bejarano_rodriguez_2013_phd_mesophotic_fishes

Deep Reef Fishes off La Parguera Insular Slope, Puerto Rico,
and their Connectivity with Shallow Reefs
By
Ivonne Bejarano Rodríguez
A dissertation submitted in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY IN MARINE SCIENCES
BIOLOGICAL OCEANOGRAPHY
UNIVERSITY OF PUERTO RICO
MAYAGÜEZ CAMPUS
2013
Approved by:
_______________________________ ____________________
Jorge R. García-Sais, PhD Date
Member, Graduate Committee
________________________________ ____________________
Paul Yoshioka, PhD Date
________________________________ ____________________
Alberto Sabat, PhD Date
________________________________ ____________________
Richard S. Appeldoorn, PhD Date
President, Graduate Committee
_______________________________ ____________________
John Kubaryk, PhD Date
Chairperson of the Department of Marine Sciences
______________________________ ____________________
Kurt Grove, PhD Date
Representative of Graduate Studies

ii
Abstract
This dissertation characterizes the fish community associated with Mesophotic
Coral Ecosystems (MCEs) of the La Parguera shelf-slope between 2007 and 2011 using
rebreather trimix technical diving. Fishes were identified, counted and lengths estimated
within belt transects (30 m2
) and roving surveys at 30, 40, 50, 60 and 70 m depth.
Vertical transects from 70 to 30 m depth helped determine depth distribution ranges.
The MCE fish assemblage (40-70 m) was distinct from shallow areas (30 m), with
taxonomic composition, abundance and the proportion of trophic guilds varying with
increasing depth. Most fishes at MCE were primarily shallow species, but others were
restricted to mesophotic depths. An additional 15 species were added to those previously
classified as indicator species of mesophotic areas in Puerto Rico. Fish abundance and
species richness within MCEs were high. A total of 103 species were identified. The
dominant trophic guild within MCEs were the zooplanktivores, while herbivores
dominated shallow reefs. Both herbivores and zooplanktivores varied markedly and
inversely to depth. The largest changes within the mesophotic fish community along the
depth gradient occurred at 60 m, similar to that reported for algae and corals, and seem to
represent both a response to reduced light and variations in herbivory. This study
represents the first quantitative in situ descriptions of fishes inhabiting MCEs at depths of
50–70 m in Puerto Rico and highlights the role of MCEs as valuable habitats for reef
fishes.
This study examined the connectivity between shallow and mesophotic depths as
expressed by the distribution and movement of fishes, and explicitly tested if MCEs
represent an additional habitat and potential refugia for shallow reef fishes, particularly

iii
for commercially targeted species. MCEs along the La Parguera shelf-edge are subject to
relatively more stable environmental conditions and a much lower impact from fisheries
than shallow reefs. Seventy-eight shallow species were present at MCEs, and six different
potential patterns of connectivity between shallow and mesophotic habitats are reported
here based on the variability in the composition, relative abundance and presence of
juveniles. Significant connectivity occured in both directions and for some ontogenetic
migrators presence within MCEs appeared to be dependent on shallow production.
Thirty-seven species were of fisheries value, and most of those are also species of
conservation concern as they were once common in shallow reefs but have markedly
declined in abundance in the area in the last 30 years likely due to high historical fishing
pressure (e.g., sharks, large groupers, snappers, parrotfish). Because connectivity allows
the dispersal of both larvae and adults, we believe that the present day spawning stock of
some species (e.g., black grouper, dog snapper) may be made up for the most part of
remnant populations within deep MCEs. This has important economic and ecological
implications, and therefore fish populations in MCEs need to be protected when
managing heavily exploited fishes, and this will enhance reef system resilience and
stability.
Lastly, this study investigated the effect of topographic complexity on the reef
fish assemblages with MCEs by assessing the association between different complexity
measurements (e.g., habitat relief, chain rugosity, slope) and fish community structure.
Statistical tests and nonmetric multidimensional scaling (NMS) identified topographic
complexity as an important parameter affecting ecological processes on MCEs, as
reflected by changes in the composition and abundance of reef fishes. Higher overall fish

iv
abundance and species richness were found at high complexity sites, as well as higher
abundance of zooplanktivores, piscivores, and large bodied vulnerable and commercially
important fishes. Gross relief characterization was the complexity variable that better
related with changes in fish assemblages at both species-specific and community levels.
Because fishes in complex MCE sites are more diverse and abundant, and include several
large bodied vulnerable commercial fishes that have virtually disappeared from shallow
reefs in the area, these sites should be considered as prime areas targeted by fisheries
management and coral reef conservation programs.
This dissertation increases our understanding of reef fish ecology, including
commercial species, and highlights the role of MCEs as valuable habitats for reef fishes.
The results emphasize the importance of incorporating the composition and distribution
of the MCE fish community when planning for the spatial management of coral reef
resources.

v
Resumen
Esta disertación caracteriza los peces asociados a los Ecosistemas Mesofóticos
Coralinos (MCEs) del veril de La Parguera, entre los años 2007 y 2011, utilizando buceo
técnico de rebreather con mezcla de gases trimix. Los peces fueron identificados y
contados, y se estimaron sus longitudes dentro de transectos (30 m2
) y otros censos
visuales a 30, 40, 50, 60 y 70 m de profundidad. Se hicieron también transectos verticales
desde los 70 hasta los 30 m de profundidad para ayudar a determinar la distribución de
las especies al aumentar la profundidad.
La comunidad de peces en los MCEs (40-70 m) fue diferente a la de aguas
someras (30 m), variando en su composición taxonómica, en su abundancia y en la
proporción de grupos tróficos al aumentar la profundidad. La mayoría de los peces en los
MCEs fueron principalmente especies someras, sin embargo otras especies estuvieron
restringidas a las profundidades mesofóticas. Quince especies fueron añadidas al grupo
de peces previamente clasificados como “indicadores de hábitats mesofóticos” en Puerto
Rico. La abundancia y riqueza de peces en los MCEs fue alta. Se identificaron en total
103 especies. El grupo trófico dominante en los MCEs fue el de los zooplanctívoros,
mientras en los arrecifes someros dominaron los herbívoros. Ambos grupos tróficos
variaron marcada y opuestamente al aumentar la profundidad. Los mayores cambios en la
comunidad de peces dentro de los MCEs a lo largo del gradiente de profundidad
ocurrieron a los 60 m, similar a los cambios reportados para las algas y los corales, y
aparentemente ocurren en respuesta a la reducción de la luz, y a cambios en el
herbivorismo. Este estudio representa la primera descripción cuantitativa in situ de los

vi
peces que habitan en MCEs a profundidades entre los 50–70 m en Puerto Rico, y resalta
el papel de los MCEs como hábitats valiosos para los peces arrecifales.
En adición, este estudio evaluó la conectividad entre profundidades mesofóticas y
someras según la distribución y el movimiento de los peces, y exploró explícitamente si
los MCEs representan un hábitat adicional y refugio potencial para las especies
arrecifales someras, particularmente para aquellas de interés comercial. Los MCEs del
veril de La Parguera tienen condiciones medioambientales relativamente más estables y
son mucho menos impactadas por las pesquerías en comparación con los arrecifes
someros. Setenta y ocho especies someras estuvieron asociadas a MCEs, y aquí se
reportan seis potenciales patrones de conectividad entre los hábitats someros y
mesofóticos, con base en los cambios en la composición taxonómica, en la abundancia
relativa de las especies y en la presencia de juveniles. La conectividad es significativa en
ambas direcciones, y la presencia de algunas especies ontogeneticamente migratorias
podría depender de la producción somera. Treinta y un especies son de interés comercial,
y de ellas la mayoría también es de interés conservacionista ya que hace treinta años
fueron peces comunes en las zonas someras pero han declinado marcadamente en el área
probablemente por su histórica sobrepesca (por ejemplo, tiburones, y meros, pargos y
loros grandes). Ya que la conectividad permite la dispersión de larvas y de adultos,
creemos que el abasto de desove de algunas especies (por ejemplo, grandes pargos, mero
negro) puede ser producida principalmente por las poblaciones remanentes que quedan en
los MCEs. Esto tiene importantes implicaciones económicas y ecológicas, y por
consiguiente, los peces de los MCEs deben ser protegidos para el manejo peces altamente
explotados. Estas medidas podrían incrementar la resiliencia y estabilidad del arrecife.

vii
Por último, este estudio investigó el efecto de la complejidad topográfica en las
comunidades de peces de los MCEs mediante la asociación de diferentes medidas de
complejidad (por ejemplo, relieve del hábitat, rugosidad, pendiente) y la estructura de la
comunidad de peces
Las pruebas estadísticas y técnicas de análisis multivariado identificaron a la
complejidad como un parámetro importante que afecta los procesos ecológicos en los
MCEs, tal como se ve reflejado en los cambios de composición y abundancia de los peces
arrecifales. La caracterización general de la complejidad fue la variable que mejor se
relacionó con los cambios en la comunidad de peces tanto a nivel específico como de
comunidad. Debido a que los peces de MCEs complejos son más diversos y abundantes,
e incluyen varias especies grandes de interés comercial y de conservación que han
desaparecido de los arrecifes someros en el área, estos lugares deben ser considerados
primordiales para el manejo de pesquerías y programas de conservación del arrecife de
coral.
Esta disertación aumenta el entendimiento de la ecología de los peces arrecifales,
incluyendo especies comerciales, y resalta el papel de los MCEs como hábitats valiosos
para los peces arrecifales. Los resultados enfatizan la importancia de incorporar la
composición y distribución de los peces que habitan en los MCEs cuando se planifica el
manejo espacial de los recursos arrecifales.

viii
Copyright ©
Ivonne Bejarano Rodríguez. April 2013

ix
Dedication
I would like to dedicate this work to my family and Daniel Mateos
Thank you for your unconditional love and support

x
Acknowledgements
I want to greatly thank my advisor Dr. Richard S. Appeldoorn for his wise guidance,
patience and support during this long process.
I specially thank my committee Dr. Jorge Garcia-Sais, Dr. Paul Yoshioka and Dr. Alberto
Sabat, for their technical and science-based comments that contributed importantly to this
research.
I deeply and sincerely thank my dive buddies Michael Nemeth, Milton Carlo, Hector
Ruiz, and Clark Sherman for their unconditional help in the field.
Very special thanks to my lab partners for their advise, support and friendship.
This work would not have been possible without the support, patience, humor, and love
of Daniel Mateos.
I deeply thank my family for being the biggest treasure in my life.

xi
Table of Contents
Abstract............................................................................................................................. ii
Resumen............................................................................................................................. v
Copyright........................................................................................................................ viii
Dedication ........................................................................................................................ ix
Acknowledgments ............................................................................................................ x
List of Tables ..................................................................................................................xiii
List of Figures................................................................................................................. .xv
Chapter 1: General Introduction ................................................................................... 1
LITERATURE CITED .......................................................................................................7
Chapter 2: Fishes associated to Mesophotic Coral Ecosystems in La
Parguera, Puerto Rico…………............................................................................... 11
ABSTRACT ……………………………………………………………………………..11
2.1 INTRODUCTION ......................................................................................................11
2.2 MATERIALS AND METHODS ................................................................................14
2.2.1 Study Area…………………………………………………………………………14
2.2.2 Data Collection…………………………………………………………………….14
2.3 RESULTS ...................................................................................................................18
2.4 DISCUSSION .............................................................................................................31
LITERATURE CITED......................................................................................................39
Chapter 3: Source-Sink and Refuge Functions of Mesophotic Coral
Ecosystems as Mediated by Connectivity with Shallow Reef Environments..45
ABSTRACT……………………………………………………………………………...45
3.1 INTRODUCTION ......................................................................................................45

xii
3.2.1 Study Area…………………………………………………………………………48
3.2.2 Data Collection…………………………………………………………………….48
3.2.3 Data Analysis………………………………………………………………………50
3.3 RESULTS ...................................................................................................................51
3.4 DISCUSSION .............................................................................................................62
Chapter 4: Topographic Complexity and Reef Fishes of Mesophotic Coral
Ecosystems along the Insular Slope off La Parguera...................................................81
ABSTRACT……………………………………………………………………………...81
4.1 INTRODUCTION ......................................................................................................81
4.2.1 Data Collection ………………………………………….…………………….......84
4.2.2 Data Analysis ...........................................................................................................86
4.3 RESULTS ...................................................................................................................87
4.4 DISCUSSION .............................................................................................................93
Chapter 5: Overall Conclusion................................................................................... 100
List of All Literature Cited ......................................................................................... 103

xiii
List of Tables
Table 2.1. Fishes recorded at six locations at mesophotic depths along the shelf-edge of
La Parguera, southwest Puerto Rico. Depth ranges (m) are based on minimum and
maximum depths recorded in vertical and horizontal transects, roving surveys, and
exploratory dives in the area to depths up to 90 m. Max. depth (m) is the maximum depth
recorded here or reported by García-Sais 2010, García-Sais et al. 2007, Nelson &
Appeldoorn 1985, Colin 1974, 1976, or www.fishbase.org. Depth classification (Depth
class, Sh:shallow, D:deep). Sighting frequency (% F) and Mean density (MD) are from
standard horizontal visual census transects along a depth gradient from 30 to 70 m. SD is
the standard deviation of the mean density........................................................................19
recorded here or reported by García-Sais et al. 2007, García-Sais 2010, Nelson &
roving surveys along a depth gradient from 30 to 70 m. SD is the standard deviation of
the mean density ...............................................................................................................22
Table 2.3. Indicator Species Analysis with Monte Carlo test of significance (p-values)
indicating primary characteristic species of shallow and mesophotic habitats pooled
across six reefs in La Parguera shelf-edge, southwest Puerto Rico...................................26
Table 2.4. Relative species richness (% R) and density (% D) of trophic guilds per 30 m2
,
along a depth gradient from 30 to 70 m pooled across six reefs in La Parguera shelf-edge,
southwest Puerto Rico. Z: zooplanktivore, MI: Mobile invertebrate feeder, H: herbivore,
O: omnivore, P: piscivore, and SI: sessile invertebrate feeder. ........................................28
Table 3.1. Fishes recorded at six reef locations, between 30 and 70 m depth, along the
shelf-edge of La Parguera, southwest Puerto Rico. Depth ranges (m) are based on
minimum and maximum depths recorded in vertical and horizontal transects, roving
surveys, and exploratory dives in the area to depths up to 90 m. Max. depth is the
maximum reported depth by García-Sais 2010, García-Sais et al. 2007, Dennis et al.
2004, Humann & Deloach 1994, Nelson & Appeldoorn 1985, Colin 1974, 1976, or
www.fishbase.org. (*) depth limit extended by more than 10 m in the Caribbean, (+)
depth limit extended by more than 10 m in Puerto Rico. Commercially important species
are highlighted in bold. (R
) rare species, recorded only once or twice. Onto.: ontogenetic
migratory species from (M) mangrove (SG) seagrass (Al) algae. Spaw.: spawning
migratory species, (Res) resident spawner ( Tran) transient spawner. Life stages: (J)
juvenile (A) adult (AA) higher abundance of adults in MCEs than shallow.
............................................................................................................................................53

xiv
Table 3.2. Sighting frequency (%) of shallow commercially important fishes recorded at
six reef locations, between 30 and 70 m depth, along the shelf-edge of La Parguera,
southwest Puerto Rico. Most vales were obtained from transect surveys, except when
better data were obtained with roving surveys (highlighted in bold). Mean density
(number of fish) and biomass (g) in 100 m2
+/_ (SD) standard deviation, are included...58
Table 4.1. Mean values and standard deviation (SD) of topographic complexity variables
and fish community parameters recorded in visual census transects at six locations at
mesophotic depths along the shelf-edge of La Parguera, southwest Puerto Rico. Sample
size n = 16, except for the measurements of slope (GIS) where n = 40. Trophic guilds
codes: (Z) Zooplanktivore (P) Piscivore (O) Omnivore (H) Herbivore (MI) Mobile
invertebrate feeders (SI) Sessile invertebrate feeders. Abundance is number of fish/100
m2
. Richness is number of species in 30 m2.
Bold values show significant differences
between high and low complexity sites (p<0.05)..............................................................84
Table 4.2. Mean species-specific abundance (number of fish/100 m2
) of the main species
changing with topographic complexity, recorded in visual census transects at six
locations at mesophotic depths along the shelf-edge of La Parguera, southwest Puerto
Rico. SD is the standard deviation of the mean density. Bold values show significant
differences between high and low complexity sites (p < 0.05).........................................90
Table 4.3. Mean species-specific sighting frequency (%) of the main species changing
with topographic complexity (p < 0.05), recorded in roving surveys at six locations at
mesophotic depths along the shelf-edge of La Parguera, southwest Puerto Rico. SD is the
standard deviation of the mean density..............................................................................91
Table 4.4. Spearman correlations among topographic complexity variables and fish
community parameters surveyed at six locations at mesophotic depths along the shelf-
edge of La Parguera, southwest Puerto Rico. In addition, species-specific correlations of
the main species responding to topography complexity were tested. Trophic guilds codes:
(Z) Zooplanktivore (P) Piscivore (O) Omnivore (H) Herbivore (MI) Mobile invertebrate
feeders (SI) Sessile invertebrate feeders. Logarithmic transformations (Log (x+1)) were
applied to fish density data. Bold values show significant correlations (p<0.05).............92

xv
List of Figures
Figure 2.1 Map of the mesophotic study sites location along the insular platform margin
(shelf-edge) south of La Parguera, Puerto Rico.................................................................15
Figure 2.2. Changes in fish community structure along a depth gradient from 30 to70 m.
(A) Mean species richness (B) Mean fish density, per 30 m2
pooled across six reefs in La
Parguera shelf-edge, southwest Puerto Rico......................................................................18
Figure 2.3. Species sighting frequency (%) along a depth gradient from 30 to 70 m,
pooled across six reefs in La Parguera shelf-edge, southwest Puerto Rico. Shallow
species are those present at <30 m, or previously reported in the area occurring at these
depths, even if are now rare due to overfishing. A) Shallow species that decreased in
frequency with depth, B) Shallow species that increased in frequency with depth, C)
Shallow species that were frequent along the entire depth range. D) Deep species
restricted to >40 m, except for the sunshine fish, longsnout butterflyfish, and greenblotch
parrotfish, which occasionally occur shallow. “Average trend” is the averaged sighting
frequency for each group...................................................................................................24
Figure 2.4. Proportion of shallow species and deep species (see Figure 1.3) per depth,
along a gradient from 20 to 70 m pooled across six reefs in La Parguera shelf-edge,
southwest Puerto Rico. Data from 20 m was taken from the CRES program (see
Methods)............................................................................................................................25
Figure 2.5. Mean trophic guild proportions observed in 30 m2
transects distributed along
a 20 to 70 m depth gradient, pooled across six reefs in La Parguera shelf-edge, southwest
Puerto Rico. Vertical line separates shallow and mesophotic zones. Data from 20 m was
taken from the CRES program (see Methods). Z: zooplanktivore, MI: Mobile invertebrate
feeder, H: herbivore, O: omnivore, P: piscivore, and SI: sessile invertebrate feeder. A)
Relative density of each trophic guild per 30 m2
. B) Relative species richness of each
trophic guild per 30 m2
......................................................................................................27
Figure 2.6. Distribution of the mean species richness of trophic guilds in roving surveys
in a 40 to 70 m depth gradient pooled across six reefs in La Parguera shelf-edge,
southwest Puerto Rico. P: piscivore, MI: Mobile invertebrate feeder, H: herbivore
Table 4. Relative species richness (% R) and density (% D) of trophic guilds per 30 m2
,
O: omnivore, P: piscivore, and SI: sessile invertebrate feeder..........................................27
Figure 2.7. A) Non-metric multidimensional scaling (NMS) plot based on Bray-Curtis
similarities of fish assemblages at six reefs in La Parguera shelf-edge, southwest Puerto
Rico, in a depth gradient from 30 to 70 m. NMS with herbivore (B) and zooplanktivore
(C) density distributions (circles size is proportional to density value), and trophic guilds
responses to ordination axes. H: herbivore, Z2: zooplanktivore excluding dominant

xvi
masked goby (Coryphopterus personatus), O: omnivore, P: piscivores depths (MRPP, p
< 0.0001)............................................................................................................................30
Figure 2.8. Mean cover (%) of coralline algae, Lobophora and turf, and relative density
(%) of herbivore fish per depth, along a gradient from 20 to 70 m at six reefs in La
Parguera shelf-edge, southwest Puerto Rico. Algal and fish data from 20 m were taken
from the CRES program (see Methods).............................................................................36
Figure 3.1. Map of the mesophotic study sites located along the insular platform margin
Figure 3.2. Mean abundance of large-bodied commercial fish in roving surveys within a
40 to 70 m depth gradient pooled across six reefs in La Parguera shelf-edge, southwest
Puerto Rico. LJOC: dog snapper, CAPE: reef shark, LCYA: cubera snapper, MBON:
black grouper.....................................................................................................................57
Figure 3.3. Mean biomass of large-bodied commercial fish in roving surveys within a 40
to 70 m depth gradient pooled across six reefs in La Parguera shelf-edge, southwest
Puerto Rico. LJOC: dog snapper, CAPE: reef shark, LCYA: Cubera snapper, MBON:
black grouper.....................................................................................................................57
Figure 3.4. Mean (A) abundance and (B) biomass (in 100 m2
), of medium-bodied
commercially important groupers in transect surveys distributed within a 30 to 70 m
depth gradient, pooled across six reefs in La Parguera shelf-edge, southwest Puerto Rico.
CCRU: graysby, CFUL: coney, EGUT: red hind..............................................................58
), of medium-bodied
commercially important snappers in transect surveys distributed within a 30 to 70 m
depth gradient, pooled across six reefs in La Parguera shelf-edge, southwest Puerto Rico.
OCHR: yellowtail snapper, LAPO: schoolmaster, LANA: mutton snapper, LMAH:
mahogany snapper. * mean abundance for this species was obtained from roving
surveys...............................................................................................................................59
), of medium-bodied
commercially parrotfishes in transect surveys distributed within a 30 to 70 m depth
gradient, pooled across six reefs in La Parguera shelf-edge, southwest Puerto Rico.
STAE: princess parrotfish, SISE: stripped parrotfish, SVIR: stoplight parrotfish............60
), of grunts in transect
surveys distributed within a 30 to 70 m depth gradient, pooled across six reefs in La
Parguera shelf-edge, southwest Puerto Rico. HFLA: French grunt, HPLU: white grunt,
HSCI: bluestriped grunt.....................................................................................................61
) of other commercial
species in transect surveys distributed within a 30 to 70 m depth gradient, pooled across
six reefs in La Parguera shelf-edge, southwest Puerto Rico. BVET: queen triggerfish,

xvii
LMAX: hogfish, LTRI: smooth trunkfish, SBAR: great barracuda, CLUG: black jack.
*mean abundance for this species was obtained from roving surveys..............................61
Figure 4.1. Map of the mesophotic study sites location along the insular platform margin
Figure 4.2. Shelf slope at the insular margin of La Parguera, southwest Puerto Rico,
displaying examples of areas with high and low complexity. Black lines are depth
contours from 20 to 100 m in 10 m intervals and the red color depicts flat areas above the
shelf break..........................................................................................................................86

1
Chapter 1: General Introduction
Reef fishes have been widely surveyed thanks to the invention and development
of scuba diving techniques, however, most studies are confined to reef areas shallower
than 40 m (Pyle 2000, Itzikowitz et al. 1991, Thresher and Colin 1986), given the
difficulties and the high cost for accessing deep reefs. Nevertheless, coral ecosystems can
extend to depths of 100 m or more, and few studies on deep reefs and their associated fish
communities have been conducted. As a consequence, our understanding of the entire
reef ecosystem, their associated communities and many reef ecological processes is
limited.
Conventional scuba technology is only suitable for depths shallower than 40 m,
and is highly restricted by bottom time (Pyle 1996, 2000). Work on mesophotic coral
ecosystems (MCE), light dependent coralline habitats occurring between 40-150 m, has
been primarily conducted using video and photography from submersibles or remotely
operated vehicles (ROVs) (Armstrong et al. 2006, Nelson and Appeldoorn 1985, Colin
1976, 1974). Although these techniques have provided broad observations (Pyle 2000),
the collection of quantitative information has been limited. However, these difficulties
can be surpassed by trimix rebreather diving, which has become the ideal technique to
study fishes within MCEs, especially given the extreme geomorphology of slope
environments.
Fish assemblages within MCEs are of great interest to scientists and managers
because they differ in taxonomic structure and abundance from that of shallower reefs
(García-Sais 2010, Brokovich et al. 2008, Feitoza et al. 2005, Colin 1976, Colin 1974).
Nevertheless, little is known on the structure, composition and ecology of these fishes in

2
the Caribbean. By studying the deep fish fauna from slopes and vertical faces, Colin
(1974, 1976) characterized the deep ichthyofauna off Jamaica, Belize and Bahamas. He
collected deep fishes using submersibles, dip-nets, fish traps, explosives, and
ichthyocides, and determined their relative abundance along transects 300 m length at 90,
105 and 120 m depth. His research resulted in a list of deep reef fishes, which includes
the relative abundance and depth ranges of species. He described fishes within MCEs as a
mixed assemblage of shallow reef fishes (>30 m), which find their lower distribution at
these depths, and true ‘‘deep-reef’’ species limited to mesophotic habitats. In addition,
Colin (1974) suggested that the vertical distribution of some reef fish species was more
related to habitat features than depth, and noted ontogenetic trends in the vertical
distribution of ‘‘deep-reef’’ species, where juvenile stages were typically observed at
shallower depths than adults.
In Puerto Rico, a preliminary evaluation of deep-water fish habitats and
abundance was performed at depths ranging from about 100–450 m, using the Johnson-
Sea-Link II submersible (Nelson and Appeldoorn 1985). This research provided the first
general observations of benthic habitats and associated fishes of the insular slope.
However to this day, the upper insular slope reefs of Puerto Rico, and their associated
fish fauna remains largely undescribed. Only recently, some studies performed by
García-Sais (2010) and García-Sais et al. (2007, 2004) in Bajo de Sico, Desecheo and
Vieques Island (Western Puerto Rico) describe the reef fishes found at depths between 30
and 50 m, including rodolith areas. These fish communities were characterized as species
rich and diverse, with most of the fishes being zooplanktivores, and Stegastes partitus
(bicolor damselfish) being by far the most abundant and common species. In addition,

3
these MCEs harbor commercially important reef fishes, such as large bodied snappers
and groupers, most of which have become locally or regionally threatened by overfishing
at shallower depths, e.g. Lutjanus analis (mutton snapper), Lutjanus cyanopterus (cubera
snapper), and Mycteroperca bonaci (black grouper).
Being potentially less exposed to fishing pressure, MCEs may represent one of
the last refugia for shallow threaten species and may serve as a potential mechanism for
reseeding shallow resources (Bongaerts et al. 2010, Riegl and Piller 2003, Glynn 1996).
This idea explicitly recognizes connectivity between MCEs and shallow areas, but the
strength, degree and direction of this connectivity is uncertain. So, MCEs may be
important genetic reservoirs with the potential of helping depleted shallow reef fish
populations to recover. Likewise, MCE populations of shallow species may represent an
ecological sink and be dependent upon inputs from shallow areas. In both cases,
connectivity would have significant ecological impacts besides refuge and reseeding.
Many physical (temperature, suspended sediments, light intensity, water motion,
pressure) and biological parameters (competition, predation) vary with depth, and most of
the spatial variation in the structure of fish communities can be explained by these depth
gradients (Brokovich et al. 2006, Srinivasan 2003, Donaldson 2002, Friedlander and
Parrish 1998, McGehee 1994). Depth distributions of reef fishes may be established at
settlement, either by physical processes or behavior influencing the vertical distribution
of fish larvae (Srinivasan 2003, Leis and Carson-Ewart 2000, Doherty and Carleton 1997,
Leis 1991), or by larvae selecting specific habitats and depths at settlement (Leis and
Carson- Ewart 2002). However, subsequent fish distributions can be either strengthened
or modified by post-settlement survival (Jones 1997) and growth (Srinivasan 2003).

4
Species with narrow depth ranges survive and grow better within their normal depth
range than outside it, while species with wide depth ranges may be less affected by depth
changes (Srinivasan 2003). Thus, some shallow reef fishes can reach deep areas, while
some fish species do not occur in shallow reefs and are confined to deep habitats.
Depth has long been considered an important factor for reef fishes, and Roberts
and Ormond (1987) showed depth to be a good predictor of species richness. However,
other aspects of reef structure such as habitat complexity may be more relevant for fish
distributions than depth (Alveizon et al. 1985, Colin 1974). Habitat rugosity is one of the
most important habitat characteristics positively correlated with reef fish abundance
(Eagle et al. 2001, Syms and Jones 2000, Friedlander and Parrish 1998, Gladfelter and
Gladfelter 1978, Luckhurst and Luckhurst 1978). The presence of structure provides
refuges and facilitates migration for most of the reef fishes (Jenkins and Southerland
1997). In particular, microhabitats provided by corals enhance net settlement and offer
refuge from predation (Hixon and Beets 1993). Thus, habitat structure and its related
abiotic parameters represent additional important agents shaping the distribution,
composition, abundance (Brokovich et al. 2006, Wantiez and Chauvet 2003, Lara and
Gonzalez 1998, McGehee 1994), and trophic groups (Nanami and Nishihira 2002,
Friedlander and Parrish 1998, Meekan et al. 1995, Carpenter et al. 1981) of coral reef
fishes.
Off La Parguera insular shelf slope, in the southwest Puerto Rico, the coral
ecosystem extends to waters deeper than 100m. The insular slope off La Parguera was
described in detail by Sherman et al. (2010). The upper slope (20-100 m depth) is divided
into six geomorphic zones based on breaks in slope, topographic features and substrate

5
type: shelf-edge reef front, shallow fore-reef slope, fore-reef spur and groove,
intermediate fore-reef slope, fore-reef terrace, and deep fore-reef slope. In general,
southeast-facing reef fronts have a gentler gradient, and westward facing areas have
steeper slopes and a more highly rugose bottom. Here, MCEs are better developed above
90 m depth (Sherman et al. 2010), in steep but not abrupt slope areas with high rugosity
and outcrops, elevated from the surrounding seafloor. This allows the sediment to be
channeled away from the substratum supporting benthic organisms (Sherman et al. 2010).
Thus, like shallow reefs the distribution of well-developed MCEs is patchy. The fact that
there are obvious differences in both the vertical and horizontal structure of the insular
slope suggests that there may be corresponding differences in the structure of reef fish
assemblages.
The purpose of this dissertation was to quantitatively describe, with in situ
observations the community composition, abundance and distribution of the fishes within
mesophotic coral ecosystems along the La Parguera insular slope. Extend reef fish
observations to a depth of ~90 m and evaluate connectivity between shallow and
mesophotic depths.
Based on these main premises, this dissertation is divided into three chapters
(Chapters 2, 3, and 4) that are consistent with the style and format of scientific
publications. The main objective of each section was to: (i) characterize the reef fish
community composition, abundance and structure off MCEs off La Parguera and their
relation to depth, (ii) characterize the connectivity between shallow and mesophotic coral
reefs, and (iii) quantify the role of substratum complexity, in structuring deep reef fish
communities. The final section of this dissertation (Chapter 5) includes an overall

6
conclusion of the results and suggestions for future research prospects for MCE fish
communities.
Chapter 2 presents the first quantitative characterization of the MCE fish
assemblage between 30 and 70m off La Parguera, Puerto Rico. Depth is evaluated as a
potential factor regulating fish community structure, and shallow (30 m depth) and
mesophotic assemblages are compared and contrasted. Characterizations are based on the
quantitative data obtained from visual census using mixed-gas rebreathers, organism
collections, and underwater photography. Observations from exploratory dives to depths
down to 91 m are included. The main null hypothesis is that reef fish assemblages
(composition, abundance, and structure) do not change with depth. To test this, fish
composition (presence/absence of species), abundance, biomass, and the proportions of
trophic guilds are compared between deep and shallow reefs, and within MCEs to
describe changes in the vertical distribution of reef fishes in the shelf slope, from 30 to 70
m depth. This study increases our understanding of reef fish ecology and highlights the
role of MCEs as valuable habitats for reef fishes.
Chapter 3 investigates the potential patterns of fish connectivity between shallow
and mesophotic reefs as expressed by the distribution and movement of fishes, and
explicitly explores whether MCEs in La Parguera represent an additional habitat and
potential refugia for shallow reef fishes, particularly for commercially targeted species.
For this, the composition, sighting frequency, relative abundance and presence of
juveniles are compared between shallow (20 and 30 m depth) and mesophotic depths (40-
70 m). In addition, results are compared with reef fish data collected 31 years ago in
shallow areas in La Parguera (Kimmel 1985) to provide a historical reference point from

7
which to compare changes in fish species occurrence. Results emphasize the importance
of incorporating the composition and distribution of the MCE fish community when
planning for the spatial management of coral reef resources.
Chapter 4 assesses the effect of topographic complexity on the reef fish
assemblages within mesophotic coral ecosystems (MCEs) by evaluating the association
between different complexity measurements, at different spatial scales, (e.g., gross
habitat relief, chain rugosity, slope) and fish community structure between 40 and 70 m
depth. To test if fish assemblage structure differs within MCEs in response to habitat
complexity, the fish taxonomic composition, abundance, biomass, and proportions of
trophic guilds are compared between high complexity (southwest facing) sites and low
complexity (southeast facing) sites. Results obtained from this study suggest that
complex MCE sites should be considered as prime areas targeted by fisheries
management and coral reef conservation programs.
In summary, this dissertation provides a quantitative characterization of the MCE
fish community that reveals important aspects of its composition, distribution, and
potential connectivity with shallow depths that may prove useful as part the management
of coral reef resources.
Literature Cited
Alveizon WR, R Richardson, P Pitts, G Serviss (1985) Coral zonation and patterns of
community structure in Bahamian reef fishes. Bull Mar Sci 36:304-318
Armstrong RA, H Singh, J Torres, RS Nemeth, A Can, C Roman, R Eustice, L Riggs and
G Garcia-Moliner (2006) Characterizing the deep insular shelf coral reef habitat
of the Hind Bank marine conservation district (US Virgin Islands) using the
Seabed autonomous underwater vehicle. Continental Shelf Research 26: 194–205

8
Bongaerts P, Ridgway T, Sampayo EM, Hoegh-Guldberg O (2010) Assessing the ‘‘deep
reef refugia’’ hypothesis: focus on Caribbean Reefs. Coral Reefs 29:309-327
Brokovich E, A Baranes, M Goren (2006) Habitat structure determines coral reef fish
assemblages at the northern tip of the Red Sea. Ecol Indic 6:494–507
Brokovich E, S Einbinder, N Shashar, M Kiflawi, S Kark (2008) Descending to the
twilight-zone: changes in coral reef fish assemblages along a depth gradient down
to 65 m. Mar Ecol Prog Ser 371:253–262
Carpenter KE, RI Miclat, VD Albaladejo, VT Corpuz (1981) The influence of substrate
structure on the local abundance and diversity of Philippine reef fishes. Proc 4th
Intl Coral Reef Symp, Manila 2:497–502
Colin PL (1974) Observations and collection of deep-reef fishes off the coasts of Jamaica
and British Honduras (Belize). Marine Biology 24:29-38
Colin PL (1976) Observations of deep-reef fishes in the Tonge-Of-The-Ocean, Bahamas.
Bull Mar Sci 26(4):603-605
Doherty PJ, JH Carleton (1997) The distribution and abundance of pelagic juvenile fish
near grub reef, central Great Barrier Reef. Proc 8th Int Coral Reef Symp 2:1155–
1160
Donaldson TJ (2002) Habitat association and depth distribution of 2 sympatric groupers
of the genus Cephalopholis (Serranidae: Epinephelinae). Ichthyol Res 49:191–
193
Eagle JG, G Jones, M McCormick (2001) A multi-scale study of the relationships
between habitat and the distribution and abundance patterns of the three coral reef
angelfishes (Pomacanthidae). Mar Ecol Pro Ser 214:253-265
Feitoza BM, RS Rosa, LA Rocha (2005) Ecology and zoogeography of deep-reef fishes
in northeastern Brazil. Bull Mar Sci 76:725–742
Friedlander AM, JD Parrish (1998) Habitat characteristics affecting fish assemblages on a
Hawaiian coral reef. J Exp Mar Biol Ecol 224:1–30
García-Sais JR, J Castro, M Sabater, M Carlo (2004) Monitoring of coral reef
communities from Isla de Vieques, Puerto Rico. Final Report submitted to the
Department of Natural and Environmental Resources (DNER), US Coral Reef
National Monitoring Program, NOAA. P 118
García-Sais JR, Castro R, Sabater J, Carlo M (2007). Characterization of benthic habitats
and associated reef communities at Bajo de Sico Seamount, Mona Passage, Puerto

9
Rico. Final Report submitted to Caribbean Fishery Management Council
(CFMC/NOAA). P 91
García-Sais JR (2010) Reef habitats and associated sessile-benthic and fish assemblages
across a euphotic–mesophotic depth gradient in Isla Desecheo, Puerto Rico. Coral
Reefs 29:277–288
Gladfelter WB, EH Gladfelter (1978) Fish community structure as a function of habitat
structure on West Indian patch reefs. Rev Biol Trop 26(1):65-84
Glynn PW (1996) Coral reef bleaching: facts, hypothesis and implications. Glob Change
Biol 2:495-509
Hixon MA, JP Beets (1993) Predation, prey-refuges, and the structure of coral-reef fish
assemblages. Ecol Monogr 63(1):77–101
Itzikowitz M, M Haley, C Otis, D Evers (1991) A reconnaissance of the deeper Jamaican
coral reef fish communities. NE Gulf Sci 12:25-34
Jenkins G, C Southerland (1997) The influence of habitat structure on nearshore fish
assemblages in a southern Australian embayment: colonization and turnover rate
of fishes associated with artificial macrophyte beds of varying physical structure.
J Exp Mar Biol Ecol 218:103-125
Jones GP (1988) Experimental evaluation of the effects of habitat structure and
competitive interactions on the juveniles of two coral reef fishes. J Exp Mar Biol
Ecol 123:115-126
Lara EN, EA Gonzalez (1998) The relationship between reef fish community structure
and environmental variables in the southern Mexican Caribbean. J Fish Biol
53:209–221
Leis JM (1991) The pelagic stage of reef fishes: the larval biology of coral reef fishes. In:
Sale PF (ed) The ecology of fishes on coral reefs. Academic Press, San Diego,
Calif. P 183–230
Leis JM, BM Carson-Ewart (2000) Behaviour of pelagic larvae of four coral-reef fish
species in the ocean and an atoll lagoon. Coral Reefs 19:247–257
Leis JM, BM Carson-Ewart (2002) In situ settlement behaviour of damselfish
(Pomacentridae) larvae. J Fish Biol 61:325–346
Luckhurst BE, K Luckhurst (1978) Analysis of the influence of substrate variables on
coral reef fish communities. Marine Biology 49: 317-323

10
McGehee MA (1994) Correspondence between assemblages of coral reef fishes and
gradients of water motion, depth, and substrate size off Puerto Rico. Mar Ecol
Prog Ser 105:243–255
Meekan MG, DL Steven, MJ Fortin (1995) Spatial patterns in the distribution of
damselfishes on a fringing coral reef. Coral Reefs 14:151–161
Nanami A, M Nishihira (2002) The structures and dynamics of fish communities in an
Okinawan coral reef: effects of coral-based habitat structures at sites with rocky
and sandy sea bottoms. Environ Biol Fishes 63:353–372
Nelson WR, RS Appeldoorn (1985) A submersible survey of the continental slope of
Puerto Rico and the US Virgin Islands, 1–23 October 1985. Cruise report, R/V
Seward Johnson, National Marine Fisheries Service, Pascagoula, Mississippi
Laboratories, P 76
Pyle RL (1996) How much coral reef biodiversity are we missing? Global biodiversity 6:
3-7
Pyle RL (2000) Assessing undiscovered fish biodiversity on deep coral reefs using
advanced self-contained diving technology. Mar Tech Soc J 34: 82-91
Riegl B, WE Piller (2003) Possible refugia for reefs in times of environmental stress.
International Journal of Earth Science 92(4):520-531
Roberts C, R Ormond (1987) Habitat complexity and coral reef fish diversity and
abundance on Red Sea fringing reefs. Mar Ecol Prog Ser 41:1-8
Sherman C, M Nemeth, H Ruíz, I Bejarano, RS Appeldoorn, F Pagan, M Sharer, E Weil
(2010) Geomorphology and benthic cover of mesophotic coral ecosystems of the
upper insular slope of southwest Puerto Rico. Coral Reefs 29:347-360
Srinivasan M (2003) Depth distributions of coral reef fishes: the influence of
microhabitat structure, settlement, and post-settlement processes. Oecologia
137:76-84
Syms C, G Jones (2000) Disturbance, habitat structure and the dynamics of a coral-reef
fish community. Ecology 81:2714-2729
Thresher RE and PL Colin (1986) Trophic structure, diversity and abundance of fishes of
the deep reef (30–300 m) at Enewetak, Marshall Islands. Bull Mar Sci 38:253–
272
Wantiez L, C Chauvet (2003) First data on community structure and trophic networks of
Uvea coral reef fish assemblages (Wallis and Futuna, South Pacific Ocean).
Cybium 27:83–100

11
Chapter 2: Fishes associated to Mesophotic Coral Ecosystems in La
Parguera, Puerto Rico
Abstract: Fishes associated with Mesophotic Coral Ecosystems (MCEs) of the La
Parguera shelf-edge were surveyed between 2007 and 2011 using rebreather trimix
technical diving. Fishes were identified and counted within belt transects (30 m2
) and
roving surveys at 30, 40, 50, 60 and 70 m depth. Vertical transects from 70 to 30 m depth
helped determine depth distribution ranges. The MCE fish assemblage (40-70 m) was
distinct from shallow areas (30 m), with taxonomic composition, abundance and the
proportion of trophic guilds varying with increasing depth. Most fishes at MCE were
primarily shallow species, but others were restricted to mesophotic depths. An additional
15 species were added to those previously classified as indicator species of mesophotic
areas in Puerto Rico. Fish abundance and species richness within MCEs were high,
though both varied greatly. Eighty-two species were identified within transects, and 37
within roving surveys, for a total of 103 species. The dominant trophic guild within
MCEs was the zooplanktivores, while herbivores dominated shallow reefs. Both,
herbivore and zooplanktivore responded strongly, and oppositely, to depth. The few
herbivores associated to deep MCEs are small-bodied species. The largest changes within
the mesophotic fish community along the depth gradient occurred at 60 m, similar to that
reported for algae and corals, and seem to represent both a response to reduced light and
variations in herbivory. The presence of commercially important fishes at MCE, many
considered to be threatened by intense fishing pressure in shallow areas, suggests that
these MCEs are important for the conservation of these species. This study represents the
first quantitative in situ observations and descriptions of fishes inhabiting MCEs at depths
of 50–100 m in Puerto Rico and highlights the role of MCEs as valuable habitats for reef
fishes. The composition and distribution of the MCE fish community should be
incorporated the when planning for the spatial management of coral reef resources.
KEY WORDS: Fish Assemblage, Mesophotic coral ecosystems (MCE), Depth,
Rebreather, Trophic guilds, Herbivores.
2.1 Introduction
Coral reef fishes have been extensively studied in many parts of the tropics, and
the composition and ecology of reef fish communities are well characterized for the
upper 30 m, largely because of accessibility provided by conventional SCUBA diving
(Pyle 2000, Itzikowitz et al. 1991, Thresher and Colin 1986). However, coral ecosystems

12
can extend to depths of 100 m or more, and research on fishes associated with
mesophotic coral ecosystems (MCEs), i.e., between 40-150 m, is very limited. Past work
at these depths has been primarily conducted using submersibles or remotely operated
vehicles (ROVs) (Nelson and Appeldoorn 1985), which has provided a general baseline
of qualitative descriptions but limited quantitative information. Due to the extreme
geomorphology of slope environments, as with shallow coral reefs, diver-based survey
techniques can provide more quantitative data, but for work in MCEs this requires the use
of trimix rebreather diving. This technique provides reasonable bottom time at
mesophotic depths to perform censuses, take photographs, collect cryptic species, or
effectively approach organisms within cracks or crevices. An added advantage is that
fishes are less evasive to the presence of divers because no bubbles are released with
close circuit diving. Nevertheless, the applicability of this technology can be limited
because it is expensive and requires extensive training.
Initial studies of MCE fish assemblages have shown them to differ in taxonomic
structure and abundance from that of shallower reefs (García-Sais 2010, Brokovich et al.
2008, Feitoza et al. 2005, Colin 1976, Colin 1974), represent the lower distribution of
many shallow species (García-Sais 2010, Brokovich et al. 2008, Tresher and Colin 1986,
Colin 1976, Colin 1974), harbor endemics (Pyle et al. 2008, Brokovich et al 2008, Pyle
2000) and species limited to deep areas (Brokovich et al. 2008, Feitoza et al. 2005).
Additionally, MCEs sometimes include large commercially-important species threatened
by overfishing at shallower depths (Fetoiza et al. 2005, García-Sais et al. 2004). Although
management of MCE fish communities is important for maintaining healthy fisheries and

13
local and regional biodiversity (Riegl and Piller 2003), little is known on this fauna
composition and ecology.
Preliminary observations of the MCE fish communities off the shelf-edge of La
Parguera, Southwest Puerto Rico, were first made using scuba in the 1970’s by Colin
(pers. comm.), with deeper observations using the Johnson-Sea-Link II submersible to
survey deep-water fish habitats and abundance at depths ranging from about 100–450 m
(Nelson and Appeldoorn 1985). However, in the last six years a detailed study of the
MCEs off La Parguera has been undertaken, under the auspices of the Coral Reef
Ecosystem Studies (CRES) program of the U.S. National Oceanic and Administration
(NOAA). This program has accessed the MCEs off La Parguera using a combination of
ROV transects for gross characterization and rebreather diving for quantitative
assessments. Sherman et al. (2010) provided a detailed description of the geomorphology,
biota and flora of these ecosystems, while a number of new species descriptions
(Ballantine et al. 2011, 2009, Ballantine & Ruíz 2011, 2010, Petrescu et al. (in review),
Petrescu et al. (in review), Pesic et al. (in review), Corgosinho & Schizas (in review)
have augmented our understanding of their biodiversity. As part of this larger study, the
main objective of this paper is to present the first quantitative characterization of the La
Parguera MCE fish assemblage between 30 and 70m. Depth was evaluated as a potential
factor regulating fish assemblage structure, and shallow (30 m depth) and mesophotic
assemblages were compared and contrasted. This study increases our understanding of
reef fish ecology and highlights the role of MCEs as valuable habitats for reef fishes.
Thus far, MCE research in Puerto Rico and USVI has focused on environments at depths
of 30–50 m. Only limited information exists on deeper MCEs at depths of 50–70 m. The

14
current study represents some of the first in situ observations and descriptions of these
deeper settings in Puerto Rico. Results emphasize the importance of incorporating the
composition and distribution of the MCE fish community when planning for the spatial
management of coral reef resources.
2.2 Materials and methods
2.2.1 Study Area
The insular slope of southwest Puerto Rico consists of an upper portion that
extends from 20 to 160 m depth, followed by the basal slope beneath (Sherman et al.,
2010). Coral ecosystems on the shelf-edge slope of La Parguera extend to waters deeper
than 100 m, but most benthic MCEs are located above 90. This zone is separated by a
near vertical break in slope (90-160 m) from a deeper insular slope zone and is
characterized by geomorphologic changes related to the prevailing wave exposure over
recent geologic time (Sherman et al. 2010). Because active down-slope sediment
transport negatively affects the growth of benthic organisms, the better-developed MCEs
are located in association with topographic highs, which are patchily distributed along the
insular slope.
2.2.2 Data collection
The distribution of MCEs along the shelf-edge slope of La Parguera, southwest
Puerto Rico, was determined coupling exploratory diving and ROV surveys with high
resolution multibeam bathymetry available from Battista and Stecher (2006). Potential
study sites were then selected from areas where the multibeam imagery indicated the

15
presence of complex topography between 40 and 100 m depth, which suggested the
presence of MCEs. Sites were visited to corroborate MCE development, and six MCE
study sites were chosen for study (Figure 2.1).
Figure 2.1 Map of the mesophotic study sites location along the insular platform margin
(shelf-edge) south of La Parguera, Puerto Rico.
Data were collected between November 2007 and December 2011 using closed-
circuit rebreather trimix technical diving, which provided enough time to perform

16
detailed fish censuses at mesophotic depths. Each site was sampled with a minimum of
four replicates taken at each of five depth intervals: 30, 40, 50, 60 and 70m. Fishes were
identified and counted during 15 minutes along 10 x 3m modified belt transects (30 m2
)
(Brock 1954). The first species recorded were those evasive at the presence of divers
(e.g., snappers) followed by the more territorial species. In addition, roving surveys were
made to quantify fishery target species outside belt transects. Six vertical roving transects
were performed at each study site, from 70 m to 30 m to help determine the maximum
depth distribution of the species in the area. Observations from three exploratory dives to
deeper depths (up to 91 m) were also considered when determining species depth range.
Species unidentified in the field were photographed and eventually collected using
quinaldine and dip nets, to identify them latter in the laboratory. Identifications followed
Nelson (2006). All species were assigned to one of six trophic guilds: planktivore (P),
herbivore (H), piscivore (P), omnivore (O), mobile invertebrate feeder (MI), sessile
invertebrate feeder (SI), based on Randall (1967), or fishbase.org. Depending on its depth
distribution, each species was classified as shallow (when present at <30 m, or previously
reported in the area occurring at these depths, even if are now rare due to overfishing) or
deep (when restricted to >40 m). The sunshine fish (Chromis insolata), longsnout
butterflyfish (Prognathodes aculeatus), and greenblotch parrotfish (Sparisoma
atomarium) were exceptions in that they were considered deep species although
occasionally occurred shallow.
2.2.3 Data Analysis
Multivariate analyses were used to examine the variations in fish species
composition and abundance along a depth gradient from 30 to 70 m. Depth patterns were

17
elucidated using nonmetric multidimensional scaling (NMS) ordination analysis in PC-
ORD version 5.0 (McCune & Mefford 2006), while a nonparametric multivariate test for
group differences (multiresponse permutation procedure, MRPP) was used to evaluate
compositional differences of fish assemblages along the depth gradient and between each
depth (pairwise comparisons) (McCune & Mefford 2006). Both ordination and MRPP
analyses used the Sorensen distance measure on Log(x+1) transformed mean fish
densities (number of fish/transect) of all observed species. This transformation was useful
for the analyses of numerical abundance because it rescaled disproportionally large and
small values, thereby reducing the range in values. Otherwise, the dominance of species
such as the masked goby (Coryphopterus personatus) would mask trends in the rest of
the fish assemblage. Joint plots were displayed on the ordination to explore the
relationship of trophic guilds responses to ordination axes. Species most frequent and
abundant at shallow reefs (30 m) and at mesophotic depths were identified using
indicator species analysis in PC-ORD (McCune & Mefford 2006) in order to allow
interpretation of single-species responses to depth. To lend more weight to our results,
data were compared with previous shallow fish data collected in the area at 20 m depth,
as part of a Coral Reef Ecosystem Studies (CRES) program supported by the National
Oceanic and Atmospheric Administration (NOAA). This data was collected with the
same methodology used in our research but with larger transects (100 m2
), as reported in
Nemeth and Appeldoorn (2009). Comparisons with our data were possible after
standardizing densities per m2
, and calculating sighting frequencies of the species, and
proportions of trophic guilds species richness in 30 m2
.

18
2.3 Results
The MCEs off the La Parguera shelf-edge between 30 and 70 m depth possess a diverse
and unique fish fauna composed of at least of 103 species from 31 families: 82 species
were identified within transects, and 37 within roving surveys (Tables 1.1, 1.2). The
mean abundance of individuals was 56 ind/30 m2
(range: 5-260 ind/30 m2
). The mean
species richness per transect was 9 (range: 4-16). Species richness at mesophotic depths
was similar among depths, but fish abundance decreased with depth (Figure 2.2 A, B).
Figure 2.2. Changes in fish community structure along a depth gradient from 30 to70 m.
(A) Mean species richness (B) Mean fish density per 30 m2
across six reefs in La
Parguera shelf-edge, southwest Puerto Rico
Most species within MCEs (76%) were common inhabitants of coral reefs above
40 m, or were previously common but are now rare likely due to overfishing. These
fishes were classified as “shallow” but extended their lower distribution to mesophotic
depths (Tables 1.1, 1.2). The main trend followed by these shallow species (by 33) was a
rapid decrease in abundance and frequency of occurrence with increasing depth (Figure
2.3 A). Several shallow species (34, 43%) do not occur below 60 m (Tables 1.1, 1.2;
Figure 2.4). The bicolor damselfish (Stegastes partitus) provides a good example for this
group. It was the most common species at 30 m (present in 85 % of the transects with a
mean of 4.2 ind/30 m2
)

19
Table2.1.Fishesrecordedatsixlocationsatmesophoticdepthsalongtheshelf-edgeofLaParguera,southwestPuertoRico.Depth
ranges(m)arebasedonminimumandmaximumdepthsrecordedinverticalandhorizontaltransects,rovingsurveys,andexploratory
divesintheareatodepthsupto90m.Max.depth(m)isthemaximumdepthrecordedhereorreportedbyGarcía-Sais2010,García-
Saisetal.2007,Nelson&Appeldoorn1985,Colin1974,1976,orwww.fishbase.org.Depthclassification(Depthclass,Sh:shallow,
D:deep).Sightingfrequency(%F)andMeandensity(MD)arefromstandardhorizontalvisualcensustransectsalongadepthgradient
from30to70m.SDisthestandarddeviationofthemeandensity

22
recorded here or reported by García-Sais et al. 2007, García-Sais 2010, Nelson &
roving surveys along a depth gradient from 30 to 70 m. SD is the standard deviation of
the mean density

23
and was frequently seen in the upper MCE (in 67% of the transects at 40 m and 31% at
50 m), but rarely reached deeper zones in La Parguera (Figure 2.3 A). Other common
shallow species (present in > 50% of the transects at 30 m) following this trend with
depth were surgeonfish (Acanthurus bahianus) and redband parrotfish (Sparisoma
aurofrenatum) (Figure 2.3 A). In contrast, another 11 shallow species followed an
opposite trend as they were more abundant at deep (Figure 3B). This group included
large predators targeted in the area such as dog (Lutjanus jocu) and cubera snappers
(Lutjanus cyanopterus), black grouper (Mycteroperca bonaci), and reef shark
(Carcharhinus perezii), as well as creole wrasse (Clepticus parrae) (Tables 1.1, 1.2).
There was a small group of fishes (9 species) that were common and abundant along the
entire depth range (30-70 m) (Figure 2.3 C). The yellowhead wrasse (Halichoeres
garnoti), the masked goby and the graysby (Cephalopholis cruentata) were present in the
majority of the transects (> 50%) at any depth, showing some variability among depths
but without a trend. Abundance of masked goby however, declined markedly at 70 m.
This species was the most abundant between 30 – 60 m, but at 70 m its frequency and
abundance declined (Table 2.1, Figure 2.3 C). The rest of the shallow species (25) were
rare or varied without a pattern in relation to depth. Finally, 25 species associated to
MCEs were considered to be “deep" species because they were only found at or below 40
m, or were scarce at shallow depths and markedly increased in frequency at mesophotic
depths (Tables 1.1, 1.2; Figure 2.4). Of these, the most frequent and abundant was the
sunshine fish, present in 87% all transects between 40-70 m, and representing on average
23% of the individuals per transect. Frequency and relative abundance (%) of this species
increased with depth. Though less abundant, the cave basslet (Liopropoma mowbrayi),

24
blackfin snapper (Lutjanus buccanella), longsnout butterflyfish and greenblotch
parrotfish increased in frequency (Figure 2.3 D) and density with depth as well. The
sargassum triggerfish (Xanthichthys ringens) was also common within MCEs up to 60 m.
This group included rare species such as the snow (Serranus chionaraia), crosshatch
(Serranus luciopercanus), and bicolor (Lipogramma klayi) basslets, and the dusky
cardinalfish (Phaeoptyx pigmentaria).
A B
C D
Figure 2.3. Species sighting frequency (%) along a depth gradient from 30 to 70 m,
pooled across six reefs in La Parguera shelf-edge, southwest Puerto Rico. Shallow
species are those present at <30 m, or previously reported in the area occurring at these
depths, even if are now rare due to overfishing. A) Shallow species that decreased in
frequency with depth, B) Shallow species that increased in frequency with depth, C)
Shallow species that were frequent along the entire depth range. D) Deep species
restricted to >40 m, except for the sunshine fish, longsnout butterflyfish, and greenblotch
parrotfish, which occasionally occur shallow. “Average trend” is the averaged sighting
frequency for each group

25
Figure 2.4. Proportion of shallow species and deep species (see Figure 2.3) per depth,
along a gradient from 20 to 70 m pooled across six reefs in La Parguera shelf-edge,
southwest Puerto Rico. Data from 20 m was taken from the CRES program (see
Methods)
Indicator species analysis (ISA) recognized 21 species as mainly responsible for
distribution differences with depth. Thirteen species were characteristic of shallow
habitats: surgeonfishes, bicolor damselfish, bluehead wrasse (Thalassoma bifasciatum),
redband parrotfish, princess (Scarus taeniopterus) and stripped parrotfish (Scarus iseri),
blue chromis (Chromis cyanea), peppermint goby (Coryphopterus lipernes), French grunt
(Haemulon flavolineatum), stoplight parrotfish (Sparisoma viride), and beaugregory
(Stegastes leucostictus). Eight species were characteristic of the mesophotic assemblage:
the sunshinefish, longsnout butterflyfish, blackfin snapper, the cave basslet, purple
reeffish (Chromis scotti), cherubfish (Centropyge argi), harlequin bass (Serranus
tigrinus), and striped grunt (Haemulon striatum) (Table 2.3).
Main trophic guilds within MCEs were zooplanktivore (Z), mobile invertebrate
feeder (MI), and piscivore (P); however some herbivores (H), sessile invertebrate feeders
(SI), and omnivores (O) were present at all depths (Figure 2.5, 1.6). Zooplanktivores
were the most diverse and abundant group (representing 36 % of all species and 52 % of
the fishes within 30 m2
transects, followed by MI’s (25 and 20%, respectively).

26
Piscivores were diverse and abundant, and their proportion increased with depth (Figure
2.6). However, this group was better characterized using data from roving surveys.
Table 2.3. Indicator Species Analysis with Monte Carlo test of significance (p-values)
indicating primary characteristic species of shallow and mesophotic habitats pooled
across six reefs in La Parguera shelf-edge, southwest Puerto Rico
The distribution of trophic guilds varied with depth (Table 2.4). Most of the fishes
at 30 m (shallow depth) were herbivores, representing 31 % of the species and 36% of the
fishes per transect. The MI and Z groups were also abundant (22 and 21% of the fishes
per transect, respectively), with the former more diverse (30 % of the species) than the
latter (16 %). Trophic groups O, SI and P were present with few species and low

27
A B
Figure 2.5. Mean trophic guild proportions observed in 30 m2
transects distributed along
a 20 to 70 m depth gradient, pooled across six reefs in La Parguera shelf-edge, southwest
Puerto Rico. Vertical line separates shallow and mesophotic zones. Data from 20 m was
taken from the CRES program (see Methods). Z: zooplanktivore, MI: Mobile invertebrate
feeder, H: herbivore, O: omnivore, P: piscivore, and SI: sessile invertebrate feeder. A)
Relative density of each trophic guild per 30 m2
. B) Relative species richness of each
trophic guild per 30 m2
Figure 2.6. Distribution of the mean species richness of trophic guilds in roving surveys
in a 40 to 70 m depth gradient pooled across six reefs in La Parguera shelf-edge,
southwest Puerto Rico. P: piscivore, MI: Mobile invertebrate feeder, H: herbivore
Table 4. Relative species richness (% R) and density (% D) of trophic guilds per 30 m2
,
O: omnivore, P: piscivore, and SI: sessile invertebrate feeder
densities. However, the dominance of H was rapidly lost with depth (Figure 2.5), except
for greenblotch parrotfish and cherubfish. From 30 to 40 m depth overall species richness
and densities decreased sharply (Figures 1.2 A, B), mostly due to the rapid reduction in

28
the number and densities of herbivores when moving from shallow to mesophotic depths
(Figure 2.5). The number of species of all trophic groups diminished in this depth range,
except for zooplanktivores, which instead increased in species number and densities
(Table 2.4, Figure 2.5). As a result, a dominance shift occurred at 40 m from herbivores
to zooplanktivores, a trend which became stronger with depth as herbivores continuously
decreased in species richness and density while zooplanktivores increased (Figure 2.5).
The other trophic groups remained relatively similar among all depths.
Table 2.4. Relative species richness (% R) and density (% D) of trophic guilds per 30 m2
,
O: omnivore, P: piscivore, and SI: sessile invertebrate feeder
Data from roving surveys unmasked a large representation of piscivores within
MCEs which was underestimated using only transect data. Roving surveys provided
information on another 22 species that were in the area but not within transects, and most
of them were piscivores (Table 1.2). The schoolmaster (Lutjanus apodus) was the most
abundant fish in roving surveys at 40 and 50 m depth, but its frequency and number

29
decreased from 60 to 70m. Dog snapper, on the contrary, became more frequent and
abundant until 60 m depth, where it was dominant in number. The upper limit of blackfin
snapper was 60 m, and its abundance increased with depth, becoming the dominant
piscivore at 70 m. In addition, roving surveys detected the presence of species now rarely
seen shallower in La Parguera, such as goliath (Epinephelus itajara), black, and
yellowfin (Mycteroperca venenosa) groupers, cubera snapper, the rainbow parrotfish
(Scarus guacamaya), sharks (e.g. reef shark, Sphyrna sp., Ginglymostoma cirratum), and
rare small species (basslets, cardinalfishes).
The NMS ordination on species composition and abundance converged on a
stable, 2-dimensional solution (final stress = 12.98, final instability = 0.000001) and
demonstrated greater structure than expected by chance (Monte Carlo procedure, p =
0.004) (Fig. 1.7 A). Ordination axes were independent (orthogonality = 93%) and
explained 90% of the fish assemblage variance. Axis 1 represented most of the variance
(59%) and was strongly associated with depth. Axis 2 represented an additional 32% of
the ordination variance, and though weaker, this axis was also influenced by depth. Thus,
shallow water sites were located in the lower right portion of Figure 2.7 A, while deep
water sites were located to the upper left. Sites with intermediate depths were located
between these extremes. The overlay of joint plots with trophic guild responses to
ordination axes showed herbivore and zooplanktivore strongly associated with axis 1 (r =
0.82, tau = 0.65, and r = 0.51, tau = 0.35, respectively) and axis 2 (r = 0.47, tau = 0.30,
and r = 0.15, tau = 0.05), and therefore with depth (Figure 2.7 B, C). Significant variation
in species composition and abundance for the entire fish assemblage was demonstrated
among depths (MRPP, p < 0.0001). Assemblages at 30 m and 70 m depth were

30
significantly different from all others (MRPP, all p < 0.01). Fishes from 40 and 50 m
were similar (MRPP, p = 0.18) but differed from deeper ones (MRPP, all p < 0.01); in
like manner fishes from 50 and 60 m were similar (MRPP, p = 0.09) but differed from
fishes at deeper and shallower depths
A
B C
Figure 2.7. A) Non-metric multidimensional scaling (NMS) plot based on Bray-Curtis
similarities of fish assemblages at six reefs in La Parguera shelf-edge, southwest Puerto
Rico, in a depth gradient from 30 to 70 m. NMS with herbivore (B) and zooplanktivore
(C) density distributions (circles size is proportional to density value), and trophic guilds
responses to ordination axes. H: herbivore, Z2: zooplanktivore excluding dominant
masked goby (Coryphopterus personatus), O: omnivore, P: piscivores depths (MRPP, p
< 0.0001).

31
2.4 Discussion
The MCE fish assemblage at 40 – 70 m off La Parguera was diverse and unique
because its taxonomic composition and abundance differed from that of shallower areas,
and varied with increasing depth. Results from NMS ordination and MRPP analysis
showed that the shallow assemblage at 30 m differed from all mesophotic assemblages.
Indicator species analysis (ISA) confirmed that 13 species were more abundant and
frequent at shallow depths than expected by chance, and 8 species were characteristic of
MCEs.
A wide variety of fishes inhabit MCEs off La Parguera, including species
common at shallow reefs (e.g., surgeonfishes), species scarce at shallow but common at
mesophotic depths (e.g., greenblotch parrotfish), and species restricted to deep areas (e.g.,
blackfin snapper). Overall, the pattern of “shallow” and “deep” species inhabiting
mesophotic depths is similar to that reported for other areas, e.g., Jamaica, Belize,
Bahamas (Colin 1976, 1974), Marshall Islands (Thresher and Colin 1986), Red Sea
(Brokovich et al. 2008), and Bajo de Sico and Isla Desecheo, Puerto Rico (García-Sais
2010, García-Sais et al. 2007). However, some differences were noted. For example,
juveniles of both blackfin snapper and striped grunt were part of the deep species
assemblage in our study (reaching up to 70 m depth); at MCEs in Jamaica and Belize
these juveniles occurred shallower and only adults were seen between 50 and 100 m
depth (Colin 1974). Mesophotic coral ecosystems are considered to be extensions of
shallower coral reefs; therefore the presence of shallow reef fishes at mesophotic depths
is possible if there is biological and physical connectivity between these ecosystems
(Hinderstein et al 2010). In our study, 25 species were classified as deep (Tables 1.1,

32
1.2), an increase of 15 species as indicators of mesophotic assemblages in Puerto Rico
over those previously described by García-Sais (2010); six of them were found in this
study because they live deeper than 50 m. Thus, similar to other localities (Khang et al
2010, McClain and Barry 2010, Rooney et al 2010), MCEs in La Parguera are valuable
habitats containing a high diversity of species that potentially connects with shallower
areas. Habitats with high diversity of species usually support high genetic variability,
functional redundancy and resiliency (Ives and Carpenter 2007, Walker and Salt 2006).
Therefore, to really understand the dynamics of reef fishes and manage their use and
conservation, a scale focus that includes mesophotic populations should be considered.
The proportion of shallow species rapidly decreased with increasing depth, and
this trend was evident at even shallower depths. Comparison of these results to data
collected at 20 m shows that the decline of abundance of shallow species within
mesophotic depths is a continuation of a trend that starts at shallow depths (Figure 4). At
the species level, those shallow species that declined the most in sighting frequency with
depth (e.g. bicolor damselfish, surgeonfishes, and redband parrotfish) were the same
species that had the highest sighting frequencies at 20 m depth (≥ 97%). Similarly, deep
species were not recorded at 20 m, except for relatively infrequent sightings of the
greenblotch parrotfish and longsnout butterflyfish. Even though decreases in the relative
abundance of shallow species with depth were gradual, the greatest decline occurred from
60 to 70 m. Several shallow species (34, out of 79) did not occur below 60 m, so that this
was also the depth range over which deep species become dominant within the
mesophotic fish assemblage (Figure 4). Despite the disappearance of shallow species
with increasing depth, overall species richness and mean densities were conserved over a

33
broad depth range due to the appearance of deep species. However at 70 m both species
richness and especially density decreased. Most deep species were solitary, and shallow
species that reached deeper zones were present in far lower densities.
Sharp changes in the fish assemblage between 60-70 meter depth were consistent
with variations in the composition and abundance of the mesophotic coral and algal
assemblages at these areas (Sherman et al. 2010). Coralline algae are more abundant at
deeper zones of MCEs in La Parguera, especially below 60 m where they are the
dominant algal taxa (Sherman et al. 2010). In contrast, abundance of non-calcareous
algae abruptly declines between 60 and 70 m, and those that do extend into deeper depths
often undergo morphological change. For example, the low-light adapted alga Lobophora
variegata (Runcie et al. 2008) has a foliose form in the upper 60 m in La Parguera but
below that depth it is only found in encrusting form (Ruíz & Ballantine pers. comm.). In
corals, the dominant species (68% of overall coral cover) at 70 m was Agaricia undata,
where it forms large plate-like colonies, but it was rare (2.5%) at shallower depths
(Sherman et al. 2010). Important changes in physical and/or biological parameters may
determine this particular depth limit for multiple taxa. For example, the low light
conditions that result from the steep decreases in light irradiance and spectral quality with
depth can be utilized by only those few autotrophic organisms that can adapt to increase
their photosynthetic efficiency (Lesser et al. 2010). To maximize light capture, some
deep corals and algae can increase their surface area by lateral spreading, growing in
flattened morphologies (Aponte & Ballantine 2001, Hanisak & Blair 1988, Kuhlmann
1983) or increasing colony size (Weil pers. comm.). However, despite the presence of
low-light adapted algae below 60 m, light levels are probably too low below this depth

34
off La Parguera to maintain sufficient production to support benthic herbivores, and
therefore changes in the food chain are expected. On the other hand, nutrients and
particulate matter that may be imported from rich deep waters to mesophotic depths by
upwelling and internal waves (Leichter & Genovese 2006, Leichter et al. 2003, 1996)
may provide the main source of energy in mesophotic habitats to favor plankton
production (Lesser 2006), and therefore benefit planktivores and filter feeders.
Additionally, changes in temperature, sedimentation, geomorphology, competition, and
predation also influence MCEs ecology (Kahng et al. 2010). Thus, the resources needed
by most common shallow species in La Parguera shelf edge were in sufficient supply up
to 60 m, but below this depth a different ichthyofauna became dominant. NMS ordination
and MRPP analysis results showed that assemblages at 70 m depth were dissimilar from
all other mesophotic assemblages (p < 0.005) (Figure 2.7 A).
As a consequence of the above transition, the trophic structure of the fish
assemblage at MCE depths differed from that of shallower reefs. Zooplanktivores were
strongly dominant at MCEs, representing more than half of the fishes within the 30 m2
transects. At shallow reefs, zooplanktivores only represented 19% of the fishes and the
predominant guild was herbivores, a group that was scarce at MCEs (Figure 2.5) and that
represented as low as 6% of the fishes within transects at 70 m depth. Herbivores (e.g.
surgeonfishes), mostly reached the upper (40-50 m) but not the deeper (60-70 m) portion
of MCEs, and their densities decreased markedly with depth. This relative scarcity of
herbivorous fish is characteristic within MCEs (e.g. Brokovich et al. 2010, 2008, García-
Sais et al. 2008, Lidell & Ohlhorst 1988, Van den Hoek et al. 1978, Gilmartin 1960). At
the same time, while algae are less diverse and grow slower in MCEs (Brokovich et al.

35
2010), they maintain high abundances (Sherman et al. 2010) due in part to low grazing
pressure (Brokovich et al. 2010, Morrison 1988, Liddell & Ohlhorst 1988, Van den Hoek
et al. 1978). Several hypotheses have been postulated to explain the low representation of
herbivores in MCEs, e.g., low algal nutritional value, palatability or digestibility. While
changes in algal nutritional characteristics with depth are still unproved (Clements et al.
2009), some of the genera occurring in high abundances at MCEs, such as Lobophora or
Halimaeda, although edible (Colin 1978), are known to be less palatable to most
herbivorous fishes (Duffy & Hay 1990). In addition, large grazers concentrate in zones of
high rates of algal turf production (Russ 2003); therefore the low number of these fish
within MCEs may be related with the slower growth and less turn-over of the algae at
these depths. Yet, some herbivorous species had their maximum occurrence at the
deepest depths surveyed, but these were only small-bodied herbivores, such as the
greenblotch parrotfish and cherubfish, which were the most abundant at 70 m. These
fishes may have found an ecological niche at these depths where the low productivity of
algae on which they feed might be sufficient for their energetic demand given their small
size. By grazing deep MCEs, these small fishes do not need to compete with larger
grazers for food.
Does the decline of herbivorous fish with depth affect the algal community within
MCEs? In this study, herbivore relative abundance decreased exponentially with depth
(R2
= 0.97, Figure 2.8). This trend is opposite to that of turf and coralline algae, which
increase from 20 to 60 m depth and 46 to 76 m, respectively (Ruíz & Ballantine pers.
comm.). Nevertheless, coralline algae were relatively abundant in 20 m reefs. The latter
results are in agreement with the relative-dominance model (Russ 2003) developed for

36
shallow reefs (20 m), which states that high grazing leads to a reduction in turf and an
increase in coralline algae (Figure 2.8). This suggests that fish grazing pressure at
shallow areas in the shelf-edge of La Parguera is at least in part shaping the algal
community. This pattern is lost with increasing depth as both the % cover of coralline
algae and turf increase as herbivores decline. Changes in turf might be reflecting lower
grazing pressure but those of coralline algae do not. Moreover, at 70 m depth, where the
minimum herbivore density occurs, turf abundance is low but coralline algae are
dominant, similar to shallow depths where grazing pressure is high. This lack of
correlation of herbivore density with changes in the algal community with depth suggests
that factors other than grazing pressure (e.g., light availability) are strongly influencing
algal communities at these deeper depths. Coralline algae are low light adapted species
(Runcie et al. 2008) and constituted the dominant algae group between 60 and 120 m
depth in Jamaica (Aponte & Ballantine 2001).
Figure 2.8. Mean cover (%) of coralline algae, Lobophora and turf, and relative density
(%) of herbivore fish per depth, along a gradient from 20 to 70 m at six reefs in La
Parguera shelf-edge, southwest Puerto Rico. Algal and fish data from 20 m were taken
from the CRES program (see Methods)

37
Thus, grazing pressure seems to partially influence algal distribution, but changes
in algal community structure and productivity with depth (Ballantine & Ruíz 2011) might
also influence the distribution of herbivorous fishes (Nemeth and Appeldoorn 2009). In
addition, temperature has been suggested as a factor potentially controlling herbivorous
distribution (Floeter et al. 2005), so that in the cooler waters of MCEs lower herbivorous
abundance (Smith 2008) and reduced grazing rates (Leichter et al. 2008) would be
expected. However, temperature differences between shallow and mesophotic depths are
not always big enough to drive herbivorous distribution (Brokovich et al. 2010). Both,
herbivores and zooplanktivores were strongly associated with NMS ordination axes,
demonstrating major and opposite changes in composition and abundance along the depth
gradient (Figure 2.7 B, C). The reduction in herbivores with depth was only compensated
by increases in zooplaktivores, and not by increases across the other trophic guilds.
Plankton thus seems to become the main source of energy for mesophotic fishes (Kahng
et al 2010) and replaces benthic primary production as the base of the food chain. These
results are consistent with descriptions of fish assemblages of MCEs from other localities
(García-Sais 2010, Fetoiza et al 2005, Thresher and Colin 1986).
The greater extent of piscivores within MCEs as evidenced in the roving surveys,
and the change in composition from medium size species (e.g., schoolmaster) shallow to
larger size species deeper (e.g., dog snapper, sharks) suggest that MCEs in La Parguera
are relatively healthy ecosystems that still sustain a balance of functional groups that
includes top predators. Top-down control in the fish population exerted by apex predators
acts to reduce the dominance of a few species, allowing a more diverse ichthyofauna. The
integrity of functional groups is a crucial component of ecosystem stability and resilience

38
(Bellwood et al. 2004). Thus, these MCEs may have sufficient resilient capacity to avoid
a lionfish (Pterois volitans) invasion and its detrimental impacts as reported in MCEs in
the Bahamas (Lesser & Slattery 2011), and potentially to contribute to shallow reef
resilience. Currently, lionfish are not common at mesophotic depths in La Parguera, and
understanding the role of a healthy predator guild on the potential for lionfish
colonization would be an interesting area of future research.
Mesophotic coral ecosystems in La Parguera are subject to a lower impact from
fisheries than shallow reefs given the difficulty of targeting a steep narrow area with
prevailing onshore winds and currents, and because the bulk of the deep commercial
fishery targets the Mona Channel. However, in shallow areas, large size snappers and
groupers have become much reduced as they have long been the main fishery targets in
Puerto Rico (Matos-Caraballo 2004). The presence of these commercial species within
MCEs suggests that these habitats play a key refuge role, and therefore are essential for
the conservation of these threatened species. Research and monitoring of MCE fish
assemblages are critical to enhance our knowledge of species composition, accurately
assess the stocks of important commercial species and to compare ecological processes
with shallow reefs.
Acknowledgements
Thanks to deep divers Michael Nemeth, Milton Carlo, Hector J. Ruíz and Clark Sherman
for assisting in field sampling. This work was supported by the National Oceanic and
Atmospheric Administration’s Center for Sponsored Coastal Ocean Research
(NOAA/CSCOR) (Grant #: NA06NOS4780190), through the Caribbean Coral Reef

39
Institute (CCRI). This project used the facilities of the University of Puerto Rico, and the
University of North Carolina – Wilmington’s Undersea Research Center oversees the
training of rebreather divers.
Literature Cited
Aponte NE, DL Ballantine (2001) Algal species distribution and composition as a factor
of depth on the deep insular slope at Lee Stocking Island, Bahamas. Deep Sea
Res. Part I 48:2185-2194
Ballantine DL, NE Aponte (2005) An annotated checklist of deep-reef benthic marine
algae from Lee Stocking Island, Bahamas II. Rhodophyta. Nova Hedwigia
80:147-171
Ballantine DL, H Ruíz, NE Aponte (2009) Notes on the benthic marine algae of Puerto
Rico IX. Aditions to the flora including reports of three species for the first time
in the Atlantic Ocean. Botanica Marina 52:229-235
Ballantine DL, H Ruíz (2010) Two new deepwater Peyssonnelia species, Peyssonnelia
iridescens and Peyssonnelia gigaspora (Peyssonneliaceae, Rhodophyta) from
Puerto Rico, Caribbean Sea. Phycologia 49(6):537–544
Ballantine DL, H Ruíz (2011) A new encrusting deep-water coral reef alga, Peyssonnelia
incomposita (Peyssonneliaceae, Rhodophyta), from Puerto Rico, Caribbean Sea.
Cryptogam Algol 32(1):19-26
Ballantine DL, A Athanasiadis, H Ruíz (2011) Notes on the benthic marine algae of
Puerto Rico. X. Additions to the flora. Botanica Marina 54:293-302
Battista TA and ML Stecher (2006) Data acquisition and processing report, NF-06-03,
S-I911-NF-06, March 21-April 2, 2006, U.S. Virgin Islands and Puerto Rico,
NOAA Ship Nancy Foster 104
Bellwood DR, TP Hughes, C Folke, N Nystrom (2004) Confronting the coral reef crisis.
Nature 429: 827–833
Brock VE (1954). A Preliminary Report on a Method of Estimating Reef Fish
Populations. The Journal of Wildlife Management 18(3):297-308
Brokovich E, S Einbinder, N Shashar, M Kiflawi, S Kark (2008) Descending to the
twilight-zone: changes in coral reef fish assemblages along a depth gradient down
to 65 m. Mar Ecol Prog Ser 371:253–262

40
Brokovich E, I Ayalon, S Einbinder, N Segev, Y Shaked, A Genin, S Kark, M
Kiflawi(2010) Grazing pressure on coral reefs decreases across a wide depth
gradient in the Gulf of Aqaba, Red Sea. Mar Ecol Prog Ser 399:69–80
Clements KD, D Raubenheimer, JH Choat (2009) Nutritional ecology of marine
herbivorous fishes: ten years on. Funct Ecol 23:79–92
Colin PL (1974) Observation and collection of deep reef fishes off the coasts of Jamaica
and Honduras. Mar Biol 24(1):29–38
Colin PL (1976) Observation of deep reef fishes in the Tongue-of-the-Ocean, Bahamas.
Bull Mar Sci 26:603–605
Colin PL (1978) Caribbean reef invertebrates and plants. TFH Publications, Neptune
City, NJ
Corgosinho PHC, NV Schizas (in review) Archeolourinia, a new genus and species of
Louriniidae Monard, 1927 (Copepoda, Harpacticoida) from a Caribbean
mesophotic zone. Journal of the Marine Biological Association of the United
Kingdom
Duffy JE, ME Hay (1990) Seaweed adaptations to herbivory. Bioscience 40:368-375
Feitoza BM, RS Rosa, LA Rocha (2005) Ecology and zoogeography of deep-reef fishes
in northeastern Brazil. Bull Mar Sci 76:725–742
Floeter SR, MD Behrens, CEL Ferreira, MJ Paddack, MH Horn (2005) Geographical
gradients of marine herbivorous fishes: patterns and processes. Mar Biol
147:1435–1447
García-Sais JR, J Castro, M Sabater, M Carlo (2004) Monitoring of coral reef
communities from Isla de Vieques, Puerto Rico. Final Report submitted to the
Department of Natural and Environmental Resources (DNER), US Coral Reef
National Monitoring Program, NOAA. P 118
García-Sais JR, Castro R, Sabater J, Carlo M (2007). Characterization of benthic
habitats and associated reef communities at Bajo de Sico Seamount, Mona
Passage, Puerto Rico. Final Report submitted to Caribbean Fishery Management
Council (CFMC/NOAA). P 91
García-Sais JR, RS Appeldoorn, T Battista, L Bauer, A Bruckner, C Caldow, L Carrubba,
J Corredor, E Diaz, C Lilyestrom, G García-Moliner, E Hernández-Delgado, C
Menza, J Morell, A Pait, J Sabater, E Weil, E Williams, S Williams (2008) The
state of coral reef ecosystems of the commonwealth of Puerto Rico. In: Waddell J,
A Clarke (eds) The state of coral reef ecosystems of the United States and Pacific

41
Freely Associated States: 2008 NOAA Technical Memorandum NOS NCCOS 78.
NOAA/NCCOS Center for Coastal Monitoring and Assessment’s Biogeography
Team, Silver Spring, MD, P 75–116
García-Sais JR (2010) Reef habitats and associated sessile-benthic and fish assemblages
across a euphotic–mesophotic depth gradient in Isla Desecheo, Puerto Rico. Coral
Reefs 29:277–288
Gilmartin M (1960) The ecological distribution of the deep water algae of Eniwetok
Atoll. Ecology 41:210–221
Hanisak MD, SM Blair (1988) Deep-water benthic macroalgal communities with
emphasis on Florida and the Bahamas. In: Agegian CR (ed) Biochemical cycling
and fluxes between the deep euphotic zone and other oceanic realms p 61-83.
National Undersea Research Program Research Report 88-1. National Oceanic
and Atmospheric Administration (NOAA), Department of Commmerce,
Washington, DC
Hinderstein LM, JCA Marr, FA Martinez, MJ Dowgiallo, KA Puglise, RL Pyle, DG
Zawada, RA Appeldoorn (2010) Theme section on “Mesophotic Coral
Ecosystems: Characterization, Ecology, and Management”. Coral Reefs 29:247-
251
Itzikowitz M, M Haley, C Otis, D Evers (1991) A reconnaissance of the deeper
Jamaican coral reef fish communities. NE Gulf Sci 12:25-34
Ives AR and SR Carpenter (2007) Stability and diversity of ecosystems. Science 317:58–
62
Kahng SE, JR García-Sais, HL Spalding, E Brokovich, D Wagner, E Weil, L Hinderstein,
RJ Toonen (2010) Community ecology of mesophotic coral reef ecosystems.
Coral Reefs 29:255–275
Kuhlmann D (1983) Composition and ecology of deep-water coral associations. Helgol
Mar Res 36:183-204
Leichter JJ, SR Wing, SL Miller, MW Denny (1996) Pulsed delivery of subthermocline
water to Conch Reef (Florida Keys) by internal tidal bores. Limnol Oceanogr
41:1490-1501
Leichter JJ, HL Stewart, SL Miller (2003) Episodic nutrient transport to Florida coral
reefs. Limnol Oceanogr 48:1394-1407
Leichter JJ, SJ Genovese (2006) Intermittent upwelling and subsidized growth of the
scleractinian coral Madracis mirabilis on the deep fore-reef slope of Discovery
Bay, Jamaica. Mar Ecol Prog Ser 316:95–103

42
Leichter JJ, MD Stokes, SJ Genovese (2008) Deep water macroalgal communities
adjacent to the Florida Keys reef tract. Mar Ecol Prog Ser 356:123–138
Lesser MP (2006) Benthic–pelagic coupling on coral reefs: feeding and growth of
Caribbean sponges. J Exp Mar Biol Ecol 328:277–288
Lesser MP, M Slattery, M Stat, M Ojimi, R Gates, A Grottoli (2010)
Photoacclimatization by the coral Montastraea cavernosa in the mesophotic zone:
light, food, and genetics. Ecology 91:990–1003
Lesser MP, M Slattery (2011) Phase shift to algal dominated communities at mesophotic
depths associated with lionfish (Pterois volitans) invasion on a Bahamian coral
reef. Biol Invasions 13:1855–1868
Liddell WD, SL Ohlhorst (1988) Hard substrata community patterns, 1–120 M, North
Jamaica. Palaios 3:413–423
Matos-Caraballo D (2004) Puerto Rico/NMFS cooperative fisheries statistics program.
April 2001-March 2004. Final report submitted to the National marine fisheries
service by the fisheries research laboratory, Department of natural and
environmental resources, San Juan
McClain CR and JP Barry (2010) Habitat heterogeneity, biogenic disturbance, and
resource availability work in concert to regulate biodiversity in deep submarine
canyons. Ecology, 91:964-976
McCune B and M Mefford (2006). PC-ORD. Multivariate analysis of ecological data.
Version 5.0. MjM Software, Gleneden Beach, Oregon, USA
Morrison D (1988) Comparing fish and urchin grazing in shallow and deeper coral reef
algal communities. Ecology 69:1367–1382
Nelson WR and RS Appeldoorn (1985) A submersible survey of the continental slope
of Puerto Rico and the US Virgin Islands, 1–23 October 1985. Cruise report, R/V
Seward Johnson, National Marine Fisheries Service, Pascagoula, Mississippi
Laboratories, P 76
Nelson JS (2006) Fishes of the world. Fishes of the world. John Wiley and Sons, Inc.
New York. 4th edition. P 601
Nemeth M and R Appeldoorn (2009) The distribution of herbivorous coral reef fishes
within fore-reef habitats: the role of depth, light and rugosity. Caribb. J. Sci..
45:247-253
Pesic V, T Chatterjee, NV Schizas (in review) A new species of Pontarachna (Acari,

bejarano_rodriguez_2013_phd_mesophotic_fishes

bejarano_rodriguez_2013_phd_mesophotic_fishes

More Related Content

bejarano_rodriguez_2013_phd_mesophotic_fishes