![Map of the study area (i.e., Ambrose Channel [entrance to New York Harbor] to Cape Henlopen Channel [entrance to Delaware Bay]) and the New Jersey trawl sampling locations defined by the New Jersey Ocean Stock Assessment Program.](https://www.researchgate.net/profile/Juan-Levesque/publication/302964645/figure/fig1/AS:360881612443648@1463052289704/Map-of-the-study-area-ie-Ambrose-Channel-entrance-to-New-York-Harbor-to-Cape_Q320.jpg)
Content uploaded by Juan C. Levesque
Author content
All content in this area was uploaded by Juan C. Levesque on May 12, 2016
Content may be subject to copyright.
RESEARCH ARTICLE
Spatio-temporal Variation in
Distribution and Relative Abundance of
Mid-Atlantic Bight Fishes and
Invertebrates off the Coast of New
Jersey (USA)
Fi sh er ie s and
Aq uacu lture Journal,
Vol. 2013: FAJ-85
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
1
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
Spatio-temporal Variation in Distribution and Relative Abundance
of Mid-Atlantic Bight Fishes and Invertebrates off the Coast
of New Jersey (USA)
Juan C Levesque1
Geo-Marine Inc., Environmental Resource Division, Marine Science Department,
2201 Avenue K, Suite A2, Plano, TX 75074, USA.
1Present Address: Cardno ENTRIX, Ecological Division, 3905 Crescent Park Drive,
Riverview, Florida 33578, USA.
Correspondence: Juan.Levesque@cardno.com, shortfin_mako_shark@yahoo.com
Accepted: May 23, 2013; Published: Jun 9, 2013
Abstract
Commercial and recreational fisheries within the Mid-Atlantic Bight are an important component of the local and national
economy. In 2010, commercial fisheries in New Jersey ranked sixth in value ($177,935,588) and eighth in landings (73,406 mt)
in the United States. The estimated number of recreational fishing trips in New Jersey ranged from 5.4 million (2009) to 7.4
million (2007) during 2003 through 2010. Because fish communities and the habitats they rely upon can be negatively impacted
by anthropogenic activities, descriptive baseline information on the coastal beach fish community beyond the surf zone must
be available to marine resource managers so they can make informed decisions. Given this management need to facilitate
objective informed marine resource decision-making, the main goal of this investigation was to provide a descriptive profile of
the New Jersey coastal fish and invertebrate community. The specific objectives were to assess fish and invertebrate population
dynamics by identifying trends (or lack thereof) in spatio-temporal variation in relative abundance and distribution for the most
numerically dominant and economically valuable fish and invertebrate species found off the coast of New Jersey. The findings
showed that annual relative abundance was stable among species, but there were temporal and spatial differences in overall
fish and invertebrate relative abundance among specific species. In addition, results demonstrated there was a seasonal
difference in species composition. Overall, butterfish and scup were the most numerically dominant species and relative
abundance generally increased from spring to summer. Most fish showed a negative binominal distribution and the highest
densities for butterfish and scup occurred at depths between 10 and 20 m.
Keywords: Commercial fisheries; ecosystem management; fisheries monitoring programs; fishery management; recreational
fisheries.
1. Introduction
Commercial and recreational fisheries within the Middle (Mid) Atlantic Bight (the area of the U.S. east coast and
continental shelf between Cape Cod, Massachusetts and Cape Hatteras, North Carolina) are an important
component of the local and national economy, especially the fisheries resources off the coast of New Jersey.
According to NMFS [1], the economic value of New Jersey commercial fishery landings over the last 10 years
(2002−2011) ranged between $112 and $214 million with a mean of $153.7 million; however, the actual economic
value was probably much greater in terms of jobs, goods, and services. In 2010, New Jersey commercial fisheries
ranked sixth in value ($177,935,588) and eighth in landings (73,406 mt) in the United States [1]. Besides the
economic value associated with commercial fisheries, recreational fisheries also have an important economic
impact to local and regional communities [2]. Recreational fishing off the coast of New Jersey is a popular year-
round hobby that consists of anglers targeting a variety of species from shore, man-made structures, and fishing
vessels (private and charter). The estimated annual number of recreational fishing trips in New Jersey ranged from
5.4 million (2009) to 7.4 million (2007) during 2003 through 2010 [3]. Besides recreational anglers fishing aboard
private fishing vessels, New Jersey’s recreational fishing fleet also consists of approximately 100 party and 300
charter vessels, which is the largest fleet of its kind on the east coast of the United States [4].
The New Jersey coastline consists of numerous nearshore (e.g., estuaries, bays, salt marshes, tidal creeks,
and coastal beaches) and offshore marine environments (e.g., shoals, sand ridges, continental shelf, canyons, hard
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
2
Research Article
bottom, and artificial reefs [e.g., ship wrecks and man-made structures]) that provide important habitat (i.e.,
feeding, spawning, and nursery grounds) to many economically valuable fish and invertebrates [5-7]. In general,
fish habitats off the coast of New Jersey are categorized as coastal beaches, pelagic, demersal (benthic), and hard
bottom (i.e., natural or artificial reef-structures) communities. Because these habitats are considered essential to a
variety of marine species, some of them have been officially designated by Fishery Management Councils (FMCs)
as Essential Fish Habitat (EFH) or Habitat Areas of Particular Concern, which are defined as discrete subsets of EFH
that provide important ecological functions that are especially vulnerable to degradation [8, 9].
To understand the importance and ecological function provided by EFH, information on the connection
between habitat and environment (e.g., temperature, salinity, and DO) is required by fishery managers. This type
of information is especially important for evaluating the population dynamics (relative abundance and distribution)
of fishes. It is well-documented in the scientific literature that relative fish abundance and distribution varies by
life-stage [10, 11] and that it is often influenced by local environmental conditions (e.g., salinity, water
temperature, and dissolved oxygen level), habitat, or physical factors that usually change with season [12]. Also,
relative fish abundance and distribution are dependent upon and influenced by life-history requirements and
migration behaviors [13]. Although there are natural patterns of variability in fish population, anthropogenic
activities, such as coastal development, pollution (e.g., nutrients and pH level), commercial and recreational
fishing, habitat alteration/destruction (e.g., sand mining) can significantly affect fish abundance [14-16]. Actually,
Jackson et al. [17] stated that nearly every estuarine species (i.e., fishes and macroinvertebrates) has already been
adversely affected to some degree by human activities; anthropogenic activities have altered the relative
abundance of many coastal fisheries (e.g., blue crabs [Callinectes sapidus], oysters, and Atlantic striped bass
[Morone saxatilis]). To assess the potential impacts of anthropogenic activities on local fish populations, baseline
population estimates and the environmental factors that influence these metrics are needed.
In general, the ideal method for assessing spatial and temporal dynamics of fishes is by using long-term
systematic fisheries data collected through a fisheries-independent monitoring program. To date, one of the oldest
ongoing fisheries independent monitoring programs in the United States is the New Jersey Ocean Stock
Assessment (OSA) program. The New Jersey Department of Environmental Protection (NJDEP), Division of Fish and
Wildlife, established the OSA program in August 1988 for several reasons including the following: to develop a
comprehensive baseline data for coastal recreational fishes and their forage items; to develop recruitment indices
for recreational fishes and documentation of annual relative abundance of juvenile fish; to provide a scientific
basis to formulate or modify existing management plans for recreational fishes; and to provide information to
complement other state and federal data for estimating populations and developing predictive models for
managing fish stocks [18, 19]. Fisheries independent monitoring data can be used in a variety of ways, but one of
the key applications is to evaluate the population dynamics of the local fish community and to identify important
fish habitats, which is generally assessed by estimating relative abundance and distribution over space and time
[20].
Various researchers have examined and described the estuarine and coastal beach fish communities off
the coast of New Jersey [5, 21], but detailed information describing the coastal fish community beyond the surf
zone is lacking [7]. Because fish communities and the habitats they rely upon can be negatively impacted by
anthropogenic activities (e.g., urban development, dredging, commercial and recreational fisheries), descriptive
baseline information on the coastal beach fish community beyond the surf zone must be available to marine
resource managers so they can make informed decisions. Given this management need to facilitate objective
informed marine resource decision-making, the main goal of this investigation was to provide a descriptive profile
of the New Jersey coastal fish and invertebrate community. The specific objectives were to assess fish and
invertebrate population dynamics by identifying trends (or lack thereof) in spatio-temporal variation in relative
abundance and distribution for the most numerically dominant and economically valuable fish and invertebrate
species found off the coast of New Jersey.
2. Methods
2.1. Study area
The New Jersey coastline is about 210 km long and consists of many beaches and islands (8−29 km) that serve as a
barrier between the Atlantic Ocean and the nearshore waters. The nearshore waters of New Jersey are connected
to the Atlantic Ocean by twelve inlets (Cape May Inlet to the South; Shark River Inlet to the North). The study area
encompassed approximately 4,600 km2 (Figure 1) and consisted of the coastal nearshore waters from Ambrose
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
3
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
Channel, New Jersey (i.e., the entrance to New York Harbor) to Cape Henlopen Channel, New Jersey (i.e., the
entrance to Delaware Bay). The study area included the waters depths between 9.1 (30 ft) and 27.4 m (90 ft).
Figure 1: Map of the study area (i.e., Ambrose Channel [entrance to New York Harbor] to Cape Henlopen Channel [entrance
to Delaware Bay]) and the New Jersey trawl sampling locations defined by the New Jersey Ocean Stock Assessment Program.
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
4
Research Article
2.2. Experimental survey design
To evaluate spatial fish dynamics, the survey area was divided into nine strata zones (15−23) that were based on
depth (0-10 m, 10-20 m, and 20-30 m) and location (latitude). To be consistent with established federal field-
sampling programs, the New Jersey OSA survey incorporated the same latitudinal boundaries defined by the
National Marine Fisheries Service (NMFS), Northeast Fishery Science Center (NEFSC), for the Northwest Atlantic
Groundfish Survey program; exceptions were those strata at the northern and southern ends of the New Jersey
coastline where NMFS extended its survey into New York or Delaware waters. The New Jersey OSA program
truncated the boundaries to exclude these northern and southern strata, confining the survey area to include only
the waters adjacent to the New Jersey coastline and the ocean waters off Delaware Bay. The longitudinal
boundaries consisted of the 9.1 (30 ft), 18.3 (60 ft), and 27.4 m (90 ft) isobaths. Because the bottom contours were
irregular, the stratum boundaries were smoothed using standard GIS techniques [22].
To reduce sampling bias, each stratum was further divided into smaller blocks and assigned a sequential
number. Mid-shore blocks (9.1−18.3 m) and offshore (18.3−27.4 m) blocks were 2.0 minute longitude by 2.5
minute latitude, whereas inshore (5.5−9.1 m) blocks were 1.0 minute longitude by 1.0 minute latitude. Inshore
block dimensions were smaller because the strata were narrower and encompassed less area than the mid and
offshore strata; thus, the smaller block size permitted a greater number of potential sampling sites than would be
possible with larger dimensions. However, it should be noted that the blocks truncated by stratum boundaries that
had less area than the whole blocks (those reduced in area by more than one-half) were generally not assigned a
number [18, 19].
2.3. Experimental sampling approach
Sampling surveys were conducted bimonthly (every two months: February, April, June, August, October, and
December) from 1988 to 1990, but in 1990, the December and February surveys were replaced by a single winter
survey in January [18, 19]. Today, this identical temporal sampling approach continues with one winter survey in
January, followed by surveys in April, June, August, and October. A total of five surveys are conducted each year
[18, 19]. Since August 1991, the overall sampling survey effort has consisted of about 39 hauls (i.e., two samples
from each strata plus one additional haul in each of the nine larger strata) per survey. A total of 186 tows are
conducted each year (39 stations per trip for spring–fall months and 30 stations per trip for winter months). Using
a depth-stratified random sampling design, a minimum of 10 tows are completed per depth interval (0–10 m, 10–
20 m, and 20–30 m).
Sampling stations (survey site location) were randomly chosen by the program leader between 1988 and
1991, but this method was replaced in 1992 by the random number selection process generated by a personal
computer. Because stratum shapes were elongate and the sampling effort was limited, a station selection
procedure was used to reduce any spatial distribution sampling bias. The station selection procedure consisted of
limiting the first station to only the top half of the block numbers and the second station to the bottom half;
however, if a third station was selected then limitations were not imposed. For instance, haul one would be
selected from blocks 1 to 25 blocks, haul two from blocks 26 to 50, and haul three from blocks 1 to 50 for a
stratum with 50 blocks. For each station, three additional alternate sites were also selected using the same
procedures described above to account for any fixed fishing gear (e.g., traps or nets), bottom obstructions, or
other impediments that have prevented sampling at the initial station [18, 19].
2.4. Field sampling gear
Field sampling was conducted with a three-to-one otter trawl that was constructed of polyethylene twine with
forward netting (i.e., wings and belly). The otter two-seam trawl was constructed with 12 cm stretch mesh and
forward netting that tapered down to 8 cm stretch mesh rear netting. The otter trawl cod-end was constructed
with 7.6 cm stretch mesh and lined with a 6.4 mm bar mesh liner. The head rope was 25 m and the footrope was
30.5 m. The trawl bridle was 36.6 m long, and the top leg consisted of a 1.27 cm wire rope. The bottom leg
consisted of a 1.91 cm wire rope covered with 6.03 cm rubber cookies. The groundline length between the bridle
and otter trawl doors was 18.3 m. The trawl doors were wooden with steel shoes and the dimensions were 2.44 m
x 1.27 m; each trawl door weighed approximately 453.5 kg (1,000 lbs) [18, 19].
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
5
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
2.5. Field sampling
All tows were conducted during daylight hours defined as the hours between sunrise and sunset. The trawl tow
duration was standardized at 20 minutes (i.e., the time the net is deployed to when the winch brakes are set to
begin haul back) and the surface ground speed was maintained between 4.0 and 4.8 km/hr (2.9 and 3.5 mph).
Based on the estimated speed, one 20 minute tow generally covered a distance of around 1.85 km. The average
wing spread of the trawl was around 13 m and each tow swept an area about 24,076 m2 (Byrne; personal
communication 26 May 2009). Tow durations that were shortened because of hangs, bottom obstructions, or
other problems were considered an adequate sample provided the tow duration was more than 15 minutes and
the net was not damaged. Tows that were less than 15 minutes were repeated unless there were extenuating
circumstances (e.g., bottom was known to be a difficult area to sample) that prevented the field crew from towing
the net. To standardize shorter tow durations (< 20 minutes), extrapolation was applied under the assumption
there was a direct relationship between catch and tow duration. To maintain a tow depth ratio of approximately
3:1, 91.5 m of wire was released during each deployment regardless of the depth. This specific tow wire depth
ratio was selected because the deepest sampling depth was 27.4 m and it ensured there was a sufficient distance
between the vessel and the net during shallow water (~ 9.1 m) tows [18, 19].
2.6. Field procedures
At each sampling station, bottom and sea surface temperature, salinity, and dissolved oxygen (DO) environmental
conditions were measured and recorded. Once the tow was completed, the gear was retrieved and the catch (fish
and macroinvertebrates) was identified to species, enumerated, and rough sorted using plastic buckets and wire
fish baskets. The total weight, or a representative sample (i.e., large catches), of each species was taken with a
hanging metric scale and the length (fork or total) was measured to the nearest cm for fish and the disk width (cm)
for stingrays. Depending on the macroinvertebrate species in the sample, various other measurements were
recorded. For example, the carapace width (mm) was measured for crabs, the carapace length (mm) for lobsters,
and mantle length (mm) for squids. Because some catches were large and mixed, a sub-sampled was weighed.
After the sub-sample was sorted, the species composition was extrapolated to determine the total catch [18, 19].
2.7. Data
To evaluate trends in fish and invertebrate populations off the coast of New Jersey, data obtained from the New
Jersey OSA program (2003−2008) were compiled, summarized, and sorted into two separate groups (i.e., top 5
species numerically collected and top 5 species having the most economic value [US$]) according to numerical
dominance and whether they were classified under a management plan or considered socially valuable (i.e.
recreational fishing). Data were individually and collectively evaluated by area (nine New Jersey OSA defined strata
[area 15−23; Figure 1]), month, and year using various univariate and multivariate procedures.
2.8. Statistical analysis
Prior to any analyses, the normality and homoscedacity of the data was evaluated using Kolmogorov-Smirnov and
Bartlett tests [23]. To ensure robustness, normality was also checked by constructing a normal probability plot of
the residuals. Also, as part of the assessment process, all outlier observations were investigated before being
rejected or retained. If the data passed the normality test, then parametric procedures were employed; otherwise,
data were transformed using an appropriate transformation process (e.g., log, square root, or arcsine square root)
to meet the underlying assumptions of normality [23]. However, if the data still failed to meet the assumptions of
normality even after transformation, then non-parametric tests were applied. For all analyses, statistical
significance was defined as P < 0.05. In the presence of significance at the 95 percent confidence level for the
omnibus analysis of variance (ANOVA) or the Kruskal–Wallis non-parametric multisample test, a post-hoc multiple
comparison test was used to perform pair wise comparisons to differentiate the differences among the population
means (parametric) or medians (non-parametric). All analyses were conducted using Microsoft Excel® and
Statgraphics Centurion XVI® Version 16.1.
First, total catch by individual species was tabulated, summarized, and graphed by year and area. Second,
data were filtered and compiled into two separate datasets ([top 5 species numerically collected] and economic
value [top 5 species collected]); descriptive statistics were generated and evaluated for each dataset. It should be
noted that the New Jersey OSA survey program collected more species and individuals than evaluated for these
analyses; but only those species having some economic (commercial fishing) or social (recreational fishing)
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
6
Research Article
importance (i.e., having an associated fishery management plan or under the jurisdiction of a state or federal
fishery organization) were considered for evaluation. In general, the most economically valuable species were
evaluated and selected according to their numerical value (dominance) and their 2010 commercial fisheries
landings rank [1]. Also, a six-year time series (2003−2008) was chosen for these analyses because the objective was
to assess the current population status of fish and invertebrate stocks. This particular time length snapshot was
selected because it was assumed that if a shorter time-series were used then it would not have been sufficient in
duration to incorporate natural annual variability and if a longer time-series had been used then it might not have
directly depicted the current status of stocks. Third, datasets were statistically evaluated using various univariate
and multivariate procedures.
Separate one-way ANOVAs were used to test the null hypothesis that annual environmental conditions
(bottom water temperature, bottom salinity, and bottom DO) were equal. A multiple Spearman rank correlation
procedure was used to evaluate the association between the relative abundance of each species in their respected
group (numerically dominant and economically valuable). Monthly and annual total catch were summarized and
analyzed by area. Separate one-way ANOVAs were used to test the null hypothesis that the monthly and annual
total number of fish collected by area was equal. Also, separate one-way ANOVAs were used to test the null
hypothesis that monthly and annual relative abundance density (number of fish collected per m2) by area was
equal. The annual and monthly relative abundance (density) of the main species collected in each group was
examined using a simple linear regression to categorize the slope of the fitted trendline as increasing (positive) or
decreasing (negative). Three separate multiple linear regression models using the forward stepwise procedures
were applied to describe the best relationship between mean density (annual, month ly, and depth) of the main
species collected in each group and three independent variables (mean bottom temperature, mean bottom DO,
and mean bottom salinity). Separate principal component analysis (PCA) procedures were used to identity patterns
(positive and negative) in the mean relative abundance of the numerically dominant and economically valuable
species and the nine areas (15−23).
3. Results
3.1. Survey effort
New Jersey DEP personnel of the OSA program completed a total of 735 tows off the coast of New Jersey within
nine strata (areas: 15−23) during 2003 through 2008. The total number of tows conducted in each area during the
six-year period ranged from 62 in area 21 to 87 in area 17 with a mean of 82 tows. The mean number of tows
completed per year was 123 and the mean number of tows completed per area was 14.
3.2. Environmental characteristics
Environmental conditions varied by season and year during 2003 through 2008. As expected, sea surface
temperatures increased from spring to summer and decreased from fall to winter. Annual sea surface
temperatures ranged from 0.7 to 27° C with a mean sea surface temperature of 14.8° C. Similarly, bottom water
temperatures ranged from 0.7 to 24.7° C with a mean bottom water temperature of 12.0° C; bottom water
temperature varied significantly among years (F [5, 729] = 4.13, P < 0.05). Salinity at the surface and bottom were
generally similar over the study period. Annual salinity at the surface ranged from 21.9 to 33.2 ppt with a mean of
30.6 ppt, while bottom salinity ranged from 28.1 to 33.4 ppt with a mean of 31.4 ppt; bottom salinity varied
significantly among years (F [5, 729] = 34.41, P < 0.05). In contrast to temperature and salinity, DO levels varied
from surface to bottom with DO levels at the bottom being slightly lower. Annual surface DO levels ranged from
4.1 to 15.4 with a mean of 8.5, while bottom DO levels ranged from 2.3 to 11.3 with a mean of 7.6; bottom DO
varied significantly among years (F [5, 729] = 7.39, P < 0.05).
3.3. Species composition
A total of 1,619,401 individuals were collected from 2003 to 2008 representing five broadly defined fish and
macro-invertebrate groups (state or federally managed [FMPs and EFH] species) consisting of 32 species (Table 1).
Overall, five species were the most numerically dominant (n = 1,332,264 or 82.3 percent) and economically
valuable (n = 691,241 or 42.7 percent) (Figure 2). The most numerically dominant species collected were butterfish
(n = 489,376 or 30.2 percent) followed by scup (n = 339,301 or 20.9 percent) and squid (n = 312,299 or 19.3
percent). The mean length of the numerically dominant species ranged from 10.3 (butterfish) to 24.4 cm
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
7
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
(weakfish). The most economically valuable species collected were scup (n = 339,301 or 25.5 percent) followed by
squid (n = 312,299 or 23.4 percent) and Atlantic mackerel (n = 16,395 or 1.6 percent). The mean length of the
economically valuable species ranged from 11.5 (squid) to 37.7 cm (summer flounder). It should be noted that
some species, such as squid and scup were both abundant (numerically dominant) and economically valuable.
Nonetheless, for these analyses the species were evaluated separately under their respected group. In general,
most of the samples (e.g., butterfish) were not collected from a normal population scenario (Shapiro-Wilk W =
0.2244; P < 0.05), but instead were modeled best under a negative binomial distribution given the variance was
greater than the mean, which suggested that schooling fishes followed a nonrandom or clumped distribution
pattern. These findings were also consistent with various Kolmogorov-Smirnov goodness-of-fit tests that
demonstrated the samples did not come from a normal distribution (P < 0.05).
Based on the estimated covariance among the numerical dominant species (densities), the Spearman rank
correlations between each pair indicated there was a statistically significant non-zero positive correlation at the
95.0 percent confidence level (n = 735) for the following pairs: butterfish and scup, butterfish an d squid, butterfish
and weakfish, scup and squid, and scup and weakfish (P < 0.05). In addition, the following pairs were also detected
to be statistically significant, but negatively correlated: butterfish and Atlantic herring, scup and Atlantic herring,
squid and Atlantic herring, squid and weakfish, and Atlantic herring and weakfish (P < 0.05). The Spearman rank
correlations between each pair of economically valuable species indicated there was a statistically significant non-
zero positive correlation at the 95.0 percent confidence level (n = 735) for the following pairs: squid and scup,
squid and summer flounder, and scup and summer flounder (P < 0.05). Also, the test showed the following pairs
were statistically significant, but negatively correlated: squid and Atlantic mackerel, squid and Atlantic menhaden,
scup and Atlantic mackerel, scup and Atlantic menhaden, Atlantic mackerel and Atlantic menhaden, and Atlantic
mackerel and summer flounder (P < 0.05).
Table 1: Broad-defined categories of fish and macro-invertebrates collected by the New Jersey Ocean Stock Assessment
Program during 2003 through 2008.
Fish/Macro-
invertebrate Group
Species
Anadromous and
Sciaenids
alewife (Alosa pseudoharengus), blueback herring (Alosa aestivalis), American shad (Alosa sapidissima),
striped bass (Morone saxatilis), Atlantic sturgeon (Acipenser oxyrinchus oxyrinchus), Atlantic croaker
(Micropogonias undulatus), spot (Leiostomus xanthurus), weakfish (Cynoscion regalis)
Flatfish and
Elasmobranchs
summer flounder (Paralichthys dentatus), winter flounder (Pseudopleuronectes americanus), witch
flounder (Glyptocephalus cynoglossus), windowpane flounder (Scophthalmus aquosus), little skate
(Leucoraja erinacea), winter skate (Leucoraja ocellata), clearnose skate (Raja eglanteria), spiny dogfish
(Squalus acanthias)
Invertebrates
American lobster (Homarus americanus), Atlantic horseshoe crab (Limulus polyphemus), squid
(Doryteuthis spp)
Gadids
Atlantic cod (Gadus morhua), red hake (Urophycis chuss), silver hake (Merluccius bilinearis), goosefish
(Lophius americanus), ocean pout (Zoarces americanus)
Schooling and Reef
Fish
Atlantic herring (Clupea harengus), Spanish mackerel (Scomberomorus maculatus), Atlantic
menhaden(Brevoortia tyrannus), butterfish (Peprilus triacanthus), scup (Stenotomus chrysops), black
seabass (Centropristis striata), tautog (Tautoga onitis), bluefish (Pomatomus saltatrix)
3.4. Annual dynamics
The total number of numerically dominant individuals collected during 2003 through 2008 ranged from 134,399
individuals in 2005 to 323,946 individuals in 2006 with a mean of 222,044 individuals per year (Figure 3). The total
number of economically valuable individuals collected ranged 58,563 individuals in 2008 to 176,845 individuals in
2006 with a mean of 115,207 individuals per year (Figure 3). Overall, the total number of numerically dominant (F
[5, 24] = 0.10, P = 0.9626) and economically valuable species (F [5, 24] = 0.23, P = 0.9457) collected were both
similar among years, but there was a significant difference in the mean annual number of individuals collected
among economically valuable species (F [5, 24] = 16.26, P < 0.05). A post-hoc Tukey multiple comparison test
demonstrated there was a significant difference detected between the total annual number of scup collected and
Atlantic mackerel, Atlantic menhaden, and summer flounder. Significant differences were also detected between
the total annual number of squid collected and Atlantic mackerel, Atlantic menhaden, and summer flounder (P >
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
8
Research Article
0.05). However, there was no significant difference detected in the mean annual number of individuals collected
among numerically dominant species (F [5, 24] = 1.05, P = 0.4019).
Figure 2: The frequency distribution of the most dominant (Top Graph: top 5 species) and economically valuable species
(Bottom Graph: top 5 species) collected by the New Jersey Ocean Stock Assessment Program off the coast of New Jersey
from 2003 to 2008. Error bars depict the standard error.
Among the numerically dominant species collected, butterfish had the greatest (0.0277 butterfish/m2)
mean relative abundance and weakfish the lowest (0.0048/m2). The total number of butterfish collected ranged
from 26,213 in 2005 to 190,685 in 2008 with a mean of 81,563 butterfish per year. The mean annual relative
abundance of butterfish ranged from 0.0089 butterfish/m2 in 2005 to 0.0649 butterfish/m2 in 2008 with a mean of
0.0277 butterfish/m2 per year (Figure 4); mean density was significantly different among years (H = 37.8963; P <
0.05). The positive correlation between mean butterfish density and time was described by the following simple
linear regression model: mean annual butterfish density = -18.7094 + 0.00934286*Year. The model as fitted
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
9
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
explained 57.87 percent of the variability in mean annual butterfish abundance; however, there was no significant
relationship between mean butterfish abundance and time at the 95.0 percent or higher confidence level (F [1, 4]
= 5.49, P = 0.0791). The results of fitting a multiple linear regression model using a forward stepwise procedure to
describe the best relationship between mean annual butterfish density and three independent variables (mean
bottom temperature, mean bottom DO, and mean bottom salinity) showed that the model as fitted explained 84.5
percent of the variability in mean annual butterfish density; there was also a significant relationship between mean
annual butterfish abundance and bottom DO at the 95.0 percent or higher confidence level (F [1, 4] = 21.88, P =
0.0095). The positive correlation between mean butterfish density and environmental conditions was described by
the following multi-linear regression model: mean annual butterfish density = 0.416729 - 0.0513006*mean bottom
DO.
Among the economically valuable species collected, scup had the greatest (0.0184 scup/m2) mean relative
abundance and summer flounder the lowest (0.005 summer flounder/m2). The total number of scup collected
ranged from 13,660 in 2008 to 95,780 in 2007 with a mean of 56,660 scup per year. The annual mean density of
scup ranged from 0.0045 scup/m2 in 2008 to 0.0311 scup/m2 in 2007 with a mean of 0.0184 scup/m2 per year
(Figure 4); mean density was significantly similar among years (H = 8.6948; P = 0.1219). A simple linear regression
showed that the model as fitted was unable to explain the variability (0.35%) in mean annual scup abundance;
there was no significant relationship between mean scup abundance and time at the 95.0 percent or higher
confidence level (F [1, 4] = 0.01, P = 0.9109). Moreover, a multiple linear regression model was unable to explain
the variability in mean scup density and three independent variables (mean bottom temperature, mean bottom
DO, and mean bottom salinity).
Figure 3: The annual frequency distribution of the most dominant (Top graph: top 5 species) and economically valuable
species (Bottom graph: top 5 species) collected by the New Jersey Ocean Stock Assessment Program off the coast of New
Jersey from 2003 to 2008.
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
10
Research Article
Figure 4: The mean annual relative abundance (number of fish/m2) of the most numerically dominant (Top Graph: top 5
species) and economically valuable species (Bottom Graph: top 5 species) collected by the New Jersey Ocean Stock
Assessment Program off the coast of New Jersey from 2003 to 2008. The error bars depict the standard error.
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
11
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
3.5. Temporal dynamics
The total number of numerically dominant individuals collected during 2003 through 2008 ranged from 1,460 in
May to 477,556 in October with a mean of 166,533 individuals per month; there was a significance difference in
total catch among months (F [7, 32] = 2.75, P = 0.0236). A post-hoc Tukey procedure demonstrated there was a
significant difference detected between the total numbers of individuals collected in February and those collected
in April, June, August, and October. There was also a significant difference found between the total number of
individuals collected in May and those in June, August, and October. Slightly different than the numerically
dominant species, the total number of economically valuable individuals collected ranged from 464 in February to
180,381 in August with a mean of 54,074 individuals per month; there was no significance difference in total catch
among months (F [7, 32] = 2.06, P = 0.078).
The mean monthly density of numerically dominant individuals collected ranged from 0.006 fish/m2 in
May to 0.1323 fish/m2 in October with a mean of 0.0532 fish/m2 per month. Similarly, the mean monthly density
of economically valuable individuals collected ranged from 0.001 fish/m2 in February to 0.0833 fish/m2 in October
with a mean of 0.312 fish/m2 per month. The mean monthly density of individuals collected among dominant
species (H = 10.8435; P = 0.1456) and economically valuable species (H = 10.4169; P = 0.1662) were both
significantly similar among months, but there was a significant difference in the monthly number of individuals
collected among numerically dominant species (H = 13.795; P = 0.03201) and economically valuable species (H =
20.4785; P = 0.0023).
Overall, the monthly number of individuals collected varied significantly among specific species. The
number of dominant individuals collected in May ranged from 12 Atlantic herring to 710 butterfish (Figure 5);
there was a significant difference in the number of individuals collected in May among numerically dominant
species (F [4, 45] = 9.27, P < 0.05). The number of economically valuable individuals collected in October ranged
from 12 Atlantic mackerel to 201,081 scup (Figure 5); there was a significant difference in the number of
individuals collected in October among economically valuable species (F [4, 745] = 95.46, P < 0.05). The positive
correlation between mean butterfish density and time was described by the following simple linear regression
model: mean monthly butterfish density = 0.00140399 + 0.00270358*month. However, the model as fitted only
explained 20.4 percent of the variability in mean monthly butterfish abundance and there was no significant
relationship between mean butterfish abundance and time at the 95.0 percent or higher confidence level (F [1, 6]
= 1.53, P = 0.2618). The results of fitting a multiple linear regression model using a forward stepwise procedure to
describe the best relationship between mean monthly butterfish density and three independent variables (mean
bottom temperature, mean bottom DO, and mean bottom salinity) showed that the model as fitted explained 54.5
percent of the variability in mean monthly butterfish density; there was a significant relationship between mean
monthly butterfish abundance and bottom DO at the 95.0 percent or higher confidence level (F [1, 6] = 7.2, P = 0.
0.0364). The positive correlation between mean butterfish density and environmental conditions was described by
the following multi-linear regression model: mean monthly butterfish density = 0.10193 - 0.0106468*bottom DO.
Similarly, the number of economically valuable individuals collected in May ranged from 1 scup to 256
Atlantic mackerel; there was a significant difference in the number of individuals collected in May among
economically valuable species (H = 10.8727, P = 0.0281). The number of economically valuable individuals collected
in October ranged from 3 Atlantic mackerel to 202,081 scup; there was a significant difference in the number of
individuals collected in October among numerically dominant species (H = 395.955, P < 0.05). The positive
correlation between mean scup density and time was described by the following simple linear regression model:
mean monthly scup density = -0.00886823 + 0.00377331*month. The model as fitted explained 52.9 percent of the
variability in mean monthly scup abundance; there was a significant relationship between mean monthly scup
abundance and time at the 95.0 percent or higher confidence level (F [1, 6] = 6.75, P = 0.0407). The results of
fitting a multiple linear regression model using a forward stepwise procedure to describe the best relationship
between mean monthly scup density and three independent variables (mean bottom temperature, mean bottom
DO, and mean bottom salinity) showed that the model as fitted was unable to explain any of the variability in
mean monthly scup density; there was no significant relationship between scup abundance and bottom DO at the
95.0 percent or higher confidence level (F [1, 6] = 0.02, P > 0.05).
3.6. Spatial dynamics
In general, the total number of fish collected by area did not vary significantly over the study period for either
dominant (H = 5.9687, P = 0.6507) or economically valuable (F [8, 36] = 0.3, P = 0.9608) species. The total number
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
12
Research Article
of numerically dominant individuals collected ranged from 58,097 in area 21 to 243,340 in area 16 with a mean of
148,028 individuals per area. Similarly, the total number of economically valuable individuals collected ranged
from 10,248 in area 18 to 152,701 in area 17 with a mean of 76,804 individuals per area.
Figure 5: The mean monthly relative abundance (number of fish/m2) of the most dominant (Top Graph: top 5 species) and
economically valuable species (Bottom Graph: top 5 species) collected by the New Jersey Ocean Stock Assessment Program
off the coast of New Jersey from 2003 to 2008.
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
13
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
The mean density of numerically dominant individuals collected ranged from 0.0378 fish/m2 in area 21 to
0.1203 fish/m2 in area 16 with a mean of 0.0742 fish/m2 per area (Figure 6). Similarly, the mean density of
economically valuable individuals collected ranged from 0.0048 fish/m2 in area 18 to 0.0707 fish/m2 in area 17 with
a mean of 0.0367 fish/m2 per area (Figure 6). Overall, the mean density of individuals collected among dominant
species (H = 4.2675; P = 0.8322) and economically valuable species (H = 2.3357; P = 0.9689) were both significantly
similar among areas, but there were significant differences in mean density among numerically dominant species
(H = 17.2321; P = 0.0017) and economically valuable species (H = 30.3909; P < 0.05).
The number of individuals collected in each area varied significantly among individual species. The
number of numerically dominant individuals collected in area 21 ranged from 1,662 Atlantic herring to 19,253
butterfish; there was a significant difference in the number of individuals collected in area 21 among numerically
dominant species (H = 23.6006; P < 0.05). In area 16, the number of numerically dominant individuals collected
ranged from 6,139 weakfish to 124,528 butterfish; there was a significant difference in the number of individuals
collected in area 16 among numerically dominant species (H = 108.244; P < 0.05). Sorting the data by depth (10,
20, and 30 m) showed there was a significant difference in mean butterfish density among depth (H = 8.7808; P =
0.01239); the greatest butterfish density was at 20 m (Figure 5). However, neither a simple linear or multi-linear
regression models were able to explain any of the variability between mean butterfish density and depth or any
environmental conditions. Slightly different than the numerically dominant species, the number of economically
valuable individuals collected in area 18 ranged from 8 Atlantic mackerel to 7,037 scup; there was a significant
difference in the number of individuals collected in area 18 among economically valuable species (H = 140.31; P <
0.05). In area 17, the number of economically valuable species collected ranged from 19 Atlantic menhaden to
92,885 scup; there was a significant difference in the number of individuals collected in area 17 among
economically valuable species (H = 174.87; P < 0.05). Sorting the data by depth (10, 20, and 30 m) showed there
was a significant difference in scup density among depth (H = 15.3143; P = 0.0004); the greatest scup density was
at 20 m. However, neither simple nor multi-linear regression models were able to explain any of the variability
between mean scup density and depth or environmental conditions.
The PCA performed on the species abundance (numerically dominant) by area produced three component
axes that explained 91.8 percent of the total variance in the dataset. Species abundance with high loading values
on Principal Component (PC) axis I were areas 22, 18, and 15; areas 23, 21, and 20 had high loading values on PC
axis II; and areas 15, 23, 19 had high loading values on PC axis III. Butterfish, scup, and squid were positively
correlated, while weakfish and Atlantic herring were negatively correlated with PC I. Squid, scup, and Atlantic
herring were positively correlated, while butterfish and weakfish were negatively correlated with PC II. Squid,
butterfish, and weakfish were positively correlated, while scup and Atlantic herring were negatively correlated
with PC III (Table 2).
The PCA performed on the species abundance (economically valuable) by area produced two component
axes that explained 98.9 percent of the total variance in the dataset. Species abundance with high loading values
on PC axis I were areas 18, 21, and 20; areas 23, 19 and 15 had high loading values on PC axis II. Scup, squid were
positively correlated, while Atlantic mackerel, Atlantic menhaden, and summer flounder were negatively
correlated with PC I. Squid was positively correlated, while scup, summer flounder, Atlantic mackerel, and Atlantic
menhaden were negatively correlated with PC II (Table 3).
Table 2: Principal Component Analysis. Species (numerically dominant) correlation loading values on the first three
components.
Species
Component I
Component II
Component III
Butterfish
2.12399
-2.26217
0.247459
Scup
1.20712
1.08323
-1.66305
Squid
0.822254
1.86996
1.58342
Atlantic Herring
-1.81735
0.666011
-0.408996
Weakfish
-2.33601
-1.35703
0.241169
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
14
Research Article
Table 3: Principal Component Analysis. Species (economically valuable) correlation loading values on the first two
components.
Species
Component I
Component II
Scup
3.3545
-1.75646
Squid
2.36084
2.16725
Atlantic mackerel
-1.93649
-0.0560474
Atlantic menhaden
-1.89938
-0.0125324
Summer flounder
-1.87948
-0.342207
Figure 6: The mean spatial relative abundance (number of fish/m2) of the most dominant (Top Graph: top 5 species) and
economically valuable species (Bottom Graph: top 5 species) collected by the New Jersey Ocean Stock Assessment Program
off the coast of New Jersey from 2003 to 2008.
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
15
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
4. Discussion
The commercial and recreational fisheries resources off the coast of New Jersey are among the most important in
the United States [1, 3]. These economically valuable fisheries rely upon a diversity of marine habitats that are
found in the nearshore and offshore waters off New Jersey [6]. Understanding the spatial and temporal population
dynamics of a community is essential for making informed conservation decisions. Given this management need to
understand these dynamics, the purpose and goal of this present study was to provide a descriptive profile of the
New Jersey coastal fish and invertebrate community.
4.1. Environmental characteristics
Readings of environmental measurements for every parameter (water temperature, salinity, and DO) recorded
during the study period varied significantly by season and year, which was expected for the Mid-Atlantic Bight
since it is a dynamic hydrological region. Although a thorough evaluation of the environmental factors and their
influence on fish relative abundance was not the goal of this study, the results did show that mean relative
abundance of butterfish and DO were positively correlated. It is difficult to speculate why temperature and salinity
did not influence overall annual relative abundance of the primary species, but these environmental factors
probably influenced some of the differences detected in monthly catch. Given the fact that many researchers have
reported how biological communities in the Mid-Atlantic Bight are affected by cold water masses that move into
the region from upwelling events, these factors should be investigated further. For instance, Moline et al. [24]
reported that strong winds from the southwest produced upwelling conditions where the nearshore pycnocline
angled upward in the water column producing an alongshore horizontal front. They also defined down-welling as
the condition where the nearshore pycnocline angled downward from the horizontal offshore pycnocline and was
often in contact with the bottom boundary layer. Overall, the researchers found that these events directed
impacted phytoplankton community structure. Episodic events also play a major role in the transport of larval
fishes and the distribution of juvenile fishes in the Mid-Atlantic Bight [24]. Epifanio and Garvine’s [25] review of the
literature found that wind and buoyancy were the principal forcing agents driving the transport of larval fish and
invertebrates on the continental shelf in the Middle Atlantic and South Atlantic Bights. Because understanding
trophic dynamics and the environmental parameters that control these dynamics is the key to understanding
community structure, it is recommended that this link be investigated further in future investigations.
4.2. Species composition
The waters off the coast of New Jersey support a variety of marine species, but the findings showed that only a few
species were common and most were uncommon or rare. In fact, only two species (butterfish and scup)
represented most (51 percent) of the catch. Nonetheless, these findings were consistent and similar to other
studies conducted previously in this region. Able et al. [26] reported collecting 51 species over five years (1995–
1999) and found that taxonomic richness varied among years ranging from 28 taxa represented in 1996 to 38 taxa
collected in 1997 and 1999. Overall, Able et al. [26] found that abundance and average density (number/100 m3) of
fish larvae from surf zone habitats were generally low for most species collected in northern New Jersey from 1995
through 1999; only a few species had relatively high densities. Similarly, Slacum et al. [6] also reported collecting
only a limited number of species off the coast of New Jersey. Overall, the researchers collected 19 species of fish
and 15 species of invertebrates from the offshore shoals. Slacum et al. [6] indicated the most abundant species for
each season were scup (32%) and butterfish (20%), which represented the greatest percentage of the total catch
over the two year study. In another study, Able et al. [7] documented over 40 taxa from 30 genera, but only six
species had densities exceeding 1.0 per 1,000 m3 in at least one season for all years combined. They found that
percent contribution (dominance) of the total larvae by individual species across habitat/season consisted of only a
few species; most species had low relative abundance. It is difficult to speculate why species diversity is relatively
low in the Mid-Atlantic Bight, but one possible explanation could be related to the classic latitudinal pattern
observed throughout the world. Generally, there is a decrease in species diversity with an increase in latitude for
many species, including marine fish [27]. The explanation for this phenomenon remains unclear, but it is probably
related to various factors, including environmental tolerance (water temperature), resource availability, trophic
dynamics, and community robustness. The trophic dynamics of the Mid-Atlantic Bight should be examined in
future studies since the findings of this present study found that some species were positively correlated (e.g.,
butterfish and scup) and others were negatively correlated (scup and Atlantic herring) by area; overall fish
community structure changed with season.
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
16
Research Article
4.3. Abundance and distribution
4.3.1. Annual dynamics
In general, the relative abundance for all species combined was stable over time, but some significant differences
were detected between relative abundance and time among individual species, which was probably related to the
fact that only a few species were dominant and most were uncommon. It’s possible that this could have driven the
statistical tests. Results showed that butterfish density, the most dominant species, was significantly different with
time and it was positively correlated. Interestingly, butterfish density was also positively correlated with bottom
DO, which might explain why there was a significant difference in annual butterfish density. Conclusive findings for
some of the other main species collected were unable to be detected using the analytical procedures used in
theses analyses. However, it is possible that the stability in annual relative abundance was more related to sample
size (low numbers) than to actual stable annual conditions. Although statistically insignificant, this did not explain
why the mean annual density for most of the species (numerically dominant and economically valuable) collected
followed a cyclical pattern of a decreasing density followed by an increasing density. Despite this apparent pattern,
weakfish and scup did not follow this pattern in 2005 and 2007, respectively. As such, these specific years should
be examined in more detail to help explain these observations for weakfish and scup. Maybe environmental
conditions or predator-prey relations affected the densities of these species during these years?
The findings of the present study showing the apparent stability of relative abundance and time appears
to be common for this region given that other researchers have reported similar findings within the Mid-Atlantic
Bight region. For instance, Able et al. [7] reported there were no significant differences in relative abundance
between years for each habitat and season combination for many of the same species collected in this present
study. Also, Slacum et al. [6] reported similar annual patterns of relative abundance for juvenile fishes in the Mid-
Atlantic Bight even though they did not statistically evaluate whether annual abundance was equal with time.
Hagan and Able [28] also reported remarkably similar annual patterns in diel differences for species richness and
total abundance between 1995 and 1996 in an estuary within the Mid-Atlantic Bight. Although the species
composition was somewhat different between the estuary and the nearshore waters of this present study, it was
interesting that the findings for annual relative abundance were still similar. Based on this present study and
previous studies, it appears the annual estimates of fish populations in the Mid-Atlantic Bight region have been
stable in the past and continue today.
4.3.2. Temporal dynamics
This study revealed that relative abundance for all species combined and for individual species varied by month,
which showed that the Mid-Atlantic region is utilized by specific species at particular times of the year. In general,
relative abundance increased from spring to summer and decreased from summer to winter. Overall, the lowest
densities occurred in May and the highest in October. Depending on the species, peak monthly abundance was
different among species. For example, butterfish abundance was lowest in May and highest in June. However, scup
abundance was lowest in the winter and highest in the fall. In contrast, Atlantic herring relative abundance was
lowest during summer through fall and highest during winter through spring.
These findings were again consistent with previous studies showing that specific species utilize the
nearshore waters of the Mid-Atlantic Bight at specific times of the year. Actually, these findings were common and
have been reported previously for many species utilizing the estuaries and nearshore waters of the Mid-Atlantic
Bight. In southern New Jersey, Szedlmayer and Able [29] reported differences in monthly abundance for many
species found in Great Bay and Little Egg Harbor, which they attributed to differences in spawning and overwinter
locations. In the same estuary, Hagan and Able [30] also reported there were strong seasonal trends in species
richness, total abundance, and total biomass with peaks in spring, summer, and fall, and very low values for these
parameters in winter. Again, the authors indicated these differences were probably related to seasonal
recruitment patterns. Although Able et al. [26] only conducted their study during summer; they too reported that
overall abundances of surf zone fish larvae were significantly different among months. Overall, they found
significantly higher abundances in May than in June or July, which demonstrated that specific species utilize the
Mid-Atlantic Bight at particular times of the year. The researchers found that there wer e monthly distinctions in
taxonomic composition of the surf zone assemblages of larval fish, which they attributed to changes in species
composition during specific months. In another study, Able et al. [7] also found statistical differences in monthly
relative abundance for specific fish species collected in the Mid-Atlantic Bight. They found that some species
peaked in abundance during summer, while others peaked during fall. Able et al. [7] suggested these monthly
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
17
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
differences in relative abundance were related to differences in seasonal migration patterns. They indicated that
some estuary species such as Bay Anchovy (A. mitchilli) migrate from estuaries to the nearshore waters of the Mid-
Atlantic Bight during the fall. Despite evident and predictable temporal patterns in relative abundance and species
composition, Able et al. [5] indicated that small-scale temporal variability in larval fish assemblages was affected by
large-scale fluctuations in water mass (i.e., short-term upwelling events) that often occurred in summer [11, 29].
4.3.3. Spatial dynamics
Examining the frequency distribution of the species collected showed that the spatial distribution of fishes off New
Jersey was more clumped than uniform. In general, the number of individuals captured per haul was either none
(zero catch) or many (> 60 individuals), which was expected for schooling fishes. Nonetheless, this particular
behavior (schooling) was problematic for making inferences about the populations using parametric approaches
since the data was unable to be transformed to meet the assumptions of normality. Because the data could not be
transformed, much of the analyses had to be performed using non-parametric procedures. Despite these analytical
challenges, evaluation of the spatial dynamics demonstrated that each area within the study area was just as
important as the other in terms of overall annual or seasonal abundance, but some particular areas were more
important to specific species at certain times. For example, the findings demonstrated there were significant
differences detected between the mean density of weakfish and butterfish in area 16 and between Atlantic herring
and butterfish in area 21. Interestingly, the findings also showed that some species were positively correlated with
some areas, while others were negatively correlated. Species abundance with high loading values were in areas 15,
18, and 22, which corresponded to the shallowest (10 m) northernmost, the shallowest (10 m) middle, and the
mid-depth (20 m) southernmost areas, respectively. The findings also showed that butterfish, scup, and squid were
positively correlated, while weakfish and Atlantic herring were negatively correlated with these areas. Based on
these findings, it was apparent that specific fishes utilize particular areas of the Mid-Atlantic Bight region, which
was probably more related to depth than to a specific habitat given that most the study area had uniform habitat;
the Mid-Atlantic Bight consists of mostly sand and sand/mud sediments. It is difficult to speculate why the greatest
density of butterfish and scup occurred at 20 m instead of 10 or 30 m, but it might be related to prey availability or
predator avoidance behavior. Further investigation is warranted.
Although it is difficult to compare these present findings to others, they were somewhat different than
Able et al. [26] reported for larval fishes for the same general area. Overall, the researchers indicated that the
distribution of larvae along the 15 km shoreline of their study area was relatively homogeneous, with no specific
areas having consistently higher or lower densities. Actually, their findings suggested that larvae did not
dependent upon specific locations, but generally utilized the entire area. The results also showed that species
composition was similar between nearshore and surf zone habitats suggesting that surf zone habitats were just as
important to larvae fish as nearshore habitats [26]. Because Able et al. [26] and others have not specifically
evaluated the spatial dynamics of fishes in the Mid-Atlantic Bight, and in particular the deeper waters (beyond the
surf zone), in any detail, it is difficult to compare the findings of this present study with others. Regardless, this
study revealed for the first time that the spatial distribution of fishes in the Mid-Atlantic Bight is evident; however,
further examination into the mechanisms that control the spatial distribution of fishes is warranted before any
conclusions can be justified.
5. Conclusion
Results from the present study showed there were not only temporal and spatial differences in overall fish
abundance among specific species, but there were seasonal differences in species composition. This assessment of
the coastal fishes off New Jersey demonstrated that there were predictable patterns in annual and seasonal
population dynamics for the fishes found off the coast of New Jersey. Given these findings, it’s apparent the
nearshore waters off the coast of New Jersey provide important habitat for many economically valuable species.
As such, it is advisable that the resource managers continue to carefully evaluate the potential impacts to fisheries
from anthropogenic activities to preserve and protect these natural resources.
Competing Interests
The author declares there are no competing interests.
http://astonjournals.com/faj Co-Publisher: OMICS Group, www.omicsonline.org
18
Research Article
Acknowledgement
I thank every individual from the New Jersey Department of Environmental Protection (NJDEP), Division of Fish and
Wildlife, that has collected fisheries data for the Ocean Stock Assessment Program since its inception. I thank C.
Han from the University of Texas at Arlington for his detailed instruction of statistical procedures. I appreciate C.
Gomez for making a significant effort digitizing and producing the study area figure, especially during his busy work
schedule. I also thank J. See for reviewing the manuscript and making valuable editorial suggestions. I especially
thank B. Muffley and G. Buchanan for providing access to these data, supporting this study, and reviewing this
article. Finally, a special admiration and gratitude goes to D. Byrne who is no longer with us. His dedication and
commitment to managing the New Jersey Ocean Stock Assessment Program is why the program is recognized as
one of the oldest and most important fisheries independent monitoring programs in the United States; he will be
truly missed as a friend and biologist. The views, opinions, conclusions, or proposals expressed are mine and do
not necessarily reflect the views of Geo-Marine, Inc., or the New Jersey Department of Environmental Protection.
The original environmental baseline study was supported by the New Jersey Department of Environmental
Protection under contract to Geo-Marine, Inc.
References
[1] NMFS (National Marine Fisheries Service), 2013. Annual commercial fishery landings.
[http://www.st.nmfs.noaa.gov/st1/commercial/landings/gc_runc.html]
[2] Preble D, 2001. The Fishes of the Sea -- Commercial and Sport Fishing in New England. Dobbs Ferry, New York: Sheridan
House.
[3] NMFS (National Marine Fisheries Service), 2011. Recreational fisheries.
[http://www.st.nmfs.noaa.gov/st1/recreational/index.html]
[4] Giordano C, Hartley SL, Kurz K, Michalski K, Figley B, 2008. New Jersey party and charter boat directory 2000: Additions
2001−2008. Trenton, New Jersey: New Jersey Department of Environmental Protection.
[5] Able KW, Wilber DH, Muzeni-Corino A, Clarke DG, 2010. Spring and summer larval fish assemblages in the surf zone and
nearshore off northern New Jersey, USA. Estuaries and Coasts, 33(1): 211−222.
[6] Slacum WH, Burton WH, Methratta ET, 2010. Assemblage structure in shoal and flat-bottom habitats on the inner
continental shelf of the Middle Atlantic Bight, USA. Marine and Coastal Fisheries: Dynamics, Management, and Ecosystem
Science, 2: 277–298.
[7] Able KW, Grothues TM, Rowe PM, Wuenschel MJ, Vasslides JM, 2011. Near-Surface larval and juvenile fish in coastal
habitats: Comparisons between the inner shelf and an estuary in the New York Bight during summer and fall. Estuaries and
Coasts, 34: 726–738.
[8] NMFS (National Marine Fisheries Service), 2002. Essential fish habitat: A marine fish habitat conservation mandate for
federal agencies, South Atlantic region. St. Petersburg, Florida: National Marine Fisheries Service, Habitat Conservation
Division, Southeast Regional Office.
[9] Code of Federal Regulations (CFR). Title 50 Wildlife and Fisheries. Chapter VI - Fishery Conservation and Management,
National Oceanic and Atmospheric Administration, Department of Commerce. Part 600 Magnuson-Stevens Act Provisions,
50 CFR 600.
[10] Cross JN, Zetlin CA, Berrien PL, Johnson DL, McBride C, 1999. Essential fish habitat source document: Butterfish, Peprilus
triacanthus, life history and habitat characteristics. NOAA Technical Memorandum, NMFS-NE-145: 1−42.
[11] Taylor DL, Nichols RS, Able KW, 2007. Habitat selection and quality for multiple cohorts of young-of the-year bluefish
(Pomatomus saltatrix): Comparisons between estuarine and ocean beaches in southern New Jersey. Estuarine, Coastal and
Shelf Science, 73: 667−679.
[12] Steimle FW, Zetlin C, 2000. Reef habitats in the Middle Atlantic Bight: abundance, distribution, associated biological
communities and fishery resource use. Marine Fisheries Review, 62(2): 24−41.
[13] Moser J, Shepherd GR, 2009. Seasonal distribution and movement of black sea bass (Centropristis striata) in the Northwest
Atlantic as determined from a mark-recaptured experiment. Journal of the Northwest Atlantic Fishery Science, 40: 17−28.
[14] Saila SB, Pratt SD, 1973. Mid-Atlantic Bight fisheries. Pages 6-1 to 6-125 in Coastal and offshore environmental inventory:
Cape Hatteras to Nantucket Shoals. University of Rhode Island: Marine Publication Series, No. 2. Kingston, RI.
[15] Malakoff D, 1997. Extinction on the high seas. Science, 277: 486−488.
[16] Lazaroff C, 2001. Historic overfishing led to modern ocean problems. Environment News Service.
[http://www.cgaux1sr.org/msep/fish.htm]
[17] Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW, Bourque BJ, et al., 2001. Historical overfishing and the recent
collapse of coastal ecosystems. Science, 293: 629−638.
Co-Publisher: OMICS Group, www.omicsonline.org http://astonjournals.com/faj
19
Fisheries and Aquaculture Journal, Vol. 2013: FAJ-85
[18] Byrne D, 2008. Ocean trawling survey. New Jersey Department of Environmental Protection, Division of Fish and Wildlife,
Marine Fisheries Administration, Bureau of Marine Fisheries.
[19] Byrne D, 1994. Stock assessment of New Jersey’s nearshore recreational fisheries resources. In Proceedings of the
Workshop on the Collection and Use of Trawl Survey Data for Fisheries Management, Editor: T. Berger, 36–42. Washington:
Atlantic States Marine Fisheries Commission.
[20] King M, 1995. Fisheries Biology, Assessment and Management. Fishing News Books, Oxford.
[21] Able KW, Fahay MP, 1998. The first year in the life of estuary fishes in the Middle Atlantic Bight. New Brunswick, New
Jersey: Rutgers University Press.
[22] ASMFC (Atlantic States Marine Fisheries Commission), 1994. Proceedings of the Workshop on Saltwater Fishing
Tournaments. U.S. Fish and Wildlife Service, Atlantic States Marine Fisheries Commission, Special Report Number 46.
[23] Zar JH, 1999. Biostatistical Analysis, 4th Edition. Prentice Hall, Upper Saddle River, NJ.
[24] Moline MA, Blackwell SM, Chant R, Oliver MJ, Bergmann T, Glenn S, et al., 2004. Episodic physical forcing and the structure
of phytoplankton communities in the coastal waters of New Jersey. Journal of Geophysical Research, 109: C12S05, DOI:
10.1029/2003JC001985.
[25] Epifanio CE, Garvin RW, 2001. Larval transport on the Atlantic continental shelf of North America: a review. Estuarine,
Coastal and Shelf Science, 52: 51–77.
[26] Able KW, Fahay MP, Witting DA, McBride RS, Hagan SM, 2006. Fish settlement in the ocean vs. estuary: Comparison of
pelagic larval and settled juvenile composition and abundance from southern New Jersey, U.S.A. Estuarine, Coastal and
Shelf Science, 66: 280−290.
[27] Floeter SR, Ferreira CEL, Dominici-Arosemena A, Zalmon IR, 2004. Latitudinal gradients in Atlantic reef fish communities:
Trophic structure and spatial use patterns. Journal of Fish Biology, 64: 1680−1699.
[28] Hagan SM, Able KW, 2009. Diel variation in the pelagic fish assemblage in a temperate estuary. Estuaries and Coasts, 31:
33–42.
[29] Szedlmayer ST, Able KW, 1996. Patterns of seasonal availability and habitat use by fishes and decapod crustaceans in a
southern New Jersey estuary. Estuaries, 19(3): 697−709.
[30] Hagan SM, Able KW, 2003. Seasonal changes of the pelagic fish assemblage in a temperate estuary. Estuarine, Coastal and
Shelf Science, 56: 15−29.
