1.03 Octocoral 1 7.37 2 2 1.14 Octocoral 3 6.88 2 2 1.06 Octocoral 7 6.5 2 2 1.06 Octocoral 7 6.26 2 2 1.06 Octocoral 18 5.8 2 2 1.07 Octocoral 14 5.8 2 2 1.05 Octocoral 9 5.1 2 2 1.06 Octocoral 22 4.69 2 2 1.04 Octocoral 10 4.89 2 2 1.05 Octocoral 16 4.6 2 2 1.04 Octocoral 16 4.48 2 2 1.08 Octocoral 14 4.6 2 2 1.06 Octocoral 23 4.1 2 2 1.04 Octocoral 10 3.94 Appendix IV. Raw data used for regression analysis (continued)

PAGE 71

63 Reef Transect Rugosity index Benthic category Frequency Depth (m) 2 2 1.02 Octocoral 20 3.75 2 2 1.03 Octocoral 22 3.88 2 2 1.02 Octocoral 14 3.62 2 2 1.02 Octocoral 15 3.55 2 2 1.04 Octocoral 10 4.89 2 2 1.05 Octocoral 16 4.6 2 2 1.04 Octocoral 6 3.8 2 2 1.05 Octocoral 6 3.82 2 2 1.02 Octocoral 12 3.64 2 2 1.02 Octocoral 15 3.55 2 2 1.03 Octocoral 17 3.63 2 2 1.03 Octocoral 24 3.8 2 2 1.01 Octocoral 12 3.59 2 2 1.07 Octocoral 16 3.53 2 2 1.08 Octocoral 17 3.48 2 2 1.07 Octocoral 15 3.84 2 2 1.09 Octocoral 8 4.31 2 2 1.03 Octocoral 8 4.55 2 2 1.05 Octocoral 8 4.6 2 2 1.08 Octocoral 4 5.14 2 2 1.25 Octocoral 7 5.3 2 2 1.28 Octocoral 2 6.66 2 2 1.16 Octocoral 7 6.58 2 2 1.07 Octocoral 1 6.92 2 2 1.03 Unidentifiable 1 7.37 2 2 1.14 Unidentifiable 12 6.88 2 2 1.06 Unidentifiable 9 6.5 2 2 1.06 Unidentifiable 2 6.26 2 2 1.06 Unidentifiable 1 5.8 2 2 1.07 Unidentifiable 6 5.8 2 2 1.05 Unidentifiable 6 5.1 2 2 1.05 Unidentifiable 2 4.6 2 2 1.04 Unidentifiable 3 4.48 2 2 1.04 Unidentifiable 6 3.94 2 2 1.03 Unidentifiable 3 3.88 2 2 1.02 Unidentifiable 8 3.62 2 2 1.04 Unidentifiable 2 3.8 2 2 1.02 Unidentifiable 11 3.64 2 2 1.02 Unidentifiable 5 3.55 2 2 1.03 Unidentifiable 4 3.63 2 2 1.03 Unidentifiable 3 3.8 2 2 1.01 Unidentifiable 4 3.59 2 2 1.07 Unidentifiable 1 3.53 2 2 1.08 Unidentifiable 8 3.48 Appendix IV. Raw data used for regression analysis (continued)

PAGE 72

64 Reef Transect Rugosity index Benthic categoryFrequency Depth (m) 2 2 1.07 Unidentifiable 5 3.84 2 2 1.09 Unidentifiable 3 4.31 2 2 1.03 Unidentifiable 2 4.55 2 2 1.05 Unidentifiable 1 4.6 2 2 1.28 Unidentifiable 8 6.66 2 2 1.15 Unidentifiable 50 6.98 2 2 1.16 Unidentifiable 2 6.58 2 2 1.07 Unidentifiable 1 6.92 2 2 1.14 Other 1 6.88 2 2 1.06 Other 1 5.8 2 2 1.05 Other 1 5.1 2 2 1.04 Other 1 4.48 2 2 1.02 Other 1 3.62 2 2 1.05 Other 1 3.82 2 2 1.03 Other 1 3.8 2 2 1.07 Other 1 3.84 2 2 1.03 Other 1 4.55 2 2 1.05 Other 1 4.6 2 2 1.25 Other 1 5.3 2 2 1.14 Boulder coral 2 6.88 2 2 1.06 Boulder coral 2 6.5 2 2 1.06 Boulder coral 3 6.26 2 2 1.07 Boulder coral 1 5.8 2 2 1.05 Boulder coral 1 4.6 2 2 1.02 Boulder coral 1 3.75 2 2 1.01 Boulder coral 2 3.59 2 2 1.16 Boulder coral 1 6.58 2 2 1.06 Branching coral1 4.1 2 2 1.02 Branching coral1 3.62 2 2 1.04 Branching coral1 3.8 2 2 1.02 Branching coral1 3.64 2 2 1.02 Branching coral3 3.55 2 2 1.03 Branching coral1 3.8 2 2 1.07 Branching coral3 3.84 2 2 1.09 Branching coral8 4.31 2 2 1.03 Branching coral1 4.55 2 2 1.02 Plate coral 1 3.62 2 2 1.03 Seagrass 13 7.37 2 2 1.14 Seagrass 1 6.88 2 2 1.04 Zoanthid 1 3.94 2 2 1.05 Zoanthid 2 3.82 2 2 1.02 Zoanthid 1 3.64 2 2 1.02 Zoanthid 1 3.55 2 2 1.03 Zoanthid 1 3.8 Appendix IV. Raw data used for regression analysis (continued)

PAGE 73

65 Reef Transect Rugosity index Benthic categoryFrequency Depth (m) 2 2 1.07 Zoanthid 1 3.53 2 2 1.08 Zoanthid 1 3.48 2 2 1.05 Zoanthid 4 4.6 3 1 1.07 Substrate 30 7.31 3 1 1.03 Substrate 37 7.4 3 1 1.03 Substrate 33 7.18 3 1 1.10 Substrate 20 6.83 3 1 1.09 Substrate 33 6.26 3 1 1.20 Substrate 32 5.18 3 1 1.09 Substrate 31 5.05 3 1 1.14 Substrate 28 4.92 3 1 1.13 Substrate 18 4.46 3 1 1.02 Substrate 26 4.38 3 1 1.01 Substrate 14 4.26 3 1 1.02 Substrate 19 4.14 3 1 1.03 Substrate 15 4.07 3 1 1.04 Substrate 27 4.22 3 1 1.05 Substrate 29 4.27 3 1 1.01 Substrate 23 4.29 3 1 1.02 Substrate 24 4.2 3 1 1.02 Substrate 20 4.34 3 1 1.05 Substrate 16 4.43 3 1 1.05 Substrate 22 4.43 3 1 1.06 Substrate 22 4.86 3 1 1.08 Substrate 26 4.65 3 1 1.30 Substrate 35 5.31 3 1 1.16 Substrate 29 6.01 3 1 1.20 Substrate 16 6.55 3 1 1.07 Macroalgae 2 7.31 3 1 1.03 Macroalgae 8 7.4 3 1 1.03 Macroalgae 17 7.18 3 1 1.10 Macroalgae 21 6.83 3 1 1.09 Macroalgae 15 6.26 3 1 1.20 Macroalgae 15 5.18 3 1 1.09 Macroalgae 15 5.05 3 1 1.14 Macroalgae 14 4.92 3 1 1.13 Macroalgae 25 4.46 3 1 1.02 Macroalgae 18 4.38 3 1 1.01 Macroalgae 24 4.26 3 1 1.02 Macroalgae 17 4.14 3 1 1.03 Macroalgae 26 4.07 3 1 1.04 Macroalgae 9 4.22 3 1 1.05 Macroalgae 10 4.27 3 1 1.01 Macroalgae 12 4.29 Appendix IV. Raw data used for regression analysis (continued)

PAGE 74

66 Reef Transect Rugosity index Benthic categoryFrequency Depth (m) 3 1 1.02 Macroalgae 15 4.2 3 1 1.02 Macroalgae 13 4.34 3 1 1.05 Macroalgae 14 4.43 3 1 1.05 Macroalgae 20 4.43 3 1 1.06 Macroalgae 19 4.86 3 1 1.08 Macroalgae 8 4.65 3 1 1.30 Macroalgae 8 5.31 3 1 1.16 Macroalgae 9 6.01 3 1 1.05 Macroalgae 10 4.27 3 1 1.07 Octocoral 1 7.31 3 1 1.10 Octocoral 7 6.83 3 1 1.09 Octocoral 1 6.26 3 1 1.20 Octocoral 2 5.18 3 1 1.09 Octocoral 4 5.05 3 1 1.14 Octocoral 8 4.92 3 1 1.13 Octocoral 5 4.46 3 1 1.02 Octocoral 6 4.38 3 1 1.01 Octocoral 11 4.26 3 1 1.02 Octocoral 13 4.14 3 1 1.03 Octocoral 8 4.07 3 1 1.04 Octocoral 7 4.22 3 1 1.05 Octocoral 9 4.27 3 1 1.01 Octocoral 14 4.29 3 1 1.02 Octocoral 5 4.2 3 1 1.02 Octocoral 15 4.34 3 1 1.05 Octocoral 12 4.43 3 1 1.05 Octocoral 8 4.43 3 1 1.06 Octocoral 5 4.86 3 1 1.08 Octocoral 12 4.65 3 1 1.30 Octocoral 5 5.31 3 1 1.16 Octocoral 9 6.01 3 1 1.20 Octocoral 6 6.55 3 1 1.10 Unidentifiable 2 6.83 3 1 1.04 Unidentifiable 1 4.22 3 1 1.01 Unidentifiable 1 4.29 3 1 1.02 Unidentifiable 3 4.2 3 1 1.05 Unidentifiable 5 4.43 3 1 1.06 Unidentifiable 4 4.86 3 1 1.08 Unidentifiable 1 4.65 3 1 1.16 Unidentifiable 1 6.01 3 1 1.20 Unidentifiable 28 6.55 3 1 1.37 Unidentifiable 50 7.65 3 1 1.02 Unidentifiable 50 8.15 3 1 1.09 Other 1 6.26 Appendix IV. Raw data used for regression analysis (continued)

PAGE 75

67 Reef Transect Rugosity index Benthic categoryFrequency Depth (m) 3 1 1.20 Other 1 5.18 3 1 1.13 Other 1 4.46 3 1 1.03 Other 1 4.07 3 1 1.05 Other 1 4.27 3 1 1.02 Other 1 4.2 3 1 1.02 Other 1 4.34 3 1 1.05 Other 1 4.43 3 1 1.08 Other 1 4.65 3 1 1.30 Other 1 5.31 3 1 1.16 Other 2 6.01 3 1 1.01 Boulder coral 1 4.26 3 1 1.04 Boulder coral 1 4.22 3 1 1.05 Boulder coral 1 4.27 3 1 1.02 Boulder coral 2 4.2 3 1 1.02 Boulder coral 1 4.34 3 1 1.05 Boulder coral 1 4.43 3 1 1.30 Boulder coral 1 5.31 3 1 1.04 Branching coral5 4.22 3 1 1.05 Plate coral 1 4.43 3 1 1.02 Sponge 1 4.14 3 1 1.08 Sponge 2 4.65 3 1 1.07 Seagrass 17 7.31 3 1 1.03 Seagrass 5 7.4 3 1 1.13 Zoanthid 1 4.46 3 2 1.02 Substrate 32 7.12 3 2 1.03 Substrate 28 6.84 3 2 1.04 Substrate 30 6.93 3 2 1.05 Substrate 28 6.51 3 2 1.03 Substrate 26 6.31 3 2 1.05 Substrate 30 5.73 3 2 1.06 Substrate 24 5.66 3 2 1.06 Substrate 14 5.12 3 2 1.04 Substrate 22 4.7 3 2 1.04 Substrate 14 4.42 3 2 1.03 Substrate 19 4.55 3 2 1.06 Substrate 25 4.26 3 2 1.09 Substrate 18 4.73 3 2 1.02 Substrate 28 4.5 3 2 1.03 Substrate 24 4.36 3 2 1.04 Substrate 15 4.37 3 2 1.04 Substrate 16 4.49 3 2 1.04 Substrate 29 4.64 3 2 1.12 Substrate 24 5.06 3 2 1.13 Substrate 29 4.78 Appendix IV. Raw data used for regression analysis (continued)

PAGE 76

68 Reef Transect Rugosity index Benthic categoryFrequency Depth (m) 3 2 1.11 Substrate 30 5.38 3 2 1.17 Substrate 22 5.65 3 2 1.23 Substrate 19 6.24 3 2 1.02 Substrate 50 8.57 3 2 1.03 Macroalgae 4 6.84 3 2 1.04 Macroalgae 7 6.93 3 2 1.05 Macroalgae 11 6.51 3 2 1.03 Macroalgae 15 6.31 3 2 1.05 Macroalgae 9 5.73 3 2 1.06 Macroalgae 10 5.66 3 2 1.06 Macroalgae 15 5.12 3 2 1.04 Macroalgae 11 4.7 3 2 1.04 Macroalgae 17 4.42 3 2 1.03 Macroalgae 18 4.55 3 2 1.06 Macroalgae 13 4.26 3 2 1.09 Macroalgae 17 4.73 3 2 1.02 Macroalgae 5 4.5 3 2 1.03 Macroalgae 16 4.36 3 2 1.04 Macroalgae 11 4.37 3 2 1.04 Macroalgae 11 4.49 3 2 1.04 Macroalgae 15 4.64 3 2 1.12 Macroalgae 10 5.06 3 2 1.13 Macroalgae 17 4.78 3 2 1.11 Macroalgae 9 5.38 3 2 1.17 Macroalgae 17 5.65 3 2 1.23 Macroalgae 6 6.24 3 2 1.04 Octocoral 2 6.93 3 2 1.05 Octocoral 9 6.51 3 2 1.03 Octocoral 5 6.31 3 2 1.05 Octocoral 2 5.73 3 2 1.06 Octocoral 11 5.66 3 2 1.06 Octocoral 11 5.12 3 2 1.04 Octocoral 14 4.7 3 2 1.04 Octocoral 14 4.42 3 2 1.03 Octocoral 9 4.55 3 2 1.06 Octocoral 6 4.26 3 2 1.09 Octocoral 12 4.73 3 2 1.02 Octocoral 13 4.5 3 2 1.03 Octocoral 8 4.36 3 2 1.04 Octocoral 20 4.37 3 2 1.04 Octocoral 14 4.49 3 2 1.04 Octocoral 4 4.64 3 2 1.12 Octocoral 5 5.06 3 2 1.13 Octocoral 2 4.78 Appendix IV. Raw data used for regression analysis (continued)

PAGE 77

69 Reef Transect Rugosity index Benthic categoryFrequency Depth (m) 3 2 1.11 Octocoral 8 5.38 3 2 1.17 Octocoral 7 5.65 3 2 1.23 Octocoral 3 6.24 3 2 1.02 Unidentifiable 3 7.12 3 2 1.03 Unidentifiable 9 6.84 3 2 1.04 Unidentifiable 10 6.93 3 2 1.05 Unidentifiable 1 6.51 3 2 1.03 Unidentifiable 1 6.31 3 2 1.05 Unidentifiable 4 5.73 3 2 1.06 Unidentifiable 3 5.66 3 2 1.06 Unidentifiable 3 5.12 3 2 1.04 Unidentifiable 3 4.7 3 2 1.04 Unidentifiable 1 4.42 3 2 1.03 Unidentifiable 1 4.55 3 2 1.06 Unidentifiable 5 4.26 3 2 1.09 Unidentifiable 2 4.73 3 2 1.02 Unidentifiable 3 4.5 3 2 1.03 Unidentifiable 2 4.36 3 2 1.04 Unidentifiable 2 4.37 3 2 1.04 Unidentifiable 7 4.49 3 2 1.04 Unidentifiable 2 4.64 3 2 1.12 Unidentifiable 11 5.06 3 2 1.13 Unidentifiable 2 4.78 3 2 1.17 Unidentifiable 2 5.65 3 2 1.23 Unidentifiable 22 6.24 3 2 1.26 Unidentifiable 50 6.49 3 2 1.05 Other 1 6.51 3 2 1.03 Other 2 4.55 3 2 1.09 Other 1 4.73 3 2 1.04 Other 1 4.37 3 2 1.03 Boulder coral 1 6.84 3 2 1.05 Boulder coral 2 5.73 3 2 1.06 Boulder coral 1 5.66 3 2 1.06 Boulder coral 3 5.12 3 2 1.04 Boulder coral 2 4.42 3 2 1.06 Boulder coral 1 4.26 3 2 1.04 Boulder coral 2 4.49 3 2 1.11 Boulder coral 1 5.38 3 2 1.17 Boulder coral 2 5.65 3 2 1.04 Branching coral1 4.42 3 2 1.04 Branching coral1 4.37 3 2 1.03 Plate coral 1 6.31 3 2 1.03 Sponge 1 6.84 3 2 1.03 Sponge 1 6.31 Appendix IV. Raw data used for regression analysis (continued)

PAGE 78

70 Reef Transect Rugosity index Benthic category Frequency Depth (m) 3 2 1.05 Sponge 2 5.73 3 2 1.06 Sponge 3 5.12 3 2 1.04 Sponge 1 4.42 3 2 1.03 Sponge 1 4.55 3 2 1.02 Sponge 1 4.5 3 2 1.02 Seagrass 15 7.12 3 2 1.03 Seagrass 7 6.84 3 2 1.04 Seagrass 1 6.93 4 1 1.07 Substrate 33 7.08 4 1 1.17 Substrate 30 6.38 4 1 1.15 Substrate 11 6.27 4 1 1.11 Substrate 31 5.68 4 1 1.05 Substrate 15 5.38 4 1 1.12 Substrate 18 4.79 4 1 1.13 Substrate 12 4.38 4 1 1.06 Substrate 4 4.2 4 1 1.03 Substrate 6 3.85 4 1 1.04 Substrate 8 3.91 4 1 1.04 Substrate 8 3.77 4 1 1.02 Substrate 12 3.88 4 1 1.02 Substrate 11 3.8 4 1 1.05 Substrate 19 3.87 4 1 1.01 Substrate 20 3.68 4 1 1.11 Substrate 27 3.68 4 1 1.02 Substrate 18 3.9 4 1 1.03 Substrate 19 4.01 4 1 1.05 Substrate 27 4.33 4 1 1.16 Substrate 14 4.37 4 1 1.11 Substrate 12 4.92 4 1 1.09 Substrate 21 5.25 4 1 1.05 Substrate 28 5.67 4 1 1.03 Macroalgae 10 3.85 4 1 1.04 Macroalgae 5 3.91 4 1 1.04 Macroalgae 7 3.77 4 1 1.02 Macroalgae 3 3.88 4 1 1.05 Macroalgae 13 3.87 4 1 1.01 Macroalgae 3 3.68 4 1 1.11 Macroalgae 5 3.68 4 1 1.02 Macroalgae 8 3.9 4 1 1.03 Macroalgae 9 4.01 Appendix IV. Raw data used for regression analysis (continued)

PAGE 79

71 Reef Transect Rugosity index Benthic category Frequency Depth (m) 4 1 1.05 Macroalgae 8 4.33 4 1 1.16 Macroalgae 8 4.37 4 1 1.11 Macroalgae 21 4.92 4 1 1.09 Macroalgae 18 5.25 4 1 1.05 Macroalgae 11 5.67 4 1 1.11 Macroalgae 5 6.43 4 1 1.07 Macroalgae 9 8.56 4 1 1.07 Octocoral 4 7.08 4 1 1.17 Octocoral 7 6.38 4 1 1.15 Octocoral 14 6.27 4 1 1.11 Octocoral 2 5.68 4 1 1.05 Octocoral 16 5.38 4 1 1.12 Octocoral 16 4.79 4 1 1.13 Octocoral 21 4.38 4 1 1.06 Octocoral 21 4.2 4 1 1.03 Octocoral 26 3.85 4 1 1.04 Octocoral 31 3.91 4 1 1.04 Octocoral 25 3.77 4 1 1.02 Octocoral 26 3.88 4 1 1.02 Octocoral 32 3.8 4 1 1.05 Octocoral 16 3.87 4 1 1.01 Octocoral 21 3.68 4 1 1.11 Octocoral 9 3.68 4 1 1.02 Octocoral 20 3.9 4 1 1.03 Octocoral 12 4.01 4 1 1.05 Octocoral 12 4.33 4 1 1.16 Octocoral 20 4.37 4 1 1.11 Octocoral 16 4.92 4 1 1.09 Octocoral 3 5.25 4 1 1.05 Octocoral 9 5.67 4 1 1.11 Octocoral 6 6.43 4 1 1.10 Octocoral 5 6.73 4 1 1.07 Unidentifiable 6 7.08 4 1 1.17 Unidentifiable 6 6.38 4 1 1.15 Unidentifiable 14 6.27 4 1 1.11 Unidentifiable 14 5.68 4 1 1.05 Unidentifiable 14 5.38 4 1 1.12 Unidentifiable 4 4.79 4 1 1.13 Unidentifiable 12 4.38 4 1 1.06 Unidentifiable 4 4.2 4 1 1.03 Unidentifiable 7 3.85 4 1 1.04 Unidentifiable 5 3.91 4 1 1.04 Unidentifiable 10 3.77 4 1 1.02 Unidentifiable 9 3.88 Appendix IV. Raw data used for regression analysis (continued)

PAGE 80

72 Reef Transect Rugosity index Benthic category Frequency Depth (m) 4 1 1.02 Unidentifiable 2 3.8 4 1 1.05 Unidentifiable 1 3.87 4 1 1.01 Unidentifiable 6 3.68 4 1 1.11 Unidentifiable 3 3.68 4 1 1.02 Unidentifiable 1 3.9 4 1 1.03 Unidentifiable 6 4.01 4 1 1.16 Unidentifiable 4 4.37 4 1 1.10 Unidentifiable 14 6.73 4 1 1.25 Unidentifiable 50 7.6 4 1 1.07 Unidentifiable 3 8.56 4 1 1.15 Other 1 6.27 4 1 1.12 Other 1 4.79 4 1 1.13 Other 1 4.38 4 1 1.03 Other 1 3.85 4 1 1.02 Other 4 3.8 4 1 1.05 Other 1 3.87 4 1 1.11 Other 2 3.68 4 1 1.03 Other 1 4.01 4 1 1.05 Other 2 4.33 4 1 1.16 Other 1 4.37 4 1 1.11 Other 1 4.92 4 1 1.05 Other 1 5.67 4 1 1.11 Other 1 6.43 4 1 1.10 Other 1 6.73 4 1 1.07 Other 1 8.56 4 1 1.07 Boulder coral 1 7.08 4 1 1.17 Boulder coral 2 6.38 4 1 1.15 Boulder coral 1 6.27 4 1 1.05 Boulder coral 2 5.38 4 1 1.12 Boulder coral 2 4.79 4 1 1.06 Boulder coral 5 4.2 4 1 1.11 Boulder coral 2 3.68 4 1 1.02 Boulder coral 1 3.9 4 1 1.09 Boulder coral 8 5.25 4 1 1.05 Boulder coral 1 5.67 4 1 1.11 Boulder coral 6 6.43 4 1 1.10 Boulder coral 1 6.73 4 1 1.07 Boulder coral 2 8.56 4 1 1.03 Branching coral 2 4.01 4 1 1.05 Branching coral 1 4.33 4 1 1.11 Branching coral 1 6.43 4 1 1.10 Branching coral 1 6.73 4 1 1.11 Plate coral 1 5.68 4 1 1.02 Plate coral 1 3.8 Appendix IV. Raw data used for regression analysis (continued)

PAGE 81

73 Reef Transect Rugosity index Benthic category Frequency Depth (m) 4 1 1.11 Sponge 1 3.68 4 1 1.02 Sponge 1 3.9 4 1 1.03 Sponge 1 4.01 4 1 1.04 Zoanthid 1 3.91 4 1 1.11 Zoanthid 1 3.68 4 1 1.02 Zoanthid 1 3.9 4 1 1.16 Zoanthid 3 4.37 4 2 1.03 Substrate 31 6.73 4 2 1.03 Substrate 40 6.83 4 2 1.06 Substrate 35 6.51 4 2 1.04 Substrate 32 6.91 4 2 1.01 Substrate 36 6.77 4 2 1.06 Substrate 32 6.76 4 2 1.08 Substrate 29 6.43 4 2 1.15 Substrate 27 6.33 4 2 1.16 Substrate 38 5.68 4 2 1.13 Substrate 38 5.74 4 2 1.18 Substrate 27 5.71 4 2 1.09 Substrate 34 5.29 4 2 1.10 Substrate 30 5.37 4 2 1.16 Substrate 34 4.78 4 2 1.11 Substrate 35 4.69 4 2 1.08 Substrate 16 4.6 4 2 1.09 Substrate 23 4.64 4 2 1.07 Substrate 20 4.43 4 2 1.07 Substrate 23 4.16 4 2 1.07 Substrate 33 3.79 4 2 1.08 Substrate 2 3.27 4 2 1.05 Substrate 10 3.59 4 2 1.02 Substrate 23 3.58 4 2 1.05 Substrate 20 3.87 4 2 1.07 Substrate 13 3.6 4 2 1.01 Substrate 24 3.81 4 2 1.11 Substrate 27 4.46 4 2 1.07 Substrate 13 4.25 4 2 1.10 Substrate 7 4.62 4 2 1.09 Substrate 15 4.39 4 2 1.09 Substrate 18 4.98 4 2 1.07 Substrate 12 5.37 4 2 1.11 Substrate 19 5.64 4 2 1.13 Substrate 27 7.62 4 2 1.02 Substrate 39 8.45 4 2 1.03 Macroalgae 4 6.83 4 2 1.06 Macroalgae 1 6.51 Appendix IV. Raw data used for regression analysis (continued)

PAGE 82

74 Reef Transect Rugosity index Benthic category Frequency Depth (m) 4 2 1.04 Macroalgae 1 6.91 4 2 1.01 Macroalgae 3 6.77 4 2 1.06 Macroalgae 7 6.76 4 2 1.08 Macroalgae 13 6.43 4 2 1.15 Macroalgae 10 6.33 4 2 1.16 Macroalgae 7 5.68 4 2 1.18 Macroalgae 10 5.71 4 2 1.09 Macroalgae 3 5.29 4 2 1.10 Macroalgae 8 5.37 4 2 1.16 Macroalgae 5 4.78 4 2 1.11 Macroalgae 8 4.69 4 2 1.08 Macroalgae 8 4.6 4 2 1.09 Macroalgae 4 4.64 4 2 1.07 Macroalgae 5 4.43 4 2 1.07 Macroalgae 3 4.16 4 2 1.07 Macroalgae 1 3.79 4 2 1.02 Macroalgae 3 3.58 4 2 1.05 Macroalgae 1 3.87 4 2 1.01 Macroalgae 7 3.81 4 2 1.11 Macroalgae 8 4.46 4 2 1.07 Macroalgae 11 4.25 4 2 1.10 Macroalgae 3 4.62 4 2 1.09 Macroalgae 7 4.39 4 2 1.09 Macroalgae 10 4.98 4 2 1.07 Macroalgae 7 5.37 4 2 1.11 Macroalgae 12 5.64 4 2 1.03 Octocoral 0 6.73 4 2 1.03 Octocoral 1 6.83 4 2 1.06 Octocoral 10 6.51 4 2 1.04 Octocoral 11 6.91 4 2 1.01 Octocoral 8 6.77 4 2 1.06 Octocoral 6 6.76 4 2 1.08 Octocoral 5 6.43 4 2 1.15 Octocoral 10 6.33 4 2 1.16 Octocoral 4 5.68 4 2 1.13 Octocoral 8 5.74 4 2 1.18 Octocoral 7 5.71 4 2 1.09 Octocoral 11 5.29 4 2 1.10 Octocoral 7 5.37 4 2 1.16 Octocoral 8 4.78 4 2 1.11 Octocoral 3 4.69 4 2 1.08 Octocoral 9 4.6 4 2 1.09 Octocoral 6 4.64 4 2 1.07 Octocoral 9 4.43 Appendix IV. Raw data used for regression analysis (continued)

PAGE 83

75 Reef Transect Rugosity index Benthic category Frequency Depth (m) 4 2 1.07 Octocoral 6 4.16 4 2 1.07 Octocoral 9 3.79 4 2 1.08 Octocoral 37 3.27 4 2 1.05 Octocoral 28 3.59 4 2 1.02 Octocoral 10 3.58 4 2 1.05 Octocoral 14 3.87 4 2 1.07 Octocoral 30 3.6 4 2 1.01 Octocoral 14 3.81 4 2 1.11 Octocoral 11 4.46 4 2 1.07 Octocoral 19 4.25 4 2 1.10 Octocoral 37 4.62 4 2 1.09 Octocoral 19 4.98 4 2 1.07 Octocoral 24 5.37 4 2 1.11 Octocoral 11 5.64 4 2 1.04 Unidentifiable 1 6.91 4 2 1.01 Unidentifiable 1 6.77 4 2 1.06 Unidentifiable 3 6.76 4 2 1.08 Unidentifiable 2 6.43 4 2 1.15 Unidentifiable 1 6.33 4 2 1.16 Unidentifiable 1 5.68 4 2 1.13 Unidentifiable 3 5.74 4 2 1.18 Unidentifiable 5 5.71 4 2 1.09 Unidentifiable 2 5.29 4 2 1.10 Unidentifiable 1 5.37 4 2 1.16 Unidentifiable 2 4.78 4 2 1.11 Unidentifiable 2 4.69 4 2 1.08 Unidentifiable 11 4.6 4 2 1.09 Unidentifiable 12 4.64 4 2 1.07 Unidentifiable 10 4.43 4 2 1.07 Unidentifiable 2 4.16 4 2 1.07 Unidentifiable 4 3.79 4 2 1.08 Unidentifiable 6 3.27 4 2 1.05 Unidentifiable 7 3.59 4 2 1.02 Unidentifiable 9 3.58 4 2 1.05 Unidentifiable 7 3.87 4 2 1.07 Unidentifiable 5 3.6 4 2 1.01 Unidentifiable 1 3.81 4 2 1.11 Unidentifiable 1 4.46 4 2 1.07 Unidentifiable 3 4.25 4 2 1.09 Unidentifiable 4 4.39 4 2 1.09 Unidentifiable 3 4.98 4 2 1.07 Unidentifiable 6 5.37 4 2 1.11 Unidentifiable 3 5.64 4 2 1.09 Unidentifiable 50 6.14 Appendix IV. Raw data used for regression analysis (continued)

PAGE 84

76 Reef Transect Rugosity index Benthic category Frequency Depth (m) 4 2 1.13 Unidentifiable 23 7.62 4 2 1.15 Unidentifiable 50 7.56 4 2 1.03 Other 2 6.73 4 2 1.06 Other 3 6.51 4 2 1.13 Other 1 5.74 4 2 1.18 Other 1 5.71 4 2 1.10 Other 1 5.37 4 2 1.16 Other 1 4.78 4 2 1.11 Other 2 4.69 4 2 1.08 Other 1 4.6 4 2 1.09 Other 3 4.64 4 2 1.07 Other 1 4.43 4 2 1.07 Other 3 4.16 4 2 1.07 Other 2 3.79 4 2 1.08 Other 2 3.27 4 2 1.05 Other 3 3.59 4 2 1.05 Other 2 3.87 4 2 1.07 Other 1 3.6 4 2 1.01 Other 2 3.81 4 2 1.11 Other 2 4.46 4 2 1.07 Other 2 4.25 4 2 1.10 Other 2 4.62 4 2 1.09 Other 24 4.39 4 2 1.07 Other 1 5.37 4 2 1.11 Other 2 5.64 4 2 1.06 Boulder coral 1 6.51 4 2 1.04 Boulder coral 4 6.91 4 2 1.01 Boulder coral 1 6.77 4 2 1.06 Boulder coral 2 6.76 4 2 1.10 Boulder coral 2 5.37 4 2 1.07 Boulder coral 1 4.43 4 2 1.07 Boulder coral 4 4.16 4 2 1.08 Boulder coral 3 3.27 4 2 1.05 Boulder coral 1 3.59 4 2 1.02 Boulder coral 2 3.58 4 2 1.05 Boulder coral 2 3.87 4 2 1.11 Boulder coral 1 4.46 4 2 1.07 Boulder coral 1 4.25 4 2 1.11 Boulder coral 2 5.64 4 2 1.08 Branching coral 3 4.6 4 2 1.09 Branching coral 1 4.64 4 2 1.07 Branching coral 3 4.43 4 2 1.07 Branching coral 3 4.16 4 2 1.02 Branching coral 2 3.58 Appendix IV. Raw data used for regression analysis (continued)

PAGE 85

77 4 2 1.01 Branching coral 2 3.81 4 2 1.07 Branching coral 1 4.25 4 2 1.04 Plate coral 1 6.91 4 2 1.01 Plate coral 1 6.77 4 2 1.08 Plate coral 1 6.43 4 2 1.15 Plate coral 2 6.33 4 2 1.10 Plate coral 1 5.37 4 2 1.07 Plate coral 6 4.16 4 2 1.07 Plate coral 1 3.79 4 2 1.05 Plate coral 4 3.87 4 2 1.08 Sponge 2 4.6 4 2 1.09 Sponge 1 4.64 4 2 1.07 Sponge 1 4.43 4 2 1.05 Sponge 1 3.59 4 2 1.02 Sponge 1 3.58 4 2 1.07 Sponge 1 3.6 4 2 1.10 Sponge 1 4.62 4 2 1.11 Sponge 1 5.64 4 2 1.03 Seagrass 17 6.73 4 2 1.03 Seagrass 5 6.83 4 2 1.02 Seagrass 11 8.45 Appendix IV. Raw data used for regression analysis (continued)