Content uploaded by Simon James Pittman
Author content
All content in this area was uploaded by Simon James Pittman
Content may be subject to copyright.
Fish assemblages and benthic habitats of Buck Island Reef National
Monument (St. Croix, U.S. Virgin Islands) and the surrounding seascape:
A characterization of spatial and temporal patterns
A cooperative investigation between NOAA and the National Park Service
May 2008
NOAA Technical Memorandum NOS NCCOS 71
Simon J Pittman
Sarah D Hile
Christopher FG Jeffrey
Chris Caldow
Matt S Kendall
Mark E Monaco
Zandy Hillis-Starr
Mention of trade names or commercial products does not constitute endorsement or recommendation for their use by the
United States government.
Citation for this Report:
Pittman, S.J., S.D. Hile, C.F.G. Jeffrey, C. Caldow, M.S. Kendall, M.E. Monaco, and Z. Hillis-Starr. 2008. Fish assemblages
and benthic habitats of Buck Island Reef National Monument (St. Croix, U.S. Virgin Islands) and the surrounding seascape:
A characterization of spatial and temporal patterns. NOAA Technical Memorandum NOS NCCOS 71. Silver Spring, MD. 96
pp.
Fish assemblages and benthic habitats of Buck Island Reef National
Monument (St. Croix, U.S. Virgin Islands) and the surrounding seascape:
A characterization of spatial and temporal patterns
Simon J Pittman
1,2
, Sarah D Hile
1
, Chris FG Jeffrey
1
, Chris Caldow
1
, Matt S Kendall
1
,
Mark E Monaco
1
and Zandy Hillis-Starr
3
1
NOAA/National Ocean Service/National Centers for Coastal Ocean Science/Center for Coastal Monitoring and
Assessment/Biogeography Branch
2
Marine Science Center, University of the Virgin Islands, St. Thomas, U.S. Virgin Islands
3
National Park Service, St. Croix, U.S. Virgin Islands
Biogeography Branch
Center for Coastal Monitoring and Assessment (CCMA)
NOAA/NOS/National Centers for Coastal Ocean Science
1305 East West Highway (SSMC-IV, N/SCI-1)
Silver Spring, MD 20910
NOAA Technical Memorandum NOS NCCOS 71
May 2008
United States Department of
Commerce
National Oceanic and
Atmospheric Administration
National Ocean Service
Carlos M Gutierrez
Secretary
Conrad C Lautenbacher, Jr.
Administrator
Jack Dunnigan
Assistant Administrator
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
About this Document
The report provides a spatial and temporal characterization of the sh and benthic communities of Buck Island Reef
National Monument and the surrounding seascapes of northeastern St. Croix, United States Virgin Islands. The project is
a component of NOAA’s Caribbean Coral Reef Ecosystem Monitoring (CREM) project of NOAA’s Coral Reef Conservation
Program (CRCP) and the National Park Service (NPS). The project integrates eld data on coral condition, living marine
resources and benthic habitats through an ongoing multi-agency collaboration between NOAA’s Center for Coastal
Monitoring and Assessment Biogeography Branch (CCMA-BB), NPS, U.S. Geological Survey and the Virgin Islands
Department of Planning and Natural Resources (VI-DPNR).
This Technical Memorandum is part one of a series of reports that focus on providing a quantitative spatial and temporal
characterization of living marine resources and benthic communities associated with marine protected areas in the U.S.
Caribbean. This project complements the National Coral Reef Ecosystem Monitoring Program’s (NCREMP) Coral Reef
Ecosystem Monitoring grants awarded to the VI-DPNR by CRCP. The integration of the NOAA/NPS lead efforts with data
generated by VI-DPNR provides robust spatial and temporal data to characterize St. Croix coral reef ecosystems. This
project was funded by NOAA’s CRCP and National Centers for Coastal Ocean Science’s CCMA and CSCOR, NPS’s
Natural Resource Preservation Program (NRPP) at Buck Island Reef National Monument and NPS’s South Florida/
Caribbean Inventory and Monitoring Program.
Related projects include:
Caribbean Coral Reef Ecosystem Monitoring
http://ccmaserver.nos.noaa.gov/ecosystems/coralreef/reef_sh.html
Development of Reef Fish Monitoring Protocols to Support the National Park Service Inventory and Monitoring Program
http://ccmaserver.nos.noaa.gov/ecosystems/coralreef/sh_protocol.html
Coral bleaching and recovery observed at Buck Island, St. Croix, U.S. Virgin Islands, October and December, 2005
http://ccmaserver.nos.noaa.gov/ecosystems/coralreef/reef_sh.html
National Coral Reef Ecosystem Montoring Program
http://ccma.nos.noaa.gov/ecosystems/coralreef/coral_grant.html
Benthic Habitat Mapping of Puerto Rico and the U.S. Virgin Islands
http://ccma.nos.noaa.gov/ecosystems/coralreef/usvi_pr_mapping.html
Seaoor Characterization of the U.S. Caribbean - R/V Nancy Foster
Missions
http://ccma.nos.noaa.gov/products/biogeography/usvi_nps/
overview.html
All photographs provided in this document were taken by NOAA/NOS/
NCCOS/Center for Coastal Monitoring Assessment Biogeography Branch
in St. Croix, USVI unless otherwise indicated.
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. vi
Project Team
Richard Berey
Laurie Bauer (NCCOS CCMA)
Craig Bonn (NCCOS CCFHR Beaufort Lab)
Chris Caldow (NCCOS CCMA)
Don Catanzaro (NPS)
John Christensen (NCCOS CCMA)
Randy Clark (NCCOS CCMA)
Michael Coyne (NCCOS CCMA)
Kimberly Foley (NCCOS CCMA)
Alan Friedlander (NCCOS-CCMA)
Ricky Grober-Dunsmore (USGS)
Sarah Hile (NCCOS CCMA)
Zandy Hillis-Starr (NPS)
Chris Jeffrey (NCCOS CCMA)
Thomas Kelley (NPS)
Matt Kendall (NCCOS CCMA)
Ian Lundgren (NPS)
Philippe Mayor (NPS)
Tom McGrath (NCCOS CCMA)
Charles Menza (NCCOS CCMA)
Jeff Miller (NPS)
Wendy Morrison (NCCOS CCMA)
Mark Monaco (NCCOS CCMA)
Shelby Moneysmith (NPS)
Brenda-Lee Phillips (NPS)
Simon Pittman (NCCOS CCMA)
Caroline Rogers (USGS)
Paige Rothenberger (VI DPNR)
Carrie Stengel (NPS)
Henry E. Tonnemacher
Joel Tutein (NPS)
Rob Waara (NPS)
Jenny Waddell (NCCOS CCMA)
Kimberly Woody (NCCOS CCMA)
p. vii
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Executive Summary
Since 1999, NOAA’s Biogeography Branch of the Center for Coastal Monitoring and Assessment (CCMA-BB) has been
working with federal and territorial partners to characterize, monitor, and assess the status of the marine environment
around northeastern St. Croix, U.S. Virgin Islands. This effort is part of the broader NOAA Coral Reef Conservation
Program’s (CRCP) National Coral Reef Ecosystem Monitoring Program (NCREMP). With support from CRCP’s
NCREMP, CCMA conducts the “Caribbean Coral Reef Ecosystem Monitoring project” (CREM) with goals to: (1) spatially
characterize and monitor the distribution, abundance, and size of marine fauna associated with shallow water coral reef
seascapes (mosaics of coral reefs, seagrasses, sand and mangroves); (2) relate this information to in situ ne-scale
habitat data and the spatial distribution and diversity of habitat types using benthic habitat maps; (3) use this information
to establish the knowledge base necessary for enacting management decisions in a spatial setting; (4) establish the
efcacy of those management decisions; and (5) develop data collection and data management protocols. The monitoring
effort in northeastern St. Croix was conducted through partnerships with the National Park Service (NPS) and the Virgin
Islands Department of Planning and Natural Resources (VI-DPNR). The geographical focal point of the research is Buck
Island Reef National Monument (BIRNM), a protected area originally established in 1961 and greatly expanded in 2001;
however, the work also encompassed a large portion of the recently created St. Croix East End Marine Park (EEMP).
Project funding is primarily provided by NOAA CRCP, CCMA and NPS.
In recent decades, scientic and non-scientic observations have indicated that the structure and function of the coral
reef ecosystem around northeastern St. Croix have been adversely impacted by a wide range of environmental stressors.
The major stressors have included the mass Diadema die off in the early 1980s, a series of hurricanes beginning with
Hurricane Hugo in 1989, overshing, mass mortality of Acropora corals due to disease and several coral bleaching events,
with the most severe mass bleaching episode in 2005. The area is also an important recreational resource supporting
boating, snorkeling, diving and other water based activities. With so many potential threats to the marine ecosystem and
a dramatic change in management strategy in 2003 when the park’s Interim Regulations (Presidential Proclamation No.
7392) established BIRNM as one of the rst fully protected marine areas in NPS system, it became critical to identify
existing marine fauna and their spatial distributions and temporal dynamics. This provides ecologically meaningful data to
assess ecosystem condition, support decision making in spatial planning (including the evaluation of efcacy of current
management strategies) and determine future information needs. The ultimate goal of the work is to better understand the
coral reef ecosystems and to provide information toward protecting and enhancing coral reef ecosystems for the benet of
the system itself and to sustain the many goods and services that it offers society. This Technical Memorandum contains
analysis of the rst six years of sh survey data (2001-2006) and associated characterization of the benthos (1999-2006).
The primary objectives were to quantify changes in sh species and assemblage diversity, abundance, biomass and
size structure and to provide spatially explicit information on the distribution of key species or groups of species and to
compare community structure inside (protected) versus outside (shed) areas of BIRNM.
Methods:
For each biannual survey mission, selection of sample sites occurred via a stratied random design (2001-2006)
using hard and soft bottom habitat types delineated in NOAA’s benthic habitat map (Menza et al., 2006). In 2003, after
implementation of the park’s Interim Regulations, sampling was also stratied by whether or not the site was located
inside or outside BIRNM to evaluate effect of the shing closure (only 2003 to 2006). Fish were surveyed during daylight
hours along 25 m long by 4 m wide belt-transects for a xed duration of 15 minutes. All species observed were identied
to the lowest possible taxonomic level and their abundance was counted and grouped by size class. To quantify benthic
habitat, ve 1 m
2
quadrats were randomly placed on the transect and used to examine the relatively ne-scale biotic and
abiotic components of the seascape (e.g., coral cover, macroalgal cover, etc.). In addition, Geographical Information
System (GIS) tools were used to quantify the seascape surrounding each transect using habitat distributions represented
in NOAA’s benthic habitat maps (e.g., amount of seagrass, number of habitat types, etc.).
Comparative analyses of biotic components inside versus outside BIRNM were conducted using a wide range of sh
variables representing community, trophic, family and individual species level data incorporating measures of abundance,
biomass and diversity. A total of 884 transects collected between 2003 and 2006 inclusively were used to examine
differences in sh metrics inside versus outside BIRNM. Benthic comparisons used 716 benthic surveys on hardbottom
habitat types conducted between 2001 and 2006. Abundance maps were used to examine species distributions for both
juveniles and adults and interpolations of point data were used to examine broad-scale spatial patterns of sh and benthic
habitat variables.
Major ndings:
Diversity hotspots
• Despite heavy impacts from disease, bleaching and hurricanes, the area around the eastern tip of Buck Island remains
ecologically distinctive having some of the highest live coral cover and rugosity in the mapped region, also with high
calcareous coralline algal cover, high sh species richness, biomass of herbivorous sh and high abundance for many
common sh species.
Executive Summary
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. viii
Executive Summary
• The linear reef (i.e., barrier reefs) and adjacent colonized pavement that extends east-west from Teague Bay to
Coakley Bay and now falls within the EEMP “no-take zone” and “recreation zone” was found to support high coral
species richness and sh species richness.
• Extensive areas with high coral species richness, high live coral cover for Montastraea cavernosa and M. annularis,
high sh species richness and high abundance for several sh species including coney (Cephalopholis fulva), rock
beauty (Holacanthus tricolor) and queen triggersh (Balistes vetula) occurred along the northernmost edge of the
benthic habitat map. This indicates that important deeper water habitat is likely to exist beyond the scope of this report,
requiring further benthic habitat mapping effort combined with visual census (diver/remotely operated vehicle [ROV])
to capture data on sh communities (see Foster Mission web site: http://ccma.nos.noaa.gov/products/biogeography/
usvi_nps/overview.htm).
• Many of the coral reefs with highest sh species richness were within 200 m of seagrass beds. Several other studies
have demonstrated links between sh distribution on coral reefs and proximity to seagrass beds suggesting that many
species may benet from complementary resources provided by seagrasses in close proximity to coral reefs. This
highlights the importance of considering mosaics of habitat types in resource management decision making.
Benthic habitat
• The benthic environment inside BIRNM was signicantly different to the outside for 75% of ne-scale variables quantied
within 1 m
2
quadrates and 78% of seascape variables quantied within 100 m
2
radius seascape units surrounding each
transect.
• Seventy-eight percent of the mapped area inside BIRNM was hardbottom habitat dominated by colonized pavement and
22% was softbottom (sand and seagrasses); outside BIRNM, 46% was hardbottom and 54% softbottom. Seascapes
inside BIRNM also had signicantly higher mean habitat richness.
• Coral cover for all major scleractinian (hard coral) families was signicantly higher inside BIRNM and coral reefs had
a signicantly higher ratio of live coral cover to macroalgal cover than outside BIRNM.
• Overall, hardbottom habitats of the study area were dominated by turf algae (37%) and macroalgae (11.4%), with
mean scleractinian coral cover of only 5.6% ranging from 12.1% on patch reefs to 2% on the less rugose reef rubble.
• Across years (2003-2006), macroalgal cover showed some indication of decline both inside and outside BIRNM.
• Filamentous cyanobacteria/macroalgal blooms were detected in the fall sampling period with mean cover as high as
18% in October 2005; a year with anomalously high summer water temperatures that also resulted in a mass coral
bleaching event.
• Peaks in mean algal turf cover (50-60%) were detected in the spring (2006) season following the mass coral bleaching
event and mean live coral cover approximately one year after the event was the lowest since this study commenced.
Fish
• A total of 201 sh species/species groups were identied from 56 families. Nine of the 10 most frequently encountered
species belonged to the families Labridae (wrasse), Acanthuridae (surgeonsh) and Scaridae (parrotsh).
• The majority of the most abundant sh across the study region were found in highest densities over hardbottom habitat
types, yet most also utilized multiple habitat types including seagrasses and sand.
• Fish metrics signicantly higher on hardbottom habitat inside BIRNM included sh biomass (all sh combined), herbivore
biomass, parrotsh biomass, shark and ray biomass, coney (C. fulva) density and biomass, blue tang (Acanthurus
coeruleus) density and biomass, and striped parrotsh (Scarus iseri) biomass.
• Fish metrics signicantly higher outside BIRNM included ecologically important predator groups such as piscivore
biomass (including sharks and rays), snapper (Lutjanidae) density, and grunt (Haemulidae) density and biomass.
• Red hind (Epinephelus guttatus) and coney (C. fulva) exhibited distinct patterns in spatial distributions, with high coney
density mostly over the contiguous colonized hardbottom areas (much of which is inside BIRNM) and high densities of
red hind found mostly to the south of Buck Island (many outside BIRNM).
• Very few of the largest (>35 cm) and very few of the smallest (<5 cm) size classes were observed for groupers
(Serranidae) and snappers. Groupers and snappers in the largest size class (>35 cm) were recorded at <1% and 3%,
of survey sites, respectively.
• Body lengths of the largest individuals of several common groupers, snappers and grunts were less than the maximum
size recorded for the species. The largest yellowtail snapper (Ocyurus chrysurus) was approximately 70% of the
maximum known adult size, schoolmaster snapper (Lutjanus apodus) 66%, bluestriped grunt (Haemulon sciurus) 65-
76%, white grunt (H. plumierii) 56-66% and red hind (E. guttatus) 60% of known maximum size.
• Highest densities of threespot damselsh (Stegastes planifrons), a potential indicator of healthy reefs with high live
coral cover, were found around the eastern tip of Buck Island within BIRNM and the fringing reef extending east-west
along the northeast coast of St. Croix.
p. ix
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Historic and recent changes in sh populations
• Synoptic overview of inter-annual differences inside and outside BIRNM showed no consistent decline for any of the
39 sh metrics inside BIRNM, but instead showed increases every year between 2003 and 2006 for mean sh density
(all species combined). Densities in 2005 and 2006 were signicantly higher than 2003. It is not yet clear if this has
resulted from initiation of NPS Interim Regulations and enforcement patrols.
• No such increases were recorded outside BIRNM, instead considerable and consecutive inter-annual decline was
apparent for grunt biomass, especially bluestriped grunt (H. sciurus), and density and biomass of stripped parrotsh
(S. iseri). Densities in 2005 and 2006 were signicantly lower than 2003.
• Only three Nassau grouper (Epinephelus striatus), three yellown grouper (Mycteroperca venenosa) and one tiger
grouper (M. tigris) were observed in the study region over the course of six years of monitoring using 1,275 samples.
Notably, these three species were completely absent from the Buck Island nearshore areas in 2001-2006, but were
present in low abundance in 1979. These grouper species are highly vulnerable to shing due to their large body
size and relatively slow maturity and this historical difference in abundance indicates that the grouper have been
overshed.
• In contrast, coney (C. fulva) and red hind (E. guttatus) were more abundant around Buck Island between 2001 and
2006 than in 1979.
• Threespot damselsh (S. planifrons), a potential indicator of healthy reefs with high live coral cover, was more abundant
around Buck Island in 1979 than in the 2001-2006 sampling period.
Macroinvertebrates
• Long-spined urchin densities (Diadema antillarum) around Buck Island have not recovered since the mass mortality in
1983. However, this study and the scientic literature indicate that some minor recovery may be occurring in lagoonal
and back reef areas along the sheltered coastline of northeastern St. Croix. Long-spined urchins were once important
ecosystem engineers controlling the abundance of algae in the region and little is known about the factors (e.g,
limitations to recruitment) that are controlling population recovery.
• Coral reef ecosystems of the study region, particularly the large expanse of seagrasses between Buck Island and St.
Croix support regionally important populations of adult and juvenile queen conch (Strombus gigas). This is important
since queen conch is an important food resource in the Virgin Islands and according to NOAA’s Ofce of Protected
Species, queen conch is declining throughout the species’ range.
Recommendations:
Additional mapping, inventory and monitoring efforts are required to explore the deeper water ecosystems within the BIRNM
that exist outside NOAA’s current benthic habitat map. In addition, acoustic tracking studies may reveal the mechanisms
underlying some of the observed temporal changes in sh communities and will determine connectivity between lagoons
and coral reefs offshore. Tracking will also provide important information on the time that individual sh spend inside and
outside the boundaries of protected areas. Very little is known about the timing of movements during the daily home range,
ontogenetic shifts and spawning migrations and spatial pathways for such movements for most species. Some targeted
surveying for specic substrate types may be required to identify the extent of suitable settlement habitat for juvenile
grouper in the study region or whether groupers are instead immigrating into the region from settlement habitat outside.
Long-term monitoring is necessary to determine the magnitude of the apparent declines and to track the trajectory of
recovery for species that exhibited an increase in density after several years of decline. Long-term monitoring effort may
also reveal direction in the change for the many species that were too highly variable from year to year to provide such
information over the four years of data used. Within the BIRNM-EEMP Marine Protected Area (MPA) complex, resource
managers and stakeholders should examine the option of closing the gap between the southern boundary of BIRNM and
the no-take zone of EEMP along the northeastern coast of St. Croix. An adjoining of the boundaries would incorporate
an extensive area of seagrass habitat thus ensuring full protection of important complementary resources that provide
food and habitat for many sh (both resident and transient species). These seagrass beds are also regionally important
habitat for queen conch and may provide important resources for Caribbean spiny lobster. Further targeted surveys are
required to assess and monitor the status of queen conch populations and to determine whether long-spined urchins are
recovering. Such information will help to determine if management intervention is needed to assist recovery of sea urchin
populations and to evaluate the conch shery. Additional work to map the distribution of juvenile and adult Caribbean
spiny lobster populations using existing survey data and to determine the factors that explain spatial distributions would
be very valuable in supporting ecosystem-based management of marine resources in the region. Benthic habitat maps
should be periodically updated due to the dynamic nature of coral reef ecosystems. This is particularly important when
linking sh seascape structure and when assessing seascape change such as quantifying gain or loss of major habitat
types.
Executive Summary
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. x
p. xi
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Contents
i Executive Summary
ii Table of Contents
iii List of Tables and Figures
1 Introduction of Study Area ................................................................................................................................1
1.1 Background ..........................................................................................................................................1
1.2 Environmental monitoring and ecosystem changes around Buck Island .............................................3
1.3 Benthic habitat mapping in the region ..................................................................................................4
2 Methods ..............................................................................................................................................................5
2.1 Field survey methods ............................................................................................................................5
2.1.1 Benthic habitat surveys ....................................................................................................5
2.1.2 Fish surveys .....................................................................................................................6
2.1.3 Macroinvertebrates count ................................................................................................7
2.1.3 Observer training .............................................................................................................7
2.1.4 Data management ...........................................................................................................7
2.2 Analyses ...............................................................................................................................................7
2.2.1 Characterizing patterns in benthic habitat cover ..............................................................7
2.2.2 Characterizing patterns in fish species communities .......................................................9
2.2.3 Comparison of sh densities and species presence between 1979 and 2001-2006 .......10
2.2.4 Characterizing patterns in macroinvertebrate abundance ...............................................10
3 Results ................................................................................................................................................................11
3.1 Benthic habitat cover .............................................................................................................................11
3.1.1 Characterization of colonized hardbottom areas ..............................................................11
3.1.2 Benthic cover inside and outside BIRNM .........................................................................13
3.1.3 Spatial patterns in benthic cover ......................................................................................14
3.1.4 Seasonal and inter-annual patterns in benthic cover .......................................................19
3.1.5 Mapping threatened Acropora spp. inside and outside BIRNM .......................................21
3.2 Fish communities, groups and species .................................................................................................23
3.2.1 Fish community metrics ....................................................................................................23
3.2.2 Fish community composition ............................................................................................23
3.2.3 Fish groups .......................................................................................................................23
3.2.4 Individual species .............................................................................................................28
3.2.5 Spatial distribution patterns and species-habitat associations .........................................31
3.2.6 Fish size frequency distributions ......................................................................................41
3.2.7 Comparison of sh densities and species presence between 1979 and 2001-2006 .......45
3.2.8 Synoptic overview of inter-annual trends in mean fish metrics (2003-2006) ....................45
3.2.9 Seasonal and inter-annual patterns in fish community metrics ........................................49
3.2.10 Seasonal and inter-annual patterns in fish groups and species .......................................51
3.3 Macroinvertebrate spatial distribution patterns and species-habitat associations ................................57
3.3.1 queen conch (Stombus gigas) ..........................................................................................57
3.3.2 Long-spined urchin (Diadema antillarum) ........................................................................59
3.3.3 Historical comparison of Diadema abundance .................................................................60
3.3.4 Caribbean spiny lobster (Panulirus argus) .......................................................................60
4 Discussion .........................................................................................................................................................61
References ...................................................................................................................................................................67
Appendix A ..................................................................................................................................................................69
Appendix B ...................................................................................................................................................................70
Appendix C ..................................................................................................................................................................71
Appendix D ...................................................................................................................................................................76
Appendix E ...................................................................................................................................................................79
Acknowledgements ......................................................................................................................................................80
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. xii
List of Tables
Table 1. Abiotic and biological variables measured to characterize benthic assemblages along sh transects
in St. Croix. ....................................................................................................................................................6
Table 2. The number of hardbottom benthic habitat sites surveyed by mapped habitat type for the St. Croix
study region as a whole and inside and outside of BIRNM. ..........................................................................7
Table 3
Number of hardbottom benthic habitat sites surveyed by mission and mapped habitat type. ......................8
Table 4.
Length at rst maturity estimates used to determine approximate size classes for juvenile/subadult
and adult sh. ................................................................................................................................................10
Table 5. Mean estimates of percent cover of selected benthic groups inside and outside BIRNM. ............................13
Table 6 Comparison of coral species richness and ratio of coral to macroalgae in major habitats inside and
outside BIRNM. .............................................................................................................................................13
Table
7. Differences in seascape composition (amount and richness of habitat types) surrounding transects
inside and outside BIRNM. ............................................................................................................................14
Table
8. Results of ANOSIM test for signicant difference in sh community composition using species
biomass between samples grouped by habitat type. .....................................................................................25
Table
9. Results of ANOSIM test for signicant difference in sh community composition using species
biomass between samples grouped by habitat type and management domain. ...........................................25
Table
10. Twenty most frequently observed species in the CREM Buck Island survey area ........................................27
Table
11. Summary information on selected species from ve key sh families showing maximum size observed
in the study region (northeastern St. Croix) compared with maximum known size for the species and
the proportion of juveniles found inside and outside BIRNM. ........................................................................41
Table
12. Comparison of mean density for a range of key sh species from 1979 and 2001-2006 monitoring
periods within 500 m surrounding Buck Island.. ............................................................................................45
Table
13. Summary statistics (mean + SE) for a range of sh variables grouped by year (2003-2006) for the
study region (northeastern St. Croix). ...........................................................................................................46
Table
14. Summary statistics (mean + SE) for a range of sh variables grouped by year (2003-2006) inside
BIRNM, northeastern St. Croix. .....................................................................................................................47
Table 15.
Summary statistics (mean + SE) for a range of sh variables grouped by year (2003-2006) outside
BIRNM, northeastern St. Croix. .....................................................................................................................48
Table 16.
Spring and fall total abundance and mean (+ SE) density for the 20 most abundant sh species
across northeastern St. Croix. .......................................................................................................................49
Table 17.
Estimates of total queen conch abundance (number of individuals) by life stage for three islands in
the U.S. Caribbean (2004-2006). ..................................................................................................................58
Table 18.
Abundance of spiny lobster (Panulirus argus) in hard and soft habitats of the study region and inside
and outside BIRNM (northeastern St. Croix) between 2003 and 2006. ........................................................60
Table 19.
Life history characteristics and vulnerability to shing for three large-bodied species and two smaller-
bodied species of grouper. ............................................................................................................................65
Table B1.USVI nsh landings as a proportion of the total nsh landings reported for the U.S. Caribbean in
1980. Listed are the most commonly landed species and species groups. ...................................................70
Table C1.Fish species list and summary data on occurrence, abundance and biomass (2001-2006) for the
study region (northeastern St. Croix) .............................................................................................................71
List of Figures
Figure 1. The island of St. Croix, USVI showing the distribution of surrounding nearshore habitat types using
NOAA’s benthic habitat map and the administrative boundary of BIRNM. .................................................1
Figure 2. Boundaries of the original BIRNM (1961) and the expanded boundary (2001). .........................................2
Figure 3. Image of bleached coral at BIRNM (October 2005). A 1 m
2
quadrat is shown for a scaling reference;
and chronology of major broad-scale stressors to coral reef ecosystem structure and function in
the Buck Island region, St. Croix since 1980. .............................................................................................2
p. xiii
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 4. Benthic habitat maps constructed for the Buck Island region since the 1960s; subset of Gladfelter
et al. (1977); Anderson et al. (1986); and subset of CCMA-BB’s digital map using a 100 m
2
MMU
based on methods described by Kendall et al. (2002). ................................................................................4
Figure 5. Ship survey tracks and bathymetric data from NOAA’s acoustic multibeam seaoor mapping
activities within and surrounding BIRNM. ...................................................................................................4
Figure 6. NOAA benthic habitat map showing hard and softbottom habitat types inside and outside BIRNM. .........5
Figure 7.
Images of NOAA trained observers recording sh species and benthic habitat composition. ....................5
Figure 8.
A selection of habitat types designated in the hierarchical classication scheme of NOAA’s benthic
habitat map for the U.S. Caribbean . ...........................................................................................................6
Figure 9.
Image of queen conch. ................................................................................................................................7
Figure 10.
Seascape sample units of 100 m radius surrounding each sh transect used in a GIS to quantify
variability in seascape composition. ............................................................................................................8
Figure 11.
Percentage cover for key benthic components across hardbottom sites in the study region
(northeastern St. Croix) from 2001-2006. ....................................................................................................11
Figure 12.
Percentage cover for key components of the benthic community across hardbottom habitat types
in the study region (northeastern St. Croix) between 2001 and 2006. .......................................................11
Figure 13. Abundance of coral genera found across hardbottom sites in the study region (northeastern
St. Croix) between 2001 and 2006. ............................................................................................................12
Figure 14. Abundance of coral genera by hardbottom habitat type in the study region (northeastern St. Croix)
between 2001 and 2006. ............................................................................................................................12
Figure 15. Spatial distributions of benthic components at all transects in the study region (northeastern St. Croix)
between 2001 and 2006 for percentage of live coral cover (hard coral including re coral), number
of coral species/groups and rugosity. .........................................................................................................15
Figure 16. Spatial distributions of benthic components at all transects in the study region (northeastern St. Croix)
between 2001 and 2006 for macroalgal cover, algal turf cover and coralline algal cover. ..........................16
Figure 17. Spatial distributions of coral cover for individual coral species at all transects in the study region
(northeastern St. Croix): Diploria strigosa, Montastraea annularis and Montastraea cavernosa. ..............17
Figure 18. Spatial distributions of coral cover for individual coral species at all transects in the study region
(northeastern St. Croix): Siderastrea siderea and Porites astreoides. ........................................................18
Figure 19.
Seasonal and inter-annual patterns of live coral cover inside and outside BIRNM over a four year
sampling period. ..........................................................................................................................................19
Figure 20.
Seasonal and inter-annual patterns of marine plant cover inside and outside BIRNM over a four
year sampling period. ..................................................................................................................................20
Figure 21.
Images of the two types of Acropora species recorded in the study region (northeastern St. Croix). .........21
Figure 22.
Spatial distribution of Acropora palmata and A. cervicornis in St. Croix, U.S. Virgin Islands. . ...................22
Figure 23.
Comparison of mean (+ SE) values inside versus outside BIRNM for biomass of all species, biomass
of all herbivores and biomass of all piscivores (including sharks and rays). ..............................................24
Figure 24.
Comparison of mean (+ SE) values inside versus outside BIRNM for number of sh species,
Shannon-Weiner diversity using abundance data and taxonomic diversity using
presence-absence data. ..............................................................................................................................24
Figure 25. Non-metric multidimensional ordination based on between site similarity in sh community
composition using species biomass data for community similarities by habitat structure, dominant
softbottom habitat type inside versus outside BIRNM and by dominant hardbottom habitat types
inside versus outside BIRNM. ......................................................................................................................25
Figure 26. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for grouper, grunt,
snapper and parrotsh. ...............................................................................................................................26
Figure 27. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for sharks and rays. .......27
Figure 28. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for two grouper
species: coney (C. fulva) and red hind (E. guttatus). ..................................................................................28
Figure 29. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for three snapper
species: yellowtail snapper (O. chrysurus), schoolmaster (L. apodus) and gray snapper (L. griseus). ......29
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. xiv
Figure 30. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for two grunt
(Haemulidae) species: French grunt (H. avolineatum) and bluestriped grunt (H. sciurus). .......................30
Figure 31.
Comparison of mean (± SE) density and biomass inside versus outside BIRNM for three numerically
dominant herbivore species: blue tang (A. coeruleus), striped parrotsh (S. iseri), redband parrotsh
(S. aurofrenatum). ........................................................................................................................................31
Figure 32.
Interpolated spatial surfaces representing number of sh species, herbivorous sh biomass and
piscivorous sh biomass. . ..........................................................................................................................32
Figure 33.
Spatial distributions of juvenile and adult coney (C. fulva) and red hind (E. guttatus) in northeastern
St. Croix. .....................................................................................................................................................33
Figure 34. Mean (+ SE) density for juvenile/subadult and adult by observer habitat type for coney (C. fulva)
and red hind (E. guttatus). .........................................................................................................................34
Figure 35. Spatial distribution of juvenile and adult for yellowtail snapper (O. chrysurus), schoolmaster
(L. apodus) and gray snapper (L. griseus) in northeastern St. Croix. .........................................................35
Figure 36. Mean (+ SE) density for juvenile/subadult and adult by observer habitat type for yellowtail snapper
(O. chrysurus), schoolmaster (L. apodus) and gray snapper (L. griseus). ..................................................36
Figure 37. Spatial distributions of juvenile and adult for French grunt (H. avolineatum), bluestriped grunt
(H. sciurus) and white grunt (H. plumierii) in northeastern St. Croix. ..........................................................37
Figure 38. Mean (+ SE) density for juvenile/subadult and adult by observer habitat type for French grunt
(H. avolineatum), bluestriped grunt (H. sciurus) and white grunt (H. plumierii). .......................................38
Figure 39. Spatial distributions of juvenile and adult for blue tang (A. coeruleus), ocean surgeonsh (A. bahianus),
redband parrotsh (S. aurofrenatum) and striped parrotsh (S. iseri) in northeastern St. Croix. ...............39
Figure 40.
Mean (± SE) density for juvenile/subadult and adult by observer habitat type for blue tang
(A. coeruleus), ocean surgeonsh (A. bahianus), redband parrotsh (S. aurofrenatum) and striped
parrotsh (S. iseri). ......................................................................................................................................40
Figure 41.
Length frequency histogram for key sh families over hardbottom sites inside and outside BIRNM for
grouper, snapper, grunts, parrotsh and surgeonsh. .................................................................................42
Figure 42. Size class frequency histogram for selected sh species over hardbottom sites inside and outside
BIRNM for coney (C. fulva), red hind (E. guttatus), yellowtail snapper (O. chrysurus), schoolmaster
(L. apodus), French grunt (H. avolineatum) and bluestriped grunt (H. sciurus). .......................................43
Figure 43. Size class frequency histogram for selected sh species over hardbottom sites inside and outside
BIRNM for blue tang (A. coeruleus), ocean surgeonsh (A. bahianus), redband parrotsh
(S, aurofrenatum) and striped parrotsh (S. iseri). ......................................................................................44
Figure 44.
Mean (+ SE) density for all species combined by sampling season for both inside and
outside BIRNM. ............................................................................................................................................49
Figure 45.
Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and
outside BIRNM for all sh biomass, herbivorous sh biomass and piscivorous sh biomass. ...................50
Figure 46. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh diversity inside and
outside BIRNM. ...........................................................................................................................................51
Figure 47. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and
outside BIRNM for groupers, snappers, grunts and parrotsh. ..................................................................52
Figure 48. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and
outside BIRNM for coney (C. fulva) and red hind (E. guttatus).. ..................................................................53
Figure 49. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and
outside BIRNM for yellowtail snapper (O. chrysurus), (b) schoolmaster (L. apodus) and
gray snapper (L. griseus). ...........................................................................................................................54
Figure 50. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and
outside BIRNM for French grunt (H. avolineatum) and bluestriped grunt (H. sciurus). .............................55
Figure 51. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and
outside BIRNM for blue tang (A. coeruleus), redband parrotsh (S. aurofrenatum) and
striped parrotsh (S. iseri). ..........................................................................................................................56
Figure 52. Mean (± SE) density for juvenile and adult queen conch (S. gigas) by habitat type, and all queen
conch inside and outside BIRNM by dominant habitat types in the study region (northeastern
St. Croix) between 2004 and 2006. ............................................................................................................57
p. xv
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 53. Spatial distributions of juvenile (immature) and adult (mature) queen conch (S. gigas) density
in the study region (northeastern St. Croix) between 2004 and 2006. .......................................................58
Figure 54.
Sighting frequency of sexually immature, mature, and all queen conch from three study sites in the
U.S. Caribbean: Southwest Puerto Rico; northeastern St. Croix and St. John. ..........................................58
Figure 55. Spatial distribution of long-spined sea urchins (Diadema antillarum) in the study region
(northeastern St. Croix) between 2005 and 2006. ......................................................................................59
Figure 56. Mean (± SE) density for long-spined urchin (Diadema antillarum) by habitat type and inside
and outside BIRNM (northeastern St. Croix) between 2005 and 2006. ......................................................59
Figure 57. The changing abundance of Diadema antillarum in (a) the lagoon and on bank barrier coral reefs
within 500 m of Buck Island and (b) Teague Bay and adjacent nearshore lagoonal environments
showing little to no recovery in over two decades since mass mortality event. ...........................................60
Figure A1.
Map of the East End Marine Park and park zoning. ...................................................................................68
Figure D1.
Spatial distributions of juvenile and adult for bluehead wrasse (T. bifasciatum), queen triggersh
(B. vetula), rock beauty (H. tricolor) and slippery dick (H. bivittatus) in northeastern St. Croix ...................76
Figure D2.
Spatial distributions of juvenile and adult princess parrotsh (S. taeniopterus) and stoplight
parrotsh (S. viride) in northeastern St. Croix. ............................................................................................77
Figure D3. Spatial distributions of juvenile and adult threespot damselsh (Stegastes planifrons), foureye
butterysh (Chaetodon capistratus), spotn butterysh (Chaetodon ocellatus), banded butterysh
(Chaetodon striatus) and great barracuda (Sphyraena barracuda) in northeastern St. Croix. ...................78
Figure E1.
Raw census data grouped by year of survey for sh metrics that exhibited an increase or decline
every year over the study period (2003-2006). ..........................................................................................79
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. xvi
1 - Introduction and Study Area
p. 1
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
1. Introduction and Study Area
1.1 Background
Buck Island Reef National Monument (BIRNM) is located on the northeastern shelf of St. Croix, in the U.S. Virgin Islands
(USVI; Figure 1) and encompasses an uninhabited island of approximately 712,000 m² and the surrounding mosaic of
coral reefs, seagrasses and sand patches. The Monument is under the jurisdiction of the U.S. National Park Service
(NPS) and was originally designated by the U.S. Department of Interior in 1961 according to Presidential Proclamation
3443, in order to preserve the island and the surrounding submerged lands which at that time included “one of the nest
marine gardens in the Caribbean Sea”. The original monument encompassed 880 acres (approximately 3.56 km
2
) and
marine areas were zoned to form a protected “Marine Garden” (259 acres or approximately 1.04 km
2
), which included
extensive stands of the now federally protected elkhorn coral (Acropora palmata) and an area with restricted shing (445
acres or approximately 1.8 km
2
; Figure 2a). The “Marine Garden” was one of the rst “no-take” marine reserves in U.S.
waters and in the Caribbean region. The boundaries were slightly modied in 1975 (Presidential Proclamation 4346), but
it was not until 2001 that the monument was greatly expanded to 19,015 acres (approximately 77 km
2
) under Presidential
Proclamation 7392 (Figure 2b). At that time, new regulations were enacted making the entire monument a no-take and
“restricted anchoring” zone. The BIRNM expansion was the rst substantial no-take area established for the island of St.
Croix and it now protects about 7.4 percent of the St. Croix shelf area. The expansion resulted in a 10-fold increase in
protection of shallow water (<30 m) hardbottom and sand habitat types and a seven-fold increase for seagrasses when
compared with the 1961 Monument (Kendall et al., 2004a). In January 2003, BIRNM became contiguous with the East
End Marine Park (EEMP) through the adjoining of the southern boundary of BIRNM and northern boundary of EEMP.
However, over 80% of EEMP is open to shing including an area that extends between the southern boundary of BIRNM
and the EEMP no-take coastal lagoon zone (see zoning map in Appendix A). In April 2003, NPS implemented the Interim
Regulations (36 CFR Part 7.73; Federal Register Volume 68, No. 65) and begun work on the General Management Plans
for BIRNM.
Figure 1. The island of St. Croix, USVI showing the distribution of surrounding nearshore habitat types using NOAA’s benthic habitat
map (Kendall et al., 2002) and the administrative boundary of BIRNM.
Study region
In recent decades, the ecological structure and function of coral reef ecosystems of the northeastern St. Croix study region
have deteriorated dramatically due to a combination of stressors including shing, anchor drops, excessive nutrient inputs,
Diadema die-off, mass coral bleaching related to anomalous sea water temperatures, the emergence of widespread coral
diseases and extensive hurricane damage (Bythell et al., 1993; Rogers and Beets, 2001). The enlarged monument now
incorporates components of the marine ecosystem, which have been impacted by shing of nsh, conch and lobster.
Currently the expanded area is being illegally shed using hand and rod shing, spear shing, sh traps, gill or trammel
nets, and long-lines in the deeper portions of the Monument. Law enforcement patrols have been active since 2003 and
compliance is increasing. These deleterious environmental changes together with signicant changes in management
strategies, such as boundary expansion, require that the ecological patterns and processes characterizing the region be
adequately inventoried, spatially characterized and continuously monitored to support the resource-management decision-
making process. Knowledge of the current status of sh communities coupled with a spatially explicit understanding of the
key resources and followed by a program of long term monitoring of sh and benthic communities will enable evaluation
of management efcacy that is required to guide future management actions. In addition, comparison between managed
versus unmanaged areas (e.g., inside and outside the protected area) allows managers to assess the impact, if any, of a
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 2
Figure 2. (a) The boundaries of the original Buck Island Reef National Monument established in 1961 and (b) the expanded
boundary established in 2001. Source: National Park Service, St. Croix.
change in regulation and evaluate the resources that are included or excluded from protection. This report represents an
evaluation and characterization of sh and their habitat both inside and outside BIRNM and summarizes the rst six years
of long-term monitoring data collected using consistent survey methods and a stratied-random sampling design.
Providing park managers with scientically validated evidence of reserve effectiveness or ineffectiveness is not only
essential to informing resource management, it is critical to building public support for management plans. In response,
NPS in partnership with NOAA’s Center for Coastal Monitoring and Assessment Biogeography Branch (CCMA-BB)
initiated their Caribbean Coral Reef Ecosystem Monitoring (CREM) project at BIRNM in February 2001. In 2003, the U.S.
Virgin Islands Department of Planning and Natural Resources joined this collaborative effort to monitor the broader region
including the East End Marine Park (EEMP). This effort supports objectives of the NPS Inventory and Monitoring Program
with an objective to “improve park management through greater reliance on scientic knowledge”, BIRNM is one of seven
National Parks forming the NPS’s South Florida/Caribbean Network (NPS-SFCN) with special requirements for producing
natural resource inventories and conducting ecosystem monitoring. This project provides data and data interpretation to
meet NPS BIRNM Government Performance Results Act Goals pertinent to Threatened and Endangered Species Ia2A,
Natural Resource Data Sets IB01 and Visitor Understanding IIb1 by providing new information on the condition of the
monuments marine resources.
Figure 3. (a) Bleached coral at BIRNM (October 2005). A 1 m
2
quadrat is shown for a scaling reference. (b) Chronology of major broad-
scale stressors to coral reef ecosystem structure and function in the Buck Island region, St. Croix since 1980.
b)
a)
2005
2000
1995
1990
1985
1980
Hurricane Hugo
Fall 1989
}
Diadema die-off, early 1980s
Mass Acroporid death
Coral diseases, early 1980s
Mass coral bleaching, Fall 1998
Hurricane Lenny, Fall 1999
Hurricane Marilyn, Fall 1995
Mass coral bleaching, Fall 2005
1 - Introduction and Study Area
p. 3
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
1.2 Environmental monitoring and ecosystem changes around Buck Island
Many marine ecological studies have been carried out in the region with long-term monitoring studies rst conducted
by scientists at the West Indies Laboratory (WIL) in the 1970s and a series of permanently marked sites for long-term
monitoring of coral reef community structure and function initiated by the NPS in the 1980s. Results of early monitoring
efforts have revealed some major changes, primarily in live coral cover and structure that have occurred across the region.
Gladfelter et al. (1977) reported greater than 50% Acropora palmata on the reef crest of the north and south bank-barrier
reefs and the northern forereef around Buck Island, but by 1984 large areas of dead A. palmata colonies encrusted with
algae and gorgonians and other dead Acropora species (including A. cervicornis) were reported and were thought to have
resulted from a mass mortality event caused by coral diseases such as white-band disease (Anderson et al., 1986). At the
same time, the widespread die-off of Diadema antillarum sea urchins in the early 1980s, a key algal grazer and ecosystem
engineer, altered the ecosystem dynamics on many shallow-water coral reefs in the region (Lessons et al., 1984). In 1989,
Hurricane Hugo passed directly over the region with reported wind speeds of 260 kph. Bythell et al. (1993) and Rogers et
al. (1982) resurveyed permanent transects and reported extensive localized damage with the southeast reef front razed
to substrate level between the surface and 7 m depth and the reef crest behind it smothered in a 1 m deep layer of broken
coral rubble. Although some coral had recovered by 1991, the community composition remained altered. In the 1990s the
region was again impacted by hurricanes with Hurricane Marilyn in 1995 and Hurricane Lenny in 1999.
Extensive bleaching was observed at BIRNM in the fall of 1998 when water temperature reached a maximum of 29.9
o
C
(Rogers and Beets, 2001). More recently, in 2005, CCMA-BB, NPS-SFCN and U.S. Geological Survey (USGS) scientists
observed widespread coral bleaching around BIRNM, which was part of a mass bleaching event that occurred throughout
the tropical western Atlantic (Clark et al., in press). During October 2005, over 90% of the coral was bleached at almost
all NPS/USGS permanent monitoring sites (J. Miller, pers. comm.). Furthermore, bleaching was observed in 91 of 94
randomly selected survey sites, with an estimated 53% of the coral cover bleached (Clark et al., in press; Figure 3).
In addition, changes in the amount and spatial arrangement of seagrasses and sand habitat types has occurred in the
BIRNM region, and have been reported in neighboring islands, yet are largely unquantied (but see Rogers and Beets,
2001; Kendall et al., 2005). The chronology of major events resulting in deterioration of coral reef ecosystem structure and
function in the St. Croix study region is depicted in Figure 4.
Various species of bleached coral. All photographs were taken during the October 2005 St. Croix mission.
Fish and sh communities respond to changes in their environment including habitat loss and extractive activities, such
as shing, but few historical monitoring surveys have focused on sh communities. Illegal commercial shing has also
been observed within BIRNM (NPS records). The Caribbean Fisheries Management Council (CFMC) reports that USVI
sheries target approximately 180 species of sh (64 species commonly caught), queen conch (Strombus gigas) and
spiny lobster (Panulirus argus; CFMC, 1985; Appendix B).
1.3 Benthic habitat mapping in the region
In 1976, the rst benthic habitat map was drawn by scientists at the WIL to depict the spatial distribution of marine habitat
types around Buck Island using both quantitative and qualitative ground-truthing of aerial photographs (Gladfelter et al.,
1977; Figure 4a). Almost a decade later, Anderson et al. (1986) updated the original benthic habitat map using 1984
ground-truthed aerial photographs (Figure 4b). In 1999, NOAA’s National Ocean Service acquired aerial photographs in
order to create benthic habitat maps in response to a need to identify Essential Fish Habitat (EFH) in the U.S. Caribbean.
CCMA-BB digitized benthic habitat for a 490 km
2
area of nearshore coral reef ecosystems in the USVI using a 1 acre
1 - Introduction and Study Area
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 4
c)
a)
b)
Figure 4. Benthic habitat maps constructed for Buck Island region since the 1960s. (a) Subset of Gladfelter et al. (1977); (b) Anderson
et al. (1986); and (c) subset of CCMA-BB’ s digital map using a 100 m
2
MMU based on methods described by Kendall et al. (2002).
(approximately 4,047 m
2
) minimum mapping
unit (MMU). Thematic accuracy around the
test area of BIRNM was assessed using 120
stratied-random benthic surveys resulting in
an overall map accuracy of 93.6%, with 100%
users accuracy for submerged vegetation, 97.2%
for hardbottom habitat types and 86.1% for
sand (Kendall et al., 2002). Data and methods
are available online: http://ccma.nos.noaa.gov/
products/biogeography/benthic/htm/data.htm
In addition, with an objective to study sh-
seascape relationships at a ner spatial scale, an
area of approximately 50 km
2
around Buck Island
was also digitized to a spatial resolution of 100 m
2
(Figure 4c).
In 2004, 2005 and 2006 CCMA-BB and NOAA’s
Ofce of Coast Survey, in collaboration with NPS,
USVI Territory and private sector partners, used
multibeam sonar and underwater video to map
bottom features (>20 m depth) and characterize
nearshore benthic structure around BIRNM
(Figure 5). These data are a component of the
Seaoor Characterization of the Caribbean project
supported by NOAA’s Coral Reef Conservation
Program (CRCP) and are available online at http://
ccma.nos.noaa.gov/products/biogeography/
usvi_nps/overview.html. Data includes 5 m point
data les, digital terrain models and mosaics of
the acoustic backscatter.
Figure 5. (a) Ship survey tracks and (b) bathymetric data from NOAA’s acoustic
multibeam seaoor mapping activities within and surrounding BIRNM.
a)
b)
1 - Introduction and Study Area
2 - Methods
Methods2.
To assist in monitoring coral reef ecosystem
resources and to achieve a better understanding
of sh-habitat relationships in the U.S. Caribbean,
CCMA-BB developed a sh and macro-
invertebrate monitoring protocol to provide
precise, shery-independent and size-structured
survey data, needed to comprehensively assess
faunal populations and communities (Menza et
al., 2006). In addition, a complementary benthic
composition survey was also developed to support
studies of sh-habitat relationships. These data
collection activities and analytical products are
core components of NOAA’s CRCP implemented
through CCMA-BB’s CREM project. CREM
protocols were created primarily to quantify long-
term changes in sh species and assemblage
diversity, abundance, biomass and size structure
and to compare these metrics between areas
inside and outside of Marine Protected Areas
(MPAs). A stratied random sampling design was
used to optimize the allocation of samples and
allow rigorous inferences to the entire study area,
as well as, the selected management domains
(e.g., inside and outside BIRNM). Two strata were
selected based upon: 1) the study objectives;
2) parsimony in the approach; and 3) results
from statistical analyses of variance (Menza
et al., 2006). The “hard” stratum comprised
bedrock, pavement, rubble and coral reefs. The
“soft” stratum comprised sand, seagrasses and
macroalgal beds. In 2003, NPS management
domains were incorporated as a second level of spatial stratication and were designated as “inside” and “outside”
BIRNM (Figure 6).
2.1 Field survey methods
This report uses underwater census data
collected in March/April and October/
November each year from 2003 to
2006. This data set is part of a broader
ongoing monitoring study that began
in year 2001, with over 1,300 transects
surveyed thus far around BIRNM. There
are two complementary components to the
biological eld methods: (1) benthic habitat
composition surveys and (2) sh surveys.
2.1.1 Benthic habitat composition surveys
To conduct benthic habitat surveys, a second
observer places a 1 m
2
quadrat divided into
100 (10 x 10 cm) smaller squares (1 square
= 1% cover) at ve randomly pre-selected locations along the transect, such that a quadrat is placed once somewhere
within every 5 m interval along the transect (Figure 7). Percent cover is estimated within the quadrat in a two-dimensional
plane perpendicular to the observer’s line of vision.
Information recorded include:
Habitat structure1) (e.g., colonized hardbottom, spur and groove, patch reef, pavement) - based on the habitat types
used in the benthic habitat maps (Kendall et al., 2002; Figure 8), until 2004, after which habitat structure was classied
only to hard, soft and mangrove.
Abiotic footprint2) - dened as the percent cover (to the nearest 1%) of sand, rubble, hardbottom, ne sediments and
other non-living bottom types within a 1 m
2
quadrat.
Biotic footprint3) - dened as the percent cover to the nearest 1% of algae, seagrass, upright sponges, gorgonians and
other biota and to the nearest 0.1% for live, bleached and recently dead/diseased coral within a 1 m
2
quadrat.
Figure 7. NOAA trained observers recording sh species abundance and body
length along a 25 x 4 m timed 15 minute belt transect (left); and benthic habitat
composition recorded within ve randomly placed 1 m
2
quadrats along the belt
transect (right).
p. 5
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 6. NOAA’s benthic habitat map showing hard and softbottom habitat
types both inside and outside BIRNM. Spatial information was used to identify
strata within which to allocate random samples for CREM sh and benthic
habitat composition surveys (gure adapted from Menza et al., 2006).
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 6
Figure 8. A selection of habitat types designated in the hierarchical classication scheme of NOAA’s benthic habitat map (Kendall et
al., 2002) for the U.S. Caribbean (clockwise from left to right): colonized pavement, patch reef, scattered coral and/or rock, linear reef,
seagrass and sand.
Transect depth prole4) - the depth at each quadrat position. Depth is measured with a digital depth gauge and rounded
up or down to the nearest foot.
Maximum canopy height5) - for each biota type, height of soft structure (e.g., gorgonians, upright sponges, seagrass,
algae) structure is recorded to the nearest 1 cm.
Hardbottom rugosity6) - measured by placing a 6-m chain at two randomly selected start positions ensuring no overlap
along the 25-m belt transect. The chain is placed such that it follows the relief along the centerline of the belt transect.
Two divers measure the straight-line horizontal distance covered by the chain.
Proximity of structure7) - on seagrass and sand sites, the habitat diver records the absence or presence of reef or hard
structure within 4 m of the belt transect.
Table 1 provides a list of measured variables.
The habitat observer also counts queen conch
(Strombus gigas), long-spined sea urchins
(Diadema antillarum) and Caribbean spiny lobster
(Panulirus argus) Further information about
macroinvertebrate data collection is described
in section 2.1.3. Conch were counted separately
as mature and immature animals based on lip
thickness and shell size.
2.1.2 Fish surveys
Fish surveys were conducted along a 25 m long
by 4 m wide belt (100 m
2
) using a xed survey
duration of 15 minutes (Figure 7). The xed
duration of 15 minutes standardizes the samples
collected to facilitate between site comparisons.
The number of individuals per species is recorded
in 5 cm size class increments up to 35 cm using
the visual estimation of fork length. Individuals
greater than 35 cm are recorded as an estimate of
the actual fork length to the nearest centimeter.
Table 1. Abiotic and biological variables measured to characterize benthic
assemblages along sh transects in St. Croix.
Benthic Biota
Measurements
% Cover Height (cm) Abund. (#)
Abiotic
Hardbottom X X
Sand X
Rubble X
Fine sediment X
Rugosity
Water depth
Biotic
Corals (by species) X
Macroalgae X X
Seagrass (by species) X X
Gorgonians
Sea rods, whips and plumes X X X
Sea fans X X X
Encrusting form X
Sponges
Barrel, tubes, vase morphology X X X
Encrusting morphology X
Other benthic macrofauna
Anemonies and hydroids X X
Tunicates and zoanthids X
Macro-invertebrates
queen conch (by sexual maturity) X
Spiny lobster X
Long-spined urchin X
2 - Methods
p. 7
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
2.1.3 Macroinvertebrates counts
Queen conch
The abundance of immature and mature queen conch (Strombus gigas) was assessed and quantied within the 25 x 4
m belt transects used for sh surveys. The maturity of each conch was determined by the presence (mature) or absence
(immature) of a ared lip (Figure 9). Conch was included in the survey protocol from 2004 onward.
Figure 9. Image of queen conch and
location of ared lip.
Caribbean spiny lobster
Abundance of Caribbean spiny lobsters (Panulirus argus) are reported for the period 2003 to
2006. Lobster sightings were recorded during sh and benthic composition surveys (i.e., within the
100 m
2
survey unit area). Lobsters were recorded if seen, but without active searches of holes or
crevices.
Long-spined sea urchins
Long-spined sea urchins (Diadema antillarum) were also counted within the 25 x 4 m belt transect
between 2004 and 2006. No measurements of size or estimates of maturity were collected.
2.1.4 Observer training
Observers were trained and tested in the identication of species/groups for both sh and habitat surveys by pairing
inexperienced and experienced observers in the water and comparing data. Fish size estimation training was carried out
in situ by estimating lengths of model sh of various shapes and sizes.
2.1.5 Data management
All sh and benthic habitat survey data were quality assessed before storage on an online relational database. All
survey data were stored with a unique identication number and a geographical coordinate to facilitate spatial analyses.
The database including metadata that provide detailed eld methods are available at: http://ccmaserver.nos.noaa.gov/
ecosystems/coralreef/reef_sh/protocols.html.
2.2 Analyses
2.2.1 Characterizing patterns of benthic
habitat cover
The benthic habitat section of the report
provides summary data from 716 benthic
in situ surveys (approximately 3,580
quadrats) on hardbottom habitat types
(e.g., linear reef, colonized pavement,
patch reefs) around BIRNM and the
northeastern shore of St. Croix between
2001 and 2006 (Tables 2 and 3). The
number of surveys conducted during any
single mission was relatively low for the
least abundant hardbottom habitat types
(i.e., reef rubble and bedrock), but high for
Table 2. The number of hardbottom benthic habitat sites surveyed by mapped habi-
tat type for the St. Croix study region as a whole and inside and outside of BIRNM.
Mapped habitat categories are from Kendall et al. (2002).
Mapped habitat types
Number of sites surveyed
Area (km
2
) % Area
Inside Outside Total
Bedrock 0.58 2% 13 14 27
Linear reef 1.56 5% 22 33 55
Patch reef 4.10 13% 98 4 102
Pavement 22.19 70% 274 199 473
Reef rubble / macroalgae 0.23 1% 2 4 6
Scattered coral and/or rock 3.22 10% 36 17 53
Total 31.88 100% 445 271 716
Flared lip
of shell
2 - Methods
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 8
colonized pavement and patch reefs (Table 3). Although many benthic variables have been measured during the surveys,
this report focuses primarily on the areal abundance (% cover) of the sessile biotic components (Table 1). In addition, we
also compared habitat composition at a broader scale by quantifying the amount of each habitat type and the number
of habitat types in the seascape surrounding each sh and benthic habitat survey site. These seascape variables were
quantied within a 100 m radius buffer using a custom-built Geographical Information Systems (GIS) tool developed
specically for this project (Diversity Calculator for ArcGIS 9.2 is freely available at http://arcscripts.esri.com). Essentially,
the buffer is analogous to a quadrat, but instead of quantifying percentage cover of ne-scale components such as coral
cover we quantify the amount of habitat type from the benthic habitat map. The selection of a 100 m radial seascape
sample unit was determined from a review of several studies that have identied the rst 100 m surroundings as most
inuential in determining sh species distributions (Kendall et al., 2003; Pittman et al., 2007; Figure 10).
Differences in the abundance of individual components of the benthos inside versus outside BIRNM were tested using
the parametric Tukey’s HSD (Honestly Signicant Difference) pairwise comparison for normally distributed data and the
nonparametric Wilcoxon test for non-normally distributed data. Broad spatial patterns in the benthic variables were determined
from visual interpretation of mapped values and simple deterministic interpolations. Interpolations were performed using
the Inverse Distance Weighting (IDW) interpolator
with a relatively small neighborhood of samples
(n=5 points) to create spatial surfaces within a
GIS. This technique makes few assumptions of
the data and estimates cell values by averaging
sample points within a neighborhood. The values
are distance weighted such that points closer to
the center of a cell are assigned more weight in the
averaging process. All data were used to construct
surfaces including samples where the measured
variable was recorded as zero. For marine algae,
the entire surface is shown (e.g., both hard and
softbottom areas) since spatial variability of algae
across seagrasses may also be informative. For
other biotic components, such as corals, softbottom
areas were masked out of the interpolated surface
since few corals exist in sand and seagrass beds.
The intent was not to create detailed and accurate
spatial predictions, but instead to show a spatially
continuous representation of ne-scale point data
to aid in interpretation of broad-scale distribution
patterns. The spatial extent of all analyses in this
report is limited to the mapped portions of the study
area.
Figure 10. Seascape sample units of 100 m radius surrounding each sh
transect were used in a GIS to quantify variability in seascape composition.
In this example, it is clear that each seascape unit is characterized by very
different patch type composition and patch richness.
Table 3. Number of hardbottom benthic habitat sites surveyed by mission and mapped habitat type. Mapped habitat categories are
from Kendall et al. (2002).
Sample period Location Bedrock Linear reef Patch reef Pavement
Reef rubble/
Macroalgae
Scattered
coral / rock Total
2001 Spring Inside 3 3 16 24 2 48
2001 Fall Inside 1 10 13 2 26
2002 Spring Inside 5 5 18 2 5 35
2002 Fall Inside 2 16 12 3 33
2003 Spring Inside 3 10 27 1 41
Outside 1 5 29 2 37
2003 Fall Inside 1 5 3 25 2 36
Outside 1 9 24 1 35
2004 Spring Inside 1 12 2 15
Outside 1 2 1 6 2 12
2004 Fall Inside 6 26 3 35
Outside 3 5 19 1 2 30
2005 Spring Inside 2 4 6 18 3 33
Outside 4 5 1 20 1 31
2005 Fall Inside 8 37 3 48
Outside 3 36 1 6 46
2006 Spring Inside 10 31 5 46
Outside 1 2 1 30 2 2 38
2006 Fall Inside 4 2 7 31 5 49
Outside 3 2 1 35 1 42
Total 27 55 102 473 6 53 716
2 - Methods
p. 9
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
2.2.2 Characterizing patterns in sh species and communities
Assessing differences in univariate metrics inside versus outside BIRNM
Differences in univariate community metrics as well as individual species/group metrics inside versus outside BIRNM
were tested using parametric and nonparametric tests as appropriate. Only data for years 2003 to 2006 were examined
because allocation of samples prior to 2003 occurred using different strata (n=431 inside BIRNM; 291 hardbottom and
140 softbottom; n=453 outside BIRNM; 303 hardbottom and 150 softbottom). Designations of habitat type (colonized
hardbottom, seagrasses and unvegetated sediments) were collected by benthic habitat observers.
Diversity was measured using the Shannon Index (H’; Equation 1). In this way, the diversity measure incorporates richness,
commonness and rarity. Although, the Shannon Index has been shown to be an effective discriminator of community
structure it is not independent of sample size (Magurran, 1988). Taxonomic indices, on the other hand are considered to
be signicantly less inuenced by sample size than the conventional species richness, evenness and diversity indices
(Warwick and Clarke, 1995) and, therefore, more appropriate for any comparative studies with unbalanced sampling effort
(Clarke and Warwick, 1998).
Taxonomic indices
Samples may differ in the way assemblages are composed at the genus, family, order, class and phylum levels of the
standard Linnean taxonomic hierarchy. For example, species diversity may be similar between two samples, yet one
may support several species belonging to the same family, while the other may support several species, all belonging
to different families and even different classes orders, etc. Quantitative taxonomic diversity indices therefore provide an
additional dimension of information that is likely to be more closely linked to functional diversity (Clarke and Warwick,
1999). The importance of this measure of diversity is that families, orders, etc. as opposed to species, represent a greater
variety of fundamentally different body plans and life histories.
As such Taxonomic diversity (∆) (Warwick and Clarke, 1995; Equation 2) was measured for all samples. Fish were
distinguished at four taxonomic levels: species, genus, family and class.
Samples were grouped by habitat type as determined by benthic habitat observers and by management domain as
determined by the mapped strata (e.g., inside and outside BIRNM).
Assessing differences in community composition inside versus outside BIRNM
Differences and similarities in the species composition of communities between samples (often referred to as assemblage
or community structure) were examined using a species biomass by site data matrix. Samples with zero sh were removed
from the data matrix. Infrequently observed species, with extreme outlying biomass were removed, including small-
bodied pelagic schooling sh (e.g., Clupeidae, Antherinidae, etc.) and large-bodied broad ranging species (e.g., sharks,
rays, barracuda). Infrequently observed sh that were not identied to species level were also removed. The matrix was
square-root transformed to ensure that intermediate biomass species, in addition to the high biomass species, played a
signicant role in determining patterns in community composition. The data was then used to construct a matrix of the
percentage similarity in community composition between all pairs of sites using the Bray-Curtis Coefcient (Equation 3).
H’ = -
S
i
p
i
( log
e
p
i
) (Equation 1)
Where H’ is a weighted combination of: total number of species (richness) and the extent to which the total
abundance is spread equally amongst the observed species (evenness). p
i
is the proportion of the total
count arising from the ith species.
(Equation 2)
Letting x
i
denote the presence or absence of the ith species and the W
ij
the “distinctness weight” given to
the path length linking species i and j in the hierarchical classication, then taxonomic diversity (D) is
dened simply as the average (weighted) path length between every pair of individuals. The null second term
in the numerator has been included to emphasize that the weight for the path linking individuals of the same
species is taken to be zero.
D =
SS
i<j
W
ij
X
i
X
j
+ S
i
0. x
i
(x
i
-1) / 2
SS
i<j
x
i
x
j
+ S
i
x
i
(x
i
-1) / 2
2 - Methods
(Equation 3)
Where x
ij
is the abundance of the ith species in the jth sample and where there are n species overall.
S’
jk
= 1
-
S
n
X
ij
-
X
ik
i=1
S
n
X
ij
+ X
ik
i=1
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 10
This algorithm is considered a robust estimator of ecological distance and has had widespread usage in ecology particularly
for comparison of biological data on community structure (Faith et al., 1987). Its robustness is in part due to its exclusion
of double zeros, that is, if two samples are missing the same species, they will not be regarded as similar based on the
same absentees (Legendre and Legendre, 1998). This similarity coefcient reduces the comparison between all pairs of
samples to single numerical values that are arranged in a secondary matrix from which pattern is examined. Sample sites
were assigned a factor representing a dominant habitat type (e.g., either colonized hardbottom, seagrasses or sand) and
a management domain (e.g., inside or outside BIRNM). Factors were used to identify pairs of treatments in order to test
for signicant differences using Analysis of Similarities (ANOSIM), a multivariate version of Analysis of Variance (ANOVA;
Primer v5; Clarke and Warwick, 1994), and for visual examination of patterns of between site similarity using a non-
metric dimensional scaling plot (nMDS). In addition,
Similarity percentages (SIMPER) were calculated and
used to identify the species which contributed most to
the differences between treatments (Primer v5; Clarke
and Warwick, 1994).
Where species groups were used, herbivores included
all species that were important consumers of marine
algae; piscivores included all sh that were important
predators of sh; snapper included all Lutjanidae
spp.; groupers included all commercially harvested
Serranidae spp.; grunts included all Haemulidae.
Juveniles/subadults were identied based on length at
maturity information provided by García-Cagide et al.
(1994) and FishBase (http://www.shbase.org, version
11/2007), whereby, juveniles/subadults were sh with
lengths less than the mean length at maturity and the
remainder were considered as adults (Table 4). If the
mean length at maturity was 14 cm then size classes <5
and 5-10 cm were considered juvenile/subadult. Where
length at maturity was unknown, 1/3 of maximum adult
size was used to segregate juveniles/subadults from
adults. For mapping of juvenile and adult distribution
all samples were used from 2001 to 2006.
2.2.3 Comparison of sh densities and species presence between 1979 and 2001-2006
Data on mean sh densities from visual surveys conducted between January and September 1979 (Gladfelter, 1980)
were used for comparison with 2001-2006 data. The 1979 surveys were conducted during the day at ve hardbottom
sites (each one of 40 x 40 m
2
) within 500 m of Buck Island. Each of the ve sites received replicate surveys (North lagoon
n= 30, SW lagoon n=30, NW leeward n=30, S forereef n=32, E forereef n=25) resulting in a total sample size of 147.
These sites were not randomly located, but were selected to represent a variety of coral reef environments across a
complexity gradient. Census involved swimming back and forth across the study site counting all sh observed. A mean
density for each species of interest was calculated from pooled data on mean densities for each of the ve plots and was
standardized to 100 m
2
for comparative purposes. To provide a comparison with similar environments using CCMA-BB
survey data from 2001-2006, only sh transects conducted over hardbottom habitat types within 500 m of Buck Island
were used. This resulted in a sample size of 184 spatially random samples. Differences in technique were clearly evident
resulting in limitations when attempting to undertake direct comparison, yet data can be usefully compared for presence
and absence of species and any large differences in density between the two time periods.
2.2.4 Characterizing patterns in macroinvertebrate abundance
For Diadema antillarum (long-spined sea urchins) and Strombus gigas (queen conch) mean (±SE) density and summary
statistics were calculated by habitat type using NOAA’s benthic habitat map. To examine spatial distribution patterns,
density data was overlayed on the benthic habitat map. In addition, for queen conch only, mean (±SE) abundance of
juvenile and adults were determined for three Caribbean islands based on the area weighted abundance for each benthic
habitat type in which queen conch surveys were conducted. Data collected between 2004 and 2006 were pooled to allow
an adequate sample size (n≥2) with which to calculate means within each habitat type. The following equation was used
to calculate total abundance estimates for each queen conch life stage:
Table 4. Length at rst maturity estimates used to determine approximate
size classes for juvenile/subadult and adult sh. Estimates are derived
from data held by FishBase (http://www.shbase.org, version 11/2007).
Species
Mean length at rst
maturity, L
m
(cm)
Juvenile/subadult
size class (cm)
Acanthurus bahianus 15.5
<15
Acanthurus coeruleus unknown
<10
Balistes vetula 25
<20
Cephalopholis fulva 16
<15
Epinephelus guttatus 25
<20
Halichoeres bivittatus unknown
<10
Haemulon avolineatum 16
<15
Haemulon plumierii 19
<15
Haemulon sciurus 18.5
<15
Holacanthus tricolor 17.4
<15
Lutjanus apodus 25
<20
Lutjanus griseus 31
<20
Ocyurus chrysurus 24.5
<20
Scarus iseri unknown
<10
Sparisoma aurofrenatum unknown
<10
Sparisoma viride 16.3
<15
Thalassoma bifasciatum unknown 0-5
A X
h
h
l
h
=
∑
1
2 - Methods
where A is the total area of each mapped habitat, is the mean density (# of queen conch/m
2
) in each habitat, and l is
the total number of mapped habitats in each island. A sampling unit was a 25 x 4 m wide belt transect or a 100 m
2
area.
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Mean densities were derived from multiple surveys that occurred within each habitat type. A range of the total abundance
was calculated with the equation:
( )
A X S E
h
h
l
h
=
∑
±
1
.
where SE is the standard error of the mean queen conch density in each mapped habitat in each island.
p. 11
3. Results
3.1 Benthic habitat cover
Colonized pavement was the
most spatially extensive habitat
type (70% of the study area) and
was therefore most intensively
surveyed, followed by patch reefs
(13%) and linear reef (5%; Table 3).
Hardbottom habitat types combined
formed a larger proportion (78%) of
BIRNM than did softbottom areas
(22%). In contrast, outside BIRNM,
softbottom habitat types formed a
larger proportion (54%) of the total
mapped area than hardbottom.
Estimates of percent cover (mean
± standard error [SE]) of selected
benthic organisms are reported
for: (1) the entire study area for all
habitat types; and (2) inside and
outside BIRNM using only three of
the most abundant habitat types
(colonized hardbottom, seagrasses,
sand sites). These comparisons
at broader thematic resolution are
intended to highlight any major
differences between inside and
outside the protected area.
3.1.1 Characterization of colonized
hardbottom types
Generally, colonized hardbottom
habitat types were dominated by
algae (36.7% ± 1.1% turf algae,
11.4% ± 0.5% macroalgae, and
1.8% ± 0.2% crustose coralline algae
[CCA]; Figure 11). Dictyota spp.,
Halimeda spp. and Sargassum spp.
were most abundant. Cyanobacteria
and lamentous algae were grouped
as a single component and had a
mean cover of 4.3% ± 0.5%. Mean
live scleractinian coral cover was
5.6% (± 0.5%) across the region.
The mean percent cover of live
scleractinian coral was highest on
patch reefs (12.1% ± 1.3%; p<0.05)
and lowest on reef rubble (2.0% ±
0.8%) and scattered coral and rock
sites (3.4% ± 0.7%; Figure 12).
Figure 11. Percentage cover for key benthic components across hardbottom sites (n=716) in
the study region (northeastern St. Croix) between 2001 and 2006. CCA= crustose coralline
algae; CB and FA= cyanobacteria and lamentous algae.
Figure 12. Percentage cover for key components of the benthic community across hardbottom
habitat types (n=716) in the study region (northeastern St. Croix) between 2001 and 2006.
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 13. Abundance of coral genera found across hardbottom sites in the study region
(northeastern St. Croix) between 2001 and 2006.
p. 12
Figure 14. Abundance of coral genera by hardbottom habitat type in the study region
(northeastern St. Croix) between 2001 and 2006.
3 - Results
Gorgonians were highest on colonized pavement and lowest on reef rubble sites. The percent cover of sponges and re
corals were similar among the habitat types surveyed (Figure 12).
Live scleractinian coral cover included at least 23 coral genera, but only nine with a mean cover greater than 0.01% (Figure
13). The three most abundant coral genus were Diploria spp. (1.2% ± 0.29%), Montastraea spp. (1.0% ± 0.09%) and
Porites spp. (0.9% ± 0.06%). Diploria spp. cover was highest on colonized bedrock, linear reef and colonized pavement;
Montastraea spp., Porites spp. and Acropora spp. was highest on patch reef and linear reef habitat types (Figure 14).
p. 13
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
3.1.2 Benthic cover inside and outside BIRNM
Benthic composition of the major habitat types (hardbottom,
seagrass and unvegetated sediments) occurring inside
BIRNM was signicantly different from those found outside.
Colonized hardbottom inside BIRNM had signicantly higher
percent cover of live coral than colonized hardbottom sites
outside and slightly lower macroalgal cover (Table 5). The
cover of living Acropora was an order of magnitude greater
inside compared with outside and percent cover of living
Montastraea and Diploria was more than twice as high
inside than outside (Table 5). Gorgonians and sponges also
had higher mean percent cover inside BIRNM compared
with outside.
Fewer signicant differences were observed in seagrass
and sand habitats. Percent cover of CCA and gorgonians in
seagrass habitats were higher inside BIRNM compared with
outside, whereas percent cover of sponges and seagrass in
seagrass habitats were higher outside BIRNM than inside. In
areas mapped as sand, only sponges and seagrass showed
signicant inside versus outside differences. Both had higher
percent cover outside BIRNM than inside (Table 5).
Mean coral species richness inside BIRNM was slightly
higher than outside, but not signicantly different (Table
6). Colonized hardbottom habitat inside BIRNM had a
signicantly higher coral:macroalgal ratio compared with
outside. This difference reects the higher percent cover of
coral and lower cover of macroalgae inside BIRNM relative
to outside and suggests BIRNM may contain coral reefs in
better condition than surrounding areas.
Furthermore, at a broader spatial scale of 100 m radius,
benthic habitat composition surrounding survey transects
were signicantly different inside versus outside BIRNM
for seven of 10 seascape variables (Table 7). The mean
area of seagrass surrounding transects outside BIRNM
was signicantly higher than inside, while seascapes inside
BIRNM had signicantly higher habitat diversity, area of
patch reefs, colonized pavement and area of sand.
3.1.3 Spatial patterns in benthic cover
Visual examination of an interpolated surface of live coral
cover indicated that areas with higher live coral cover were
more extensive inside BIRNM than outside. For instance, live
coral cover between 15-50% existed outside BIRNM in only
one relatively small area (southeast of Buck Island along
the north shore of St. Croix), whereas several areas inside
BIRNM had live coral ranging from 15-50% (Figure 15a). The
eastern tip and northwest end of Buck Island had the largest
areas with live coral cover exceeding 15%. These two areas
were also the most topographically rugose (Figure 15c) in
the region forming a mosaic of branching coral dominated
patch reefs interspersed among a matrix of massive coral
dominated colonized pavement. Furthermore, two small
areas inside the BIRNM had live coral cover exceeding 50%
(Figure 15a). The number of hard coral species groups were
relatively evenly distributed inside and outside BIRNM, with
several areas in both domains having 9-14 coral species
(Figure 15b). Overall, the existing boundary of BIRNM
encompassed the majority of the most topographically
complex hardbottom habitat types, with intermediate to high
coral cover relative to that of the study region.
Table 5. Mean estimates of percent cover of selected benthic
groups inside and outside BIRNM. Asterisks (*) indicate signicant
differences (p<0.05).
Mean percent cover (+ SE)
Habitat Benthic Taxa Inside Outside
Colonized
n = 310 n = 292
hardbottom
Live coral* 4.9 (0.3) 2.4 (0.2)
Acropora* 0.3 (0.1) 0.03 (0.02)
Montastraea* 1.0 (0.2) 0.5 (0.1)
Diploria* 1.8 (0.2) 0.6 (0.1)
Agaricia 0.2 (0.02) 0.2 (0.03)
Algae 51.9 (1.6) 53.6 (1.8)
Macroalgae* 14.5 (0.9) 15.7 (0.8)
Turf algae 35.7 (1.6) 36.8 (1.7)
Crustose algae 1.8 (0.3) 1.6 (0.2)
Gorgonians* 2.1 (0.2) 0.7 (0.08)
Sponge* 1.4 (0.2) 1.7 (0.1)
Seagrass* 0.3 (0.2) 1.4 (0.5)
Seagrass n = 55 n = 115
Live coral 0.5 (0.21) 0.19 (0.08)
Acropora 0 0
Montastraea 0.04 (0.03) 0.01 (0.01)
Diploria 0.32 (0.1) 0.06 (0.03)
Agaricia 0.01 (<0.01) 0.03 (0.02)
Algae 14.89 (3.4) 9.46 (1.6)
Macroalgae 6.92 (1.1) 5.73 (0.6)
Turf algae 7.52 (2.9) 3.3 (1.4)
Crustose algae* 0.45 (0.3) 0.44 (0.3)
Gorgonians* 0.14 (0.06) 0.04 (0.03)
Sponge* 0.07 (0.04) 0.26 (0.07)
Seagrass* 19.84 (2.9) 40.32 (2.7)
Unvegetated
n = 67 n = 41
sediments
(sand)
Live coral 0.2 (0.1) 0.01 (<0.01)
Acropora 0 0
Montastraea 0.02 (0.02) 0
Diploria 0.12 (0.1) 0
Agaricia 0.04 (0.04) 0
Algae 4.1 (1.5) 2.1 (0.3)
Macroalgae 2.7 (0.7) 2.0 (0.3)
Turf algae 1.0 (0.5) 0.1 (0.1)
Crustose algae 0.44 (0.4) 0
Gorgonians 0.05 (0.05) 0
Sponge* 0.001 0.1 (0.05)
Seagrass 0.7 (0.16) 2.3 (1.7)
3 - Results
Table 6. Comparison of coral species richness and ratio of coral
to macroalgae in major habitats inside and outside BIRNM.
Asterisks (*) indicate signicant differences (p<0.05).
Mean % cover (+ SE)
Variable Habitat type
Inside Outside
n = 310 n = 292
Number of
coral species
Colonized hardbottom
5.3 (0.2) 5.2 (0.2)
Seagrass
0.47 (0.16) 0.56 (0.13)
Unvegetated sediments
0.3 (0.1) 0.1 (0.06)
Coral : Macro-
algae ratio
Colonized hardbottom*
2.4 (0.5) 0.6 (0.1)
Seagrass
0.04 (0.02) 0.04 (0.02)
Unvegetated sediments
0.04 (0.03) 0.01 (0.01)
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 14
Intermediate to high levels of macroalgal cover (>20-
95%) and algal turf was evident over a broad expanse of
hardbottom area, both within and outside BIRNM (Figure
16a,b). Particularly high macroalgal cover was evident
at deeper water sites along the northern boundary of the
mapped/unmapped area and a large region along the
eastern boundary of the study region (Figure 16a). Most of
the hardbottom habitat inside BIRNM supported a high cover
of algal turf (>30-96%; Figure 16b). In contrast, fewer areas
had crustose coralline algae exceeding 30% cover (Figure
16c). These areas of high CCA cover were relatively localized
and occurred mostly over colonized pavement within BIRNM
north and east of Buck Island (Figure 16b). Given that CCA
is known to facilitate settlement of coral larvae, the observed
spatial distribution of CCA suggests that hardbottom areas
inside BIRNM may have greater potential for coral settlement
and recruitment than hardbottom areas outside BIRNM.
Although the study region had relatively low mean coral
cover (approximately 5%; Table 2), interpolations of live cover for the ve most abundant coral species show that several
areas inside BIRNM had higher coral cover for some species in comparison with areas outside (Figures 17 and 18). Live
cover of Diploria strigosa exceeded 5% in several areas inside BIRNM including two areas north of Buck Island which
ranged between 10 and 26% (Figure 17a). Outside BIRNM, maximum live cover of Diploria strigosa was 10% in only one
area along the eastern tip of St. Croix. The spatial distribution of Montastraea annularis was similar to that of D. strigosa.
Three areas north and two areas east-southeast of Buck Island had live cover of M. annularis ranging from 15-33%,
whereas live cover of M. annularis exceeded 15% in only one area outside BIRNM directly south of Buck Island (Figure
17b). Live cover of Montastraea cavernosa exceeded 5% in two areas inside BIRNM: (1) east of Buck Island and (2)
north-northeast of Buck Island toward the edge of the study area; and one area outside BIRNM, southeast of Buck Island
toward Point Udal on the island of St. Croix (Figure 17c). The highest cover of Siderastrea siderea ranged from 5.1-9.7%
in two areas: (1) inside BIRNM north east of Buck Island and (2) outside BIRNM directly south of Buck Island close to St.
Croix (Figure 18a). The highest cover of Porites astreoides ranged from 5-7% in two areas inside BIRNM (east of Buck
Island) and one area outside BIRNM southeast of Buck Island (Figure 18b).
Table 7. Differences in seascape composition (amount and
richness of habitat types) surrounding transects inside and
outside BIRNM. Seascape composition was quantied within 100
m radius seascape units surrounding each sh transect using
the NOAA benthic habitat map. Asterisks (*) indicate signicant
differences (p<0.05).
Habitat type
Mean area m
2
(+ SE)
Inside Outside
Colonized hardbottom* 17232 (419) 15351 (576)
Colonized pavement* 11495 (394) 9910 (574)
Patch reef* 1341 (108) 395 (64)
Linear reef 1802 (147) 1650 (208)
Reef rubble 296 (70) 382 (82)
Scattered coral and/or rock* 1776 (136) 3015 (259)
Seagrass* 5812 (329) 10674 (556)
Sand* 6532 (284) 4061 (373)
Number of habitat types* 3.5 (0.06) 2.6 (0.06)
3 - Results
Diploria strigosa colony
p. 15
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 15. Spatial distributions of benthic components at all transects in the study region (northeastern St. Croix) between 2001 and
2006. (a) Percentage live coral cover (hard coral including re coral); (b) number of coral species/groups; and (c) rugosity. White areas
inside the mapped region denote softbottom habitats (sand and seagrasses). Small hotspots of high coral cover area are encircled.
b)
c)
a)
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 16
Figure 16. Spatial distributions of benthic components at all transects in the study region (northeastern St. Croix) between 2001 and
2006. (a) Macroalgal cover (including lamentous algae/cyanobacteria; (b) algal turf cover; and (c) crustose coralline algal cover.
b)
c)
a)
3 - Results
p. 17
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
a)
Figure 17. Spatial distributions of coral cover for individual coral species at all transects in the study region (northeastern St. Croix)
between 2001 and 2006. (a) Diploria strigosa, (b) Montastraea annularis and (c) Montastraea cavernosa. White areas inside the
mapped region denote softbottom habitats (sand and seagrasses). Small hotspots of high coral cover are encircled.
b)
c)
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 18
Figure 18. Spatial distributions of coral cover for individual coral species at all transects in the study region (northeastern St. Croix)
between 2001 and 2006. (a) Siderastrea siderea and (b) Porites astreoides. White areas inside the mapped region denote softbottom
habitats (sand and seagrasses). Small hotspots of high coral cover are encircled.
b)
a)
3 - Results
p. 19
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
3.1.4 Temporal patterns in benthic cover
Mean live coral cover was higher inside than outside
BIRNM for all sampling seasons and years (Figure 19).
A dramatic decline in live coral cover due to a mass coral
bleaching event in October 2005 was recorded outside
BIRNM during the October 2005 sampling season and
the subsequent April 2006 and October 2006 seasons.
The decline inside BIRNM was not detected until the
following year (October 2006).
Examination of differences in mean values for selected
dominant plant biota inside and outside BIRNM over
eight eld missions from 2003 to 2006 (e.g., since the
sampling design included the two management domains)
also revealed some distinct temporal trends. Mean
macroalgal cover was higher outside BIRNM for six of
eight sampling periods (Figure 20). Spring macroalgal
cover declined in abundance from 2003 to 2006 both
inside and outside BIRNM, but exhibit higher cover in
the fall. Algal turf cover was generally lower in the fall
than spring, yet appeared similar in abundance inside
and outside across seasons and years (Figure 20). In
contrast, lamentous cyanobacterial/algal cover was
markedly higher in the fall than in spring and was highly
variable across years with greatest abundance observed
in October 2005 (Figure 20).
Further more detailed examination of temporal trends
in benthic components will be the focus of a separate
future report.
Inside BIRNM
Outside BIRNM
Figure 19. Seasonal and inter-annual patterns of live coral cover in-
side and outside BIRNM over a four year sampling period. Error bars
indicate + SE.
Healthy coral
Recently bleached coral Turf and lamentous algae over bleached coral
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 20
Inside BIRNM
Outside BIRNM
a)
b)
c)
Figure 20. Seasonal and inter-annual patterns of marine plant cover inside and outside BIRNM over a four year sampling period: (a)
macroalgae, (b) lamentous algae/cyanobacteria and (c) algal turf. Fil. algal/cyano.= lamentous algal/cyanobacterial. Error bars indi-
cate + SE.
3 - Results
p. 21
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
3.1.5 Mapping threatened Acropora species inside and outside BIRNM
Since 1980, populations of Acropora cervicornis (staghorn coral) and A. palmata (elkhorn coral) have declined by up to
98% throughout their range and localized extirpations have occurred due to combinations of stressors including disease,
hurricanes and coral bleaching (Figure 21). In BIRNM, white-band disease and several major hurricanes have reduced
live elkhorn coral cover by over 80 percent since the 1970s and 1980s. In 1999, NOAA National Marine Fisheries Service
(NMFS) added elkhorn coral to the candidate species-list of the Endangered Species Act (ESA), but it was not until May
2006 that staghorn coral and elkhorn coral were formally listed as threatened species under the ESA. According to the
Act, a species is considered endangered if it is in danger of extinction throughout all or a signicant portion of its range
or if it is likely to become an endangered species within the foreseeable future. In response to the designation, NMFS
proposed in February 2008 to designate critical habitat areas for Acropora species throughout the U.S territories based
on best available information on species distributions and habitat parameters (Federal Register 50 CFR Parts 223 and
226, February 6, 2008). Critical habitat was dened by Section 3 of the ESA (and further by 50 CFR 424.02(d)) and is
paraphrased here as: (i) specic areas essential to the conservation of the species; and (ii) areas which may require
special management considerations or protection; and (iii) specic areas outside the geographical area occupied by the
species that are determined essential for the conservation of the species.
Within the BIRNM, NPS staff identied 2,492 A. palmata colonies greater than 1 m in size at 455 of 617 random survey
sites (Mayor, 2005). In addition, CCMA-BB documented the presence of A. palmata at 32 of 815 hardbottom sites within
the BIRNM and at 11 of 430 sites within the EEMP. The distribution of A. palmata is almost entirely conned to relatively
shallow waters (<12 m or approximately 35 ft) with most colonies observed in waters less than 10 m on the exposed
seaward side of Buck Island within BIRNM. The presence and absence of A. palmata in the study region of northeastern
St. Croix is shown in Figure 22. A. cervicornis is considerably rarer in the study region than was A. palmata. CCMA-BB
CREM surveys recorded the presence of A. cervicornis at 12 of 815 hardbottom sites within the BIRNM and at two of 430
sites within the EEMP, but they did not observe any colonies at 39 other sites visited in northeast St. Croix. In general,
A. cervicornis has received much less attention by researchers than A. palmata, although for both species more data
are required to adequately assess the distribution and occurrence of the species in the USVI. Due to the relatively well
dened environmental conditions that support A. palmata establishment and growth (i.e., wave exposed shallow water
hardbottom areas with good circulation and low exposure to sedimentation and fresh water incursions) it should be
possible to develop predictive models that will help ll in the data gaps by providing species distribution maps.
Figure 21. Two types of Acropora species recorded in the study region (northeastern St. Croix): Acropora palmata (left) and A.
cervicornis (right).
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 22
Figure 22. Spatial distribution of Acropora palmata (red circles) and A. cervicornis (yellow circles) in St. Croix, U.S. Virgin Islands. Open
circles indicate survey sites where Acropora corals were not observed. Source of Data: Mayor (2005) and NOAA Biogeography Branch
database http://www8.nos.noaa.gov/biogeo_public/query_main.aspx
3 - Results
p. 23
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Blue tang (A. coeruleus) and
A. palmata
Bar jacks (C. ruber)
Red hind (E. guttatus)
3.2 Fish communities, groups and species
3.2.1 Fish community metrics
Fish biomass over colonized hardbottom habitat was signicantly higher inside BIRNM for all (1) sh species combined
and (2) all herbivorous sh (Figure 23). In contrast, mean biomass of piscivorous sh was lower inside BIRNM for all
habitat types, although only signicantly so over unvegetated sandy sediments (Figure 23). No signicant difference was
detected between inside and outside BIRNM for sh biomass over seagrasses.
Fish diversity (number of species, Shannon diversity and Taxonomic diversity) was highest over colonized hardbottom
and lowest over unvegetated sediments and no differences were found between inside and outside BIRNM (Figure 24).
In the study area overall, however, more sh species/species groups have been observed inside BIRNM than outside
(201 inside and 182 outside).
3.2.2 Fish community composition
nMDS plots and ANOSIM tests indicated that sh community composition between hard and soft habitat types was
signicantly different and well separated (Figure 25). Dissimilarities between sand and seagrasses were less distinct
with considerable overlap (Table 8). Pairwise comparisons between the individual hardbottom habitat types revealed that
sh communities were barely separable with substantial overlap. Highest dissimilarity existed between linear reefs and
scattered coral. Fish community composition of aggregated patch reefs, linear reefs and colonized pavement were not
signicantly different (Table 8).
Much overlap was also found in sh community composition when also considering management domains. “R” values
were very low (e.g., high similarity) when comparing the dominant softbottom habitat types inside versus outside BIRNM
and even lower when comparing like hardbottom habitat types inside versus outside BIRNM (Figure 26 and Table 9).
Although the null hypotheses of no difference was rejected (p=<0.05) for pairwise hardbottom habitat types, this is likely
to be indicative of the high sample size rather than ecologically meaningful differences. The R value is a better relative
indicator of the amount of dissimilarity between groups and is thus given greater emphasis here.
3.2.3 Fish groups
For all selected sh species groups and families (all Lutjanidae. [snapper], all Haemulidae. [grunts], all Scaridae [parrotsh],
large-bodied Serranidae [groupers] species), mean density and biomass were highest on colonized hardbottom sites
(Figure 26). Highest density and biomass were recorded for parrotsh. The most frequently observed species of parrotsh,
with highest biomass were the redband parrotsh (Sparisoma aurofrenatum), the stoplight parrotsh (Sparisoma viride)
and the striped parrotsh (Scarus iseri; Table 10). When management domains were considered, parrotsh exhibited
signicantly higher mean biomass inside BIRNM over colonized hardbottom.
Assemblage of Acanthuridae species
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 24
Inside BIRNM
Outside BIRNM
*
*
*
a)
b)
c)
Figure 23. Comparison of mean (+ SE) values inside versus
outside BIRNM for: (a) biomass of all species; (b) biomass of
all herbivores; and (c) biomass of all piscivores including sharks
and rays. Asterisks (*) indicate a statistically signicant difference
between inside and outside.
Inside BIRNM
Outside BIRNM
a)
b)
c)
Figure 24. Comparison of mean (+ SE) values inside versus
outside BIRNM for: (a) number of sh species; (b) Shannon-
Weiner diversity (H’) using abundance data; and (c) taxonomic
diversity using presence-absence data. Asterisks (*) indicates a
statistically signicant (p=<0.05) difference between inside and
outside.
3 - Results
p. 25
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 25. Non-metric multidimensional ordination based on
between site similarity in sh community composition using
species biomass data. (a) Community similarities by habitat
structure; (b) community similarities by dominant softbottom
habitat type inside versus outside BIRNM; and (c) community
similarities by dominant hardbottom habitat types inside versus
outside BIRNM.
a)
b)
c)
Table 8. Results of ANOSIM test for signicant difference in sh
community composition using species biomass between samples
grouped by habitat type. R<0.25 = barely separable; R<0.05 =
overlapping, but clearly different. Asterisks (*) indicates null
hypothesis of no difference rejected at p=<0.05. Agg= aggregated,
Ind= individual.
Habitat pairs ANOSIM R
Between major habitat types
Global - all pairs* 0.80
Sand and Seagrass* 0.19
Sand and Colonized Hard* 0.83
Seagrass and Colonized Hard* 0.84
Amongst Colonized Hard
Global - all pairs* 0.19
Linear reef and Scattered coral/rock* 0.31
Colonized pavement and Patch reef (Ind)* 0.29
Colonized pavement and Linear reef* 0.18
Colonized pavement and Scattered coral/rock* 0.28
Patch reef (Agg) and Linear reef <0.01
Colonized pavement and Patch reef (Agg)* 0.13
Patch reef (Ind) and Linear reef* 0.14
Patch reef (Agg) and Scattered coral/rock* 0.17
Patch reef (Ind) and Scattered coral/rock* 0.11
Patch reef (Agg) and Patch reef (Ind) 0.04
Table 9. ANOSIM test for signicant difference in sh community
composition using species biomass between samples grouped
by habitat type and management domain (inside/outside BIRNM).
R<0.25 = barely separable; R<0.05 = overlapping, but clearly
different. Asterisks (*) indicates null hypothesis of no difference
rejected at p=<0.05.
Habitat pairs
ANOSIM
R
Sand Inside and Sand Outside 0.06
Seagrass Inside and Seagrass Outside 0.07
Col. pavement Inside and Col. pavement Outside*
0.04
Patch reef Inside and Patch reef Outside 0.05
All colonized hard Inside and All Colonized hard Outside* 0.04
3 - Results
Grouper were almost entirely found over colonized hardbottom, with very few individuals over seagrasses (Figure 26).
Neither mean density nor mean biomass were signicantly different inside versus outside BIRNM. The most frequently
observed species of grouper were the coney (Cephalopholis fulva) at 32.4% of transects, red hind (Epinephelus guttatus)
at 18.1% of transects and graysby (Cephalopholis cruentata) at 4.2% of transects (Appendix C). All other species of
large-bodied serranids were relatively rare within the surveyed region. For example, only one tiger grouper (Mycteroperca
tigris), three Nassau grouper (Epinephelus striatus) and three yellown grouper (Mycteroperca venenosa) were observed
in 1,275 surveys over six years (Appendix C).
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 26
*
*
*
d)
Inside BIRNM
Outside BIRNM
*
*
a)
b)
c)
Figure 26. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for: (a) grouper, (b) snapper, (c) grunt and
(d) parrotsh. Asterisks (*) indicate a statistically signicant difference between inside and outside.
3 - Results
p. 27
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Table 10. Twenty most frequently observed species in the CREM Buck Island survey area. For full species list see Appendix C. Fish
surveys from http://ccmaserver.nos.noaa.gov/ecosystems/coralreef/reef_sh/protocols.html.
Species name Common name
Total
occurrence
%
occurrence
Total
abundance
Mean
abundance
(+ SE)
Total
biomass,
kg
Mean
biomass, kg
(+ SE)
Halichoeres bivittatus Slippery dick 933 73.2 24752
19.4 (1.8)
97.96
0.08 (<0.01)
Thalassoma bifasciatum Bluehead wrasse 777 60.9 32001
25.1 (1.2)
46.06
0.04 (<0.01)
Acanthurus bahianus Ocean surgeonsh 762 59.8 8601
6.7 (0.36)
540.37
0.42 (0.04)
Sparisoma aurofrenatum Redband parrotsh 669 52.5 4887
3.8 (0.18)
240.21
0.19 (0.01)
Stegastes partitus Bicolor damelsh 624 48.9 10202
8.0 (0.44)
17.76
0.01 (<0.01)
Acanthurus coeruleus Blue tang 595 46.7 8597
6.7 (0.65)
904.14
0.71 (0.08)
Halichoeres garnoti Yellowhead wrasse 586 46.0 4658
3.7 (0.19)
34.69
0.03 (<0.01)
Sparisoma viride Stoplight parrotsh 460 36.1 2456
1.9 (0.11)
331.90
0.26 (0.02)
Scarus iseri Striped parrotsh 434 34.0 4761
3.7 (0.28)
74.27
0.06 (<0.01)
Halichoeres maculipinna Clown wrasse 432 33.9 2176
1.7 (0.11)
10.19
0.01 (<0.01)
Cephalopholis fulva Coney 413 32.4 1391
1.1 (0.07)
193.13
0.15 (0.01)
Stegastes leucostictus Beaugregory 378 29.6 3991
3.1 (0.25)
12.88
0.01 (<0.01)
Carangoides ruber Bar jack 364 28.5 1735
1.4 (0.17)
57.87
0.05 (<0.01)
Haemulon avolineatum French grunt 357 28.0 1944
1.5 (0.26)
107.19
0.08 (<0.01)
Serranus tigrinus Harlequin bass 349 27.4 713
0.56 (0.03)
5.55
0.00 (<0.01)
Holocentrus rufus Longspine squirrelsh 321 25.2 574
0.45 (0.03)
65.19
0.05 (<0.01)
Scarus taeniopterus Princess parrotsh 318 24.9 1713
1.3 (0.10)
72.30
0.06 (<0.01)
Coryphopterus glaucofraenum Bridled goby 302 23.7 1670
1.3 (0.11)
1.49
0.00 (<0.01)
Pseudupeneus maculatus Spotted goatsh 297 23.3 915
0.72 (0.08)
72.22
0.06 (<0.01)
Microspathodon chrysurus Yellowtail damselsh 272 21.3 1158
0.91 (0.08)
77.58
0.06 (<0.01)
Inside BIRNM
Outside BIRNM
Figure 27. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for sharks and rays. Asterisks (*) indicate a
statistically signicant difference between inside and outside.
Yellowtail snapper (Ocyurus chrysurus) was the most commonly observed lutjanid recorded at 20.4% of transects;
mahogany snapper (Lutjanus mahogoni) at 3.4% and schoolmaster snapper (Lutjanus apodus) at 3.3% (Appendix C).
Thirteen species of grunt (Haemulidae) were identied, with French grunt (Haemulon avolineatum) the most commonly
observed at 28% of transects followed by bluestriped grunt (Haemulon sciurus) at 4.3% of transects and tomtate (Haemulon
aurolineatum) at 4.2% of transects (Appendix C).
Snapper and grunt density were higher outside BIRNM in all habitat types, although the difference was only statistically
signicant for colonized hardbottom. Mean biomass of snapper over colonized hardbottom and seagrasses was slightly
higher inside BIRNM, but was not signicantly different to the outside
(Figure 26). Biomass of grunts, however, was signicantly higher over
colonized hardbottom outside BIRNM (Figure 26). Sharks and rays
exhibited highest mean density and biomass over unvegetated sandy
sediments (Figure 27). Although, only observed infrequently across
the region, mean shark and ray density was higher over sand inside
BIRNM than outside, but mean biomass was higher outside than inside
BIRNM. None of the differences were signicantly different. In total,
40 rays were recorded: three spotted eagle rays (Aetobatus narinari),
37 southern stingrays (Dasyatis americana), and nine nurse sharks
(Ginglymostoma cirratum; Appendix C).
Nurse shark (Ginglymostoma cirratum)
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 28
3.2.4 Individual species
In the northeastern St. Croix study region, 201 sh species from 56 families have been positively identied using visual
census and underwater photography. A further 26 sh types were identied to family only. Only selected species are
examined at the individual species level herein, including species of special interest to NPS, those that were potentially
threatened by overshing and those that were dominant components of the sh community across the region.
Mean biomass for the two most abundant grouper species was higher inside BIRNM, with biomass and abundance of
coney being signicantly higher over colonized hardbottom habitat types inside (Figure 28).
None of the three most abundant snapper species were signicantly different when comparing biomass and density
inside versus outside BIRNM (Figure 29). Although not statistically signicant, mean biomass of yellowtail snapper was
markedly higher inside than outside, particularly over colonized hardbottom. Of note, was that yellowtail snapper were
observed in all three major habitat types. In contrast, schoolmaster snapper and gray snapper (Lutjanus griseus) were
found primarily over colonized hardbottom habitat types.
In general, French grunt and bluestriped grunt density and biomass were higher outside BIRNM (Figure 30). This was
statistically signicant for French grunt biomass and bluestriped grunt density. Assessment of differences for grunt,
however, is hampered by the fact that 3,419 grunts were identied only to family; primarily small juveniles that can be very
similar in appearance between species.
For abundant herbivore species, blue tang (Acanthurus coeruleus) and striped parrotsh biomass were signicantly
higher inside BIRNM over colonized hardbottom (Figure 31). Redband parrotsh density and biomass was very similar
inside and outside BIRNM. Comparatively few individuals of the abundant herbivores were observed over seagrasses or
sand.
*
Inside BIRNM
Outside BIRNM
a)
b)
Figure 28. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for two grouper species: (a) coney (C. fulva)
and (b) red hind (E. guttatus). Asterisks (*) indicate a statistically signicant difference between inside and outside.
3 - Results
p. 29
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Inside BIRNM
Outside BIRNM
a)
b)
c)
Figure 29. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for three snapper species: (a) yellowtail
snapper (O. chrysurus), (b) schoolmaster (L. apodus) and (c) gray snapper (L. griseus). Asterisks (*) indicate a statistically signicant
difference between inside and outside.
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 30
*
Inside BIRNM
Outside BIRNM
a)
b)
*
Figure 30. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for two grunt (Haemulidae) species: (a)
French grunt (H. avolineatum) and (b) bluestriped grunt (H. sciurus). Asterisks (*) indicate a statistically signicant difference between
inside and outside.
French grunts (Haemulon avolineatum)
3 - Results
p. 31
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Inside BIRNM
Outside BIRNM
a)
b)
c)
*
*
*
Figure 31. Comparison of mean (± SE) density and biomass inside versus outside BIRNM for three numerically dominant herbivore
species: (a) blue tang (A. coeruleus), (b) striped parrotsh (S. iseri) and (c) redband parrotsh (S. aurofrenatum). Asterisks (*) indicate
a statistically signicant difference between inside and outside.
3.2.5 Spatial distribution patterns and species habitat associations
Twenty-ve or more sh species were recorded at 100 sites, 62 of these high sh species richness sites were located
within BIRNM and 38 were located outside BIRNM, but within 2 km of the current boundary. Fish species richness
(number of species) hotspots were associated with shallow-water hardbottom habitat types and were located both within
and outside BIRNM (Figure 32). The largest continuous area of high sh species richness existed around the eastern
end of the Buck Island fringing reef comprising an interspersion of patches of linear reef, patch reef, colonized pavement
and scattered coral. This included much of the area of the original “marine garden” zones. A second region of high sh
species richness, however, also existed to the south of Buck Island immediately south of BIRNM, along the northern shore
of St. Croix. This area comprised a similar range of hardbottom habitat types including linear reef, patch reef, colonized
pavement and scattered coral, extending east to west alongside large expanses of seagrasses on both the north and
south sides. Overlaying these high sh species richness sites on the NOAA benthic habitat map and measuring proximity
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 32
Figure 32. Interpolated spatial surfaces representing (a) number of sh species; (b) herbivorous sh biomass; and (c) piscivorous sh
biomass using 1,275 samples.
b)
c)
a)
3 - Results
p. 33
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 33. Spatial distributions of juvenile and adult (a) coney (C. fulva) and (b) red hind (E. guttatus) in northeastern St. Croix.
from sampling point to nearest seagrass beds revealed that approximately 80% of all high (=>30 species) sh species
richness sites on coral reefs were within 200 m of seagrass beds.
The spatial distribution of herbivore biomass as represented by interpolation of transect data revealed a distinctive area
of high herbivore biomass over much of the hardbottom habitat in the region, but especially high at the eastern end of the
Buck Island fringing reef and to the north of Buck Island over scattered coral and branching coral dominated pavement
(Figure 32). Lowest herbivore biomass was over sand and seagrasses farthest from hardbottom, creating an interior effect
in the spatial surface of herbivore biomass over softbottom habitat types. High herbivore biomass was also estimated for
the coral reefs along the northern shore of St. Croix, south of Buck Island.
Piscivore biomass was biased toward high biomass hotspots created by sharks (e.g. large-bodied sh predators).
Otherwise, no distinctive hotspots were seen. Large areas of low piscivore biomass existed around the north shore of
Buck Island and at several inshore locations along the north shore of St. Croix (Figure 32).
Spatial distributions for juveniles and adults of specic sh species show some individualistic species-specic and life-
stage specic patterns. Coney juveniles and adults were widespread across hardbottom habitats, with both utilizing
similar habitat types (Figure 33). Highest mean density for juveniles and adults was recorded for the most structurally
complex coral reef habitat types including colonized pavement (0.75 and 1.83/100 m
2
), linear reef (0.63 and 0.63/100
m
2
) and aggregated patch reef (0.31 and 0.54/100 m
2
; Figure 34). Juveniles and adults also were observed over less
rugose reef rubble (0.52 and 0.61/100 m
2
) and scattered coral and rock habitat types (0.45 and 0.53/100 m
2
). Very few
coney were observed over seagrasses and sand (0 and <0.01/100 m
2
). Red hind were less abundant than coney over
the large expanse of coral reef north of Buck Island, with distributions more closely associated with complex coral reef
habitat types fringing Buck Island, and areas south and east of BIRNM (Figure 34). Highest mean density for adults was
recorded for colonized pavement (0.31/100 m
2
), patch reefs (0.23/100 m
2
) and scattered coral and rock (0.19/100 m
2
).
Juvenile densities were highest for colonized pavement (0.36/100 m
2
), scattered coral/rock (0.27/100 m
2
) and reef rubble
(0.23/100 m
2
; Figure 34).
a)
b)
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 34
b)
a)
Juveniles/Subadults
Adults
Figure 34. Mean (+ SE) density for juvenile/subadult and adult by observer habitat type for (a)
coney (C. fulva) and (b) red hind (E. guttatus).
3 - Results
p. 35
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 35. Spatial distribution of juvenile and adult: (a) yellowtail snapper (O. chrysurus), (b) schoolmaster (L. apodus) and (c) gray
snapper (L. griseus) in northeastern St. Croix.
a)
b)
c)
3 - Results
Juvenile yellowtail snapper were associated with all benthic habitat types, with highest mean density over patch reefs
(0.94/100 m
2
) and seagrasses (0.88/100 m
2
). Spatially, juveniles were most abundant in shallow inshore waters around
Buck Island and along the north coast of St. Croix outside BIRNM (Figure 35 and Figure 36). Adults used a similar range
of habitat types, but were less frequently observed and occurred in lower densities. Highest adult densities were recorded
for aggregated patch reefs (0.33/100 m
2
). Juvenile schoolmaster snapper and gray snapper were infrequently observed in
the study region (Figure 36 and Appendix C). Adults were slightly more abundant than juveniles, but relatively localized in
distribution to the north shore of Buck Island and several locations along the St. Croix shoreline (Figure 35). Schoolmaster
snapper adults were observed in all hardbottom habitat types, with highest density in aggregated patch reefs (0.29/100
m
2
) and linear reefs (0.18/100 m
2
; Figure 36). Mean density of adult gray snapper was highest in patch reefs (0.01/100
m
2
).
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 36
b)
a)
c)
Juveniles/Subadults
Adults
Figure 36. Mean (+ SE) density for juvenile/subadult and adult by observer habitat type
for: (a) yellowtail snapper (O. chrysurus), (b) schoolmaster (L. apodus) and (c) gray
snapper (L. griseus).
3 - Results
p. 37
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 37. Spatial distributions of juvenile and adult: (a) French grunt (H. avolineatum), (b) bluestriped grunt (H. sciurus) and (c) white
grunt (H. plumierii) in northeastern St. Croix.
a)
b)
c)
3 - Results
French grunt were the most widely distributed of the haemulids (Figure 37), found in all hardbottom habitat types (Figure
38). Similar habitat associations and spatial distributions were observed for both juveniles and adults, which exhibited
highest densities over aggregated patch reefs (5.13 and 0.9/100 m
2
), patch reefs (3.1 and 1.4/100 m
2
), scattered coral/
rock (2.67 and 0.57/100 m
2
) and linear reef (2.60 and 1.08/100 m
2
). Highest occurrence was recorded around the
shallow fringing mosaic of habitats surrounding Buck Island (Figure 37). The majority of bluestriped grunt and white grunt
(Haemulon plumierii) were found either in close proximity to Buck Island or to the south of Buck Island in shallow water
hardbottom habitats that exist in close proximity to seagrasses. Very few individuals were found in the northern portion
of the mapped BIRNM (Figure 37). Highest densities of juvenile and adult bluestriped grunt (0.60 and 0.67/100 m
2
) and
white grunt (4.35 and 0.51/100 m
2
) were observed over patch reefs (Figure 38).
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 38
b)
a)
c)
Juveniles/Subadults
Adults
Figure 38. Mean (+ SE) density for juvenile/subadult and adult by observer habitat
type for: (a) French grunt (H. avolineatum), (b) bluestriped grunt (H. sciurus) and (c)
white grunt (H. plumierii).
3 - Results
p. 39
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure 39. Spatial distributions of juvenile and adult: (a) blue tang (A. coeruleus), (b) ocean surgeonsh (A. bahianus), (c) redband
parrotsh (S. aurofrenatum) and (d) striped parrotsh (S. iseri) in northeastern St. Croix.
a)
b)
c)
d)
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 40
Juvenile and adult blue tang, ocean surgeonsh (Acanthurus bahianus) and redband parrotsh were distributed widely
across the study area, sharing the same habitat types (Figure 39 and Figure 40). Although present in all habitat types
except macroalgal beds, highest densities of juvenile and adult blue tang were associated with structurally complex
hardbottom habitats including aggregated patch reefs (4.08 and 17.88/100 m
2
), linear reef (3.56 and 20.04/100 m
2
) and
patch reef (2.46 and 10.6/100 m
2
). Adult and juvenile ocean surgeonsh were also very widespread across hardbottom
habitat types, with highest density of adults recorded for colonized pavement (3.49/100 m
2
) and aggregated patch reefs
(3.3/100 m
2
) and highest density of juveniles was recorded for linear reef (12.86/100 m
2
), colonized pavement (7.79/100
m
2
), aggregated patch reefs (5.21/100 m
2
), reef rubble (7.2 /100 m
2
) and scattered coral/rock (7.03/100 m
2
; Figure 40).
Visual comparison of adult to juvenile ratios for the two acanthurids revealed that adult blue tang were more abundant
than juveniles, while juvenile ocean surgeonsh were more abundant than adult ocean surgeonsh.
Highest densities of juvenile and adult redband parrotsh were associated with aggregated patch reef (6.83 and 2.72/100
m
2
), linear reef (3.70 and 2.12/100 m
2
), patch reef (3.41 and 2.59/100 m
2
) and colonized pavement (3.50 and 2.38/100 m
2
;
Figure 40). Juvenile and adult striped parrotsh were more restricted in distribution with highest densities of adults and
juveniles close to Buck Island and also with high densities of juveniles along the nearshore fringing reef parallel to Teague
Bay (Figure 39). Although observed in all habitat types, densities of juveniles and adults were highest over aggregated
patch reef (11.64 and 3.06/100 m
2
), linear reef (9.47 and 1.59/100 m
2
) and patch reef (6.55 and 0.84/100 m
2
; Figure 40).
Spatial distributions for other species are shown in Appendix D.
a)
d)
b)
c)
Juveniles/Subadults
Adults
Figure 40. Mean (± SE) density for juvenile/subadult and adult by observer habitat type for: (a) blue tang (A. coeruleus), (b) ocean
surgeonsh (A. bahianus), (c) redband parrotsh (S. aurofrenatum) and (d) striped parrotsh (S. iseri).
3 - Results
p. 41
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Table 11. Summary information on selected species from ve key sh families showing maximum size observed in the study region
(northeastern St. Croix) compared with maximum known size for the species and the proportion of juveniles found inside and outside
BIRNM based on n=453 samples outside BIRNM and 431 inside BIRNM from 2003-2006. Maximum known sh size and size at rst
maturity are from FishBase (http://www.shbase.org). TL= total length; FL= fork length
Species Common name
Approx. size
class at rst
maturity*
Max. known
size, TL
Max. size
observed St.
Croix, FL
% juveniles
Inside BIRNM Outside BIRNM
Serranidae
C. fulva Coney 15-20 41 30-35 35.1 35.2
E. guttatus Red hind 20-25 76 45 47.4 61.2
Lutjanidae
O. chrysurus Yellowtail snapper 20-25 86.3 60 56.6 81.4
L. apodus schoolmaster 20-25 67.2 45 35.3 22.9
Haemulidae
H. avolineatum French grunt 15-20 30 30-35 65.7 73.4
H. plumierii White grunt 15-20 53 30-35 12.0 20.5
H. sciurus Bluestriped grunt 15-20 46 30-35 0.0 55.6
Acanthuridae
A. bahianus Ocean surgeonsh 15-20 38.1 30-35 73.8 77.8
A. chirurgus Doctorsh 10-15 39 25-30 33.2 38.6
A. coeruleus Blue tang 10-15 39 20-25 15.4 43.9
Scaridae
S. iseri Striped parrotsh 15-20 35 30-35 81.9 91.9
S. taeniopterus Princess parrotsh 15-20 35 30-35 50.3 64.2
S. aurofrenatum Redband parrotsh 15-20 28 30-35 64.7 63.9
S. viride Stoplight parrotsh 15-20 64 50 74.7 87.4
3 - Results
3.2.6 Fish size class frequency distributions and maximum lengths
The body lengths of the largest individuals of several common groupers, snappers and grunts observed in the study
region were less than the maximum size recorded for the species (Table 11). For example, the largest red hind was
approximately 60% of the maximum known adult size for that species. The largest yellowtail snapper was approximately
70% of the maximum known, schoolmaster snapper 66%, white grunt 56-66%, and bluestriped grunt 65-76%. Even when
factoring in the relatively small difference between fork length and total length these individuals were between 20-40%
less than maximum size. In contrast, the longest parrotsh and surgeonsh were estimated at or near maximum size.
Size frequency distribution for all grouper species on hardbottom habitat types combined was approximately normal (e.g.,
a bell shaped distribution) both inside and outside BIRNM, with very few newly settled individuals (<5 cm FL) and very
few large adults (>35 cm FL; Figure 41). Snappers exhibit a slightly atter distribution slightly skewed towards a higher
frequency of small and medium length sh. A higher frequency of large snapper were found inside BIRNM (Figure 41).
Grunts and parrotsh exhibited a more strongly skewed size frequency distribution towards a higher frequency of the
smallest size classes, this pattern was particularly strong for grunts outside BIRNM (Figure 41). Comparatively few large
adults were observed either inside or outside BIRNM. Surgeonsh were more normally distributed inside BIRNM, with
size frequency distribution outside skewed towards a higher frequency of smaller size classes.
At the species level, highest frequency of coney and red hind individuals occurred for subadults and small mature adults,
with very few small juveniles or large adults. This pattern was similar to the size-frequency distribution for all grouper
combined (Figure 42). Yellowtail snapper outside BIRNM showed a more skewed distribution than inside, with a higher
frequency of juveniles and subadults and a greater decrease in frequency with larger size classes of subadults and
adults (Figure 42). Schoolmaster snapper distribution was skewed in the opposite direction towards a higher frequency of
small mature adults both inside and outside BIRNM. Very low frequency of the smallest juvenile (<5 cm) or largest adult
(>35) schoolmaster snapper were recorded in the study area (Figure 42). French grunt and bluestriped grunt exhibited
very different distributions inside versus outside BIRNM, with French grunt slightly skewed towards higher frequency of
large juveniles and subadults inside BIRNM and towards higher frequency of small juveniles outside (Figure 42). Overall,
blue tang and ocean surgeonsh showed similar near-normal size class distributions. Bluestriped grunt exhibited higher
frequency of mature adults inside and higher frequency of juveniles and subadults outside.
Blue tang exhibit a peak in frequency for small adults and ocean surgeonsh exhibit a peak for subadults (Figure 43).
In contrast, redband parrotsh and striped parrotsh showed a strongly skewed distribution, with high frequency of the
smallest juveniles (<5 cm) and gradual decline with size with very few of the largest adults (Figure 43). Similar patterns
existed for populations inside and outside BIRNM.
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 42
Inside BIRNM
Outside BIRNM
e) surgeonsh
d) parrotshc) grunts
b) snappera) grouper
Figure 41. Length frequency histograms for key sh families over hardbottom sites inside and outside BIRNM: (a) grouper, (b) snapper,
(c) grunts, (d) parrotsh and (e) surgeonsh.
3 - Results
p. 43
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
a) coney
Inside BIRNM
Outside BIRNM
c) yellowtail snapper
e) French grunt
b) red hind
d) schoolmaster
f) bluestriped grunt
Figure 42. Size class frequency histogram for selected sh species over hardbottom sites inside and outside BIRNM. (a) Coney (C.
fulva), (b) red hind (E. guttatus), (c) yellowtail snapper (O. chrysurus), (d) schoolmaster (L. apodus), (e) French grunt (H. avolineatum)
and (f) bluestriped grunt (H. sciurus).
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 44
Inside BIRNM
Outside BIRNM
c) redband parrotsh
d) striped parrotsh
a) blue tang b) ocean surgeonsh
Figure 43. Size class frequency histogram for selected sh species over hardbottom sites inside and outside BIRNM. (a) Blue tang (A.
coeruleus), (b) ocean surgeonsh (A. bahianus), (c) redband parrotsh (S. aurofrenatum), and (d) striped parrotsh (S. iseri).
3.2.7 Comparison of sh densities and species presence between 1979 and 2001-2006
The most conspicuous difference between sh surveyed in 1979 and those surveyed from the same general area (within
500 m of Buck Island) over two decades later was the absence of Nassau grouper (E. striatus), tiger grouper (M. tigris)
and yellown grouper (M. venenosa) from the 2001-2006 data (Table 12). In contrast, smaller species of grouper such as
red hind (E. guttatus) and coney (C. fulva) had increased in density since 1979. Overall, parrotsh also have increased
in density since 1979 as have French grunt (Haemulon avolineatum) and white grunt (H. plumierii), but not bluestriped
grunt (H. sciurus). Snappers showed very mixed differences between the two survey periods, with a decrease in school-
master snapper (L. apodus) and mahogany snapper (L. mahogoni) and an increase in yellowtail snapper (O. chrysurus)
and mutton snapper (Lutjanus analis; Table 12 ). Other commercially targeted species such as queen triggersh (Balistes
vetula), spotted goatsh (Pseudupeneus maculatus), yellow goatsh (Mulloidichthys martinicus) and bar jacks (Caran-
goides ruber) also increased since 1979. Threespot damselsh (Stegastes planifrons), a potential indicator of live coral
cover, was substantially lower in 2001-2006 than in 1979.
3 - Results
Fish assemblages around dead Acropora palmata and
Millepora colony
p. 45
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Table 12. Comparison of mean density for a range of key sh species from 1979 (n=147)
and 2001-2006 (n=184) monitoring periods within 500 m surrounding Buck Island. Source:
Gladfelter and Gladfelter, 1980; CCMA-BB reef sh database
Species
Mean density 1979 Mean density 2001-2006
Change
/100m
2
/100m
2
Serranidae
Epinephelus striatus 0.01 Absent -0.01
Epinephelus guttatus 0.08 0.18 +0.1
Mycteroperca venenosa 0.001 Absent -0.001
Mycteroperca tigris 0.02 Absent -0.02
Cephalopholis. fulva 0.01 0.30 +0.3
Lutjanidae
Lutjanus apodus 0.28 0.16 -0.12
Lutjanus griseus 0.01 0.08 +0.07
Ocyurus chrysurus 0.17 0.24 +0.07
Lutjanus mahogoni 0.33 0.29 -0.04
Lutjanus analis 0.004 0.06 +0.06
Haemulidae
Haemulon avolineatum 2.49 2.89 +0.4
Haemulon sciurus 0.28 0.14 -0.14
Haemulon plumierii 0.21 0.22 +0.01
Scaridae
Scarus vetula 0.38 1.21 +0.83
Sparisoma aurofrenatum 0.23 1.50 +1.27
Sparisoma viride 0.92 0.79 -0.13
Scarus guacamaia 0.01 0.07 +0.06
Sparisoma rubripinne 0.13 0.24 +0.11
Other
Stegastes planifrons 4.09 2.18 -1.19
Balistes vetula 0.01 0.02 +0.01
Carangoides ruber 1.12 2.07 +0.95
Mulloidichthys martinicus 0.53 0.98 +0.45
Pseudupeneus maculatus 0.13 0.44 +0.31
3 - Results
3.2.8 Synoptic overview of inter-annual trends in mean sh metrics (2003-2006)
Presented here is a synoptic overview of inter-annual changes in summary statistics (mean and SE) for 39 sh metrics
at the level of species, family, trophic group and community using data from both the whole study area and inside and
outside BIRNM separately. The intention is to assist in highlighting potential trends that may be emerging even within the
relatively short term monitoring data currently available. The synopsis is presented for: (1) the entire study area; (2) inside
BIRNM; and (3) outside BIRNM.
Study area
Across the entire study area, mean biomass of bluestriped grunt, density of striped parrotsh and biomass of yellowtail
snapper exhibited a year after year decline across three consecutive years (Table 13). Striped parrotsh density and
bluestriped grunt biomass was signicantly (p=<0.01) lower in 2003 than 2006. Although mean yellowtail snapper
biomass declined over the study period too the difference was not signicant. In contrast, mean herbivore density and
biomass of redband parrotsh increased year after year across three consecutive years, with 2003 signicantly higher
than 2006. Other metrics were less consistent in the directionality of change. For instance, ve metrics exhibited at least
two consecutive years of increase, 15 exhibited at least two consecutive years of decrease and 19 showed no distinctive
direction in inter-annual change (Table 13).
Inside BIRNM
Inside BIRNM, no metric exhibited three consecutive years of decline. Declines for at least two consecutive years, however,
were documented for coney biomass, gray snapper density and biomass, all grunt density and biomass including French
grunt density and biomass. In contrast, year after year increases for three consecutive years were documented for sh
density (all species combined), with 2005 and 2006 density signicantly higher than 2003 (Table 14 and Appendix E).
Parrotsh biomass was also higher but not signicantly (p=>0.05) different between years. In addition, increases inside
BIRNM for at least two consecutive years were documented for herbivore density, parrotsh biomass, including redband
parrotsh biomass and striped parrotsh biomass. Red hind density and all snapper density also increased. Overall,
seven metrics exhibited at least two years of consecutive decline, seven exhibited increases and 25 showed no clear
directionality (Table 14).
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 46
Table 13. Summary statistics (mean + SE) for a range of sh variables grouped by year (2003-2006) for the study region (northeastern
St. Croix). Blue arrow = two consecutive years of increase; Orange arrows = two consecutive years of decrease. Double arrows indi-
cate three consecutive years. Multidirectional change is indicated by the label “variable”.
Fish variable
2003 2004 2005 2006
ChangeMean SE Mean SE Mean SE Mean SE
Community metrics
Number of species 14.2 0.5 13.4 0.6 14.2 0.5 14.4 0.4
Shannon diversity 1.8 0.04 1.7 0.05 1.8 0.04 1.8 0.0 Variable
Taxonomic diversity (pres/abs) 57.4 0.7 57.4 0.8 58.0 0.7 59.6 0.7
Density (all species combined) 115.5 5.8 115.2 7.4 132.9 6.7 148.8 7.6
Biomass 4371.1 497.8 5705.6 704.7 3444.7 273.1 5160.3 476.8 Variable
Herbivore density 44.8 2.6 45.1 3.7 53.7 3.3 56.2 3.6
Herbivore biomass 1773.5 206.1 3156.7 582.7 1806.5 182.0 2326.6 358.7 Variable
Piscivore density 4.5 0.5 4.3 1.2 4.0 0.4 5.5 0.7
Piscivore biomass 1015.9 162.7 1044.7 288.7 759.0 167.2 1173.5 231.9 Variable
Serranidae
Grouper density 1.5 0.2 0.9 0.1 1.9 0.2 1.8 0.2 Variable
Grouper biomass 317.0 68.2 243.9 42.6 219.6 30.8 263.6 28.1
Coney (C. fulva) density 1.2 0.2 0.7 0.1 1.5 0.2 1.4 0.1 Variable
Coney (C. fulva) biomass
231.3 63.3 173.8 35.4 131.0 20.5 175.3 22.3
Red hind (E. guttatus) density 0.2 0.05 0.2 0.0 0.4 0.1 0.4 0.1 Variable
Red hind (E. guttatus) biomass 79.5 15.2 59.6 21.9 75.2 18.1 77.1 13.7 Variable
Lutjanidae
Snapper density 0.9 0.2 0.7 0.1 0.6 0.1 1.2 0.3
Snapper biomass 193.2 47.0 198.8 71.3 136.4 30.6 195.5 50.8 Variable
Yellowtail snapper (O. chrysurus) density 0.7 0.2 0.4 0.1 0.5 0.1 0.8 0.2 Variable
Yellowtail snapper (O. chrysurus) biomass 77.0 16.1 57.6 21.2 50.1 18.7 48.7 9.9
Schoolmaster snapper (L. apodus) density 0.1 0.0 0.09 0.05 0.04 0.01 0.1 0.0
Schoolmaster snapper (L. apodus) biomass 20.9 13.8 40.6 23.8 16.2 7.3 22.4 9.6 Variable
Gray snapper (L. griseus) density 0.04 0.03 0.01 0.01 0 0 0.1 0.1 Variable
Gray snapper (L. griseus) biomass 12.2 9.1 2.5 1.8 0 0 25.7 25.6
Haemulidae
Grunt density 10.3 2.9 8.1 3.8 2.1 0.6 4.5 1.9
Grunt biomass 418.7 93.9
283.2 67.1 123.1 45.1 202.4 41.4
French grunt (H. avolineatum) density 2.7 0.9 1.1 0.3 1.0 0.3 2.0 0.9
French grunt (H. avolineatum) biomass 135.9 28.4 103.1 18.9 44.1 8.5 86.1 15.8
Bluestriped grunt (H. sciurus) density 0.5 0.2 0.1 0.0 0.1 0.0 0.0 0.0
Bluestriped grunt (H. sciurus) biomass 82.9 36.5 42.7 22.0 11.3 6.2 7.8 4.1
Scaridae
Parrotsh density 16.1 1.1 13.8 1.2 14.8 1.1 13.1 0.9 Variable
Parrotsh biomass 646.2 70.4 837.1 118.4 718.0 80.6 827.5 115.4 Variable
Redband parrotsh (S. aurofrenatum) density 3.7 0.4 2.4 0.3 5.4 0.5 4.7 0.4 Variable
Redband parrotsh (S. aurofrenatum) biomass 181.3 22.0 199.6 32.8 203.5 22.1 254.9 31.8
Striped parrotsh (S. iseri) density 4.4 0.5 3.4 0.6 3.1 0.5 2.0 0.3
Striped parrotsh (S. iseri) biomass 59.9 10.9 50.6 11.1 31.2 5.8 45.7 13.8
Other species
Blue tang (A. coeruleus) density 4.1 0.8 8.3 1.8 6.8 1.0 6.9 1.9 Variable
Blue tang (A. coeruleus) biomass 443.8 134.0
1199.5 308.5 483.9 106.0 788.5 248.8 Variable
Sharks and Rays density 0.03 0.01 0.01 0.01 0.03 0.01 0.05 0.02 Variable
Sharks and Rays biomass 435.9 381.9 41.2 39.6 43.4 31.4 148.6 88.3 Variable
3 - Results
p. 47
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Table 14. Summary statistics (mean + SE) for a range of sh variables grouped by year (2003-2006) inside BIRNM, northeastern St.
Croix. Blue arrow = two consecutive years of increase; Orange arrows = two consecutive years of decrease. Double arrows indicate
three consecutive years. Multidirectional change is indicated by the label “variable”.
Fish variable
2003 2004 2005 2006
Change
Mean In SE In Mean In SE In Mean In SE In Mean In SE In
Community metrics
Number of species 14.65 0.74 13.13 0.81 14.74 0.66 14.91 0.59 Variable
Shannon diversity 1.80 0.07 1.58 0.07 1.77 0.06 1.84 0.06 Variable
Taxonomic diversity (pres/abs) 55.91 1.27 55.35 1.53 57.14 1.16 60.00 0.84 Variable
Density (all species combined) 111.75 7.34 128.63 10.68 147.95 10.14 150.64 10.24
Biomass 4520.20 542.35 6188.42 1261.80 4106.78 379.82 5985.86 773.23 Variable
Herbivore density 46.34 4.38 55.51 6.91 64.22 5.43 61.43 5.42
Herbivore biomass 2567.12 390.94 4641.46 1177.32 2690.23 332.21 3520.99 697.34 Variable
Piscivore density 4.68 0.79 3.27 0.92 3.70 0.40 5.72 1.12 Variable
Piscivore biomass 970.10 201.34 479.55 102.22 594.25 105.95 831.62 141.42 Variable
Serranidae
Grouper density 1.84 0.38 0.92 0.19 2.21 0.33 1.90 0.23 Variable
Grouper biomass 426.91 135.70 198.97 58.64 206.06 37.86 324.03 47.55 Variable
Coney (C. fulva) density 1.65 0.37 0.74 0.18 1.84 0.30 1.63 0.22
Variable
Coney (C. fulva) biomass 358.96 128.08 158.96 56.73 122.72 21.28 231.44 38.87
Red hind (E. guttatus) density 0.15 0.04 0.18 0.06 0.31 0.07 0.24 0.04
Red hind (E. guttatus) biomass 60.01 18.80 38.06 14.17 82.24 26.62 82.46 22.03 Variable
Lutjanidae
Snapper density 0.56 0.10 0.27 0.07 0.58 0.14 0.75 0.21
Snapper biomass 154.03 45.73 76.84 34.09 159.13 52.69 228.58 81.98 Variable
Yellowtail snapper (O. chrysurus) density 0.39 0.08 0.20 0.06 0.44 0.13 0.31 0.07 Variable
Yellowtail snapper (O. chrysurus) biomass 75.49 25.11 44.77 28.39 58.72 32.31 54.53 14.93 Variable
Schoolmaster snapper (L. apodus) density 0.03 0.02 0 0 0.08 0.03 0.05 0.02 Variable
Schoolmaster snapper (L. apodus) biomass 16.13 11.36 0 0 32.86 14.93 33.93 18.85 Variable
Gray snapper (L. griseus) density 0.04 0.04 0.01 0.01 0 0 0.01 0.01
Gray snapper (L. griseus) biomass 18.00 18.00 1.89 1.89 0 0 0.26 0.26
Haemulidae
Grunt density 2.49 0.61 1.15 0.28 0.99 0.23 2.42
1.20
Grunt biomass 180.51 53.39 119.77 31.46 77.53 21.42 221.49 58.89
French grunt (H. avolineatum) density 1.16 0.34 0.85 0.24 0.77 0.19 1.81 1.06
French grunt (H. avolineatum) biomass 63.91 10.93 57.58 15.05 44.39 10.61 114.66 23.44
Bluestriped grunt (H. sciurus) density 0.06 0.02 0.04 0.02 0.05 0.03 0.02 0.01 Variable
Bluestriped grunt (H. sciurus) biomass 14.12 5.70 22.71 12.56 12.44 8.25 10.27 7.45 Variable
Scaridae
Parrotsh density 13.73 1.39 11.54 1.37 16.55 1.83 15.44 1.35 Variable
Parrotsh biomass 909.42 125.44 997.91 214.23 1039.05 145.55 1261.63 225.00
Redband parrotsh (S. aurofrenatum) density 3.24 0.50 1.67 0.28 5.22 0.73 5.61 0.58 Variable
Redband parrotsh (S. aurofrenatum) biomass 175.43 30.79 136.37 37.07 221.33 30.47 304.39 55.36
Striped parrotsh (S. iseri) density 3.41 0.61 2.37 0.63 4.52 0.97 2.97 0.58 Variable
Striped parrotsh (S. iseri) biomass 72.06 19.86 38.14 14.57 42.93 10.05 83.89 27.88
Other species
Blue tang (A. coeruleus) density 6.29 1.57 13.65 3.60 11.18 1.99 10.92 3.76
Variable
Blue tang (A. coeruleus) biomass 818.37 272.97 2091.38 638.73 874.25 210.59 1305.35 504.49 Variable
Sharks and Rays density 0.03 0.02 0 0 0.03 0.02 0.07 0.03 Variable
Sharks and Rays biomass 100.50 71.75 0 0 84.21 64.25 126.96 85.60 Variable
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 48
Table 15. Summary statistics (mean + SE) for a range of sh variables grouped by year (2003-2006) outside BIRNM, northeastern St.
Croix. Blue arrow = two consecutive years of increase; Orange arrows = two consecutive years of decrease. Double arrows indicate
three consecutive years. Multidirectional change is indicated by the label “variable”.
Fish variable
2003 2004 2005 2006
Change
Mean Out SE Out Mean Out SE Out Mean Out SE Out Mean Out SE Out
Community metrics
Number of species 13.79 0.67 12.43 0.78 13.68 0.69 13.78 0.66 Variable
Shannon diversity 1.82 0.06 1.72 0.07 1.76 0.05 1.67 0.06 Variable
Taxonomic diversity (pres/abs) 58.79 0.83 58.51 0.93 58.70 0.79 59.31 1.03 Variable
Density (all species combined) 119.08 8.99 100.73 11.89 119.08 8.65 146.80 11.21 Variable
Biomass 3540.57 412.35 4918.35 806.26 2793.13 387.51 4284.39 561.44 Variable
Herbivore density 43.71 2.88 33.39 3.64 43.89 3.81 50.84 4.71 Variable
Herbivore biomass 1051.65 135.51 1607.58 307.11 961.77 124.16 1158.73 199.61 Variable
Piscivore density 4.25 0.74 5.40 2.38 4.33 0.62 5.34 0.79 Variable
Piscivore biomass 1070.27 256.58 1704.56 624.65 913.07 313.03 1503.03 431.69 Variable
Serranidae
Grouper density 1.20 0.21 0.70 0.15 1.61 0.22 1.78 0.24
Grouper biomass 213.22 36.57 297.83 71.38 224.39 47.82 207.83 31.45
Coney (C. fulva) density 0.81 0.16 0.47 0.12 1.09 0.18 1.11 0.18
Coney (C. fulva) biomass 110.21 24.02 194.08 51.59 140.02 34.87 125.58 23.13
Red hind (E. guttatus) density 0.33 0.08 0.19 0.09 0.39 0.08 0.57 0.11
Red hind (E. guttatus) biomass 98.15 23.84 91.41 46.23 67.08 24.99 69.78 16.88 Variable
Lutjanidae
Snapper density 1.32 0.39 1.09 0.28 0.58 0.12 1.57 0.46
Snapper biomass 233.31 81.46 181.72 64.75 105.80 31.36 156.75 62.33
Yellowtail snapper (O. chrysurus) density 0.94 0.30 0.61 0.19 0.44 0.12 1.18 0.39
Yellowtail snapper (O. chrysurus) biomass 79.74 21.03 49.39 28.51 32.34 17.57 37.97 12.03
Schoolmaster snapper (L. apodus) density 0.07 0.06 0.18 0.11 0.01 0.01 0.06 0.02 Variable
Schoolmaster snapper (L. apodus) biomass 25.76 24.96 83.09 52.40 0.29 0.29 8.79 5.53 Variable
Gray snapper (L. griseus) density 0.04 0.03 0.01 0.01 0 0 0.21 0.21
Gray snapper (L. griseus) biomass 6.82 5.62 3.53 3.53 0 0 49.88 49.88
Haemulidae
Grunt density 17.87 5.62 15.10 8.33 3.22 1.08 6.52 3.59
Grunt biomass 650.30 174.85
354.92 96.83 166.04 86.32 156.63 51.25
French grunt (H. avolineatum) density 4.23 1.72 1.11 0.40 1.27 0.53 2.16 1.56 Variable
French grunt (H. avolineatum) biomass 206.13 54.14 131.15 34.57 42.22 13.23 58.95 21.42
Bluestriped grunt (H. sciurus) density 0.89 0.37 0.05 0.04 0.09 0.07 0.01 0.01 Variable
Bluestriped grunt (H. sciurus) biomass 149.13 70.93 27.31 20.55 10.28 9.23 1.99 1.99
Scaridae
Parrotsh density 18.58 1.61 14.40 2.09 13.25 1.22 10.82 1.19
Parrotsh biomass 407.44 63.93 600.81 122.72 409.99 64.85 406.83 57.24
Redband parrotsh (S. aurofrenatum) density 4.06 0.56 2.57 0.47 5.60 0.77 3.86 0.57 Variable
Redband parrotsh (S. aurofrenatum) biomass 189.60 31.85 213.25 51.02 184.58 32.24 212.55 33.74 Variable
Striped parrotsh (S. iseri) density 5.37 0.80 3.90 1.08 1.79 0.38 1.02 0.28
Striped parrotsh (S. iseri) biomass 49.33 10.28 44.36 16.26 19.90 5.90 6.00 2.27
Other species
Blue tang (A. coeruleus) density 2.09 0.34 2.98 0.77 2.62 0.43 2.93 0.67 Variable
Blue tang (A. coeruleus) biomass 97.01
28.60 371.83 116.34 112.93 22.53 260.31 84.41 Variable
Sharks and Rays density 0.02 0.01 0.01 0.01 0.02 0.01 0.03 0.02 Variable
Sharks and Rays biomass 13.70 8.58 3.51 3.51 4.53 4.01 171.17 152.96 Variable
3 - Results
p. 49
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Outside BIRNM
Outside BIRNM, declines over three consecutive years were recorded for striped parrotsh and for grunt biomass
including bluestriped grunt biomass, with 2005 and 2006 values signicantly lower (p=<0.05) than 2003 (Table 15 and
Appendix E). Unlike inside BIRNM, no metric exhibited three consecutive years of increase. Declines recorded for at
least two consecutive years were considerably more widespread, with decreases in mean metric value for all major sh
families, including: all grouper biomass; coney biomass; red hind density and biomass; all snapper density and biomass;
yellowtail snapper density and biomass; gray snapper density and biomass; grunt density and biomass; French grunt and
bluestriped grunt biomass; parrotsh density and biomass, including striped parrotsh density and biomass (Table 15).
Overall, 17 metrics exhibited at least two consecutive years of decline, three metrics increased over two years, and 19
showed no clear directionality (Table 15). Interestingly, many of the sh metrics that showed a declining trend between
2003 and 2005, then showed a substantial increase in 2006.
3.2.9 Seasonal and inter-annual patterns in sh community metrics
Mean sh density was markedly lower
over hardbottom habitat types during the
spring sampling season than during the fall
sampling season (Figure 44). Table 16 shows
that all of the most abundant sh species in
the study region were more abundant in fall
than spring. Greatest differences were found
for the three most abundant sh species
bluehead wrasse (Thalassoma bifasciatum),
slippery dick (Halichoeres bivittatus) and
bicolor damselsh (Stegastes partitus).
As a result of this distinct seasonal
difference, temporal change was examined
separately for: (1) March/April; and (2)
October sampling periods.
Figure 44. Mean (+ SE) density for all species combined by sampling season for
both inside and outside BIRNM.
Table 16. Spring and fall total abundance and mean (+ SE) density for the 20 most abundant sh species across
the study region (northeastern St. Croix).
Species Common name
Spring total
abundance
Spring mean
density (+SE)
Fall total
abundance
Fall mean
density (+SE)
Thalassoma bifasciatum Bluehead wrasse 9525 15.1 (0.9) 22476 34.9 (2.2)
Halichoeres bivittatus Slippery dick 9710 15.4 (3.4) 15042 23.4 (1.3)
Stegastes partitus Bicolor damselsh 3500 5.5 (0.5) 6702 10.4 (0.7)
Acanthurus bahianus Ocean surgeonsh 3670 5.8 (0.4) 4931 7.7 (0.5)
Acanthurus coeruleus Blue tang 3737 5.9 (0.8) 4860 7.5 (1.1)
Sparisoma aurofrenatum Redband parrotsh 1820 2.9 (0.2) 3067 4.8 (0.3)
Scarus iseri Striped parrotsh 2212 3.5 (0.3) 2549 4.0 (0.4)
Halichoeres garnoti Yellowhead wrasse 1660 2.6 (0.2) 2998 4.7 (0.3)
Stegastes leucostictus Beaugregory 1537 2.4 (0.3) 2454 3.8 (0.4)
Chromis cyanea Blue chromis 1426 2.3 (0.4) 1895 2.9 (0.5)
Haemulon aurolineatum Tomtate 1052 1.7 (0.9) 1581 2.5 (1.9)
Xyrichtys martinicensis Rozy razorsh 1231 2.0 (0.6) 1351 2.1 (0.4)
Sparisoma viride Stoplight parrotsh 968 1.5 (0.1) 1488 2.3 (0.2)
Decapterus macarellus Mackerel scad 820 1.3 (0.5) 1499 2.3 (0.9)
Halichoeres maculipinna Clown wrasse 743 1.2 (0.1) 1433 2.2 (0.2)
Clepticus parrae Creole wrasse 827 1.3 (0.5) 1333 2.1 (1.2)
Haemulon avolineatum French grunt 687
1.1 (0.2) 1257 2.0 (0.5)
Stegastes diencaeus Longn damselsh 930 1.5 (0.2) 940 1.5 (0.2)
Scarus taeniopterus Princess parrotsh 711 1.1 (0.1) 1002 1.6 (0.1)
Cephalopholis fulva Coney 604 1.0 (0.1) 787 1.2 (0.1)
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 50
a)
b)
c)
Inside BIRNM
Outside BIRNM
Figure 45. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and outside BIRNM for: (a) all sh
biomass; (b) herbivorous sh biomass; and (c) piscivorous sh biomass.
3 - Results
Between 2003 and 2006, mean density increased in both sampling seasons. Mean sh biomass was higher inside BIRNM
for six of eight sampling seasons and mean herbivore biomass was higher inside BIRNM in all seasons and years.
Piscivore biomass was higher outside in six out of eight sampling seasons (Figure 45). Furthermore, spring piscivore
decreased inside, while increasing outside BIRNM between 2003 and 2006.
In contrast, the number of sh species and value of diversity indices varied very little across years except for a slight
decrease during 2004, although mean values were marginally higher in fall 2005 and 2006 than fall 2003 and 2004
(Figure 46).
p. 51
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Inside BIRNM
Outside BIRNM
Figure 46. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh diversity inside and outside BIRNM.
3 - Results
3.2.10 Seasonal and inter-annual patterns in sh groups and species
Mean grouper biomass was higher in spring than fall sampling seasons, particularly within BIRNM, although spring
biomass declined inside BIRNM from 2003 to 2005 (Figure 47a). Spring snapper biomass also was greater than fall
and increased gradually from 2003 to 2006 inside BIRNM, but appeared to decrease outside over the same time period
(Figure 47b). Grunt and parrotsh biomass also was greater in spring than fall, but not for all years. In general, grunt
biomass decreased outside BIRNM between 2003 and 2006 and remained relatively consistently low inside (Figure
47c). Parrotsh biomass was higher inside BIRNM for all sampling seasons and all years (Figure 47d). Spring biomass
appeared fairly consistent across years, but fall biomass increased year upon year inside BIRNM.
Coney mean biomass was higher inside BIRNM for ve of eight sampling seasons, but showed a dramatic decline
from 2003 to 2005 in the spring sampling season (Figure 48a). Fall biomass was lower than spring and highly variable
within season. Red hind biomass was lower than coney biomass and highly variable across seasons and years, with
similar inside/outside biomass patterns (Figure 48b). Yellowtail snapper mean biomass in spring was higher inside than
outside for all years (Figure 49a). Other snapper species were highly variable and not abundant enough to determine any
meaningful temporal patterns between season and years (Figure 49).
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 52
Inside BIRNM
Outside BIRNM
a)
b)
c)
d)
Figure 47. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and outside BIRNM for: (a) groupers, (b)
snappers, (c) grunts and (d) parrotsh.
3 - Results
p. 53
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Inside BIRNM
Outside BIRNM
a)
b)
Figure 48. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and outside BIRNM for (a) coney (C.
fulva) and (b) red hind (E. guttatus).
Red hind (Epinephelus guttatus)
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 54
Inside BIRNM
Outside BIRNM
a)
b)
c)
Figure 49. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and outside BIRNM for: (a) yellowtail
snapper (O. chrysurus), (b) schoolmaster (L. apodus) and (c) gray snapper (L. griseus).
Yellowtail snapper (Ocyurus chrysurus)
3 - Results
p. 55
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Inside BIRNM
Outside BIRNM
a)
b)
Figure 50. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and outside BIRNM for (a) French grunt
(H. avolineatum) and (b) bluestriped grunt (H. sciurus).
Mean biomass of French grunts and bluestriped grunts outside BIRNM declined year after year from 2003 to very low levels
in 2006 (Figure 50). Inside BIRNM, biomass increased for French grunt and varied little between years for bluestriped
grunt.
Herbivore mean biomass in spring was higher inside BIRNM for all species (blue tang, striped parrotsh and redband
parrotsh) and almost all years (Figure 51). In addition, spring biomass for all species examined was higher in spring than
in fall. For both seasons biomass was highly variable among years and no obvious trend was observed.
Assemblage of juvenile, subadult and adult grunts (Haemulidae) on a patch reef.
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 56
Inside BIRNM
Outside BIRNM
a)
b)
c)
Figure 51. Seasonal and inter-annual (2003-2006) change in mean (+ SE) sh biomass inside and outside BIRNM for: (a) blue tang (A.
coeruleus), (b) redband parrotsh (S. aurofrenatum) and (c) striped parrotsh (S. iseri).
3 - Results
p. 57
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
3.3 Macroinvertebrate spatial distribution patterns and species-habitat associations
3.3.1 queen conch (Strombus gigas)
A total of 736 queen conch were observed in the study region (northeastern St. Croix) between 2004 and 2006, of which
72.3% were juveniles (i.e., no ared lip). Highest mean density of both juveniles and adults were recorded in seagrass
beds inside BIRNM (Figure 52). The maximum number of individuals at any one survey site (100 m
2
) was 59, recorded
from a seagrass bed inside BIRNM. Overlay of distributions on the benthic habitat map showed a concentration of juveniles
in seagrasses directly in the sheltered leeward side of Buck Island. Adults showed a similar distribution (Figure 53), albeit
less concentrated and at lower densities revealing that adult and juvenile S. gigas were not spatially segregated. Sixty
percent of sites with juveniles present also had adults present. Comparison of estimates of sighting frequency (Figure
54) show that S.gigas juveniles and adults were markedly more common in northeastern St. Croix than for St. John or
southwestern Puerto Rico and comparison of area-weighted abundance for the three islands highlights the importance of
the St. Croix seagrass beds in supporting a queen conch population of regional signicance.
Comparison of estimates of sighting frequency (Figure 54) show that S.gigas juveniles and adults were markedly more
common in northeastern St. Croix than for St. John or southwestern Puerto Rico and comparison of area-weighted
abundance for the three islands highlights the importance of the St. Croix seagrass beds in supporting a queen conch
population of regional signicance (Table 17).
Juveniles
Adults
Inside BIRNM
Outside BIRNM
3 - Results
Figure 52 Mean (± SE) density for (a) juvenile and adult queen conch (S. gigas) by
habitat type, and (b) all queen conch inside and outside BIRNM by dominant habitat
types in the study region (northeastern St. Croix) between 2004 and 2006.
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 58
Figure 53. Spatial distributions of (a) juvenile (immature) and (b) adult (mature) queen conch (S. gigas) density in the study region
(northeastern St. Croix) between 2004 and 2006.
3 - Results
Figure 54. Sighting frequency of sexually immature (juvenile), mature (adult),
and all queen conch from three study sites in the U.S. Caribbean: Southwest
Puerto Rico; northeastern St. Croix and St. John. Sighting frequency was
calculated as the percentage of sampled sites where at least one juvenile or
adult conch was observed. Source: Jeffrey and Monaco, 2007.
Table 17. Estimates of total queen conch abundance (number of individuals) by life stage for three
islands in the U.S. Caribbean (2004-2006). Source: Jeffrey and Monaco, 2007.
Island
Size of study
area (ha)
% of area
sampled
# of
surveys Life stage
Estimated
abundance Range of estimate
Puerto Rico 157,348 < 0.01 394
Immature 1,100,248 236,943 - 1,963,553
Mature 204,645 34,698 - 374,591
Total 1,304,893 271,641 - 2,338,144
St. Croix 32,014 0.02 624 Immature 1,933,950 1,025,084 - 2,842,815
Mature 835,005 493,204 - 1,176,805
Total 2,768,954 1,518,288 - 4,019,620
St. John 4,697 0.11 505 Immature 169,838 39,019 - 300,656
Mature 72,832 26,576 - 119,087
Total 242,669 65,596 - 419,743
p. 59
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
3.3.2 Long-spined sea urchin (Diadema antillarum)
D. antillarum was observed at approximately 10%
of sites (38 out of 364) surveyed between October
2005 and November 2006. Mean density across the
study region was 0.03 (± 0.01 SE) per 1 m
2
. Maximum
density recorded at any individual site was 4.2 m
2
.
Visual assessment of spatial distributions of abundance
revealed that very few D. antillarum were using the
coral reef ecosystems around Buck Island between
2005 and 2006 (Figure 55) and D. antillarum densities
were considerably higher on hardbottom habitat types
outside BIRNM than inside (Figure 55). Highest densities
were observed in the nearshore environments within
the EEMP on colonized bedrock, colonized pavement
and macroalgal beds in close proximity to extensive
seagrass beds (Figure 56).
Figure 55. Spatial distribution of long-spined urchins (Diadema
antillarum) in the study region (northeastern St. Croix) between 2005
and 2006.
Juveniles
Adults
Inside BIRNM
Outside BIRNM
Figure 56. Mean (± SE) density for long-spined sea urchin (Diadema antillarum) by (a)
habitat type and (b) inside and outside BIRNM (northeastern St. Croix) between 2005
and 2006.
3 - Results
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 60
3.3.3 Historical comparison of Diadema abundance
In 1979, Gladfelter (1980) carried out daytime surveys
and recorded a peak density of 10.6 per m
2
on pavement
areas at the northwest end of Buck Island ,between 5
and 10 per m
2
in the lagoonal patch reef south of island
and 5-8.3 m
2
on the bank barrier reef at the eastern tip
of the island. A few years earlier, Ogden et al. (1972)
recorded mean D. antillarum densities of between 0.81
and 4.08 per m
2
on patch reefs in Teague Bay (Figure
57).
In contrast, between 2005 and 2006, mean density of
D. antillarum was recorded at 0.03 m
2
across the study
region and presence of D. antillarum. Of particular
importance was the observation of only ve D. antillarum
individuals over hardbottom habitats in the lagoon and
bank barrier reef areas (n=43 transects or 4,300 m
2
surveyed within 500 m of Buck Island) between October
2005 and November 2006. However, sea urchin
densities were higher in lagoonal and back reef areas
outside BIRNM along the northeastern coastline of St.
Croix.
D. antillarum, October 2005
Figure 57. The changing abundance of Diadema antillarum in (a) the
lagoon and on bank barrier coral reefs within 500 m of Buck Island
and (b) Teague Bay and adjacent nearshore lagoonal environments
showing little to no recovery in over two decades since the mass
mortality event. Source: Ogden, 1972; Gladfelter, 1980; CCMA-BB.
3.3.4 Caribbean spiny lobster (Panulirus argus)
A total of 24 spiny lobster (Panulirus argus) were recorded over
hardbottom areas from 2003 to 2006, including two in 2003
from two sites; eight in 2005 from six sites and 14 in 2006 from
ve sites (Table 18). There were no lobsters observed in 2004.
Fifteen spiny lobsters were observed inside BIRNM and nine
outside. The highest densities at individual sites were observed
in patch reef and colonized pavement habitat types dominated
by branching corals (three and nine lobsters respectively). Five
lobsters were observed over scattered coral/rock in sand habitat
type. No lobsters were observed on softbottom sites. However,
the abundance of lobsters detected using existing techniques is
very likely to be an underestimate
of abundance. Lobsters are cryptic
and crevice dwelling animals that
are best surveyed using dedicated
lobster census techniques and
supplemented with night time surveys
when some lobsters are more active
and therefore more visible. These
data should not be used to estimate
spiny lobster populations in the study
region.
P. argus, March 2007
Table 18. Abundance of spiny lobster (Panulirus argus) in
hard and soft habitats of the study region and inside and
outside BIRNM (northeastern St. Croix) between 2003 and
2006.
Habitat
Location
Year Types
Inside Outside Total
2003
Hard 0 2 2
Soft 0 0 0
Overall 0 2 2
2004
Hard 0 0 0
Soft 0 0 0
Overall 0 0 0
2005
Hard 4 4 8
Soft 0 0 0
Overall 4 4 8
2006
Hard 11 3 14
Soft 0 0 0
Overall 11 3 14
Total 15 9 24
3 - Results
4 - Discussion
4. Discussion
This report demonstrates quantitatively and spatially that the benthic environment inside BIRNM was signicantly different
to the outside (Tables 5, 6 and 7). The results show conclusively that the abundance of key benthic components, at both
the 1 m
2
spatial scale (quadrat sampling unit) and at the 100 m radius spatial scale (e.g., surrounding seascape unit), were
signicantly different inside. For instance, nine of 14 biotic variables measured within quadrats set over hardbottom habitat
sites and seven of nine seascape metrics (amount and richness of habitat types) were signicantly different inside versus
outside BIRNM. Seascapes inside BIRNM were more diverse and had on average a higher area of colonized hardbottom
and lower area of seagrasses surrounding transects. Furthermore, visual examination of interpolated distribution maps
clearly showed that a greater spatially continuous area of high coral cover, coral species richness and rugosity was
contained within the boundary of BIRNM consistent with objectives of the marine protected area designation. The existing
BIRNM boundary broadly follows the boundary between medium-high coral cover and medium-low coral cover as depicted
in the spatial interpolation. In addition, higher coral cover for all major scleractinian families was recorded inside BIRNM
and coral reefs inside also had a higher ratio of live coral cover to macroalgal cover than coral reefs outside BIRNM.
Such information may serve as a useful indicator for change detection, with the coral-macroalgal ratio being an important
benthic relationship relevant to faunal communities and indicative of the functional status of the environment.
Although coral reef structure has been modied by several major
events since the original National Monument designation and the
special recognition of the “marine gardens” in 1961, the area around
the eastern tip of Buck Island remains distinctive in both biological and
geomorphological structure. The spatial data contained in the report
characterizes this area as having high live coral cover, high rugosity,
high coralline algal cover, high sh species richness, high biomass of
herbivorous sh and high density for many sh species. The relevance
of coralline algae is related to its function in providing an important
food source for some sh including parrotsh and physical structure
and chemical cues promoting the settlement and metamorphosis of
many invertebrates including some coral larvae. The distinct biological
features of the eastern tip of Buck Island suggest that the area could
function as refugia for several benthic organisms. If true, then because
of prevailing northeasterly winds and circulatory patterns, the area may
be an important source of larvae for corals and other benthic organisms in down stream areas. Hardbottom areas inside
BIRNM exhibited a signicantly higher ratio of coral to macroalgae than did hardbottom areas outside. This may be
indicative of greater top-down control of macroalgae due to grazing pressure since hardbottom areas inside BIRNM also
have signicantly higher herbivore biomass, particularly parrotsh and surgeonsh than similar outside areas.
In addition to the well-recognized and extensive hardbottom areas north and east of Buck Island within BIRNM, an
additional area was found to support high coral species richness and sh species richness along the northeast coastline
of St. Croix. The linear reef and adjacent colonized pavement extends east-west from Teague Bay to Coakley Bay and
now falls within the EEMP no-take zone and recreation zone (Appendix A, Figure A1). This extensive fringing coral reef
may also offer important habitat to sh moving from nearshore seagrass and patch reef environments to coral reefs
as part of ontogenetic transitions in habitat use. More focused research including acoustic tracking may elucidate on
the connectivity between nearshore lagoonal environments and coral reefs both inside and outside BIRNM. Very little
is known about sh movement patterns in the study region and data from an acoustic tracking project would provide
important information on connectivity. For example, is the reason for low abundance of the large size classes of grouper,
snapper and grunt due to emmigration out from the study region to deeper waters or mortality? A multi-year broadscale
acoustic tracking study has the capability to answer this question.
Extensive areas with high coral species richness, high cover for Montastraea cavernosa and M. annularis, high sh
species richness and high abundance for several sh species including
coney (C. fulva), rock beauty (Holacanthus tricolor) and queen triggersh
(Balistes vetula) occurred along the northernmost edge of the benthic
habitat map. This indicates that important deeper water habitat is likely
to exist beyond the scope of this report, requiring further benthic habitat
mapping effort combined with deeper water visual census to capture
data on sh communities.
In comparative investigation of management domains for the purpose of
evaluating efcacy (e.g., inside versus outside protected areas), caution
is required since clear differences are evident in benthic habitat that will
likely explain some of the patterns in faunal distributions independently
of management practices. However, this does not obviate the usefulness
of comparative analyses, since differences and similarities that exist
within or between management domains provide valuable information
p. 61
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Crustose coralline algae
queen triggersh (Balistes vetula)
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 62
in support of local decision-making processes for the selected regions of interest. Future assessments could, however,
usefully attempt to partition out the relative inuence of differences in habitat from the effects of management actions.
Fish species, family/trophic group and communities inside and outside BIRNM
As expected, sh community composition was most dissimilar between linear reefs and scattered coral or rock, habitat
types that are known to differ markedly in structural complexity and areal extent. Unexpectedly, sh communities
associated with aggregated patch reefs, linear reefs and colonized pavement were so similar as to be barely separable.
These habitat types were considered distinct to the human observer when classifying the benthic structure, yet for the
sh communities the differences were not signicant. This may have implications for the level of thematic accuracy
required when delineating benthic habitat in the construction of benthic maps for sh-habitat studies. Considerable cost
savings may result from delineation of fewer classes of hardbottom. Further studies using high resolution Light Detection
and Ranging (LiDAR) bathymetry should provide useful information on the structural variability within existing NOAA
benthic habitat types and will help to understand the implications of benthic structural types for sh. Furthermore, sh
communities associated with the same habitat types inside and outside BIRNM were very similar and indistinguishable.
Examination of more subtle differences in community composition between strata may require additional multivariate
analytical techniques.
Several individual sh metrics were signicantly higher on colonized
hardbottom habitats inside than outside BIRNM including sh
biomass (all sh combined), herbivore biomass, parrotsh biomass,
shark and ray biomass, coney density and biomass, blue tang (A.
coeruleus) density and biomass and striped parrotsh (S. iseri)
biomass. Comparatively fewer sh metrics were signicantly higher
outside, but included ecologically important predator groups such as
piscivore biomass, snapper density and grunt density and biomass.
The greater biomass of piscivorous sh outside BIRNM probably
relates more to habitat preferences of relatively infrequently occurring
sharks and rays (over unvegetated sediments) than any indication
of MPA performance. Similarly, higher grunt density and biomass
outside BIRNM likely relates more to the abundance of shallow
nearshore hardbottom environments in close proximity to seagrass beds than MPA performance. Interpretation of these
patterns highlights the importance of using spatially explicit census data.
Size frequency histograms revealed some differences in size structure inside versus outside BIRNM for grunts and
snapper, with: (1) a higher frequency of large bodied snapper inside BIRNM; (2) a greater decrease in frequency of large
subadult and adult yellowtail snapper (O. chrysurus) with increasing size outside BIRNM; (3) a higher frequency of large
juvenile and subadult French grunt inside BIRNM; and (4) higher frequency of small juveniles outside (Figures 42 and
43). These differences may reect differences in environment or management/shing pressure or both, but nevertheless
data indicate that a larger proportion of the mature adult snapper and grunts are occurring inside the protected area than
outside.
Interestingly, very few of the largest and very few of the smallest size
classes were observed for grouper and snapper. For small juveniles,
this suggests that either: (1) the study region is not a primary
settlement area for groupers and snappers; or (2) newly settled
juveniles exist, but are not being detected by the belt transect survey
technique due to cryptic coloration or hiding behavior thus providing
false absences. For large adults, this suggests that either: (1) large
adults do not use the study area and may perhaps inhabit deeper
unsurveyed waters by day; (2) large adults exist, but are not being
detected by the belt transect survey technique due to avoidance
behavior thus providing false absences; or (3) large adults are being
removed from the system by shing. Additional targeted survey work
may be required to determine the locally important areas for both
newly settled juveniles and large-bodied mature adults.
Spatial distributions and sh-habitat associations
Although mean densities varied widely between habitat types, with many of the most abundant sh across the study
region found in highest densities over hardbottom habitat types, most utilized multiple habitat types including seagrasses
and sand. This has implications for the way species-habitat relationships are understood and managed and further study
may provide insight into sh resource use patterns and predictions on resilience to change, with many species having
evolved sufcient plasticity to exploit a diverse range of structurally distinct habitat types. In addition, many of the juveniles
of the most abundant species used the same habitat types as adults and the two life-stages were often found to coexist in
the same areas of the study area. This pattern is in contrast to results for the same species elsewhere (e.g., southwestern
Southern stingray (Dasyatis americana)
Stoplight parrotsh (Sparisoma viride) and longn damselsh
(Stegastes diencaeus)
4 - Discussion
p. 63
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Puerto Rico) that have shown a cross-shelf or water depth related size distribution or have reported juveniles using
spatially discrete areas as nursery before undertaking ontogenetic habitat shifts to preferred adult habitat. This difference
in pattern observed in northeastern St. Croix further highlights the exibility or plasticity in resource utilization patterns for
sh using heterogeneous coral reef ecosystems.
Visual assessment of density distributions for the two abundant species
of grouper (red hind [E. guttatus] and coney) around northeastern St.
Croix suggest different habitat use patterns, with high coney density
being widespread over the contiguous colonized hardbottom areas
(much of which is inside BIRNM) and high densities of red hind found
mostly to the south of Buck Island (many outside BIRNM). These areas
of highest red hind density are also close to the interface between
seagrasses and extensive areas of colonized hardbottom which
may relate to direct or indirect utilization of seagrasses presence of
seagrasses or some other covarying environmental variable such as
wave exposure (Kendall et al., 2004b). Almost all species examined
here were found at higher densities on hardbottom habitat types than
seagrasses or other softbottom, however, surveys were conducted
during daylight hours and many of these species may use seagrasses
during nocturnal foraging excursions. Thus, the low densities of sh associated with seagrasses should not be used
in evaluation of relative habitat importance without complimentary information on resource use activity within the daily
home range that includes examination of diel migrations. This is particularly relevant to many grunts which are known to
make use of seagrasses adjacent to coral reefs as a nighttime feeding ground. Furthermore, most of the coral reefs with
high species richness were within 200 meters of seagrass beds. Several studies have demonstrated links between sh
distribution on coral reefs and proximity to seagrass beds suggesting that many species may benet from complementary
and supplementary resources provided by seagrasses in close proximity (Grober-Dunsmore, 2007; Pittman et al., 2007)
to coral reefs. However, not all coral reef sites in close proximity to seagrasses supported high sh richness, thus it is
likely that the interaction between multiple environmental variables including surface rugosity determine such complex
spatial patterns.
Spatial distribution data on selected species may also provide valuable information that can be used as indicators for
habitat structure, health, and in combination with temporal data, as a tool in change detection. For example, the threespot
damselsh (Stegastes planifrons), which exhibits a strong preference for select taxa of live coral (e.g., Agaricia spp.,
Acropora spp. and Montastraea spp.) and through feeding promotes high algal diversity, may function as a useful indicator
of healthy and structurally complex coral reefs. In the northeastern St. Croix study region, highest densities of threespot
damselsh were found around the eastern tip of Buck Island within BIRNM and the fringing reef outside BIRNM extending
east-west along the northeast coast of St. Croix (Appendix D).
Temporal trends in sh and benthic habitat
With respect to temporal trends, hardbottom benthic habitat exhibited a
higher cover of lamentous algae/cyanobacteria and macroalgae, and
lower turf algae in fall than in spring. A peak in lamentous cyanobacteria/
algae was recorded for October 2005, a season with anomalously high
water temperature that also resulted in a mass coral bleaching event.
Highest algal turf was recorded for spring 2006, which may be linked to
the algal colonization of dead coral colonies that is commonly observed
after a bleaching event. Across years, macroalgae cover shows some
indication of decline both inside and outside BIRNM between 2003 and
2006. This was especially evident for spring sampling seasons and
may indicate increased grazing from a concurrent increase in density
of herbivorous sh in the region as a whole and particularly larger (e.g.
higher biomass) parrotsh inside BIRNM (see below). Another important
observation that emerged from a seasonal comparison of sh density
was the markedly higher abundance of sh in fall than in spring. This
seasonal pattern was also noted by Simpson (1979) and attributed to the inuence of summer recruitment. Still, little is
known about the spatio-temporal characteristics of the life-history patterns for even the most common sh in the study
region and Simpson’s (1979) statement that “knowledge of spawning and recruitment of shes is still rudimentary” is still
pertinent today.
Synoptic overview of inter-annual differences in mean metric values both inside and outside BIRNM showed no consistent
decline for any of the 39 sh metrics inside BIRNM, but increases every year between 2003 and 2006 were recorded for
density of all sh and mean biomass of parrotsh. In contrast, increases for the entire sampling period were not evident
outside BIRNM, instead considerable and consecutive decline was apparent for grunt biomass, especially bluestriped
grunt (H. sciurus) and density and biomass of stripped parrotsh. Grunts and parrotsh are all readily captured in baited
trap and net sheries, and thus are highly susceptible to extraction from the ecosystem. Results here suggest that target
Seagrasses and macroalgae habitat
4 - Discussion
Bleached Diploria strigosa with turf and cyanobacteria
overgrowth.
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 64
species may receive some protection within BIRNM and that may account for the absence of continuous decline and the
apparent increases during the sampling period between 2003 and 2006. If BIRNM retains sh species within its boundary
and the existing legislated protection is enforced effectively, then biomass and abundance of sh would be expected to
increase since the commencement of no-take regulations in 2003. Accumulation of biomass at detectable levels, however,
may require longer term monitoring data. This is particularly likely for grouper since many grouper are comparatively slow
growing and some such as the Nassau grouper are late maturing (i.e., approximately 50 cm length). This together with
the fact that many grouper and snapper aggregate to spawn make them particularly vulnerable to shing pressure and
slower in recovery of viable populations.
Interestingly, many of the sh variables that showed a decline between
2003 and 2005 then showed an increase (sometimes substantial) in
2006, for example, coney biomass, piscivore density, grunt biomass
and French grunt (H. avolineatum) biomass inside BIRNM. This may
be indicative of: (1) the beginning of a response to increased protection;
(2) an artifact of random sampling; or (3) part of variable uctuations
driven by other ecological factors (e.g., predation pressure, changing
habitat quality etc.). It is known that greater voluntary compliance began
in 2003 and this was then supplemented with law enforcement patrols
from 2004. Thus, if shing pressure was an important determinant of
sh abundance and biomass the recovery would not be expected to
be detectable for several years after compliance. Further long term
monitoring will be required to reveal the direction of apparent upturns
reported in 2006 at the end of this portion of the monitoring data.
When comparing inside with outside BIRNM, neither groupers nor snappers exhibited any consistent increases or declines
over the entire sampling period (2003-2006), yet at the scale of the entire study area, yellowtail snapper biomass declined
year upon year. This trend warrants concern and requires further monitoring since yellowtail snapper mean is a highly
valuable commercial species and important ecological component of the ecosystem, being found in all habitat types in
the study region. Furthermore, breakout of data by spring season revealed apparent downward directional trends inside
BIRNM for density of yellowtail snapper, coney, gray snapper (L. griseus) and piscivore biomass (2003-2005). No obvious
trend for the same species existed during fall and this may relate to seasonal differences in the inux and outow of sh
to BIRNM. Acoustic tracking may provide useful information on the seasonal sh movement patterns.
The multi-level and multi-resolution analyses in this report showed that direction of trend and subsequent interpretation of
results varied with temporal grouping or resolution. An apparent inter-annual trend for a single season and management
domain may not be apparent when both seasons are combined, or when management domains are combined. In
addition, density and biomass may exhibit different patterns over time. It is important, therefore, that future temporal
characterizations using summary statistics be undertaken at multiple levels of temporal resolution (season, year), as well
as biological resolution (species, family, trophic group, community) and management domain resolution (inside, outside,
region).
Overall, the majority of sh biomass was highly variable between years and therefore long term monitoring is required to
elucidate further on the direction of change and particularly to track the declining trends in several key sh species and
groups.
Connection between life history and vulnerability to shing
Fisheries management organizations and shers themselves need to be more aware of the relationship between sh life
history characteristics and vulnerability to shing. The apparent decline of three species of large-bodied and late maturing
grouper (tiger, yellown and Nassau grouper) in the study region highlights the vulnerability of some species that were
once of signicant commercial value to the local shery. Species with larger body size, higher longevity, higher age at
maturity, and lower growth rate are generally considered to have higher vulnerability to shing (Jennings et al., 1999;
Dulvy and Reynolds, 2002; Dulvy et al., 2003).
Evidence that shing pressure may have been responsible for some
observed declines in abundance comes from analytical methods used
to estimate extinction vulnerability to shing which have classied
tiger grouper and Nassau grouper as having HIGH to VERY HIGH
vulnerability and yellown grouper as MODERATE to HIGH vulnerability
based on body size and other biological characteristics (FishBase:
http://www.shbase.org; Cheung et al., 2005; Table 19). These species
are also thought to mature late, with tiger grouper estimated to mature
at approximately half its maximum size at an age of between 6.5-9.5
years. As such all three species have low resilience to shing with a
minimum population doubling time calculated at between 4.5-14 years
(Table 19). In contrast, the smaller-bodied grouper (red hind and coney)
Diver collecting sh data
E. striatus (Nassau grouper)
4 - Discussion
p. 65
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
which are now more abundant in the study region have LOW to MODERATE vulnerability and medium resilience to shing
with an estimated population doubling time of between 1.4-4.4 years. For instance, the maximum body size for red hind
is 76 cm TL, yet maturity has been recorded for sh at 25 cm FL (three years of age).
Distribution of macroinvertebrates
Long-spined sea urchins
Dramatic shifts in the distribution and abundance of the long-spined sea urchin (Diadema antillarum) were evident when
comparing present day distributions (2005-2006) with historical distributions recorded in the 1970s. Comparison with
historical density estimates indicated that 1970s/early 1980s densities were substantially higher than in 2005-2006.
Gladfelter (1980) stated “Diadema antillarum is a conspicuous member of the Buck Island fauna” and Ogden et al. (1972)
note that “on certain patch reefs in the vicinity of Teague Bay Reef, the grazing activities of this sea urchin have all but
eliminated the growth of large benthic algae”. Furthermore, high abundance of sea urchins was thought to have been
responsible for the high abundance of sea urchin predators observed within the original BIRNM, including black margate
(Anisotremus surinamensis), Spanish grunt (Haemulon macrostomum), caesar grunt (Haemulon carbonarium) and queen
triggersh; Gladfelter et al., 1977). In contrast, between 2001 and 2006, only two black margate were observed across the
study region, however, caesar grunt, Spanish grunt and queen triggersh were relatively abundant.
Even though differences occurred in the survey methods between studies in 1970s and 2005-2006, the evidence is
clear that D. antillarum no longer plays such an important role as an “ecosystem engineer” in the coral reef ecosystems
of northeastern St. Croix. This decline is likely to have contributed in part to the status of macroalgal/cyanobacterial/
turf dominated coral reefs that currently exist over much of the study region. The
likely cause of these changes was the reported sea urchin die-off that occurred in
1983 and 1984 throughout the Caribbean region resulting in an estimated 95-99%
mortality rate (Lessios et al., 1984; Carpenter, 1988). Carpenter (1988) reported
that ve days after the mass mortality event in St. Croix, algal biomass increased
by 20% and herbivorous removal of algal biomass decreased by 50%, although an
increase in the rate of grazing by herbivorous shes was also observed suggesting
that exploitative competition for food was occurring between D. antillarum and
some herbivorous sh species (Carpenter, 1988). In areas of the Caribbean where
shing had reduced the number of herbivorous sh, the growth and persistence
of macroalgae increased (Lessios, 1988). It appears that recovery to pre-disease
levels of density have not yet occurred in the BIRNM region even after more than
two decades since the mass mortality event.
In other areas of the Caribbean (e.g., Jamaica) populations of sea urchins appear to be showing signs of a recovery
(Edmunds and Carpenter, 2001). In St. Croix, however, the high amounts of macroalgal cover on hardbottom areas and
the small number of D. antillarum survivors (some of which may be relatively isolated from one another) may impede
population recovery in the region. The ecological consequences of such low sea urchin abundance have not been evaluated
for the coral reef ecosystems of BIRNM and surrounding areas. Furthermore, the dynamics of larval connectivity and
other factors affecting recruitment are not well known for the region. However, at several shallow water sites (< 3 m) in
Teague Bay lagoon, sea urchin densities were relatively high and this may be related to water depth and substratum type,
with high densities observed at two colonized bedrock sites. Studies in St. Croix and elsewhere indicate that lagoonal
and sheltered back reef areas may function to promote recovery in D. antillarum populations (Miller et al., 2003; Debrot
and Nagelkerken, 2006) Further investigation is needed to determine the environmental factors that correlate with the
observed spatial patterns in sea urchin occurrence and abundance and to determine if a recovery is occurring. Lessios
(2005) recommended that assessment of sea urchin recovery should be are conducted using consistent survey methods
that also include use of permanent transects due to the highly clustered distributions that are typically observed.
It is likely that 1970s populations were a result of explosive release due to predator removal by the shery and were
ecologically unsustainable thereby facilitating the exceptionally rapid spread of disease throughout the Caribbean.
Table 19. Life history characteristics and vulnerability to shing for three large-bodied species and two smaller-bodied species of
grouper. Vulnerability is based on maximum size and the von Bertalanffy growth parameter (K) used by Cheung et al., 2005. Population
doubling time is based on calculations by Musick et al., 2000 as reported in FishBase.
M. tigris M. venenosa E. striatus C. fulva E. guttatus
Tiger grouper Yellown grouper Nassau grouper Coney Red hind
Max. size (cm) 101 TL 100 TL 122 TL 41 TL 76 TL
Length rst maturity (cm) 46 - 55 TL 51 FL 48 FL 16 FL 25 FL
Age rst maturity (Y) 6.5-9.5 ? ? ? 3
Pop. doubling time (Y) 4.5-14 4.5-14 4.5-14 1.4-4.4 1.4-4.4
Vulnerabilty (index) High-Very High (74.9) Mod.-High (49.7) High-Very High (71.9) Low-Mod. (32.5) Moderate (41.5)
D. antillarum
4 - Discussion
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 66
Biological populations are naturally dynamic, yet questions regarding the optimum densities of sea urchin populations for
a healthy coral reef ecosystem remain unanswered and future changes in sea urchin abundance must be monitored in
order to determine if recovery is occurring and whether management intervention may be required.
Queen conch
Due to overshing in many regions of the Caribbean and Florida, queen conch (Strombus
gigas) has been listed in Annex II of the Cartagena Convention’s Protocol Concerning
Specially Protected Areas and Wildlife (SPAW Protocol) as a species that may be used
on a rational and sustainable basis and that requires protective measures. Because of
this recognition, the United States proposed queen conch for listing in Appendix II of
the Convention on International Trade in Endangered Species of Wild Fauna and Flora
(CITES) in 1992; this proposal was adopted, and queen conch became the rst large-
scale sheries product to be regulated by CITES (see http://www.nmfs.noaa.gov/pr/
species/invertebrates/queenconch.htm). In the U.S. Caribbean, the queen conch shery
is regulated under the auspices of the CFMC.
The coral reef ecosystems of BIRNM and adjacent seascapes support regionally
signicant populations of juvenile and adult queen conch (S. gigas). The large expanse
of seagrasses between Buck Island and St. Croix coastline are key resources supporting
conch populations. However, very little is known about the historical densities to determine if present day populations
are relatively large or small. This is important since according to the NOAA Ofce of Protected Resources, queen conch
abundance is declining throughout the species’s range as a result of overshing. Conch are an important food resource
for the Virgin Islands and are known to be harvested from the Buck Island region, yet the extent and impact of the local
conch shery is undocumented. Inlling of this substantial knowledge gap may require collaboration with local shers to
ascertain levels of exploitation and to manage for ecologically sustainable levels of extraction.
S. gigas looking out of shell
4 - Discussion
References
p. 67
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
References
Anderson, M., H. Lund, E.H. Gladfelter, and M. Davies. 1986. Ecological community type maps and biological community descriptions
for Buck Island Reef National Monument and proposed marine park sites in the British Virgin Islands. U.S. National Park Service and
Virgin Islands Resource Management Cooperative Report. 236 pp.
Bythell, J.C., E.H. Gladfelter, W.B. Gladfelter, K.E. French, and Z. Hillis. 1989. Buck Island Reef Monument- changes in modern reef
community structure. pp. 145-153. In: D.K. Hubbard (ed.). Terrestrial and marine geology of St. Croix, U.S. Virgin Islands. Special
Publication No. 8. Teague Bay, St. Croix, West Indies Laboratory. 213 pp. http://www.aoml.noaa.gov/general/lib/CREWS/Cleo/St.%20
Croix/salt_river164.pdf.
Bythell, J.C., M. Bythell, and E.H. Gladfelter. 1993. Initial results of a long-term coral reef monitoring program: Impact of Hurricane Hugo
at Buck Island Reef National Monument, St. Croix, U.S. Virgin Islands. J. Exp. Mar. Biol. Ecol. 172(1-2): 171-183.
Caribbean Fishery Management Council (CFMC). 1985. Fishery Management Plan, Final Environmental Impact Statement, and Draft
Regulatory Impact Review, for the Shallow-Water Reefsh Fishery of Puerto Rico and the U.S. Virgin islands. Caribbean Fishery
Management Council and National Marine Fisheries Service. 178 pp. http://www.caribbeanfmc.com/SCANNED%20FMPS/REEF%20
FISH/REEF-FISH%20FMP.htm.
Carpenter, R.C. 1988. Mass Mortality of a Caribbean Sea Urchin: Immediate Effects on Community Metabolism and Other Herbivores.
Proceedings of the National Academy of Sciences. 85(2): 511-514.
Cheung, W.W.L., T.J. Pitcher, and D. Pauly, 2005. A fuzzy logic expert system to estimate intrinsic extinction vulnerabilities of marine
shes to shing. Biol. Conserv. 124: 97-111.
Clark, R.D., C.F.G. Jeffrey, K. Woody, Z. Hillis-Starr, and M.E. Monaco. In press. Spatial and temporal patterns of coral bleaching
around Buck Island Reef National Monument, St. Croix, U.S. Virgin Islands. Bull. Mar. Sci.
Clarke, K.R. and R.M. Warwick. 1994. Change in marine communities: an approach to statistical analysis and interpretation, 1st edition.
Plymouth Marine Laboratory, Plymouth. 144 pp.
Clarke, K.R. and R.M. Warwick. 1998. A taxonomic distinctness index and its statistical properties. J. Appl. Ecol. 35: 523-531.
Clarke, K.R. and R.M. Warwick. 1999. The taxonomic distinctness measure of biodiversity: weighting of step lengths between hierarchial
levels. Mar. Ecol. Prog. Ser. 194: 21-29.
Debrot, A.O. and I. Nagelkerken. 2006. Recovery of the long-spined sea urchin Diadema antillarum in Curçao (Netherlands Antilles)
linked to lagoonal and wave sheltered shallow rocky habitats. Bull. Mar. Sci. 79(2):415-424.
Dulvy, N.K. and J.D. Reynolds. 2002. Predicting extinction vulnerability in skates. Conserv. Biol.16 (2): 440-450.
Dulvy, N.K., Y. Sadovy, and J.D. Reynolds. 2003. Extinction vulnerability in marine populations. Fish and Fisheries 4: 25-64.
Edmunds, P.J. and R.C. Carpenter. 2001. Recovery of Diadema reduces macroalgal cover and increases the abundance of juvenile
corals on a Caribbean reef. Proceedings of the National Academy of Science. 98: 5067–5071.
Faith, D.P., P.R. Minchin, and L. Belbin. 1987. Compositional dissimilarity as a robust measure of ecological distance. Vegetatio 68:
57-68.
Gladfelter, E.H. 1980. Aspects of population dynamics and ecological impact of the sea urchin Diadema antillarum. In: E.H. Gladfelter
and W.B. Gladfelter (eds.). Chapter 5: Environmental studies of Buck Island Reef National Monument III, St. Croix, U.S. Virgin Islands.
West Indies Laboratory, Fairleigh Dickinson University. St. Croix, U.S. Virgin Islands. 116 pp.
Gladfelter, W.B., E.H. Gladfelter, R.K. Monahan, J.C. Ogden, and R.F. Dill. 1977. Environmental studies of Buck Island Reef National
Monument. National Park Service Report. Washington, DC. 144 pp. http://www.aoml.noaa.gov/general/lib/CREWS/Cleo/St.%20Croix/
salt_river164.pdf.
Grober-Dunsmore, R., T.K. Frazer, W.J. Lindberg, and J. Beets. 2007. Reef sh and habitat relationships in Caribbean seascapes: the
importance of reef context. Coral Reefs 26(1): 210-216.
Hughes, T.P. 1994. Catastrophes, Phase-Shifts, and Large-Scale Degradation of a Caribbean coral reef. Science 265(5178): 1547-
1551.
Jeffrey, C. and M. Monaco. 2007. Domain (island) wide estimates of queen conch (Strombus gigas) abundance for three U.S.
Caribbean Islands based on habitat-derived densities. SEDAR Report SEDAR14-AW3. Prepared by Biogeography Branch, Center for
Coastal MOnitoring and Assessment, NOAA National Centers for Coastal Ocean Science. Prepared for Southeast Fisheries Sicence
Center, NOAA National Marine Fisheries Service. St. Thomas, U.S. Virgin Islands. 11 pp. http://www.sefsc.noaa.gov/sedar/download/
S14AW03%20conch%20Habitatbasedanalysis.pdf?id=DOCUMENT
Jennings, S., J.D. Reynolds, and N.V.C. Polunin. 1999. Predicting the vulnerability of tropical reef shes to exploitation with phylogenies
and life histories. Conserv. Biol. 13(6): 1466-1475.
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 68
Kendall, M.S., C.R. Kruer, K.R. Buja, J.D. Christensen, M. Finkbeiner, R. Warner, and M.E. Monaco. 2002. Methods used to map the
benthic habitats of Puerto Rico and the U.S. Virgin Islands. NOAA Technical Memorandum 152. Silver Spring, MD. http://ccma.nos.
noaa.gov/ecosystems/coralreef/usvi_pr_mapping.html.
Kendall, M.S., J.D. Christensen, and Z. Hillis-Starr. 2003. Multi-scale data used to analyze the spatial distribution of French grunts,
Haemulon avolineatum, relative to hard and soft bottom in a benthic landscape. Environ. Biol. Fish. 66: 19-26.
Kendall, M.S., T. Battista, and Z. Hillis-Starr. 2004b. Long term expansion of a deep Syringodium liforme meadow in St. Croix, U.S.
Virgin Islands: the portential role of hurricanes in the dispersal of seeds. Aquat. Bot. 78: 15-25.
Kendall, M.S., J.D. Christensen, C. Caldow, M. Coyne, C.F.G. Jeffrey, M.E. Monaco, W.Morrison, and Z. Hillis-Starr . 2004a. The
inuence of bottom type and shelf position on biodiversity of tropical sh inside a recently enlarged marine reserve. Aquatic. Conserv.
Mar. Freshw. Ecosyst. 14: 113-132.
Legendre, P. and L. Legendre. 1998. Numerical Ecology, 2nd English edition. Developments in Environmental Modelling 20 Series .
Elsevier Science Publishing. Amsterdam, The Netherlands. 870 pp.
Lessios, H. 2005. Diadema antillarum populations in Panama twenty years following mass mortality. Coral Reefs 24(1):125-127.
Lessios, H.A. 1998. Mass mortality of Diadema antillarum in the Caribbean: what have we learned? Ann. Rev. Ecol. Syst. 19: 371-
393.
Lessios, H., D.R. Robertson, and J.D. Cubit. 1984. Spread of Diadema mass mortality through the Caribbean. Science 226(4672):
335-337.
Magurran, A.E. 1988. Ecological diversity and its measurement. Princeton University Press. Princeton, NJ. 192 pp.
Mayor, P. 2005. Distribution and abundance of elkhorn coral, Acropora palmata, at Buck Island Reef National Monument in 2004. Buck
Island Reef National Monument, National Park Service, U.S. Department of the Interior. 28 pp.
Menza, C., J. Ault, J. Beets, C. Bohnsack, C. Caldow, J. Christensen, A. Friedlander, C. Jeffrey, M. Kendall, J. Luo, M.E. Monaco, S.
Smith, and K. Woody. 2006. A guide to monitoring reef sh in the National Park Service’s South Florida/Caribbean Network. NOAA
Technical Memorandum NOS NCCOS 39. Silver Spring, MD. 166 pp. http://ccma.nos.noaa.gov/news/feature/FishMonitoring.html.
Miller, J. South Florida / Caribbean Network, National Park Service. St. John, U.S. Virgin Islands. Personal communication.
Miller, R.J., A.J. Adams, N.B. Ogden, J.C. Ogden, and J.P. Ebersole. 2003. Diadema antillarum 17 years after mass mortality: is
recovery beginning on St. Croix? Coral Reefs 22: 181-187.
Musick, J.A., M.M. Harbin, S.A. Berkeley, G.H. Burgess, A.M. Eklund, L. Findley, R.G. Gilmore, J.T. Golden, D.S. Ha, G.R. Huntsman,
J.C. McGovern, S.J. Parker, S.G. Poss, E. Sala, T.W. Schmidt, G.R. Sedberry, H. Weeks, and S.G. Wright. 2000. Marine, estuarine,
and diadromous sh stocks at risk of extinction in North America (exclusive of Pacic salmonids). Fisheries 25: 6-30.
Ogden, J.C., D. Helm, J. Peterson, A. Smith, and S. Weisman (eds.). 1972. An ecological study of Teague Bay Reef, St. Croix, U.S.
Virgin Islands. West Indies Laboratory, St. Croix, U.S. Virgin Islands. Special Publication 1: 50 pp.
Ogden, J.C., D.P. Abbott, and I.A. Abbott. 1973. Studies on the activity and food of the echinoid Diadema antillarum Philippi on a West
Indian patch reef. West Indies Lab, St. Croix, U.S. Virgin Islands. Special Publication 2: 96 pp.
Pittman, S.J., C. Caldow, S.D. Hile, and M.E. Monaco. 2007. Using seascape types to explain the spatial patterns of sh in the
mangroves of SW Puerto Rico. Mar. Ecol. Prog. Ser. 348: 273-284
Rogers, C.S., T. Suchanek, and F. Pecora. 1982. Effects of Hurricanes David and Federic (1979) on shallow Acropora palmata reef
communities: St. Croix, USVI. Bull. Mar. Sci. 32: 532-548.
Rogers, C., V. Garrison, and R. Grober-Dunsmore. 1997. A shy story about hurricanes and herbivory: seven years of research on a
reef in St John, US Virgin Islands. pp. 555-560. In: H.A. Lessios and I.G. Macintyre (eds.). Proceedings of the 8
th
International Coral
Reef Symposium, Vol. 1. Panama City, Panama. 1040 pp.
Rogers, C.S. and J. Beets. 2001. Degradation of marine ecosystems and decline of shery resources in marine protected areas in the
U.S. Virgin Islands. Environ. Conserv. 84: 312-322.
Warwick, R.M. and K.R. Clarke. 1995. New ‘biodiversity’ measures reveal a decrease in taxonomic distinctness with increasing stress.
Mar Ecol. Prog. Ser. 129: 301-305.
References
p. 69
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Appendix A
Figure A1. Map of the East End Marine Park and park zoning.
Source: http://www.stxeastendmarinepark.org/about.htm
Appendix A
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 70
Appendix B
Table B1. USVI nsh landings as a proportion of the total nsh landings
reported for the U.S. Caribbean in 1980. Listed are the most commonly
landed species and species groups. Data from the Caribbean Fisheries
Management Council (CFMC, 1985).
Species/Species group Fish Family
USVI % of
total landings
U.S. Caribbean
Grunts Haemulidae 0.47
Groupers Serranidae 13.91
Goatsh Mullidae 0.99
Parrotsh Scaridae 5.83
Lane snapper (L. synagris) Lutjanidae 0.03
Yellowtail snapper (O.chryusurus) Lutjanidae 2.89
Triggershes Balistidae 29.68
Squirrelshes Holocentridae 4.84
Mutton snapper (L. analis) Lutjanidae 0.13
Other snappers Lutjanidae 1.04
Hogsh Labridae 1.06
Trunksh Ostraciidae 0.08
Appendix B
p. 71
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Appendix C
Appendix C
Table C1. Fish species list and summary data on occurrence, abundance and biomass (2001-2006) for the study region (northeastern
St. Croix).
Family
Species name
Common name
%
occurrence
Total
occurrence
Total
abundance
Mean abundance
Total
biomass, g
Mean biomass, g
(+ SE) (+ SE)
Acanthuridae
Acanthurus bahianus Ocean surgeonsh 59.8 762 8601 6.7 (0.36) 540368.5 423.8 (37.0)
Acanthurus chirurgus Doctorsh 11.8 150 652 0.51 (0.06) 48646.1 38.2 (5.7)
Acanthurus coeruleus Blue tang 46.7 595 8597 6.7 (0.65) 904137.1 709.1 (83.8)
Acanthurus UNK SURGEONFISH sp 0.1 1 1 <0.01 (<0.01) 0.8 <0.01 (<0.01)
Apogonidae
Apogon binotatus Barred cardinalsh 0.1 1 1 <0.01 (<0.01) 7.7 <0.01 (<0.01)
Apogon maculatus Flamesh 0.2 2 2 <0.01 (<0.01) 8.1 <0.01 (<0.01)
Apogon quadrisquamatus Sawcheek cardinalsh 0.5 6 22 0.02 (<0.01) 10.1 <0.01 (<0.01)
Apogon townsendi Belted cardinalsh 0.5 7 16 0.01 (<0.01) 29.0 0.02 (0.02)
Apogon UNK CARDINALFISH sp 0.3 4 7 <0.01 (<0.01) 46.8 0.04 (0.03)
Astrapogon puncticulatus Blackn cardinalsh 0.1 1 1 <0.01 (<0.01) 0.5 <0.01 (<0.01)
Astrapogon stellatus Conchsh 0.2 3 4 <0.01 (<0.01) 2.0 <0.01 (<0.01)
Aulostomidae
Aulostomus maculatus Trumpetsh 6.3 80 107 0.08 (0.01) 9163.9 7.2 (1.0)
Balistidae
Balistes vetula Queen triggersh 8.5 108 208 0.16
(0.02) 175978.6 138.0 (19.6)
Melichthys niger Black durgon 4.4 56 158 0.12 (0.02) 118235.5 92.7 (21.4)
Belonidae
Ablennes hians Flat needlesh 0.2 2 51 0.04 (0.04) 1891.0 1.5 (1.5)
Blenniidae
Ophioblennius macclurei Redlip blenny 9.0 115 368 0.29 (0.04) 1442.1 1.1 (0.18)
Parablennius marmoreus Seaweed blenny 0.1 1 4 <0.01 (<0.01) 1.2 <0.01 (<0.01)
Bothidae
Bothus lunatus Peacock ounder 2.3 29 32 0.03 (<0.01) 5391.4 4.2 (1.5)
Bothus ocellatus Eyed ounder 0.6 8 9 <0.01 (<0.01) 224.7 0.18 (0.10)
Bothus UNK FLOUNDER sp 0.5 6 6 <0.01 (<0.01) 13.4 0.01 (<0.01)
Callionymidae
Paradiplogrammus bairdi Lancer dragonet 1.1 14 20 0.02 (<0.01) 49.2 0.04 (0.02)
Carangidae
Carangoides bartholomaei Yellow jack 0.5 6 22 0.02 (<0.01) 33344.3 26.2 (17.8)
Caranx crysos Blue runner 12.8 163 1064 0.83 (0.12) 369852.4 290.1 (53.4)
Caranx hippos Crevalle jack 0.2 2 2 <0.01 (<0.01) 2218.8 1.7 (1.3)
Caranx latus Horse-Eye jack 0.2 2 7 <0.01 (<0.01) 1822.9 1.4 (1.1)
Carangoides ruber Bar jack 28.5 364
1735 1.4 (0.17) 57872.9 45.4 (8.6)
Caranx UNK JACK sp 0.1 1 1 <0.01 (<0.01) 389.1 0.31 (0.31)
Decapterus macarellus Mackerel scad 2.9 37 2319 1.8 (0.51) 136084.3 106.7 (30.6)
Decapterus UNK SCAD sp 0.3 4 244 0.19 (0.13) 6799.1 5.3 (2.9)
Selar crumenophthalmus Bigeye scad 0.1 1 24 0.02 (0.02) 2182.9 1.7 (1.7)
Chaenopsidae
Acanthemblemaria aspera Roughhead blenny 0.2 2 3 <0.01 (<0.01) 0.6 <0.01 (<0.01)
Acanthemblemaria maria Secretary blenny 0.2 2 2 <0.01 (<0.01) 0.4 <0.01 (<0.01)
Acanthemblemaria spinosa Spinyhead blenny 0.1 1 1 <0.01 (<0.01) 0.2 <0.01 (<0.01)
Acanthemblemaria UNK TUBE BLENNY sp 0.1 1 6 <0.01 (<0.01) 1.2 <0.01 (<0.01)
Chaenopsis limbaughi Yellowface pikeblenny 2.0 26 123 0.10 (0.03) 266.2 0.21 (0.09)
Chaenopsis ocellata Bluethroat pikeblenny 2.0 26 54 0.04 (0.01) 241.0 0.19 (0.06)
Chaenopsis UNK PIKEBLENNY sp 0.4 5 8 <0.01 (<0.01) 21.3 0.02 (<0.01)
Emblemaria pandionis Sailn blenny 0.3 4 7 <0.01 (<0.01) 1.4 <0.01 (<0.01)
Chaetodontidae
Chaetodon capistratus Foureye butterysh
15.4 196 385 0.30 (0.02) 9481.2 7.4 (1.2)
Chaetodon ocellatus Spotn butterysh 0.7 9 14 0.01 (<0.01) 283.1 0.22 (0.14)
Chaetodon sedentarius Reef butterysh 1.2 15 22 0.02 (<0.01) 624.8 0.49 (0.17)
Chaetodon striatus Banded butterysh 7.0 89 147 0.12 (0.01) 4940.0 3.9 (0.79)
Prognathodes aculeatus Longsnout butterysh 0.5 7 9 <0.01 (<0.01) 151.4 0.12 (0.06)
Cirrhitidae
Amblycirrhitus pinos Redspotted hawksh 1.3 17 22 0.02 (<0.01) 24.8 0.02 (<0.01)
Clupeidae
Clupeidae UNK HERRING sp 0.1 1 70 0.05 (0.05) 3.4 <0.01 (<0.01)
Jenkinsia UNK HERRING sp 0.1 1 50 0.04 (0.04) 2.4 <0.01 (<0.01)
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 72
Table C1 cont...
Family
Species name
Common Name
%
occurrence
Total
occurrence
Total
abundance
Mean abundance
Total
biomass, g
Mean biomass, g
(+ SE) (+ SE)
Congridae
Heteroconger longissimus Brown garden eel 1.5 19 1711 1.3 (0.66) 46379.3 36.4 (18.8)
Dactylopteridae
Dactylopterus volitans Flying gurnard 0.1 1 3 <0.01 (<0.01) 330.6 0.26 (0.26)
Dasyatidae
Dasyatis americana Southern stingray 2.2 28 37 0.03 (<0.01) 10064.2 7.9 (2.3)
Diodontidae
Chilomycterus antennatus Bridled burrsh 0.1 1 1 <0.01 (<0.01) 395.5 0.31 (0.31)
Diodon holocanthus Balloonsh 0.7 9 10 <0.01 (<0.01) 1196.3 0.94 (0.44)
Diodon hystrix Porcupinesh 0.7 9 10 <0.01 (<0.01) 9898.9 7.8 (3.9)
Echeneidae
Echeneis naucrates Sharksucker 0.7 9 11 <0.01 (<0.01) 8529.7 6.7 (2.5)
Engraulidae
Engraulidae UNK 0.1 1 1 <0.01 (<0.01) 0.2 <0.01 (<0.01)
Gerreidae
Eucinostomus gula Silver jenny 0.2 3 10 <0.01 (<0.01) 616.2 0.48 (0.32)
Eucinostomus melanopterus Flagn mojarra 0.2 2 6 <0.01 (<0.01) 115.1 0.09 (0.07)
Eucinostomus UNK MOJARRA sp 0.1 1 13 0.01 (0.01) 668.4 0.52 (0.52)
Gerres cinereus Yellown mojarra 6.9 88 162 0.13 (0.02) 7597.4 6.0 (1.1)
Ginglymostomatidae
Ginglymostoma cirratum Nurse shark 0.5 7
9 <0.01 (<0.01) 37663.9 29.5 (12.1)
Gobiidae
Coryphopterus dicrus Colon goby 1.3 16 19 0.01 (<0.01) 12.5 <0.01 (<0.01)
Coryphopterus eidolon Pallid goby 0.1 1 1 <0.01 (<0.01) 0.7 <0.01 (<0.01)
Coryphopterus glaucofraenum Bridled goby 23.7 302 1670 1.3 (0.11) 1490.4 1.2 (0.12)
Coryphopterus lipernes Peppermint goby 0.2 2 5 <0.01 (<0.01) 3.3 <0.01 (<0.01)
Coryphopterus personatus/hyalinus Masked/Glass goby 1.7 22 639 0.50 (0.28) 418.8 0.33 (0.19)
Ctenogobius saepepallens Dash goby 1.1 14 68 0.05 (0.03) 44.6 0.03 (0.02)
Elacatinus chancei Shortstripe goby 0.5 7 20 0.02 (<0.01) 5.0 <0.01 (<0.01)
Elacatinus evelynae Sharknose goby 6.4 81 143 0.11 (0.02) 35.8 0.03 (<0.01)
Elacatinus multifasciatus Greenbanded goby 0.1 1 1 <0.01 (<0.01) 0.3 <0.01 (<0.01)
Elacatinus prochilos Broadstripe goby 1.8 23 45 0.04 (0.01) 11.3 <0.01 (<0.01)
Elacatinus saucrus Leopard goby 0.1 1 1 <0.01 (<0.01) 0.3 <0.01 (<0.01)
Elacatinus UNK GOBY sp 0.1 1 1 <0.01 (<0.01) 0.3 <0.01 (<0.01)
Gnatholepis thompsoni Goldspot goby 14.6 186 690 0.54
(0.07) 298.7 0.23 (0.07)
Gobiidae UNK GOBIES 0.2 2 2 <0.01 (<0.01) 1.3 <0.01 (<0.01)
Microgobius carri Seminole goby 0.1 1 1 <0.01 (<0.01) 0.2 <0.01 (<0.01)
Nes longus Orangespotted goby 0.4 5 8 <0.01 (<0.01) 68.9 0.05 (0.04)
Priolepis hipoliti Rusty goby 0.1 1 1 <0.01 (<0.01) 0.4 <0.01 (<0.01)
Grammatidae
Gramma loreto Fairy basslet 4.0 51 154 0.12 (0.02) 131.7 0.10 (0.03)
Haemulidae
Anisotremus surinamensis Black margate 0.2 2 2 <0.01 (<0.01) 2048.1 1.6 (1.3)
Anisotremus virginicus Porksh 0.2 3 3 <0.01 (<0.01) 850.7 0.67 (0.38)
Haemulon album White margate 0.2 2 5 <0.01 (<0.01) 2840.8 2.2 (2.1)
Haemulon aurolineatum Tomtate 4.2 53 2633 2.1 (1.1) 109702.8 86.0 (32.2)
Haemulon carbonarium Caesar grunt 1.8 23 130 0.10 (0.03) 22569.4 17.7 (6.5)
Haemulon chrysargyreum Smallmouth grunt 1.2 15 201 0.16 (0.07) 5190.1 4.1 (2.1)
Haemulon avolineatum French grunt 28.0 357 1944 1.5 (0.26) 107188.4 84.1 (7.5)
Haemulon macrostomum Spanish grunt 1.0 13 54 0.04 (0.02) 2954.0
2.3 (1.5)
Haemulon melanurum Cottonwick 0.5 6 209 0.16 (0.16) 15919.8 12.5 (11.5)
Haemulon parra Sailors choice 0.2 2 3 <0.01 (<0.01) 377.4 0.30 (0.27)
Haemulon plumierii White grunt 3.5 45 468 0.37 (0.26) 64316.4 50.4 (17.7)
Haemulon sciurus Bluestriped grunt 4.3 55 199 0.16 (0.04) 42646.6 33.4 (8.4)
Haemulon striatum Striped grunt 0.1 1 1 <0.01 (<0.01) 43.9 0.03 (0.03)
Haemulon UNK GRUNT sp 4.1 52 3419 2.7 (0.78) 9538.7 7.5 (3.8)
Holocentridae
Holocentrus adscensionis Squirrelsh 11.8 150 349 0.27 (0.05) 48639.6 38.1 (6.3)
Holocentrus rufus Longspine squirrelsh 25.2 321 574 0.45 (0.03) 65191.6 51.1 (4.2)
Myripristis jacobus Blackbar soldiersh 3.5 45 127 0.10 (0.04) 13899.2 10.9 (4.3)
Sargocentron vexillarium Dusky squirrelsh 0.6 8 13 0.01 (<0.01) 448.2 0.35 (0.16)
Appendix C
p. 73
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Table C1 cont...
Family
Species name Common name
%
occurrence
Total
occurrence
Total
abundance
Mean abundance
Total
biomass, g
Mean biomass, g
(+ SE) (+ SE)
Inermiidae
Inermia vittata Boga 0.5 7 490 0.38 (0.19) 6365.2 5.0 (2.9)
Kyphosidae
Kyphosus sectator Chub (Bermuda/Yellow) 0.5 6 20 0.02 (<0.01) 10739.8 8.4 (4.6)
Labridae
Bodianus rufus Spanish hogsh 6.4 81 149 0.12 (0.02) 14514.0 11.4 (2.1)
Clepticus parrae Creole wrasse 3.5 45 2160 1.7 (0.67) 65470.5 51.3 (17.7)
Halichoeres bivittatus Slippery dick 73.2 933 24752 19.4 (1.8) 97962.0 76.8 (5.8)
Halichoeres cyanocephalus Yellowcheek wrasse 0.7 9 9 <0.01 (<0.01) 656.0 0.51 (0.29)
Halichoeres garnoti Yellowhead wrasse 46.0 586 4658 3.7 (0.19) 34689.8 27.2 (1.8)
Halichoeres maculipinna Clown wrasse 33.9 432 2176 1.7 (0.11) 10193.3 8.0 (0.78)
Halichoeres pictus Rainbow wrasse 1.8 23 164 0.13 (0.04) 483.3 0.38 (0.12)
Halichoeres poeyi Blackear wrasse 16.7 213 650 0.51 (0.05) 3592.1 2.8 (0.33)
Halichoeres radiatus Puddingwife 21.1 269 503 0.39 (0.03) 5361.9 4.2 (1.2)
Lachnolaimus maximus Hogsh 0.4 5 6 <0.01 (<0.01) 115.5 0.09 (0.05)
Thalassoma bifasciatum Bluehead wrasse 60.9 777 32001 25.1 (1.2) 46055.3 36.1
(1.7)
Xyrichtys martinicensis Rosy razorsh 13.6 174 2582 2.0 (0.34) 12572.0 9.9 (2.4)
Xyrichtys novacula Pearly razorFish 0.3 4 6 <0.01 (<0.01) 40.6 0.03 (0.02)
Xyrichtys splendens Green razorsh 11.8 150 598 0.47 (0.08) 3050.0 2.4 (0.54)
Xyrichtys UNK RAZORFISH sp 0.2 3 23 0.02 (0.01) 6.2 <0.01 (<0.01)
Labrisomidae
Malacoctenus aurolineatus Goldline blenny 0.9 11 23 0.02 (<0.01) 15.6 0.01 (<0.01)
Malacoctenus boehlkei Diamond blenny 0.1 1 1 <0.01 (<0.01) 0.2 <0.01 (<0.01)
Malacoctenus gilli Dusky blenny 0.4 5 9 <0.01 (<0.01) 2.2 <0.01 (<0.01)
Malacoctenus macropus Rosy blenny 4.7 60 129 0.10 (0.02) 87.4 0.07 (0.01)
Malacoctenus triangulatus Saddled blenny 10.4 132 265 0.21 (0.02) 77.7 0.06 (0.01)
Malacoctenus UNK LABRISOMIDS 0.5 6 8 <0.01 (<0.01) 2.9 <0.01 (<0.01)
Malacoctenus versicolor Barn blenny 1.1 14 14 0.01 (<0.01) 10.4 <0.01 (<0.01)
Lutjanidae
Lutjanus analis Mutton snapper 3.2 41 49 0.04 (<0.01) 81134.4 63.6 (15.1)
Lutjanus apodus Schoolmaster 3.3 42 81 0.06 (0.02) 30045.2 23.6
(6.0)
Lutjanus buccanella Blackn snapper 0.1 1 1 <0.01 (<0.01) 1.5 <0.01 (<0.01)
Lutjanus griseus Gray snapper 1.2 15 53 0.04 (0.02) 12823.5 10.1 (5.3)
Lutjanus jocu Dog snapper 0.2 2 2 <0.01 (<0.01) 7404.6 5.8 (5.4)
Lutjanus mahogoni Mahogany snapper 3.4 43 129 0.10 (0.03) 13467.7 10.6 (2.4)
Lutjanus synagris Lane snapper 2.3 29 60 0.05 (0.01) 5485.8 4.3 (1.2)
Lutjanus UNK SNAPPER sp 0.1 1 1 <0.01 (<0.01) 0.4 <0.01 (<0.01)
Ocyurus chrysurus Yellowtail snapper 20.4 260 741 0.58 (0.06) 73934.6 58.0 (7.0)
Malacanthidae
Malacanthus plumieri Sand tilesh 10.7 137 232 0.18 (0.02) 66683.8 52.3 (6.3)
Megalopidae
Megalops atlanticus Tarpon 0.1 1 1 <0.01 (<0.01) 37379.9 29.3 (29.3)
Microdesmidae
Ptereleotris helenae Hovering goby 0.8 10 29 0.02 (0.01) 39.5 0.03 (0.02)
Monacanthidae
Aluterus scriptus Scrawled lesh 0.2 3 3 <0.01 (<0.01) 153.5 0.12 (0.11)
Cantherhines macrocerus Whitespotted lesh 0.4 5 6 <0.01 (<0.01) 734.9 0.58 (0.35)
Cantherhines pullus Orangespotted lesh 1.7 22 26 0.02 (<0.01) 1157.9
0.91 (0.36)
Monacanthus ciliatus Fringed lesh 1.1 14 14 0.01 (<0.01) 23.2 0.02 (<0.01)
Monacanthus tuckeri Slender lesh 1.5 19 24 0.02 (<0.01) 44.4 0.03 (0.01)
Monacanthus UNK FILEFISH sp 0.2 3 3 <0.01 (<0.01) 1.5 <0.01 (<0.01)
Mullidae
Mulloidichthys martinicus Yellow goatsh 7.3 93 344 0.27 (0.05) 69054.0 54.2 (16.5)
Pseudupeneus maculatus Spotted goatsh 23.3 297 915 0.72 (0.08) 72224.9 56.6 (6.8)
Muraenidae
Enchelycore nigricans Viper moray 0.1 1 1 <0.01 (<0.01) 253.1 0.20 (0.20)
Gymnothorax funebris Green moray 0.1 1 1 <0.01 (<0.01) 52.9 0.04 (0.04)
Gymnothorax miliaris Goldentail moray 0.2 2 2 <0.01 (<0.01) 6.7 <0.01 (<0.01)
Gymnothorax moringa Spotted moray 0.2 3 3 <0.01 (<0.01) 2063.0 1.6 (1.2)
Gymnothorax UNK MORAY EEL sp 0.1 1 1 <0.01 (<0.01) 0.6 <0.01 (<0.01)
Gymnothorax vicinus Purplemouth moray 0.2 2 2 <0.01 (<0.01) 373.1 0.29 (0.22)
Muraenidae UNK 0.1 1 1 <0.01 (<0.01) 18.6 0.01 (0.01)
Appendix C
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 74
Table C1 cont...
Family
Species name
Common name
%
occurrence
Total
occurrence
Total
abundance
Mean abundance
Total
biomass, g
Mean biomass, g
(+ SE) (+ SE)
Myliobatidae
Aetobatus narinari Spotted eagle ray 0.2 3 3 <0.01 (<0.01) 26717.4 21.0 (16.1)
Ogcocephalidae
Ogcocephalus nasutus Shortnose batsh 0.1 1 1 <0.01 (<0.01) 0.5 <0.01 (<0.01)
Ophichthidae
Myrichthys breviceps Sharptail eel 0.1 1 1 <0.01 (<0.01) 1000.0 0.78 (0.78)
Myrichthys ocellatus Goldspotted eel 0.3 4 4 <0.01 (<0.01) 130.5 0.10 (0.06)
Ophichthus ophis Spotted Snake eel 0.1 1 2 <0.01 (<0.01) 42.4 0.03 (0.03)
Opistognathidae
Opistognathus aurifrons Yellowhead jawsh 3.5 44 145 0.11 (0.03) 710.6 0.56 (0.13)
Opistognathus macrognathus Banded jawsh 0.3 4 7 <0.01 (<0.01) 16.2 0.01 (<0.01)
Ostraciidae
Acanthostracion polygonius Honeycomb cowsh 0.5 6 7 <0.01 (<0.01) 1620.8 1.3 (0.65)
Acanthostracion quadricornis Scrawled cowsh 0.2 2 3 <0.01 (<0.01) 199.8 0.16 (0.14)
Lactophrys bicaudalis Spotted trunksh 0.5 7 7 <0.01 (<0.01) 1566.0 1.2 (0.62)
Lactophrys trigonus Trunksh 1.1 14 15 0.01 (<0.01) 4200.4 3.3 (1.2)
Lactophrys triqueter Smooth trunksh 4.1 52 55 0.04 (<0.01) 7529.1 5.9 (1.1)
Paralichthyidae
Syacium UNK SAND FLOUNDER sp 0.2 3 3
<0.01 (<0.01) 113.5 0.09 (0.08)
Pempheridae
Pempheris schomburgkii Glassy sweeper 0.3 4 40 0.03 (0.02) 609.0 0.48 (0.30)
Pomacanthidae
Holacanthus ciliaris queen angelsh 0.6 8 10 <0.01 (<0.01) 2662.5 2.1 (1.3)
Holacanthus tricolor Rock beauty 6.0 76 94 0.07 (<0.01) 9185.6 7.2 (1.5)
Pomacanthus arcuatus Gray angelsh 1.0 13 16 0.01 (<0.01) 8538.6 6.7 (4.5)
Pomacanthus paru French angelsh 2.5 32 45 0.04 (<0.01) 15188.9 11.9 (3.9)
Pomacentridae
Abudefduf saxatilis Sergeant major 5.2 66 202 0.16 (0.03) 7497.3 5.9 (1.5)
Abudefduf taurus Night sergeant 0.2 2 2 <0.01 (<0.01) 256.9 0.20 (0.14)
Chromis cyanea Blue chromis 17.3 220 3321 2.6 (0.30) 18487.0 14.5 (2.0)
Chromis multilineata Brown chromis 6.4 81 812 0.64 (0.11) 7800.1 6.1 (1.4)
Microspathodon chrysurus Yellowtail damselsh 21.3 272 1158 0.91 (0.08) 77577.9 60.8 (6.8)
Stegastes adustus Dusky damselsh 16.8 214 1544 1.2 (0.12) 11375.4 8.9 (1.1)
Stegastes diencaeus Longn damselsh 20.7 264 1870 1.5 (0.13) 18510.1 14.5 (1.8)
Stegastes leucostictus Beaugregory 29.6 378 3991 3.1
(0.25) 12877.4 10.1 (0.86)
Stegastes partitus Bicolor damselsh 48.9 624 10202 8.0 (0.44) 17761.2 13.9 (1.2)
Stegastes planifrons Threespot damselsh 15.9 203 1221 0.96 (0.10) 12604.7 9.9 (1.1)
Stegastes variabilis Cocoa damselsh 8.9 114 343 0.27 (0.04) 1788.7 1.4 (0.33)
Priacanthidae
Heteropriacanthus cruentatus Glasseye snapper 0.2 2 2 <0.01 (<0.01) 314.9 0.25 (0.19)
Scaridae
Cryptotomus roseus Bluelip parrotsh 18.1 231 1968 1.5 (0.17) 13233.5 10.4 (1.6)
Scarus guacamaia Rainbow parrotsh 0.3 4 14 0.01 (<0.01) 3907.1 3.1 (1.8)
Scarus iseri Striped parrotsh 34.0 434 4761 3.7 (0.28) 74268.8 58.3 (5.8)
Scarus taeniopterus Princess parrotsh 24.9 318 1713 1.3 (0.10) 72301.0 56.7 (5.3)
Scarus UNK PARROTFISH sp 0.5 6 103 0.08 (0.04) 466.9 0.37 (0.22)
Scarus vetula Queen parrotsh 13.4 171 489 0.38 (0.04) 133281.4 104.5 (12.0)
Sparisoma atomarium Greenblotch parrotsh 13.7 175 845 0.66 (0.08) 1900.2 1.5 (0.46)
Sparisoma aurofrenatum Redband parrotsh 52.5 669 4887 3.8 (0.18) 240212.9 188.4 (10.0)
Sparisoma chrysopterum Redtail parrotsh 5.5 70 118 0.09 (0.01)
16052.8 12.6 (2.6)
Sparisoma radians Bucktooth parrotsh 17.5 223 1181 0.93 (0.11) 3764.1 3.0 (0.61)
Sparisoma rubripinne Yellowtail parrotsh 6.8 87 198 0.16 (0.03) 36043.3 28.3 (6.1)
Sparisoma UNK PARROTFISH Genus 0.4 5 7 <0.01 (<0.01) 9.9 <0.01 (<0.01)
Sparisoma viride Stoplight parrotsh 36.1 460 2456 1.9 (0.11) 331900.0 260.3 (24.1)
Sciaenidae
Equetus lanceolatus Jackknife-sh 0.1 1 1 <0.01 (<0.01) 2.5 <0.01 (<0.01)
Equetus punctatus Spotted drum 0.4 5 5 <0.01 (<0.01) 1078.1 0.85 (0.59)
Pareques acuminatus Highhat 0.3 4 8 <0.01 (<0.01) 2.3 <0.01 (<0.01)
Scombridae
Scomberomorus regalis Cero 0.9 11 11 <0.01 (<0.01) 13666.4 10.7 (4.2)
Appendix C
p. 75
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Table C1 cont...
Family
Species name
Common name
%
occurrence
Total
occurrence
Total
abundance
Mean abundance
Total
biomass, g
Mean biomass, g
(+ SE) (+ SE)
Scorpaenidae
Scorpaena plumieri Spotted scorpionsh 0.3 4 4 <0.01 (<0.01) 1015.8 0.80 (0.44)
Scorpaena UNK SCORPIONFISH sp 0.2 2 2 <0.01 (<0.01) 113.6 0.09 (0.09)
Serranidae
Alphestes afer Mutton hamlet 0.3 4 4 <0.01 (<0.01) 896.1 0.70 (0.50)
Cephalopholis cruentata Graysby 4.2 53 81 0.06 (0.01) 9010.6 7.1 (1.3)
Cephalopholis fulva Coney 32.4 413 1391 1.1 (0.07) 193131.3 151.5 (15.0)
Epinephelus adscensionis Rock hind 0.4 5 7 <0.01 (<0.01) 6562.3 5.1 (4.2)
Epinephelus guttatus Red hind 18.1 231 379 0.30 (0.02) 89106.8 69.9 (6.9)
Epinephelus striatus Nassau grouper 0.2 2 3 <0.01 (<0.01) 1267.8 0.99 (0.81)
Hypoplectrus chlorurus Yellowtail hamlet 1.2 15 19 0.01 (<0.01) 235.5 0.18 (0.06)
Hypoplectrus guttavarius Shy hamlet 0.1 1 1 <0.01 (<0.01) 4.1 <0.01 (<0.01)
Hypoplectrus indigo Indigo hamlet 0.1 1 1 <0.01 (<0.01) 19.5 0.02 (0.02)
Hypoplectrus nigricans Black hamlet 1.1 14 16 0.01 (<0.01) 146.5 0.11 (0.05)
Hypoplectrus puella Barred hamlet 1.4 18 31 0.02 (<0.01) 234.7 0.18
(0.06)
Hypoplectrus unicolor Butter hamlet 0.7 9 12 <0.01 (<0.01) 88.7 0.07 (0.03)
Hypoplectrus UNK HAMLET sp 0.4 5 5 <0.01 (<0.01) 43.5 0.03 (0.02)
Mycteroperca tigris Tiger grouper 0.1 1 1 <0.01 (<0.01) 2200.4 1.7 (1.7)
Mycteroperca venenosa Yellown grouper 0.1 1 3 <0.01 (<0.01) 684.7 0.54 (0.54)
Rypticus saponaceus Greater soapsh 0.1 1 1 <0.01 (<0.01) 29.1 0.02 (0.02)
Serranus baldwini Lantern bass 3.7 47 112 0.09 (0.02) 187.2 0.15 (0.04)
Serranus tabacarius Tobaccosh 3.5 45 75 0.06 (0.01) 796.5 0.62 (0.20)
Serranus tigrinus Harlequin bass 27.4 349 713 0.56 (0.03) 5553.1 4.4 (0.59)
Serranus tortugarum Chalk bass 0.2 3 9 <0.01 (<0.01) 3.2 <0.01 (<0.01)
Sparidae
Calamus calamus Saucereye porgy 0.1 1 2 <0.01 (<0.01) 101.4 0.08 (0.08)
Sphyraenidae
Sphyraena barracuda Great barracuda 4.5 58 61 0.05 (<0.01) 263227.7 206.5 (37.9)
Sphyraena picudilla Southern sennet 0.1 1 300 0.24 (0.24) 344915.6 270.5 (270.5)
Syngnathidae
Acentronura dendritica Pipehorse 0.1 1 1 <0.01 (<0.01) 0.2 <0.01
(<0.01)
Cosmocampus elucens Shortn pipesh 0.2 3 5 <0.01 (<0.01) 1.5 <0.01 (<0.01)
Hippocampus reidi Longsnout seahorse 0.1 1 1 <0.01 (<0.01) 0.6 <0.01 (<0.01)
Hippocampus UNK PIPEFISH sp 0.1 1 2 <0.01 (<0.01) 1.3 <0.01 (<0.01)
Synodontidae
Synodus intermedius Sand diver 2.9 37 41 0.03 (<0.01) 3906.8 3.1 (0.89)
Tetraodontidae
Canthigaster rostrata Sharpnose puffer 15.1 192 287 0.23 (0.02) 1059.6 0.83 (0.12)
Sphoeroides spengleri Bandtail puffer 2.5 32 36 0.03 (<0.01) 181.3 0.14 (0.04)
Sphoeroides testudineus Checkered puffer 0.2 2 5 <0.01 (<0.01) 70.4 0.06 (0.04)
Appendix C
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 76
Appendix D
a)
b)
d)
c)
Figure D1. Spatial distributions of juvenile and adult: (a) bluehead wrasse (T. bifasciatum), (b) queen triggersh (B. vetula), (c) rock
beauty (H. tricolor) and (d) slippery dick (H. bivittatus) in northeastern St. Croix.
Appendix D
p. 77
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Figure D2. Spatial distributions of juvenile and adult (a) princess parrotsh (S. taeniopterus) and (b) stoplight parrotsh (S. viride) in
northeastern St. Croix.
a)
b)
Appendix D
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 78
Appendix D
e)
a)
c)
b)
d)
Figure D3. Spatial distributions of juvenile and adult: (a) threespot damselsh (Stegastes planifrons), (b) foureye butterysh (Chaetodon
capistratus), (c) spotn butterysh (Chaetodon ocellatus), (d) banded butterysh (Chaetodon striatus) and (e) great barracuda
(Sphyraena barracuda) in northeastern St. Croix.
p. 79
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
Appendix E
Appendix E
Fish density / 100 m
2
Inside BIRNM
a)
Grunt biomass (g/100 m
2
)
Outside BIRNM
c)
S. iseri density (100 m
2
)
Outside BIRNM
e)
Parrotsh biomass (g/100 m
2
)
Inside BIRNM
b)
H. sciurus biomass (g/100 m
2
)
Outside BIRNM
d)
S. iseri biomass (g/100 m
2
)
Outside BIRNM
f)
Figure E1. Raw census data grouped by year of survey for sh metrics that exhibited an increase or decline every year over the study
period (2003-2006). The horizontal blue line connects the mean (+ SE) for each year. (a) Fish density (all species) increased gradu-
ally inside BIRNM, with 2005 and 2006 densities signicantly higher than 2003; (b) Mean parrotsh biomass increased inside BIRNM
from 2003 to 2006, but with no signicant difference between years; (c) grunt biomass decreased each year outside BIRNM, with 2005
and 2006 biomass signicantly lower than 2003; (d) bluestriped grunt (H. sciurus) biomass decreases each year outside BIRNM, with
2005 and 2006 biomass signicantly lower than 2003; (e) striped parrotsh (S. iseri) density decreased each year outside BIRNM, with
2005 and 2006 biomass lower than 2003; and (f) striped parrotsh (S. iseri) biomass decreased each year outside BIRNM, with 2006
biomass signicantly lower than 2003 and 2004.
Fish assemblages and benthic habitats of the Buck Island Reef National Monument and the surrounding seascape
p. 80
Acknowledgements
Acknowledgements
Many thanks to our partners in the eld including the many NPS scientists that contributed to data collection and logistical
support enabling many successful missions over six years and to the NOAA Coral Reef Conservation Program, NCCOS
CSCOR and NCCOS CCMA for funding the Biogeography Branch’s CREM project. A special thanks to Mr. John Christensen
on the initial formulation of sample design and data analysis in support of CREM. In addition, we thank our colleagues at
the U.S. Virgin Islands Department of Planning and Natural Resources for their contributions to the monitoring study.
United States Department of Commerce
Carlos M Gutierrez
Secretary
National Oceanic and Atmospheric Administration
Vice Admiral Conrad C Lautenbacher, Jr. USN (Ret.)
Under Secretary of Commerce for Ocean and Atmospheres
National Ocean Service
Jack H Dunnigan
Assistant Administrator
