|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
INFLUENCE OF HIGH-RESOLUTION SPATIAL INFORMATION ON RESOURCE
EXPLOITATION: AN EXAMPLE FROM ANGLER IMPACTS ON ARTIFICIAL
STEPHEN J. LARSEN
A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF
FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
Stephen J. Larsen
I thank my advisor Dr. William J Lindberg for introducing me to the Gulf of
Mexico ecosystems and encouragement to see and think about the big picture. This
research and the completion of my graduate career would not have been possible without
his guidance, optimism, and perspective on the project at hand. I thank Dr. Debra Murie
for serving on my committee and for her guidance and input throughout my graduate
experience. I thank Dr. Tom Frazer for serving on my committee as well as providing
perspective on what the important things are and not to lose sight of them.
Enormous thanks go out to Dr. Ken Portier for his help with the statistical aspects
of this project. Thanks also to Dr. Mike Allen for statistical guidance and the
encouragement to complete the degree for there are bigger and better things following
graduation. Special thanks go to lab-mates Doug Marcinek, Mark Butler, and Brian
Nagy who collectively welcomed me into the Lindberg Lab, made my transition to
graduate school easy and provided a great sounding board for ideas, discussions, and
frustrations. Additional thanks go to graduate students Rick Kline, Mark Rogers, and
Paul Anderson for always being willing to give time even when they had none to give. In
fact every graduate student who passed through the carrels of the Fisheries and Aquatic
Sciences Department from August, 2001, through Fall, 2005, contributed to a great work
environment and community of friends that is regrettably so temporary in nature.
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ......... .................................................................................... iii
L IST O F T A B L E S ..................................................................... .....................
L IST O F FIG U R E S .... .............................. ....................... .......... ............... vi
ABSTRACT .............. ..................... .......... .............. vii
1 IN TRODU CTION ................................................. ...... .................
2 M E TH O D S .................................................................9
Experim mental System .................. .............................. ....... ...............
Study Organism ....................................................................... ......... 10
S ou rces of D ata ............................................................................... 1 1
Suitability of D ata...................................................... 12
E xperim mental D design ........................ ................ .. ......... ....... 13
S statistic al A n aly se s .....................................................................................................13
3 R E S U L T S ........................................................................................................1 6
L egal G ag .....................................16
Pre-publication Time Period (Initial Conditions) .............................................16
Transition from "Pre" to "Acute" Time Periods .............................................17
W within "A cute" .................................. .......................... ..... ............ 17
Transition from "A cute" to "Chronic" ............ ................................ ...............18
W within "Chronic" .................. ................................... .... ........... .18
Sub-L egal G ag ................................................................ .. .... ......... 19
4 DISCUSSION .................. .................................... ........... .... .......... 27
REFERENCES ................... ......... .. ...... ... ..................37
B IO G R A PH IC A L SK E TCH ..................................................................... ..................44
LIST OF TABLES
1. Type 3 tests of fixed effects for legal gag ................................. .......... ...... ......... 21
2. Type 3 tests of fixed effects for sub-legal gag.....................................................22
LIST OF FIGURES
1. Design of the Suwannee Regional Reef System ................................ ..................... 15
2. Pre-publication distribution of legal gag grouper on the SRRS. Data are drawn
from not-to-be-published arrays due to the influence of pre-publication
perturbations on to-be-published arrays. ........................................ ............... 23
3. Chronology of impacts to legal gag on 16-cube arrays of the SRRS by array
architecture and publication status. ............................................... ............... 23
4. Chronology of impacts to legal gag on four-cube arrays of the SRRS by array
architecture and publication status. ............................................... ............... 24
5. Abundance of legal gag per array within the "acute" period by array architecture
and publication status. .................................... ...............................24
6. Abundance of legal gag per array within the "chronic" period by array architecture
and publication status. .................................... ...............................25
7. Abundance of sub-legal gag per array, by array architecture and publication status,
averaged across all tim e periods........................................ ........................... 25
8. Abundance of sub-legal gag per array, by array architecture on (A) published and
(B ) unpublished arrays. ...................... .. ...... .............. ...................... .....26
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
INFLUENCE OF HIGH-RESOLUTION SPATIAL INFORMATION ON RESOURCE
EXPLOITATION: AN EXAMPLE FROM ANGLER IMPACTS ON ARTIFICIAL
Stephen J. Larsen
Chair: William J. Lindberg
Major Department: Fisheries and Aquatic Sciences
Exploitation of natural resources is dependent on knowing where to find those
resources. Increasingly specific and detailed information regarding the location and
relative value of many resources is being continually developed by rapidly advancing
technologies in remote sensing. It is generally assumed that such information allows
more efficient and directed exploitation of those resources. I evaluated impacts of the
public release of detailed geographic information regarding artificial reefs designed for
gag grouper (Mycteroperca microlepis). Reefs were constructed of concrete cubes
approximately 90 cm on a side. Size of reef patches (4-cube and 16-cube, small and large
respectively), and distances among patches (25 m and 225 m), were examined.
Approximately half of the reef locations were published to the angling community. Reefs
were monitored from two years prior to publication to two years-post publication and
again six to eight years post-publication.
Impacts on published reef arrays were highest on large, closely spaced reef arrays,
and least on small, widely spaced reef arrays. In contrast to pre-publication trends, large
reef patches, once published, no longer held more legal gag (>508 mm) than small
patches. Six to eight years after publication sub-legal gag (<508 mm) were reduced on
the largest, most closely spaced arrays, indicating that these reefs received the highest
fishing effort. Large unpublished reefs as well as small, widely spaced reefs showed
effects of discovery by anglers in the period six to eight years after other reefs in the
study system were made publicly known. The disclosure of locations of previously un-
targeted, high-quality habitat led to immediate exploitation of legal gag, with the
exception of apparent refuges of small habitat patches, which are inherently more
difficult to locate and to fish. Resource managers should take into account the improved
efficiency of exploitation made possible by the collection and dissemination of high-
resolution geographic habitat data.
Accurate and rapid collection of spatially explicit data regarding the natural
environment has been made possible over recent decades by the development of new
technologies. Advances in satellite imaging and sensing systems as well as sonic and
laser technologies are providing unprecedented detail and information. The utility of
high-resolution spatial data to both resource managers and user groups has been clear. It
has made possible the quantification of changes in land use and land cover to describe
processes such as deforestation and desertification (Pamo, 2004; Leimgruber et. al.,
2005) and shoreline changes (Morton et. al., 2005) with much greater accuracy than prior
techniques, provided for far more proficient and directed use of natural resources (Pickrill
and Todd, 2003), and has opened lines of scientific inquiry previously limited by the
technology available (Wright, 1999; Andersen et. al., 2005; Leimgruber et. al., 2005).
In the sea, detailed surveys have been conducted through the use of multi-beam
sonar systems (Verlaan, 1997; Pickrill and Todd, 2003) as well as the LIDAR (Laser
Imaging Detection and Ranging) optical system (Kincade, 2003). Coupled with
Geographic Positioning Systems (GPS), data can be placed very precisely in a spatial
context. Such detailed information has already demonstrated its utility in the
management, exploration, and understanding of the marine environment. In coastal
environments, decisions regarding location of sewage outfalls as well as cable and pipe-
line routes have been well founded (Pickrill and Todd, 2003). Side-scan sonar has been
used to identify mineral deposits on the ocean floor (Lee and Kim, 2004) and investigate
the geo-physical properties of hydrothermal vents and areas of sea-floor spreading
(Wright, 1999). In fisheries, LIDAR has been combined with video technology for the
remote sensing of schools of capelin (Mallotus villosus) and zooplankton densities
(Brown et. al. 2002). Detailed maps were created of Brown's Bank in the Northern
Atlantic, which were then coupled with biological data to identify key scallop habitats
(Kostylev et. al., 2003). As a consequence, commercial trawlers were able to reach
quotas in as little as one quarter the time while avoiding fragile or unproductive regions
(Pickrill and Todd, 2003).
The availability of such detailed geographic data is clearly a boon to resource
managers and those working in the natural environment. Mack (1990) noted, however,
that there are unintended uses of new technologies that may only become apparent once a
program is well underway. For example, fishery resources are inherently patchy in space,
and detailed geographic information collected to other ends has the potential to facilitate
their exploitation. In the example of the commercial scallop fishery on Brown's Bank
(Kostylev et. al., 2003; Pickrill and Todd, 2003), a quota limited the harvest. New
technologies provided information facilitating more efficient harvest without increasing
the overall effect on the fishery, and possibly minimized impacts to benthic habitats by
reducing the area trawled. Had there been no quota, continued efficient and directed
efforts could have led to much greater yields and thus potential overexploitation in the
fishery. The development of information increasing the potential for efficient
exploitation of natural resources, if not coupled with proper management strategies,
creates the possibility of over-exploitation of those resources.
Like the scallop, many marine fishery species are associated with specific habitat
types. The current efforts in sea-floor mapping hold the potential of identifying and
describing those habitats in great detail, leading to far better understanding of population
distributions and identification of essential fish habitats. While clearly useful to
managers, to date, there are few documented examples of how fishermen use detailed
geographic information of fish habitat. Identification of densely populated habitat among
recreational fishermen is commonly achieved through large expenditures of effort in
exploring new areas and searching with the limited bottom mapping information and
technologies available to the recreational sector. Due to the investment in locating
fishing sites, once found, such areas are rarely made public. Although maps are created
and bathymetric data are available to recreational anglers, they are typically of coarse
scale and limited scope.
The Ideal Free Distribution (IFD), as applied to the distribution of angling effort
(Walters and Martell, 2004), predicts that effort will be directed to the most profitable
fishing grounds. One assumption of the IFD is complete knowledge of fishing grounds
and the relative profitability of each. Increase in awareness of available fishing grounds,
as is made possible by detailed bottom mapping, will likely lead to a distribution of
angling effort closer to that predicted by the IFD. The predicted consequences are an
increase in catch-per-unit-effort (CPUE) in the short-term, and hyper-stability in the long-
term as detailed geographic information facilitates exploitation in the face of declining
Due to the expanse of marine environments and limited ability of recreational
anglers to accurately perceive the sea floor, habitat extent is likely a key factor in
determining its probability of discovery. Small, isolated habitat patches that support
dense local populations of fish are difficult to locate by conventional means, rarely shared
if they are discovered, and are likely key habitat features providing refuge for targeted
marine species. The uncontrolled dissemination of fine-scale bathymetry could reveal
such locations in detail, leading to better-directed and more efficient exploitation of
stocks that are patchy in space.
The importance of such refuges is largely unknown, but is potentially critical in the
maintenance of a viable fishery. The impacts of both recreational and commercial fishing
on populations and assemblages of marine fish have been well documented (Haedrich
and Barnes, 1997; Wantiez et. al., 1997; Bianchi et. al., 2000; Roberts et. al., 2001;
Bremner et. al., 2003; Coleman et. al., 2004). Fishing generally leads to fewer and
smaller fish of the targeted species (Haedrich and Barnes, 1997; Grandcourt, 2003) as
well as reduced overall diversity (Albaret and Lae, 2003; Ley et. al., 2002). Protection
from fishing, in some cases, has reversed these trends (Roberts, 1994; Wantiez et. al.,
1997) and resulted in increased abundance and size of targeted fish as well as increased
biodiversity in protected areas. Refuges provide fish the opportunity to become older and
larger. Such individuals are of great importance to the welfare of a fishery (Birkeland
and Dayton, 2005). For some species, larger and older individuals have been shown to
have exponentially higher fecundity (Berkeley et. al., 2004a) and to produce healthier,
faster-growing larvae (Conover and Munch, 2002; Berkeley et. al., 2004b), as well as
preserving genetic diversity (Hauser et. al., 2002).
As large-scale sea-floor mapping programs are relatively new, studies of such
natural, small, isolated, but high quality, habitat features are rare. The use of artificial
reef structures to create such locales provides the opportunity to control for habitat
features (size, shape, and distribution) that may influence the resulting assemblage of
species as well as the exploitation of the resident species by fishers.
The specific characteristics of artificial habitat along with reef placement (Coll et.
al., 1998) strongly influence the resulting assemblage of marine life. There are two
characteristics of artificial reef use by marine fish that have been widely studied: as
shelter and as a food source. Available refuge has been demonstrated to be the dominant
structuring influence on the species resident on a reef (Williams and Sale, 1981; Doherty
and Sale 1985; Sweatman, 1985; Hixon and Beets, 1989; Eklund, 1997; Lindberg in
press) although resident species clearly use the reef as a forage base (Herrera et. al., 2002;
Szedlmayer and Lee, 2004). Many top-level predators are transient groups (e.g., jacks,
sharks, tunas, mackerels) and would be less-frequently observed at artificial reefs despite
their potential for playing a major role in the structuring processes of the resident fish
assemblage (Bohnsack et. al., 1994; Carr et. al., 2002).
The use of artificial reefs as refuge is dependent on the shelter characteristics of the
structure provided. Many studies have investigated the effects of reef size and design on
resulting fish assemblages. Several general trends have been observed. Larger reef
structures support higher abundances of fish (Rountree, 1989; Lindberg and Loftin, 1998)
though not necessarily at higher densities (per unit reef area) (Bohnsack et. al., 1994;
Lindberg and Loftin, 1998). Bohnsack et. al. (1994) found that although there are lower
densities at larger reefs, the resident fish are larger on average than on smaller reefs. In
contrast, Lindberg et. al. (in press) observed higher growth rates by gag, Mycteroperca
microlepis, on smaller artificial reef patches. These patterns are likely dependent on the
species and system being studied. Size of refuges within a reef has been shown to
positively influence the abundance of fish suited to the size of refuge provided (Frazer
and Lindberg, 1994; Hixon and Beets, 1989; Hart, 2002). Fish tend to prefer openings
close to their own body size (Shulman, 1984). Within reefs, higher habitat complexity
increases diversity of the resulting assemblage as adequate shelter is provided for a larger
range of species (Charbonnel et. al., 2002; Sherman et. al., 2002; Gratwicke and Speight,
2005). Although resident species also use the reef and surrounding area as a forage base
(Herrera et. al., 2002; Szedlmayer and Lee, 2004), several studies have concluded that
refuge is the over-riding factor determining abundance of resident fish on artificial reefs
(Ecklund, 1997; Lindberg et. al. in press).
For recreational anglers, artificial reefs are widely-known and publicized, easily
locatable angling sites. In public surveys, 6.4% (McGlennon and Branden, 1994) to 87%
(Ditton and Graefe, 1978) of anglers reported fishing on artificial reefs. The lower end of
the spectrum resulted in a fishing intensity (angler hours/unit area) 92-171 times greater
than observed on natural substrates (McGlennon and Branden, 1994). While
expectations of high catch rates are a motivating factor for selecting artificial reefs
(Milon, 1989) only 5 of 27 comparisons by McGlennon and Branden (1994) yielded
significantly higher catch rates on artificial habitats, and these were all for pelagic
species. Solonsky (1985) demonstrated the significant impact that anglers can have on
recreationally targeted game fish on an artificial reef in Monterey Bay. One central reef
amidst several others was marked for recreational fishing. After three years, there were
significantly fewer rockfish on the marked reef relative to the unmarked reefs
surrounding it. Furthermore, tagged fish on the surrounding reefs were observed to move
onto the marked reef, but none were observed leaving the marked reef, demonstrating the
function of a known artificial reef to attract fish thus increasing their vulnerability to
fishing. Turpin and Bortone (2002), who observed artificial reefs before and after they
were displaced by hurricanes, also observed a significant increase in lengths of gag, and
red snapper, Lutjanus campechanus, on these reefs following displacement as anglers no
longer knew their locations. The use of artificial reefs in the study reported here offers
the additional dimension of high attraction for recreational anglers.
The purpose of this study was to test how the public release of detailed location
information regarding patchy habitat impacted the standing stock of gag, a dominant reef-
associated, recreationally-targeted species. The specific objectives were: (1) to determine
how publication of reef locations to anglers differentially affected abundances of both
legal and sub-legal gag across different reef architectures (sizes and spacings), and (2) to
determine the impacts of angling over a broad time-scale, including the discovery and
exploitation of unpublished reef locations. This evaluation of angling impacts as a
function of habitat patchiness and location information provides fishery managers with
experimental results pertinent to the management of fisheries in light of the ever-
increasing efficiencies that anglers garner through new and developing technologies.
Publication was predicted to lead to rapid reductions of legal gag. A refuge effect
on more-widely spaced patches was predicted as these are more difficult to fish as a
single unit. Angling impacts were predicted to be greatest on 16-cube arrays as anglers
are likely to favor these under the conception that larger habitat patches hold more fish.
Discovery of unpublished arrays was expected to be a function of both patch size and
spacing. Larger and more-widely spaced patches provide the greatest probability of
encounter for searching anglers. Discovery of a single patch should lead to discovery of
the entire array as anglers were aware of the reef layouts within the experimental system.
Several mechanisms likely affect sub-legal gag abundance. Compensatory responses to
predicted decreases in legal gag abundances would lead to an increase in sub-legal gag
abundance. Induced immigration and mortality from catch and release would lead to
declines in sub-legal gag abundance. Large, closely-spaced arrays were predicted to have
the highest fishing effort, thus they were expected to demonstrate the relative importance
of these mechanisms.
The Suwannee Regional Reef System (SRRS) was constructed from 1990 to 1993
in the northeastern Gulf of Mexico, approximately 20-25 km from the mouth of the
Suwannee River (Figure 1A). The building blocks of the SRRS were 89 cm x 89 cm x 89
cm concrete cubes with 60 cm diameter, cylindrical horizontal passages through the
centers and 10 cm diameter, horizontal, right-angle tunnels through each of the eight
corners (Figure 1B). Cubes were arranged in either 4-cube or 16-cube square reef
patches, such that central passages all run in the same direction. Patches were arranged
hexagonally, and were spaced 25, 75 or 225 m apart, constituting a reef array (Figure
1B). Reef arrays were spaced one to two km apart along the 13 m depth contour.
Twenty-two arrays comprise the entire SRRS, which was designed as a 3 x 2 factorial
experiment with three spacing treatments (25m, 75m, 225m) and two patch size
treatments (4-cube and 16-cube).
The SRRS was twice affected by perturbations in its initial years. In 1992,
locations of four arrays were inadvertently published. In 1995, all fish on the five
southernmost arrays were eliminated by a red tide event. This fish-kill effectively reset
the fish colonization process on those arrays. Two arrays were both published
inadvertently and experienced the fish-kill.
In November of 1996, locations of two to three arrays per treatment group were
published for access by the general public. No locations of the 75 m spaced arrays were
published and thus the 75 m treatment was excluded from this study. The early
perturbations and requirements of other on-going research projects warranted restrictions
on randomization when selecting arrays for publication, e.g. inadvertently published
arrays were assigned to the 'published' treatment.
Gag, Mycteroperca microlepis, is a dominant reef-associated, demersal fish species
in the Western Atlantic ranging from New England to Brazil, including the Gulf of
Mexico (McErlean, 1963). As protogynous hermaphrodites, female gag mature at 3-6
years old and 538 to 795 mm TL (Hood and Schlieder 1992; Harris and Collins 2000). In
two studies, no males younger than five or smaller than 875 mm TL were observed
(Hood and Schlieder 1992; Harris and Collins 2000). By age 11, males comprised 50%
of the age class. Transition from female to male is hypothesized to be socially mediated
(Koenig et. al. 1999; Coleman et. al. 2002) but is debated (Kenchington, 1999). Gag
reach sizes between 1100 and 1200 mm and ages ranging into the early twenties (Hood
and Schlieder 1992; Collins et. al. 1998).
Gag form spawning aggregations of 10's to 100's of individuals (Coleman et. al.,
1996). Aggregations have been observed along the shelf edge break in the Northwest
Gulf of Mexico in 50-120 m (Coleman et. al. 1996; Koenig et. al., 1999), on deep (70-
100 m) reefs off the Atlantic coast of Florida (Gilmore and Jones, 1992), and have been
recently reported on the Campeche Bank, Yucatan, Mexico (Brule et. al., 2003). Gag
spawn from late December through April with peak spawning in February and March
(Hood and Schlieder 1992; Collins et. al. 1998). Currents carry gag larvae inshore where
they settle in seagrass beds and oyster bars (Ross and Moser, 1995). Average larval
duration is approximately 43 days (Keener et. al. 1988). Gag spend their first summer
growing in inshore habitats where their diet is composed primarily of crustaceans
(Bullock and Smith 1991; Weaver 1996). In the late summer and fall, gag move to
deeper water and take up residency on near-shore reefs and live-bottom where their diet
shifts to decapod crustaceans and fishes (Bullock and Smith 1991; Weaver 1996). It is in
these habitats that gag reach legally harvestable size and are first targeted by fishers. Gag
continue to grow and mature in these environments before moving to spawning grounds
as mature adults. The natural annual mortality rate is unknown but is assumed to be
M=0.15 for stock assessment purposes (Turner et. al. 2001). It is during their transition
across the shallow continental shelf that juvenile to sub-adult pre-reproductive females
take up residency on the SRRS
Sources of Data
Gag present on each SRRS patch were counted once each Summer from 1995 to
1998 and 2002 to 2004 by trained SCUBA divers (see Lindberg et. al., in press for
details). Types of data collected changed over the years of sampling as the research
questions being addressed changed. I worked with a historical data set (1995 to 1998)
and participated in data collection during the 2002 to 2004 period. From 1995 to 1996
the total number of gag on each patch reef were counted and divers visually estimated the
length of the largest and smallest gag as well as the average size of gag. In 1997 and
1998, counts were divided into total number of legal ( >20" or 508 mm) and sub-legal
(<20" or 508 mm) gag. In 2002 through 2004, gag were individually assigned to 10 cm
size groups. The 50-59 cm size group was further divided into gag from 50-55 cm and
56-59 cm in order to determine the number of legal and sub-legal gag on a patch. This
coincided with an increase in the legal size of recreationally harvestable gag from 20"
(508 mm) to 22" (559 mm) that occurred on June 19, 2000 (Turner et. al., 2001). For this
investigation, all data were converted to counts of legal and sub-legal gag by array.
Legal and sub-legal counts were estimated for 1995 and 1996 from percentages of legal
and sub-legal gag observed on unpublished arrays of the same architecture in 1997. This
is based on a necessary assumption that the proportions of legal and sub-legal gag on
patches of like architecture and presumed level of fishing effort is consistent over time
(1995-1997) and space (among unpublished arrays). These approximations were further
adjusted based on the size estimates for the largest and smallest gag (i.e., if the largest
gag observed was smaller than the legal limit, all gag were considered sub-legal). Patch
reef counts were summed for each array, as the arrays rather than the patches are the
proper replicates for this study.
Suitability of Data
Preliminary analyses were conducted to establish the suitability of data to be used.
Reef arrays were constructed from 1990-1993. Within arrays, data from 1995 and 1996
were compared for significant differences. None were found and thus variable reef age
was determined not to be a significant factor influencing gag abundances. In 1995 a fish-
kill event eliminated all fish on the five southern-most reef arrays. The fish-kill was
investigated as a treatment effect within reef architectures. Data were excluded from the
final model for all architecture/year combinations in which the fish-kill treatment was
significant. This resulted in the exclusion of data from all fish-killed arrays in 1995. In
1996 legal counts on four-cube arrays were considered recovered and included in further
analyses. Analysis indicated all affected arrays had recovered by 1997. In 1992, several
array locations were inadvertently published. Due to low replication and the impacts of
the fish-kill, I was unable to evaluate this effect.
Three variables were investigated in a repeated measures 2 x 2 x 2 x 3 factorial
design for their effects on gag abundance on the SRRS: patch spacing (25 m or 225 m
between patches within an array), patch size (4 or 16 cubes per patch), publication of
locations for public fishing (published or not), and time period. Annual counts were
nested within three different time periods (time relative to the publication of array
locations). "Pre" consists of data collected in 1995 and 1996, prior to publication.
"Acute" consists of data collected in 1997 and 1998, the period immediately following
publication. "Chronic" consists of data taken in 2002 through 2004, six to eight years
All analyses were conducted independently on legal and sub-legal gag count data.
To satisfy the assumptions of normality and homogeneity of variances, all count data
were square root transformed.
Data not excluded by the preliminary analyses were entered in a four-way repeated-
measures, mixed-model Analysis of Variance (ANOVA) to determine significant
interactions and main effects. Alternate correlation structures for the effect of year nested
within period were tested, but none improved the model over a zero-correlation model.
The Kenward-Rogers method was used to estimate the degrees of freedom for the
appropriate F-test for fixed effects. Follow-up analyses were done by pair-wise
comparisons of least square means provided for each treatment by the model. Residuals
were plotted in a Q-Q plot and against predicted values to examine model fit. Akaike's
information criterion as well as mean square error values were compared between
treatment levels to investigate error structure in the model. All statistical analyses were
performed in SAS version 8 (SAS Institute Inc. 2002, Cary, NC). An alpha value of 0.05
was used for determining statistical significance. Due to the low replication and thus low
suspected power of these tests, the alpha value was not adjusted to maintain an
experiment-wise error rate of 0.05.
30' 20 10' 83' 50' 40'
Figure 1. Design of the Suwannee Regional Reef System (A) Location of SRRS in the
Big Bend region of Florida (inset). (B) Design of a component artificial reef
block and layout of an SRRS array.
The four-way interaction of patch size, patch spacing, publication and time period
was significant (p=0.0083, Table 2) for legal-sized gag abundances on the SRRS. The
results of follow-up comparisons below explain this four-way interaction. The
interaction is partitioned chronologically with steady-state comparisons within time
periods and transitional comparisons between time periods. Mean square errors were
examined between treatment levels and determined not to be significantly different
between levels of any treatment. Residual analysis indicated good model fit. The
significant model was run using untransformed data in order to generate least square
means and error estimates for use in figures to better convey the results.
Pre-publication Time Period (Initial Conditions)
Arrays were not selected randomly for publication and thus there were pre-existing
differences between arrays selected for the published and unpublished treatments. Arrays
that experienced the fish-kill and those that were inadvertently published were given
priority as 'published' treatments in order to maintain a valid 'unpublished' treatment to
compare against. To-be-published 4 x 225 m arrays had significantly fewer legal gag
than their not-to-be-published counterparts even before their publication (p=0.0008).
Beyond the perturbations, there was a distributional pattern on the unpublished
arrays dependent on the size and spacing of arrays (Figure 2). Within spacing treatments,
16-cube arrays had a higher abundance of legal gag than 4-cube arrays (25 m p=0.0002;
225 m p=0.0009). Within 4-cube arrays, there was a higher abundance of legal gag on
225 m arrays than 25 m arrays (p=0.0314).
Transition from "Pre" to "Acute" Time Periods
The publication of array locations significantly impacted the abundances of legal-
sized gag for 16-cube arrays (Figure 3) for both spacing treatments (25 m p<0.0001; 225
m p<0.0001). Four-cube published arrays, however, did not change significantly
following publication (25 m p=0.3378; 225 m p=0.3898), (Figure 4). There was a
significant increase in legal gag abundance on unpublished 4 x 25 m arrays (p=0.0071),
and thus no difference in the "acute" period between unpublished 4-cube arrays based on
their spacing (p=0.6264). This was the only significant change on any of the unpublished
The effect of publication on legal gag was most pronounced for arrays with larger
reef patches (Figure 5). The number of legal gag on published 16-cube arrays, however,
was not significantly different than the number of legal gag on published 4-cube arrays
(25 m p=0.5977; 225 m p=0.2462) or unpublished 4-cube arrays (25 m p=.0004 [greater
abundance on four-cube arrays]; 225 m p=0.2380). Unpublished arrays maintained the
size-dependent difference of more legal gag on 16-cube than 4-cube arrays (25 m
p=0.0005; 225 m p<0.0001). In the "acute" period a pattern emerged of more legal gag
on unpublished 4-cube arrays relative to published 4-cube arrays for both spacing
treatments (25 m p=0.0044; 225 m p=0.0249). More widely spaced arrays were
hypothesized to be more difficult to fish and although there were more legal gag on
published 16-cube arrays at the 225 m spacing, the difference was not significant
Transition from "Acute" to "Chronic"
Following the initial impacts of publication observed in the "acute" period, none of
the published arrays experienced significant changes (Figures 3 and 4). On unpublished
arrays, decreases indicative of discovery by anglers were observed. Although 4-cube, 25-
meter arrays did not change (p=0.2764), 4-cube arrays at the 225-meter spacing declined
significantly (p=0.0483) (Figure 4). Sixteen-cube arrays experienced a significant
reduction in legal gag between the "acute" and "chronic" periods at both spacing
treatments (25m p=0.0010; 225m p<0.0001) (Figure 3).
Despite the decrease on unpublished 16-cube, 25 m arrays, these arrays still held
significantly more legal gag in the "chronic" period than did the published 16 cube, 25 m
arrays (p=0.0003) (Figure 6). Reductions on unpublished arrays lead to there being no
difference in legal gag abundances between 16-cube, 225 m arrays and their published
counterparts (p=0.0872). Sixteen-cube arrays, both published and unpublished, did not
show any difference by spacing (pub, p=0.1246; unpub, p=0.4132).
Four-cube arrays showed a different pattern of gag abundance in the "chronic"
period than the "acute." Although there were still more legal gag on the unpublished 4-
cube, 25 m than the published 4 cube, 25 m arrays (p=0.0214), 4-cube, 225 m arrays
showed no difference between published and unpublished treatments (p=0.7808) (Figure
6). As with the 16-cube arrays, this was the result of a significant decrease on the
unpublished arrays. Within publication treatments, spacing was not significant on 4-cube
arrays (published p=0.4519; unpublished p=0.0872)
In the "chronic" period, patch size was no longer significant for any treatment.
Effects of publication continued to be significant for 25 m arrays (4-cube p=0.0214; 16-
cube p=0.0003), however, 225 m arrays no longer showed any differences by publication
(4-cube p=0.7808; 16-cube p=0.1812)
There was a significant main effect of time period (p<0.0001) for sub-legal gag
(Table 4). The three-way interaction between publication, patch size, and patch spacing
was also significant (p=0.0513) (Table 4) given the low replication and thus low expected
power of this test. Mean square error was examined between treatment levels and
determined not to be significantly different between levels of any treatment. Residual
analysis indicated good model fit. The significant model was run using untransformed
data in order to generate least square means and error estimates for use in figures to better
convey the results.
The significant main effect of time period was manifested as a trend of decreasing
abundance of sub-legal gag. Although the decrease was not significant from the "pre" to
the "acute" period (p=0.5617), it was significant from "acute" to "chronic" (p=0.0001).
The significant effect of publication in the three-way interaction was only on 16-
cube, 25 m arrays. There were more sub-legal gag on the unpublished than published
arrays (p=0.0151) (Figure 7).
In most cases, there were more sub-legal gag on 16-cube arrays than on 4-cube
arrays. This held true for unpublished arrays (25 m p=0.0101; 225 m p=0.0017) as well
as published 225 m arrays (p=0.0003), but not for published 25 m arrays (p=0.6966).
Patch spacing had the most involved interaction determining sub-legal gag
abundance (Figure 8). There was a significant interaction with publication as spacing had
no significant effect on unpublished arrays (4-cube p=0.3868; 16-cube p=0.5679). Patch
spacing and patch size interacted to affect patterns of sub-legal gag abundance on
published arrays. There were more sub-legal gag on published 16-cube arrays at the 225
m spacing than the 25 m spacing (p=0.0125). The opposite was true for the published 4-
cube arrays with more sub-legal gag on 25 m arrays than 225 m arrays (p=0.0188).
Table 1. Type 3 tests of fixed effects for legal gag
Covariance Parameters Subject Estimate Std. Err. Z value Pr Z
Reef 1.1787 0.6847 1.72 0.0462
Year (Period) Reef 2.3078 0.3555 6.49 <0.0001
Four-way interaction Num DF Den DF F-value Pr>F
Pd*Fishing*Size*Dist 2 85.3 5.08 0.0083
Three-way interactions Num DF Den DF F-value Pr>F
Pd*Fishing*Size 2 85.3 15.85 <0.0001
Pd*Fishing*Dist 2 85.3 4.01 0.0217
Pd*Size*Dist 2 85.3 2.72 0.0716
Fishing* Size*Dist 1 10.5 1.59 0.2344
Two-way interactions Num DF Den DF F-value Pr>F
Pd*Fishing 2 85.3 27.51 <0.0001
Pd*Size 2 85.3 36.64 <0.0001
Pd*Dist 2 85.3 6.21 0.0030
Fishing*Size 1 10.5 3.22 0.1014
Fishing*Dist 1 10.5 0.60 0.4549
Size*Dist 1 10.5 4.39 0.0613
Main Effects Num DF Den DF F-value Pr>F
Period 2 85.3 39.43 <0.0001
Fishing 1 10.5 48.20 <0.0001
Size 1 10.5 43.84 <0.0001
Distance 1 10.5 4.19 0.0664
Table 2. Type 3 tests of fixed effects for sub-legal gag.
Covariance Parameters Subject Estimate Std. Err. Z Value PrZ
Reef 1.0481 0.7444 1.41 0.0795
Year (Period) Reef 3.6065 0.5661 6.37 <0.0001
Four-way interaction Num DF Den DF F-value Pr>F
Pd*Fishing*Size*Dist 2 83.9 1.53 0.2222
Three-way interactions Num DF Den DF F-value Pr>F
Pd*Fishing*Size 2 83.9 2.75 0.0695
Pd*Fishing*Dist 2 83.9 1.25 0.2928
Pd*Size*Dist 2 83.9 1.24 0.2937
Fishing* Size*Dist 1 10.5 4.83 0.0513
Two-way interactions Num DF Den DF F-value Pr>F
Pd*Fishing 2 83.9 0.50 0.6088
Pd*Size 2 83.9 2.88 0.0615
Pd*Dist 2 83.9 0.24 0.7852
Fishing*Size 1 10.5 1.34 0.2719
Fishing*Dist 1 10.5 0.01 0.9252
Size*Dist 1 10.5 13.23 0.0042
Main Effects Num DF Den DF F-value Pr>F
Period 2 83.9 12.44 <0.0001
Fishing 1 10.5 4.15 0.0675
Size 1 10.5 38.63 <0.0001
Distance 1 10.5 0.02 0.8831
Distance Between Patches (m)
Figure 2. Pre-publication distribution of legal gag grouper on the SRRS. Data are drawn
from not-to-be-published arrays due to the influence of pre-publication
perturbations on to-be-published arrays.
150 -.,.-Pub, 25m
T--ime- Pub, 225m
S architecUnpub, 25m
8 100oo ,,
S1 0---x--- Unpub, 225m
Pre Acute Chr
Figure 3. Chronology of impacts to legal gag on 16-cube arrays of the SRRS by array
architecture and publication status.
Pre Acute Chr
- -- Pub, 25m
- -- Pub, 225m
--X- --Unpub, 225m
Figure 4. Chronology of impacts to legal gag on four-cube arrays of the SRRS by array
architecture and publication status.
0 O] Unpublished
4x25m 4x225m 16x25m 16x225m
Figure 5. Abundance of legal gag per array within the "acute" period by array
architecture and publication status.
CD ] Unpublished
4x25m 4x225m 16x25m 16x225m
Figure 6. Abundance of legal gag per array within the "chronic" period by array
architecture and publication status.
4x25m 4x225m 16x25m 16x225m
Figure 7. Abundance of sub-legal gag per array, by array architecture and publication
status, averaged across all time periods.
A. Published Arrays
o 100 225m
B. Unpublished arrays
100 0 225m
Figure 8. Abundance of sub-legal gag per array, by array architecture on (A) published
and (B) unpublished arrays.
The SRRS was designed to examine how spatial components of habitat structure
affect gag abundance. The subsequent publication of several array locations introduced
directed fishing effort as an additional factor. Although no data were collected regarding
fishing effort or angler behavior, differences between published and unpublished
treatments are assumed to be due to directed fishing. Further, inferences are possible
regarding angler behavior based on observed gag abundances on the SRRS. In a survey
of recreational anglers in Dade County, FL, over 28% used artificial reef sites, motivated
by expectations of high catch rates (Milon, 1989). Fishing effort usually is positively
related with fish abundance that "represents a cumulative statistical summation of the
effects of individual fisher choices" (Walters and Martell, 2004, p.205). For recreational
anglers this leads to a roughly linear relationship between fish abundance and angling
effort (Walters and Martell, 2004). On the SRRS this relationship is complicated by the
rapid depletion of localized standing stocks of legal gag. The reduction in abundance
lowers the benefit to anglers of selecting particular arrays and offers feedback to their
decision-making process, but must have a lag as new experiences are likely slow to
modify initial impressions and expectations.
This experiment effectively demonstrated the ability of anglers to remove the vast
majority of legally harvestable fish from published artificial reefs over the course of only
a few years. Once published, size of reef patches had no impact on abundance of legal
gag, indicating both large and small patches were fished down to a similar level. Over
the six to eight years following publication and the initial rapid reductions in legal gag,
effort was apparently redirected to locating unpublished reef arrays. Changes on
unpublished reef arrays can likely be attributed to discovery by anglers as they were
proportional to predicted impacts based on architecture. Large and widely-spaced
patches (those most prone to discovery) experienced declines in abundance, while reef
arrays composed of small closely-spaced patches did not change. Patterns in sub-legal
gag abundance are driven by significant declines on the most heavily fished arrays
(published, 16-cube, 25 m spaced).
Several studies have reported higher abundances of fish on larger habitat patches or
patches providing more refuge (de Boer 1978; Roberts and Ormond 1987; Hixon and
Beets, 1989; Rountree, 1989; Kuwamura et. al. 1994, Eggleston et. al. 1998; Abelson and
Shlesinger 2002). The initial conditions observed on the SRRS with regard to array
architecture are thus not surprising. Studies of gag home range on the SRRS (Kiel 2004)
and residency (Lindberg et. al. in press) have indicated that reef patches spread 25 m
apart are collectively used as a single home-site whereas 225 m spaced patches are used
The fact that fishing had both an immediate and significant impact on the largest
aggregations of gag is not surprising. Reductions in catch assumed to be reflective of
similar reductions in abundance were observed by Grandcourt (2003) in monitoring the
exploitation of a virgin stock of crimson jobfish, Pristipomoidesfilamentous, a shelf-edge
hook and line fishery. There are also numerous examples in the literature of comparisons
between both recreationally and commercially fished and un-fished areas in which there
are more fish and larger fish in protected areas than unprotected ones (Roberts, 1994;
Jennings and Polunin, 1996; Wantiez et. al., 1997, Guidetti et. al., 2005, Kamukuru et.
al., 2005) as well as declines over time of exploited stocks (Haedrich and Barnes, 1997;
Fogarty and Murawski 1998; McClanahan and Mangi, 2001; Albaret and Lae 2003;
Laurans et. al., 2004). Solonsky (1985) demonstrated the significant impact of fishing on
artificial reefs in Monterey Bay, CA. A single marked reef was constructed amidst other
unmarked reefs. Observed abundance of legal-sized rockfish on the marked reef was
reduced after three years of fishing effort relative to the unmarked reefs. Furthermore,
tagged fish were observed to move from unmarked reefs to the marked reef, but not vice
versa, indicating the potential 'sink' quality of a fished artificial reef. Similar movements
of fish (from natural to artificial reefs, but not back) were observed in Puerto Rico (Fast
and Pagan, 1974). On the SRRS we had the unique opportunity to examine the manner in
which variable spatial qualities of reefs interact with fishing effort to determine the
abundance of a targeted species.
Although 25 and 225 m published 16-cube arrays experienced significant declines
in abundance of legal gag following publication, the predicted refuge effect of more
widely spaced arrays was only marginally significant in the "acute" period. There are
several potential mechanisms for such a refuge effect. Since gag treat 25 m arrays as a
single home-site, fishing effort, even when focused on a single patch, is effectively
fishing the entire array due to the movement of gag among patches. There is also a
significant margin of error for non-anchored fishing techniques (drift-fishing or trolling)
on 25 m arrays. Should the boat's course not take it over the intended patch, it is far
more likely to encounter gag on another patch, or between patches. Furthermore, 25 m
spaced arrays are likely to be more appealing to spear-fishermen who can easily swim
between patches in search of legal gag. All of these factors likely contribute to higher
attractiveness to anglers of 25 m arrays relative to 225 m arrays. An additional
consideration is that 225 m spaced patches may recruit more gag as a result of their larger
footprint, simply by an increased chance of encounter by gag, however, immigration rates
and residence times under fished conditions, relative to removal rates by anglers are
While the publication of array locations led to a rapid depletion of legal gag on 16-
cube patches, 4-cube patches did not change significantly with publication. Both their
smaller size and lower abundances of gag make them more difficult to locate and fish
effectively. Furthermore, the four cube arrays were likely less attractive to anglers based
on the common conception that smaller reefs hold fewer fish. Although there was no
significant change on the published four-cube arrays between the "pre" and "acute"
periods, there was a significant effect of publication within the "acute" period (more legal
gag on unpublished arrays). Keeping in mind that there were randomization restrictions
in assignment of publication treatments because of earlier perturbations, the effect of
publication on four-cube arrays may have been to maintain low abundances on these
arrays, rather than to depress them below their pre-publication levels.
The immediacy of the decrease in gag abundances following publication is further
emphasized by the fact that none of the published treatments demonstrated any
significant further decreases between the "acute" and "chronic" periods. A new
equilibrium abundance appears to have been reached in which legally harvestable fish
were maintained at low levels by fishing activities. The increase in the minimum size of
recreationally harvestable gag between the "acute" and "chronic" periods had the
potential to further decrease abundances of legal gag as those between 508 and 550 mm
would be considered sub-legal in the "chronic" period. There was, however, a significant
decrease of sub-legal gag between these periods, indicating that this was a minimal effect
in relation to the other processes affecting abundances of legal and sub-legal gag on the
Because the SRRS represents only a way-point in the spatially stage-structured life
history of gag, there is a constant immigration to and emigration from the reefs. In
studies done prior to introducing the publication treatment, mean residence time for an
individual gag was 9.8 months (Lindberg et. al. in press). It is assumed that newly
arriving gag make a decision as to whether to stay or not based on current conditions on
the patch. Although density-dependent habitat selection has been confirmed for gag
(Lindberg et. al., in press), the exact influence of the standing stock of gag on an array in
the decision making process of newly arriving gag is unknown. On arrays depleted by
fishing, it seems reasonable that newly arrived legal-sized gag would find themselves
among the dominant fish present and be more likely to stay on the reef than if there was a
full compliment of un-fished legal gag present. In this manner there may be a higher
percentage of newly arrived gag deciding to stay on published patches, thus making
themselves vulnerable to fishing, than on unpublished patches or patches prior to
publication. Fully understanding the rate at which gag arrive, and the proportion of those
taking up residence, as well as criteria important in making such a decision, would be a
big step forward in understanding the implications of fishing on habitat selection and
habitat-specific mortality risk.
The tendency of gag to aggregate in specific habitats known to recreational anglers
fits the definition of an 'ecological trap' (Schlaepfer et. al., 2002). The concept of
ecological traps has most widely been applied to terrestrial environments. Ecological
traps result when an organism makes a mal-adaptive habitat selection, due most
frequently to an anthropogenic change or manipulation of the environment, based on cues
that over evolutionary time were correlated with improved fitness. In the marine
environment, the tendency of many game-fish species to associate with structure can be
interpreted as an ecological trap. Once discovered by anglers, high quality habitat (both
artificial and natural) will continue to attract and aggregate both fish and the anglers
targeting them. In the case of the SRRS, the artificial reef patches are high quality habitat
(Lindberg et. al. in press) providing gag both refuge and a prey base. The outcome,
however, is an aggregation accurately predictable in space, leading to the removal of the
majority of legally harvestable gag. The decision to take up residence on published
patches of the SRRS seems to meet the criteria of an ecological trap (Schlaepfer et. al.
2002), but could only be conclusively defined as such with complete knowledge of the
overall reproductive success of gag taking up residence on the SRRS relative to those
inhabiting natural hard-bottom habitats.
One of the most telling results is the lack of any significant difference between size
treatments of both published and unpublished arrays during the "chronic" period. While
other factors such as catch-ability of legal gag and by-catch of under-sized gag factor into
the benefit to anglers of fishing an array, the primary focus is assumed to be harvest of
legal gag, and each size of reef provides similar potential harvest at its new equilibrium
abundance. This indicates that the additional shelter on 16-cube patches does not provide
gag any greater protection from fishing relative to the four-cube arrays. Also, when
constructing reefs for the benefit of the gag, the additional investment in construction
materials is not justified by its potential to house more legal gag under conditions where
recreational fishing will occur.
The predictable impacts of fishing on the SRRS extend beyond the published
arrays. Ultimately the unpublished arrays of the SRRS are all predicted to become
discovered. Thus the rate of progression from their un-fished state in the pre-publication
period and "acute" period to their moderately-fished state in the "chronic" period is an
indication of their combined ease of discovery and subsequent value as fishing sites.
Between the "acute" and "chronic" periods, unpublished 16-cube arrays, in both spacing
treatments, experienced significant declines in legal gag. This is believed to be due to
increased fishing effort, most likely from directed efforts in response to declining catches
on published arrays, but also possibly due to simple chance, to locate sites that were
known to exist. Sixteen-cube arrays would be the most susceptible to discovery as they
have a larger footprint than four-cube arrays, and larger aggregations of gag, thus
providing the largest targets for searching anglers. Of the four-cube arrays, only the 225-
meter spaced treatments showed significant effects of discovery. The probability of a
searching angler encountering an array is higher on more-widely and systematically
spaced arrays where there is likely less overlap of the areas used by gag on different
patches. As the types of array architectures within the SRRS were publicly known,
discovery of one patch within an array should be interpreted as discovery of the array as a
Impacts of directed fishing extended beyond the removal of legally-harvestable gag
to the sub-legal gag. All comparisons involving published, 16-cube, 25-meter arrays
demonstrated the impacts of fishing pressure on sub-legal gag. This treatment apparently
received the highest level of fishing effort and thus was the first to demonstrate effects of
that angling pressure trickling down to the sub-legal gag. First, there were more sub-
legal gag on unpublished arrays of this architecture than published arrays. Second, there
was no difference between this treatment and 4-cube, published arrays at the 25-meter
spacing. In all other treatment combinations, 16-cube arrays held more sub-legal gag
than 4-cube arrays as would be expected by the amount of shelter provided. Finally,
published, 25-meter spaced, 16-cube arrays had fewer sub-legal gag than the more
widely-spaced, published, 16-cube arrays, a pattern that was not observed among the
unpublished arrays. Hooking mortality, illegal harvest, and induced emigration are all
potential mechanisms by which sub-legal gag may be affected. No data were collected to
determine the relative importance of these mechanisms. Catch and release associated
mortality from recreational anglers has been measured to be as high as 20 % from
headboat fisheries (R. Dixon, NMFS, personal communication). These gag, however,
were caught from a minimum depth of 25 m where there is a greater effect of swim
bladder expansion than occurs on the SRRS in 13 m of water, and they were not followed
beyond release. Both depth-related effects as well as hooking effects contribute to
release mortality. In a study of depth-related capture-release mortality on gag, where gill
or gut-hooked fish were not included, no gag caught from 20 m of water showed any
adverse depth-related effects of being caught (C. Koenig, FSU, unpublished data).
Hooking mortality on another serranid, Epinephelus quoyanus, in shallow waters (<2m)
of the Great Barrier Reef, where depth related effects were not a factor, was found to be
highest at 5.1% when bait-fishing with single hooks (Diggles and Ernst, 1997). Actual
catch and release mortality on the SRRS is likely close to this figure, as bottom-fishing
with natural bait is a very common technique. It is likely that many legal gag are hooked
several times before being landed and that sub-legal gag are both hooked and landed
without being harvested from the system. Beyond hooking mortality, the role these
experiences play in modifying behavior and possibly residence times of surviving gag
may be of importance.
The main effect of time period on sub-legal gag abundance manifested itself in a
decreasing trend over time. The only significant change, however, was from the "acute"
period to the "chronic" period. The lack of an interaction with publication makes
interpretation somewhat troublesome. It is possible that there were simply several years
of smaller year-classes of gag colonizing the SRRS. The continued directed angling
efforts both on published and unpublished arrays following the "acute" period, however,
certainly contributed to the decline.
This study has revealed an interaction of spatial qualities of habitat and angling
pressure, mediated by detailed information of those habitat qualities, in determining local
abundances of a targeted reef-dependent species. The relative benefit to gag that occupy
small patches of artificial habitat are clear. Widely-spaced patches within an array
increase the probability of a single patch being discovered. Discovery of a single patch,
in this case, should lead to discovery of the array as a whole as anglers are familiar with
the layout of patches in the SRRS. While the end results have been demonstrated there is
still uncertainty in the precise mechanisms, and their relative importance, combining to
produce the observed patterns. Identifying and quantifying these mechanisms would be
important for applying the results of this study to management decisions for gag, other
fish and other systems. Additional information regarding the usage and opinions of the
SRRS in the angling community could validate several of the assumptions and inferences
made herein as well as increase the applicability of conclusions drawn in this study. The
publication of detailed geographic information in this case allowed anglers to select or
disregard fishing sites according to their expectations of success based on habitat
architecture. The continued development of such detailed information regarding natural
habitats will enhance efficiencies in the exploitation of natural resources.
Ableson, A. and Shlesinger, Y. (2002). Comparison of the development of coral and fish
communities on rock-aggregated artificial reef in Eilat, Red Sea. ICES Journal of
Marine Science. 59, S122-S126.
Albaret, J. and Lae, R. (2003). Impact of fishing on fish assemblages in tropical lagoons:
the example of the Ebrie lagoon, West Africa. Aquatic Living Resources. 16, 1-9.
Andersen, H., McGaughey, R.J., and Reutebuch, S.E. (2005). Estimating forest canopy
fuel parameters using LIDAR data. Remote Sensing the Environment. 94, 441-449.
Berkeley, S.A., Hixon, M.A., Larson, R.J. and Love, M.S. (2004a). Fisheries
sustainability via protection of age structure and spatial distribution of fish
populations. Fisheries. 29, 23-32.
Berkeley, S.A., Chapman, C. and Sogard, S.M. (2004b). Maternal age as a determinant of
larval growth and survival in a marine fish, Sebastes melanops. Ecology. 85, 1258-
Bianchi, G., Gislason, H., Graham, K., Hill, L., Jin, X., Koranteng, K., Manickchand-
Heileman, S., Paya, I., Sainsbury, K., Sanchez, F. and Zwanenburg, K. (2000).
Impact of fishing on size composition and diversity of demersal fish communities.
ICES Journal of Marine Science. 57, 558-571.
Birkeland, C. and Dayton, P.K. (2005). The importance in fishery management of leaving
the big ones. Trends in Ecology and Evolution. 20, 356-358.
Bohnsack, J.A., Harper, D.E., McClellan, D.B. and Hulsbeck, M. (1994). Effects of reef
size on colonization and assemblage structure of fishes at artificial reefs off
southeastern Florida, U.S.A. Bulletin of Marine Science. 55, 796-823.
Bremner, J., Frid, C.L.J. and Rogers, S.I. (2003). Assessing marine ecosystem health:
The long-term effects of fishing on functional biodiversity in North Sea benthos.
Aquatic Ecosystem Health and Management. 6, 131-137.
Brown, E.D., Churnside, J.H., Collins, R.L., Veenstra, T., Wilson, J.J. and Abnett, K.
(2002). Remote sensing of capelin and other biological features in the North Pacific
using lidar and video technology. ICES Journal of Marine Science. 59, 1-11.
Brule, T., Deniel, C., Colas-Marrufo, T. and Renan, X. (2003). Reproductive biology of
gag in the southern Gulf of Mexico. Journal of Fish Biology. 63, 1505-1520.
Bullock, L.H. and Smith, G.B. (1991). Seabasses (Pisces: Serranidae). Memoirs of the
Hourglass Cruises. Marine Research Laboratory, Florida Department of Marine
Resources, St. Petersburg, FL. 8, 1-205.
Carr, M.H., Anderson, T.W., and Hixon, M.A. (2002). Biodiversity, population
regulation, and the stability of coral-reef fish communities. Proc. Nat. Acad. Sci.
Charbonnel, E., Serre, C., Ruitton, S., Harmelin, J. and Jensen, A. (2002). Effects of
increased habitat complexity on fish assemblages associated with artificial reef
units (French Mediterranean coast). ICES Journal of Marine Science. 59, S208-
Coleman, F.C., Figueira, W.F., Ueland, J.S. and Crowder, L.B. (2004). The impact of
United States recreational fisheries on marine fish populations. Science. 305, 1958-
Coleman, F.C., Koenig, C.C. and Chapman, R. (2002). Review of Kenchington
Document "Management of the Gulf of Mexico Gag Grouper Fisheries: A
Reconsideration". Tallahassee, Department of Biological Science, Florida State
Coleman, F.C., Koenig, C.C. and Collins, L.A. (1996). Reproductive styles of shallow-
water groupers (Pisces: Serranidae) in the eastern Gulf of Mexico and the
consequences of fishing spawning aggregations. Environmental Biology of Fishes.
Coll, J., Moranta, J., Renones, O., Garcia-Rubies, A. and Moreno, I. (1998). Influence of
substrate and deployment time on fish assemblages on an artificial reef at
Formentera Island (Balearic Islands, western Mediterranean). Hydrobiologia. 385,
Collins, L.A., Johnson, A.G., Koenig, C.C. and Baker, M.S. Jr. (1998). Reproductive
patterns, sex ratio, and fecundity in gag, Mycteroperca, microlepis (Serranidae), a
protogynous grouper from the northeastern Gulf of Mexico. Fishery Bulletin. 96,
Conover, D.O. and Munch, S.B. (2002). Sustaining fisheries yields over evolutionary
time scales. Science. 297, 94-96.
De Boer, B.A. (1978). Factors influencing the distribution of the damselfish Chromis
cyanea (Poey), Pomacentridae, on a reef at Curacao, Netherlands Antilles. Bulletin
of Marine Science. 28, 550-565.
Diggles, B.K. and Ernst, I. (1997). Hooking mortality of two species of shallow-water
reef fish caught by recreational angling methods. Marine and Freshwater Research.
Ditton, R.B. and Graefe, A.R. (1978). Recreational fishing use of artificial reefs on the
Texas Gulf Coast. Texas Agricultural Experiment Station, Texas A&M University,
College Station. 155pp.
Doherty, P.J. and Sale, P.F. (1985). Predation on juvenile coral reef fishes: an exclusion
experiment. Coral Reefs. 4, 225-234.
Eggleston, D.B., Etherington, L.L. and Elis, W.E. (1998). Organism response to habitat
patchiness: species and habitat-dependent recruitment of decapod crustaceans.
Journal of Experimental Marine Biology and Ecology. 223, 111-132.
Eklund, A.M. (1997). The importance of post-settlement predation and reef resources
limitation on the structure of reef fish assemblages. Proceedings of the 8th
International Coral Reef Symposium. 2, 1139-1142.
Fast, D.E. and Pagan F.A. (1974). Comparative observations of an artificial tire reef and
natural patch reefs off southwestern Puerto Rico. Pages 49-50 in L. Colunga and R.
Stone, eds. Proceedings: artificial reef conference. Texas A&M University,
Fogarty, M.J. and Murawski, S.A. (1998). Large-scale disturbance and the structure of
marine systems: fishery impacts on Georges Bank. Ecological Applications. S6-
Frazer, T.K. and Lindberg W.J. (1994). Refuge spacing similarly affects reef-associated
species from 3 phyla. Bulletin on Marine Science. 55, 388-400.
Gilmore, R.G. and Jones, R.S. (1992). Color variation and associated behavior in the
epinepheline groupers, Mycteroperca microlepis (Goode and Bean) and M. phalanx
Jordan and Swain. Bulletin of Marine Science. 51, 83-103.
Grandcourt, E.M. (2003). The effect of intensive line fishing on the virgin biomass of a
tropical deepwater snapper, the crimson jobfish (Pristopomoidesfilamentosus).
Fisheries Bulletin. 101. 305-311.
Gratwicke, B. and Speight, M.R. (2005). Effects of habitat complexity on Caribbean
marine fish assemblages. Marine Ecology Progress Series. 292, 301-310.
Guidetti, P., Verginella, L., Viva, C., Odorico, R. and Boero, F. (2005). Protection effects
on fish assemblages and comparison of two visual-census techniques in shallow
artificial rocky habitats in the northern Adriatic Sea. Journal of the Marine Biology
Association of the U.K. 85, 247-255.
Haedrich, R.L. and Barnes, S.M. (1997). Changes over time of the size structure in an
exploited shelf fish community. Fisheries Research. 31, 229-239.
Harris, P.J. and Collins, M.R. (2000). Age, growth and maturity of gag, Mycteroperca
microlepis, from the southeastern United States during 1994-1995. Bulletin of
Marine Science. 66, 105-117
Hart, M. (2002). Habitat-mediated direct and indirect effects among three Serranid fishes.
Master's Thesis. University of Florida, Gainesville, Florida.
Hauser, L., Adcock, G.J., Smith, P.J., Ramirez, J.H.B., and Carvalho, G.R. (2002). Loss
of microsatellite diversity and low effective population size in an overexploited
population of New Zealand snapper (Pagrus auratus). Proc. Natl. Acad. Sci.
U.S.A. 99, 11742-11747.
Herrera, R., Espino, F., Garrido, M., Haroun, R.J. (2002). Observations on fish
colonization and predation on two artificial reefs in the Canary Islands. ICES
Journal of Marine Science. 59, S69-S73.
Hixon, M.A. and Beets, J.P. (1989). Shelter characteristics and Caribbean fish
assemblages: experiments with artificial reefs. Bulletin of Marine Science. 44, 666-
Hood, P.B. and Schleider, R.A. (1992). Age, growth, and reproduction of gag,
Mycteroperca microlepis (Pisces: Serranidae), in the eastern Gulf of Mexico.
Bulletin of Marine Science. 51, 337-352.
Jennings, S. and Polunin, N.V.C. (1996). Effects of fishing effort and catch rate upon the
structure and biomass of Fijian reef fish communities. Journal of Applied Ecology.
Kamukuru, A.T., Hecht, T. and Mgaya, Y.D. (2005). Effects of exploitation on age,
growth and mortality of the blackspot sanpper, Lutjanusfulviflamma, at Mafia
Island, Tanzania. Fisheries Management and Ecology. 12, 45-55.
Keener, P., Johnson, G.D., Stender, B.W., Brothers, E.B. and Beatty, H.R. (1988).
Ingress of postlarval gag, Mycteroperca microlepis, (Pisces: Serranidae), through a
South Carolina barrier island inlet. Bulletin of Marine Science. 42, 376-396.
Kenchington, T.J. (1999). Management of the Gulf of Mexico gag grouper fisheries: A
reconsideration (revised version). Gadus Associates report prepared for the
Southeastern Fisheries Association Inc., Musquodoboit Harbor, Nova Scotia: 47pp.
Kiel, B.L. (2004). Homing and spatial use of gag grouper, Mycteroperca microlepis.
Masters Thesis. University of Florida, Gainesville, Florida.
Kinkade, K. (2003). Sensors and lasers map ebb and flow of ocean life. Laser Focus
World. 39, 91-96.
Koenig, C.C., Chapman, R.W., Collins, M.R., Harris, P., McGovern, J., Sedberry, G.R.,
Wyanski, D.M., and A.G. Johnson. (1999). The effects of shelf-edge fishing on the
demographics of the gag, Mycteroperca microlepis, population of the southeastern
United States. Tallahassee, FL, Department of Biological Science, Florida State
Kostylev, V.E., Courtney, R.C., Robert, G. and Todd, B.J. (2003). Stock evaluation of
giant scallop (Placopecten magellanicus) using high-resolution acoustics for
seabed mapping. Fisheries Research. 60, 479-492.
Kuwamura, T., Yogo, Y. and Nakashima, Y. (1994). Population-dynamics of goby
Paragobiodon Echinocephalus and host coral Stylophora-pistillata. Marine Ecology
Progress Series. 103, 17-23.
Laurans, M., Gascuel, D., Chassot, E. and Thiam, D. (2004). Changes in the trophic
structure of fish demersal communities in West Africa in the last three decades.
Aquatic Living Resources. 17, 163-173.
Lee, S.H. and Kim, K. (2004). Side-scan sonar characteristics and manganese nodule
abundance in the Clarion-Clipperton Fracture Zones, NE equatorial Pacific. Marine
Georesources and Geotechnology. 22, 103-114.
Leimgruber, P., Christen, C.A., Laborderie, A. (2005). The impact of Landsat satellite
monitoring on conservation biology. Environmental Monitoring and Assessment.
Ley, J.A., Halliday, I.A., Tobin, A.J., Garrett, R.N. and Gribble N.A. (2002). Ecosystem
effects of fishing closures in mangrove estuaries of tropical Australia. Marine
Ecology Progress Series. 245, 223-238.
Lindberg, W.J. and Loftin, J.L. (1998) Effects of artificial reef characteristics and fishing
mortality on gag (Mycteroperca microlepis) productivity and reef fish community
structure. Final Project Report, Florida Marine Resources Grants Number MR-073.
Lindberg, W.J., Frazer, T.K., Portier, K.P., Vose, F., Loftin, J, Murie, D.J., Mason, D.M.,
Nagy, B., and Hart, M.K. Density-dependent habitat selection and performance by
a large mobile reef fish. Ecological Applications. In Press.
Mack, P.E. (1990). Viewing the Earth: The social construction of the Landsat Satellite
System. MIT Press, Cambridge, USA.
McClanahan, T.R. and Mangi, S. (2001). The effect of a closed area and beach seine
exclusion on coral reef fish catches. Fisheries Management and Ecology. 8, 107-
McGlennon, D. and Branden, K.L. (1994). Comparison of catch and recreational anglers
fishing on artificial reefs and natural seabed in Gulf St. Vincent, South Australia.
Bulletin of Marine Science. 55, 510-523.
McErlean, A.J. (1963). A study of the age and growth of the gag, Mycteroperca
microlepis Goode and Bean (Pisces: Serranidae) on the west coast of Florida.
Florida Board of Conservation Marine Laboratory, Technical Series. 41, 29pp.
Milon, J.A. (1989). Artificial marine habitat characteristics and participation behavior by
sport anglers and divers. Bulletin of Marine Science. 44, 853-862.
Morton, R.A., Miller, T. and Moore, L. (2005). Historical shoreline changes along the US
Gulf of Mexico: A summary of recent shoreline comparisons and analyses. 21,
Pamo, E.T. (2004). Community production practices and desertification in the Sahelo-
Sudanian region of Cameroon at the turn of the millennium. Environmental
Monitoring and Assessment. 99, 197-210.
Pickrill, R.A., Todd, B.J. (2003). The multiple roles of acoustic mapping in integrated
ocean management, Canadian Atlantic continental margin. Ocean and Coastal
Management. 46, 601-614.
Roberts, C.M. (1994). Rapid build-up of fish biomass in a Caribbean marine reserve.
Conservation Biology. 9, 815-826.
Roberts, C.M., Bohnsack, J.A., Gell, F., Hawkins, J.P. and Goodridge, R. (2001). Effects
of marine reserves on adjacent fisheries. Science. 294, 1920-1923.
Roberts, C.M. and Ormond, R.F.G. (1987). Habitat complexity and coral reef fish
diversity and abundance on Red Sea fringing reefs. Marine Ecology Progress
Series. 41, 1-8.
Ross, S.W. and Moser, M.L. (1995). Life history of juvenile gag, Mycteroperca
microlepis, in North Carolina estuaries. Bulletin of Marine Science. 56, 222-237.
Rountree, R.A. (1989). Association of fishes with fish aggregation devices: effects of
structure size on fish abundance. Bulletin of Marine Scienec. 44, 960-972.
Schlaepfer, M.A., Rungs, M.C. and Sherman P.W. (2002). Ecological and evolutionary
traps. Trends in Ecology & Evolution. 17, 474-480.
Sherman, R.L., Gilliam, D.S. and Spieler, R.E. (2002). Artificial reef design: void space,
complexity and attractants. ICES Journal of Marine Science. 59, S196-S200.
Shulman, M.J. (1984). Resource limitation and recruitment patterns in a coral reef
assemblage. Journal of Experimental Marine Biology. 74, 85-109.
Solonsky A.C. (1985). Fish colonization and the effect of fishing activities on two
artificial reefs in Monterey Bay, California. Bulletin of Marine Science. 37, 336-
Sweatman, H.P.A. (1985). The influence of adults of some coral reef fishes on larval
recruitment. Ecological Monographs. 55, 469-485.
Szedlmayer, S.T. and Lee, J.D. (2004). Diet shifts of juvenile red snapper (Lutjanus
campechanus) with changes in habitat and fish size. Fishery Bulletin. 102, 366-375.
Turner, S.C., Porch, C.E., Heinemann, D., Scott, G.P. and Ortiz, M. (2001). Status of gag
in the Gulf of Mexico, Assessment 3.0. NMFS Southeast Fisheries Center, Miami
Laboratory, Miami SFD 01/02 -134. 33p.
Turpin, R.K. and Bortone, S.A. (2002). Pre- and post-hurricane assessment of artificial
reefs: evidence for potential use as refugia in a fishery management strategy. ICES
Journal of Marine Science. 59, S74-S82.
Verlaan, P.A. (1997). New seafloor mapping technology and Article 76 of the 1982
United Nations Convention on the Law of the Sea. Marine Policy. 21, 425-434.
Walters, C.J. and Martell, S.J.D. (2004). Fisheries Ecology and Management. Princeton
University Press, Princeton, New Jersey, USA.
Wantiez, L., Thollot, P. and Kulbicki, M. (1997). Effects of marine reserves on coral reef
fish communities from five islands in New Caledonia. Coral Reefs. 16, 215-224.
Weaver, D.C. (1996). Feeding ecology and ecomorphology of three sea basses (Pisces:
Serranidae) in the northeastern Gulf of Mexico. Thesis. University of Florida,
Williams, D.McB., and Sale, P.F. (1981). Spatial and temporal patterns of recruitment of
juvenile coral reef fishes to coral habitats within "One Tree Lagoon," Great Barrier
Reef. Marine Biology. 65, 245-253.
Wright, D.J. (1999). Getting to the bottom of it: Tools, techniques, and discoveries of
deep ocean geography. Professional Geographer. 51, 426-439.
The author was born in Washington, D.C. During his early years he was fascinated
by the fish, frogs, turtles and insects that he caught and raised and watched in his
neighborhood streams and ponds. As early as pre-school he was the tender of the class
aquarium. His first exposure to the scientific method came from Stanley Edwards, his
science teacher from fourth through sixth grades. Mr. Edwards effectively conveyed his
own passion for the sciences and natural world.
During summer vacation following his sophomore year of high school, he began
volunteering for Dr. Marjorie Reaka-Kudla, a coral reef biologist at the University of
Maryland. Being invited on a two-week research trip to the Florida Keys as a 16-year-
old who was half way through high-school convinced him he was in the right field.
Involvement in her research continued through his high school years.
He attended the University of California at Santa Barbara from which he graduated in
1998 with a Bachelor of Science degree in aquatic sciences. After graduation he returned
to the University of Maryland, this time working with Dr. David Secor of the
University's Chesapeake Biological Laboratory. Following three years as a faculty
research assistant he decided to return to school in pursuit of a master degree. The
opportunity to again work beneath the water drew him to the University of Florida where
he studied under Dr. William J Lindberg.
Following graduation he has accepted a position in Gainesville, FL, with Golder
Associates, an environmental consulting firm.