Fisheries Research 172 (2015) 17–25

Contents lists available at ScienceDirect

Fisheries Research
journal homepage: www.elsevier.com/locate/fishres

Population structure of red snapper (Lutjanus campechanus) in U.S.

waters of the western Atlantic Ocean and the northeastern Gulf of
Mexico
Christopher M. Hollenbeck a,∗ , David S. Portnoy a,b , Eric Saillant c , John R. Gold a,b
a
Harte Research Institute, Marine Genomics Laboratory, Texas A&M University-Corpus Christi 6300 Ocean Drive, Corpus Christi, Texas 78412-5869, USA
b
Department of Life Sciences, Texas A&M University-Corpus Christi, 6300 Ocean Drive, Corpus Christi, Texas 78412-5869, USA
c
Department of Coastal Sciences, University of Southern Mississippi, Gulf Coast Research Laboratory, 703 East Beach Drive, Ocean Springs, MS 39564, USA

a r t i c l e i n f o a b s t r a c t

Article history: Population structure of adult red snapper (Lutjanus campechanus) from the southeastern coast of the
Received 18 November 2014 United States (Atlantic) and the northeastern Gulf of Mexico (Gulf) was assessed using genotypes at 16
Received in revised form 15 June 2015 nuclear-encoded microsatellites and mitochondrial (mt)DNA haplotypes of the NADH dehydrogenase
Accepted 17 June 2015
4 (ND4) gene. Initial tests (FST -based, hierarchical AMOVA) of spatial genetic homogeneity within and
Available online 8 July 2015
between regions were non-significant, consistent with a single population or stock of red snapper in
the Atlantic and Gulf. Inferences derived from other statistical approaches were consistent with genetic
Keywords:
and/or demographic differences within and between the two regions. The estimated, average, long-term
Red snapper
Lutjanus campechanus
migration rate between the two regions (0.27%) was well less than the 10% rate below which populations
Atlantic Ocean can respond independently to environmental perturbation. Comparisons of global estimates of average,
Gulf of Mexico long-term effective size (NeLT ) with estimates from individual sample localities indicated genetic hetero-
Population structure geneity within both the Atlantic and Gulf. These results paralleled those of prior genetic studies of red
snapper from the Gulf. Future genetics studies and other work on red snapper in both the Atlantic and
Gulf should include approaches to identify demographically independent units within each region and
assess their size, patterns of connectivity, and contribution to the fishery. Monitoring global and/or local
effective size also should be considered.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction should be managed as a single stock (SEDAR, 2010), and in part

because different management councils are responsible for fishery
Red snapper (Lutjanus campechanus) have supported impor- resources in each region. The focus of red snapper management
tant commercial and recreational fisheries in U.S. waters of in U.S. waters primarily has been the Gulf where the stock was
the western Atlantic Ocean, including the Atlantic coast of the considered overfished and to be undergoing overfishing since at
southeastern United States (hereafter Atlantic) and the northern least the late 1980s (Strelcheck et al., 2007; NOAA, 2012); the Gulf
coast of the Gulf of Mexico (hereafter Gulf) for several decades stock is still considered overfished but not undergoing overfishing
(Hood and Strelcheck, 2007). Commercial landings of red snapper (SEDAR, 2013). The most recent assessments (SEDAR, 2009, 2010)
in the Gulf, for example, averaged 1630 metric tons between of red snapper in the Atlantic indicate that the stock in that region
2010 and 2012, with an average dockside value of $11.7 million is overfished and that overfishing is occurring at several times the
(http://www.st.nmfs.noaa.gov/commercial-fisheries/commercial- sustainable level. Factors impacting red snapper decline in both
landings/annual-landings/index); while recreational fishing in the regions include mortality due to directed fisheries and habitat
Gulf in 2011 involved >375,000 targeted trips and $52.8 million alteration and degradation. Mortality of juveniles taken as bycatch
in economic impact (GMFMC, 2013). Red snapper in the two in the shrimp fishery and seasonal hypoxic zones may also have
regions are currently managed as separate stocks, in part because been important factors in the Gulf (Christman, 1997; Schirripa and
there is no evidence that red snapper in the Atlantic and Gulf Legault, 1999; SEDAR, 2013).
Understanding the nature of stock structure is important in
management of exploited marine fishes. Failure to recognize
∗ Corresponding author. cryptic stocks can lead to over-exploitation and depletion of an
E-mail address: [email protected] (C.M. Hollenbeck). undetected stock, potentially resulting in loss of unique genetic

http://dx.doi.org/10.1016/j.fishres.2015.06.020
0165-7836/© 2015 Elsevier B.V. All rights reserved.
18 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25

resources and compromised long-term sustainability of the fish- North Carolina

ery (Carvalho and Hauser, 1994; Begg et al., 1999; Hilborn et al.,
2003). Knowledge of population or stock structure also is critical
for rebuilding and restoring depleted stocks, in part to anticipate South Carolina
patterns of recruitment (Ruzzante et al., 1999), and in part to •NC
inform management decisions if the future might involve restora-
tion enhancement (Bell et al., 2006), and/or establishment of a
Alabama Georgia
• SC
protected or restricted fishing areas, e.g., Marine Protected Areas
or MPAs (Botsford et al., 2003).
Prior studies of population structure of red snapper in U.S.
• GA
Atlantic Ocean
waters largely have involved genetic markers and samples of adults
from the directed fishery in the Gulf. The early work (Camper PC
• • DA
et al., 1993; Gold et al., 1997; 2001; Heist and Gold, 2000 Heist
•
Florida
MG
and Gold, 2000) utilized homogeneity tests of spatial distribu-
tions of microsatellite alleles and genotypes and/or haplotypes of
• ML
mitochondrial (mt)DNA) and found little evidence of population
subdivision. Saillant and Gold (2006), however, found signifi-
SA •
cant differences in contemporary (variance) effective size among Gulf of Mexico
red snapper sampled from offshore of Alabama, Louisiana, and
Texas, and Saillant et al. (2010) found significant heterogeneity in
microsatellite allele and genotype distributions among age-0 red
snapper sampled at small spatial scales in the western and central
Gulf. They (Ibid) also found a significant, positive spatial autocorre- 0 100 200 km

lation of microsatellite genotypes in age 0 fish within a geographic

range of 50–100 km. The lone genetic study of red snapper from the
Fig. 1. Approximate sampling localities for red snapper (Lutjanus campechanus) in
Atlantic (Garber et al., 2004) involved a comparison of sequences the U.S. Atlantic and eastern Gulf of Mexico. PC: Panama City, MG: Middle Grounds,
of the hypervariable, mtDNA control region among five geographic SA: Sarasota, ML: Melbourne, DA: Daytona, GA: Georgia, SC: South Carolina, NC:
localities, one in the Atlantic (35 fish) and four in the Gulf (140 fish); North Carolina.
no differences in mtDNA haplotype frequencies were detected, con-
sistent with the null hypothesis that red snapper in the Atlantic and
Gulf comprise a single, genetic population. 2. Methods
Studies of life history have shown differences in demograph-
ics between red snapper in the Atlantic and Gulf (Brown-Peterson A total of 669 adult red snapper were sampled dockside between
et al., 2009) and among red snapper in the Gulf (Fischer et al., 2004; 2008 and 2011 from boats fishing offshore of North Carolina (NC),
Jackson et al., 2007). Tagging studies to assess movement of red South Carolina (SC), Georgia (GA), Daytona, Florida (DA), and Mel-
snapper between the Atlantic and the Gulf are limited to studies bourne, Florida (ML) in the Atlantic, and offshore of Sarasota, Florida
by Burns et al. (2004, 2008) where more than 5200 red snapper (SA), the Florida Middle Grounds (MG), and Panama City, Florida
were tagged and released in the Atlantic and Gulf over a 13-year (PC) in the Gulf. Approximate fishing localities are indicated in
period. Roughly ∼40% of the fish were released in the Atlantic. The Fig. 1. Tissue samples (fin clips) were obtained by personnel from
majority of recaptures were taken within 3 km of the tagging site several state or federal agencies and fixed in 10% DMSO buffer
and only one fish (tagged in the Florida panhandle and recaptured (Seutin et al., 1991). DNA was extracted following a modified
on the Atlantic coast of Florida) moved between regions. Studies chelex extraction protocol (Estoup et al., 1996); following final cen-
based on tagging and/or ultrasonic telemetry of red snapper in the trifugation, 1 ␮L of the supernatant was used as the template in
Gulf are equivocal; some have indicated relatively high site fidelity subsequent polymerase-chain-reaction (PCR) amplification.
(Szedlmayer, 1997; Schroepfer and Szedlmayer, 2006), while oth- All fish were genotyped at 16 nuclear-encoded microsatellites,
ers (Patterson et al., 2001; Patterson and Cowan 2003) have not. following multiplex PCR protocols described in Renshaw et al.
The low site fidelity reported, however, may have been due to sea- (2006) and using PCR primers described in Bagley and Geller
sonal conditions and hurricanes (Watterson et al., 1998; SEDAR, (1998) and Gold et al. (2001). Amplicons were electrophoresed
2010). Two tagging studies carried out within the Atlantic indicated and visualized on 6% polyacrylamide gels, using an ABI Prism 377
relatively limited movement within that region (SEDAR, 2009). automated sequencer (Applied Biosystems). Allele-calling was con-
In this study, we used nuclear-encoded microsatellites and ducted manually, using Genescan v.3.1.2 (Applied Biosystems Inc.,
sequences of mtDNA to assess genetic population structure of red Warrington, UK) and Genotyper v.2.5 (PerkinElmer). A fragment
snapper sampled from five localities in the Atlantic and three local- (590 base pairs) of the mitochondrially-encoded NADH dehydro-
ities in the northeastern Gulf. Our primary interests were to test genase subunit 4 (ND4) gene was amplified from 20 individuals
the (null) hypothesis that red snapper in the Atlantic and Gulf are at each locality, using primers ND4LB (Bielawski and Gold, 2002)
genetically homogeneous and to assess patterns of genetic diversity and NAP2 (Arèvalo et al., 1994). Thirty microliter PCR reactions
of red snapper in the Atlantic. We also assessed genetic connectivity consisted of 1x reaction buffer, 1.45 mM MgCl2 , 0.25 mM of each
between the two regions and genetic effective size at each sample dNTP, 30 pmol of each primer, 0.1 U/␮L Taq polymerase, and 2 ␮L
locality and within each region. Information regarding connectivity of DNA template. Reaction conditions consisted of an initial denat-
is important to understanding and managing marine biota whose uration at 95 ◦ C for 3 min, followed by 35 cycles of 95 ◦ C for 30 s,
biogeographic distributions cross political boundaries (Kough et al., 50 ◦ C for 30 s, and 72 ◦ C for 45 s, and a final extension at 72 ◦ C for
2013), and effective size is important because it provides infor- 10 min. Amplified products were purified with ExoSAP-ITTM PCR
mation regarding standing levels of neutral genetic diversity and cleanup kit (GE Healthcare, Piscataway, NJ, USA) and sequenced bi-
a population’s capacity to respond to changing or novel environ- directionally, using BigDye Terminator v.3.1Cycle Sequencing Kit
mental pressures (Frankham, 1995; Allendorf and Waples, 1996; (Applied Biosystems). Five microliter sequencing reactions con-
Higgins and Lynch, 2001). sisted of 10–50 ng of template, 0.5 ␮L of BigDye master mix,
 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25 19

0.875 ␮L of BigDye 5× reaction buffer, and 32 pmol of forward or framework (Beerli and Palczewski, 2010); different models can thus
reverse primer. Sequencing conditions consisted of denaturation be explored and the relative probability of each model compared.
at 96 ◦ C for 1 min followed by 25 cycles of 96 ◦ C for 10 s, 50 ◦ C for Both stock models were run in five random subsamples of 50 indi-
5 s, and 60 ◦ C for 4 min. Amplifications were electrophoresed on viduals, each with two replicates (ten total runs), with a burn-in
an ABI 3100 Sequencer (Applied Biosystems) through 50 cm cap- period of 100,000 steps, followed by 1 × 107 steps, and with trees
illaries. Sequence chromatograms were aligned and trimmed to a sampled every 200 steps; a total of 50,000 trees were sampled.
common 590 base pair region, using Sequencher 4.8 (Gene Codes We estimated the average, long-term migration rate (m)
Corporation). between the Atlantic and Gulf, using the Bayesian approach in
Conformance to expectations of Hardy–Weinberg equilibrium Migrate-n to estimate the parameter M (mutation-scaled migra-
(HWE) was tested for each microsatellite in each locality, using tion rate), where M = m/ and  is the modal mutation rate across
exact tests as implemented in Genepop v.4.0.7 (Raymond and all microsatellites per generation. A random subsample (n = 50; the
Rousset, 1995; Rousset, 2008). Parameters of the Markov Chain smallest individual sample size) was drawn from pooled samples
employed in estimation were 10,000 dememorizations, 1000 from the Atlantic and pooled samples from the Gulf and replicate
batches, and 10,000 iterations per batch. Sequential Bonferroni runs were combined to generate parameter estimates of M. Esti-
correction (Rice, 1989) was applied for multiple tests. Possi- mates of  were obtained using the Bayesian coalescent approach
ble occurrence of scoring error due to stuttering, large allele implemented in Msvar v1.3 (Beaumont, 1999; Storz and Beaumont,
dropout, and/or null alleles was evaluated for each sample, using 2002); Boa (Smith, 2005) was used to calculate the 95% highest pos-
Microchecker (Van Oosterhout et al., 2004). Tests of genotypic terior density (HPD) intervals for the modal value of . Lower and
disequilibrium between pairs of microsatellites within each local- upper bounds of m were estimated using 95% HPD intervals of M
ity were carried out using Arlequin v.3.5 (Excoffier and Lischer, generated by Migrate-n.
2010). Number of alleles, allelic richness, unbiased gene diversity Estimates of the effective number of breeders (Nb ) were gen-
(expected heterozygosity), and the inbreeding coefficient FIS (Weir erated for each sample locality, using microsatellite data and the
and Cockerham, 1984) were generated for each microsatellite in linkage disequilibrium method implemented in LDNe (Waples,
each locality, using Fstat v.2.9.3.2 (Goudet, 2001). Homogeneity of 2006; Waples and Do, 2008). Rare alleles below a frequency of
allelic richness and unbiased gene diversity among localities was 0.02 were excluded from calculations; confidence intervals were
assessed using Friedman rank tests, as implemented in R (R Core obtained by jackknifing. Estimates of average, long-term effective
Team, 2013). For mtDNA sequences, number of haplotypes, haplo- population size (NeLT ) for each sample locality were generated
type diversity (h), and nucleotide diversity () were obtained for using microsatellite data and Migrate-n. A random subsample
each sample, using Arlequin. (n = 50) was drawn from each sample locality and replicate runs
The degree of divergence in microsatellites and mtDNA between were combined to generate parameter estimates of theta (), where
pairs of localities was estimated as FST and ФST , respectively,  = 4Ne ; Ne is the average, long-term effective population size
using Arlequin. For mtDNA, ФST values were estimated under a (NeLT ) and  is the modal mutation rate across all microsatellites
Tamura-Nei substitution model (Tamura and Nei, 1993) with a per generation. Estimates of NeLT for the Atlantic (localities pooled),
gamma shape parameter of 0.198, as selected by jModelTest v.2.1.1 Gulf (localities pooled), and overall (all localities pooled) also were
(Guindon and Gascuel, 2003; Darriba et al., 2012). Significance of generated from random subsamples (n = 50) where replicate runs
FST and ФST values between pairs of localities was assessed by were combined to generate parameter estimates of . Estimates of
permuting individuals between localities 10,000 times. Hierarchi-  were obtained as above. Lower and upper bounds for NeLT were
cal analysis of molecular variance (AMOVA), as implemented in estimated using 95% HPD intervals of  generated by Migrate-n and
Arlequin, was conducted for both microsatellites and mtDNA by the modal value of .
grouping Atlantic localities (NC, SC, GA, DA, and ML) and Gulf local-
ities (SA, MG, and PC) separately; significance of the between-group
component of variance was assessed by permuting sample locali- 3. Results
ties between groups 10,000 times.
Mantel tests, implemented in Arlequin, were used to test for Significant deviations from expectations of HWE equilibrium
a correlation between genetic distance and geographic distance prior to Bonferroni correction were found in 13 of 128 tests;
for both microsatellite genotypes and mtDNA haplotypes. Dis- no significant deviations were found following correction. Pos-
tance matrices contained pairwise measures of genetic distance, sible null alleles, as inferred by Microchecker, were detected at
coded as Cavalli-Sforza/Edwards chord distance (Cavalli-Sforza Lca107 (NC), Ra6 (SA), Lca43 (MG), and Prs221 (PC); no evidence
and Edwards, 1967 microsatellites) or ФST /1-ФST (mtDNA), and of gel scoring errors caused by stuttering or large allele dropout
approximate coastline geographic distance, and were permuted was detected. Following Bonferroni correction, two pairwise tests
10,000 times to assess significance. Hudson’s (2000) Snn approach of genotypic disequilibrium were significant: Lca20-Lca107 and
was used to determine whether ‘nearest neighbor’ mtDNA hap- Lca20-Prs328, both in SC. A summary of general statistics for
lotypes (in terms of sequence identity) were sampled within a microsatellites and mtDNA is given in Appendix A. For microsatel-
region (Atlantic and Gulf) more often than would be expected in lites, mean (± SE) number of alleles across sample localities ranged
a randomly mating population. Last, a median-joining network of from 7.44 ± 1.11(PA) to 9.69 ± 1.12 (ML); mean (± SE) allelic rich-
mtDNA haplotypes was constructed using Network v.4.6.11 (http:// ness ranged from 7.40 ± 1.11 (PC) to 8.12 ± 0.97 (ML); and mean
www.fluxus-engineering.com/) in order to visualize spatial rela- (± SE) gene diversity ranged from 0.583 ± 0.05 (PC) to 0.596 ± 0.05
tionships among mtDNA haplotypes. (ML). Friedman rank tests of homogeneity across sample localities
The program Migrate-n v.3.2.16 (Beerli and Felsenstein, 2001; in allelic richness (AR ) and gene diversity (HE ) were non-significant
Beerli, 2006) was used to evaluate the marginal likelihood (relative (AR : 2 [7,15] = 4.01, P = 0.778; HE : 2 [7,15] = 3.12, P = 0.874). For
probability) of two different, population-structure models: (i) sam- mtDNA, 39 haplotypes were found among the 160 individuals sur-
pled localities represent a single admixed population (one stock), veyed. The distribution of haplotypes in the eight sample localities
and (ii) samples were drawn from separate Gulf and Atlantic stocks, (Appendix B) consisted primarily of two common haplotypes (#2
with limited migration (gene flow) between the regions. Migrate-n and #4) and numerous rare haplotypes. A total of 21 haplotypes
computes the marginal likelihood of pre-defined population mod- were unique to the Atlantic, while 12 haplotypes were unique
els, using factor analysis implemented under a Bayesian coalescent to the Gulf. Previously undescribed haplotypes were submitted
20 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25

region were more closely related in sequence to other haplotypes

from the same region than would be expected in a randomly mat-
ing population. The median-joining network of mtDNA haplotypes
(Fig. 2) grouped by region (Gulf and/or Atlantic) also was consistent
with slight differences between regions as at least four putative
clades of Atlantic-only haplotypes were recovered, whereas no
clades of haplotypes were evident in the Gulf. The difference in
number of putative clades may be in part a result of the difference in
sample size (60 in the Gulf vs. 100 in the Atlantic), as more unique,
rare haplotypes (21 vs. 12) were sampled in the Atlantic. Bayes
factor analysis (microsatellite data) yielded mixed results; eight
of ten runs indicated that a two-stock model (Atlantic and Gulf)
had the highest marginal likelihood (P > 0.999) when compared to
a one-stock model (all sample localities pooled).
The estimate of average, long-term migration rate (m) from the
Atlantic to the Gulf was 0.21% (95% CI: 0.04%–0.77%), while the rate
from the Gulf to the Atlantic was 0.33% (95% CI: 0.07%–1.11%). The
parameter m is defined as the proportion of individuals in a pop-
ulation that are of immigrant origin and is often expressed as the
parameter mNe , the effective number of migrants entering a sub-
population each generation (Mills and Allendorf, 1996). Because
the estimates of m and NeLT (see below) for the two regions did not
Fig. 2. Median-joining network of ND4 mtDNA haplotypes: gray, haplotypes found
differ significantly, we used average estimates of m (0.0027) and
in the Atlantic; black, haplotypes found in the Gulf. Each node (small circle)
represents a unique haplotype; sizes of nodes are scaled to reflect the relative fre-
NeLT (4022) to generate an average, long-term estimate of 10.86
quency of each haplotype. Lengths of lines connecting haplotypes reflect number of (mNe ), the effective number of migrants per generation moving in
single-base substitutions between haplotypes; the shortest line is one base-pair sub- either direction from one region to the other.
stitution. Small nodes indicated by small, unnumbered circles are inferred mtDNA Mean estimates of the effective number of breeders (Nb ) for
haplotypes. Dotted lines surround putative clades found in the Atlantic.
four of the eight sample localities were infinite, as were upper
bounds for seven of the eight (data not shown but available from
CMH), suggesting a uniformly large Nb across the localities sam-
to GenBank (NCBI), with accession numbers KT201529-KT201542.
pled (Waples and Do, 2010). However, estimates of Nb generated
Estimates of haplotype (h) and nucleotide () diversity (Appendix
from adult samples are complex and difficult to interpret because
A) ranged from 0.658 (PC) to 0.905 (ML) and 0.050 (MG) to 0.185
they are influenced by the effective number of breeders that gener-
(ML), respectively.
ated each cohort in a sample (Waples, 2005). Red snapper can live
Pairwise estimates of FST (microsatellites) and ˚ST (mtDNA)
for over 50 years (Wilson and Nieland, 2001) and mature sexually
are given in Table 1; none of the pairwise comparisons of FST or
between age two and four (Schirripa and Legault, 1999; Fitzhugh
˚ST values differed significantly from zero. Results of hierarchical
et al., 2004). Because virtually all of the red snapper sampled in this
AMOVA (Table 2) indicated that the component of molecular vari-
study were adults, fish in each sample were potentially a mix of
ance attributable to differences between regions (Atlantic versus
overlapping sets of parents in different years, making interpreta-
Gulf) and among localities within regions was non-significant
tion of Nb estimates problematic (Waples, 2010).
for microsatellites and mtDNA. These results are consistent with
Estimates of NeLT and 95% confidence intervals for each sam-
previous studies of red snapper adults in the Gulf (Gold et al.,
ple locality, for the Atlantic (localities pooled) and Gulf (localities
1997, 2001; Heist and Gold, 2000) where significant genetic dif-
pooled), and overall (all localities pooled) are given in Table 3.
ferences in microsatellite allele/genotype and mtDNA haplotype
Estimates of NeLT for the eight sample localities ranged from
distributions, using similar analytical approaches, have not been
826.1 (PC) to 2111.3 (SA) and did not differ significantly from one
detected.
another. The global estimate of NeLT for the Atlantic (3930.1; 95% CI:
Mantel tests for correlation between genetic distances and
3104.3–5,023.76.02) was significantly larger than estimates of NeLT
geographic distances were significant for both microsatellites
for each of the five sample localities from the Atlantic; while the
(P = 0.040) and mtDNA (P = 0.023), indicating that both microsatel-
global estimate of NeLT for the Gulf (4114.1; 95% CI: 3204.5–4,923.5)
lite genotypes and mtDNA haplotypes are not distributed randomly
was significantly larger than estimates of NeLT for two (MG and PC)
across the sampled range. Hudson’s nearest-neighbor test, based
of the three sample localities in the Gulf (Table 3). The overall global
on mtDNA haplotypes, was significant (Snn = 0.566, P = 0.038) when
estimate of NeLT (6267.1; 95% CI: 5474.3–7,076.5) was significantly
pooled samples from the Atlantic were compared with pooled sam-
larger than global estimates of NeLT for either region.
ples from the Gulf, indicating that haplotypes sampled from either

Table 1
Estimates of FST (microsatellites, upper diagonal) and ˚ST (mtDNA, lower diagonal) for pairwise comparisons of sample localities: North Carolina (NC), South Carolina (SC),
Georgia (GA), Daytona, FL (DA), Melbourne, FL (ML), Sarasota, FL (SA), Florida Middle Grounds (MG), and Panama City, FL (PC).

NC SC GA DA ML SA MG PC

NC 0 0.001 0.001 0.001 0 0 0.002 −0.001

SC 0.046 0 −0.001 −0.001 0 -0.001 0 0.001
GA −0.023 0.036 0 0.001 −0.001 -0.001 0 0.002
DA 0.005 −0.018 −0.004 0 0 0.002 0 0.002
ML −0.024 0.02 −0.036 -0.076 0 0 0 −0.001
SA 0.015 0.027 0.011 −0.023 −0.025 0 0.001 -0.001
MG 0.012 -0.028 0.018 −0.012 −0.087 −0.04 0 0
PC 0.092 0.056 0.059 −0.04 −0.025 −0.019 -0.025 0
 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25 21

Table 2
Hierarchical analysis of molecular variance (AMOVA), based on 16 microsatellite loci. Variance components attributed to between regions and among localities within regions
were non-significant (˛ = 0.05); d.f. = degrees of freedom; F = fixation index; P = probability that F = 0.

Source of Variation d.f. Sum of squares Variance component % of variation F P

Microsatellites

Between regions (Atlantic and Gulf) 1 5.12 0.0007 0.02 0.0002 0.303
Among localities within regions 6 28.41 0.0003 0.01 0.0001 0.505
Within localities 1330 6227.98 4.6827 99.98 0.0002 0.444

Mitochondrial DNA
Between regions (Atlantic and Gulf) 1 3.11 0.02 1.21 0.012 0.106
Among localities within regions 6 8.68 -0.02 -1.05 0.002 0.7
Within localities 153 278.73 1.83 99.84 -0.01 0.598

Table 3 this study assessed the spatial distribution of adults sampled from
Estimates and 95% confidence intervals of average, long-term effective population
the directed fishery. In general, isolation by distance occurs when
size (NeLT) for each sample locality, the Atlantic and Gulf (localities in each region
pooled), and overall (all localities pooled). the rate of gene exchange between distinct subpopulations is a
function of geographic distance or when dispersal within a con-
NeLT
tinuously distributed population becomes restricted (Hardy and
2.5 Mode 97.5 Vekemans, 1999). The studies of red snapper adults from across
NC 717.7 1343.5 1969.4 the northern Gulf have not detected an isolation-by-distance effect
SC 567.5 1143.3 1719.1 in either microsatellites or mtDNA sampled over a distance of
GA 433.9 1009-7 1585.5 ∼1,600 km (Gold et al., 1997; 2001; Saillant and Gold, 2006 Saillant
ML 901.3 1510.4 2153
and Gold, 2006), suggesting that genetic divergence in adult red
DA 884.6 1443.7 2019.5
SA 1168.3 2111.3 3304.6 snapper is not necessarily a function of geographic distance. Con-
MG 901.3 1477.1 2036.2 sequently, we interpret the isolation-by-distance effect observed
PC 333.8 826.1 1285.1 here to largely reflect limited gene flow between the two regions,
Atlantic 3104.3 3930.48 5023.7 given that the distance sampled within each region (∼835 km in
Gulf 3204.5 4114.1 4923.5 the Atlantic and ∼475 km in the Gulf) is considerably less than that
All 5474.3 6267.1 7076.5 sampled previously across the Gulf.
Our primary interests in the study were to test the (null)
hypothesis that red snapper in the Gulf and Atlantic are geneti-
4. Discussion cally homogeneous and to assess patterns of genetic diversity of
red snapper in the Atlantic. While there were no significant fix-
Initial tests (FST -based, hierarchical AMOVA) of spatial genetic ation indices between individual sample locations, we did find
homogeneity within and between regions were non-significant, genetic evidence of different populations in each region, consis-
consistent with a single population or stock of red snapper along the tent with current management of red snapper resources in U.S.
U.S. southeast Atlantic coast and in the northeastern Gulf of Mexico; waters (SEDAR 2010, 2013) and with numerous lines of non-genetic
however, inferences derived from other statistical approaches indi- evidence. The latter include tag recoveries indicating very limited
cated genetic and/or demographic differences among red snapper movement between the regions (Burns et al., 2004; Burns et al.,
within and between the two regions. The significant correlation 2004) and low abundance in the Florida Keys, especially along the
between genetic and geographic distances for both microsatellites southeast coast of Florida (Moe, 1963). Levels of genetic variability
and mtDNA indicated an isolation-by-distance effect, and Hudson’s (allelic richness, gene and haplotype diversity) among the samples
nearest-neighbor test indicated mtDNA haplotypes sampled within from both regions were essentially the same as found previously
a region (Atlantic or Gulf) were more closely related in sequence to for adults in the Gulf (Gold et al., 2001; Pruett et al., 2005; Saillant
other haplotypes from the same region than would be expected in a and Gold, 2006), and estimates of average, long-term effective size
randomly mating (panmictic) population. Small, mtDNA haplotype (NeLT ) did not differ among the sample localities. However, the
clades that were unique to a region were found only in the Atlantic. global estimate of NeLT for the Atlantic was significantly larger than
Bayes factor analysis (microsatellite data) also supported a two- estimates of NeLT for the five localities sampled from the Atlantic.
stock (Atlantic/Gulf) model, and the estimated, average, long-term Similar results were obtained from the Gulf (the global estimate
migration between the two regions (0.27%) was well less than the was significantly larger than estimates for two of the three local-
10% rate below which populations can respond independently to ities sampled) and overall (the global estimate was significantly
environmental perturbation (Hastings, 1993; Hauser and Carvalho, larger than the estimates for the Atlantic and Gulf). These compar-
2008; Waples, 2010). Finally, differences between local and global isons are inconsistent with the null hypothesis that samples from
estimates of NeLT were consistent with genetic and/or demographic the Atlantic, Gulf, or overall were drawn from a single, well-mixed
heterogeneity across the sampled area (see below). population, given that Ne in samples drawn from a single, panmictic
The above results are very similar to those of prior genetic unit should approximate global Ne , while Ne from a subdivided pop-
studies of red snapper adults from the Gulf where inferences ulation should be less than the global Ne (Waples 2010). Inferences
of spatial genetic structure have relied on historical patterns of from prior genetic studies (Pruett et al., 2005; Saillant et al., 2010)
mtDNA diversification (Pruett et al., 2005), differences in contem- have suggested that genetic patterns among red snapper in the Gulf
poraneous genetic effective size (Saillant and Gold, 2006), or tests are consistent with the metapopulation-type structure envisaged
of the spatial distribution of microsatellite alleles, including an by Kritzer and Sale (2006) where connectivity between local pop-
isolation-by-distance effect, in age-0 fish (Saillant et al., 2010). The ulations is variable and where there is geographic asynchrony in
isolation-by-distance effect observed by Saillant et al. (2010), how- demographics across the metapopulation. Data from this study do
ever, differs from that found in this study. Their study (Ibid) assessed not refute the existence of this type of metapopulation structure.
the spatial distribution of age-0 juveniles, presumably reflecting A conservative management strategy might be to account for this
average movement from nursery areas across the region, whereas
22 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25

type population structure, particularly given the economic impor- et al., 2010), should be informative in better understanding popu-
tance of the species. lation dynamics and maintaining genetic and phenotypic diversity
The issues associated with management of exploited marine in exploited species such as red snapper. One final point to note
species with a metapopulation-type structure are beyond the scope is that genetic studies, using microsatellites and mtDNA, of four
of this paper and are discussed in Grimm et al. (2003) and Kritzer other, exploited snapper species in the Atlantic and Gulf and in U.S.
and Sale (2006). Two points, however, warrant mention: (i) the waters of the Northern Antilles also relied on isolation by distance
viability and sustainability of a metapopulation depends on the and historical demographics rather than statistical heterogeneity
included local populations and their interactions with one another in FST -based tests to assess population subdivision in each species
(Akçakaya et al., 2007), and (ii) unless isolation is essentially com- (Tringali and Higham, 2007; Gold et al., 2009; Carson and Saillant,
plete, the loss of genetic diversity within a metapopulation via 2011; Saillant et al., 2012). Snappers (Lutjanidae) are found in trop-
genetic drift is determined primarily by the global or metapop- ical and subtropical regions over much of the world (Allen, 1985)
ulation effective size (Ne ) rather than the effective sizes of local and constitute an important food resource, especially in develop-
subpopulations or demes (Waples, 2010). This suggests that future ing countries (Russ and Alcala, 1989; Blaber et al., 2005). A better
genetics studies and other work on red snapper in both the Atlantic understanding of population dynamics in red snapper in U.S. waters
and Gulf should include approaches to identify demographically may thus be useful in conserving marine resources elsewhere.
independent units within the fishery and assess their demographics
(e.g., size, patterns of connectivity, and contribution to the fishery) Acknowledgments
and to continually monitor global and/or local effective size. The
former will not be an easy task because well designed, replicative We thank C. Collier and M. Duval (North Carolina Department of
spatial/temporal sampling will be needed; the importance of doing Environment and Natural Resources), M. Reichert (South Carolina
so is in identifying spatial and temporal units (local populations) Department of Natural Resources), K. Knowlton and S. Woodward
that are disproportionally contributing to recruitment and to the (Georgia Department of Natural Resources), J. Bogdan, B. Kalmeyer,
fishery. The latter (monitoring global or local Ne ) is more straight- K. Kowal, B. McMichael, and E. Sanders (Florida Wildlife Research
forward but will require sampling of individuals from the same Institute), and K. Brennan and B. Walling (National Marine Fisheries
cohort in different local populations across time in order to utilize Service) for providing tissue samples, D. Kulaw and J. Puritz for
recent statistical approaches (Waples et al., 2014) that generate helpful comments on a draft of the manuscript, and B. Sauls for
single-sample estimates of contemporaneous Ne . providing information on red snapper catches in southeast Florida.
Genetic and other data obtained to date are consistent with exis- Work was supported by the Marfin Program of the US Department
tence of distinct populations of red snapper in U.S. waters (Atlantic of Commerce (Grant NA10NMF4330114) and Texas AgriLife under
and Gulf) which exchange small numbers of migrants over genera- Project H-6703. The paper is contribution 7 of the Marine Genomics
tional time scales. Further identifying discrete subunits within red Laboratory at Texas A&M University-Corpus Christi and Number
snapper populations, using microsatellites and mtDNA, will likely 101 in the series ‘Genetic Studies in Marine Fishes.’
be ineffective, in part because of insufficient resolution related
to too few markers, but also because of incomplete lineage sort-
Appendix A
ing of ancestral polymorphisms in a species that only recently
in evolutionary time colonized its current habitat (Pruett et al.,
Summary statistics for red snapper (Lutjanus campechanus)
2005; Orozco-Terwengel et al., 2011). Next-generation sequencing
microsatellite and mtDNA loci at each locality. For microsatel-
approaches (Davey et al., 2011) that allow genome-wide surveys
lites: n = number of individuals sampled, #A = number of
of thousands of genetic markers, including those that may be asso-
alleles, AR = allelic richness, HE = expected heterozygosity,
ciated with fitness on local scales (Nielsen et al., 2009; Allendorf
PHW = probability of conformance to Hardy–Weinberg expec-

Sample NC SC GA DA ML SA MG PC

Microsatellite
Lca20 n 93 84 101 98 101 48 97 46
#A 4 5 5 4 5 5 5 3
AR 3.8 4.05 3.72 3.3 4.25 4.93 4.03 2.98
HE 0.19 0.179 0.208 0.162 0.177 0.196 0.174 0.198
PHW 0.009 1 0.063 0.115 0.116 0.41 0.101 0.852
FIS 0.152 −0.067 0.094 0.182 0.049 0.15 −0.009 −0.099

Lca22 n 93 84 101 96 100 47 97 46

#A 14 12 12 12 13 9 11 8
AR 12.06 10.35 11.03 11.1 10.9 8.95 9.45 7.98
HE 0.751 0.726 0.727 0.783 0.725 0.693 0.752 0.761
PHW 0.265 0.629 0.21 0.077 0.304 0.032 0.23 0.661
FIS 0.055 −0.034 0.087 0.082 0.062 −0.043 0.04 −0.028

Lca43 n 93 85 101 98 101 48 97 46

#A 7 8 9 9 9 7 8 7
AR 6.4 7 7.61 7.58 7.8 6.99 7.34 6.94
HE 0.491 0.581 0.539 0.526 0.583 0.56 0.56 0.589
PHW 0.597 0.127 0.123 0.486 0.366 0.745 0.011 0.196
FIS 0.08 −0.013 0.026 −0.029 −0.018 0.033 0.19 −0.107

Lca64 n 93 84 101 98 101 48 97 46

#A 11 12 11 12 9 8 9 6
AR 9.17 10.23 8.67 9.48 7.8 7.81 7.62 5.98
HE 0.789 0.791 0.782 0.801 0.774 0.713 0.77 0.736
PHW 0.075 0.792 0.88 0.953 0.808 0.468 0.177 0.384
FIS −0.077 0.007 0.013 −0.032 -0.036 0.065 0.023 −0.094
 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25 23

Sample NC SC GA DA ML SA MG PC

Lca91 n 93 85 101 98 101 47 97 46

#A 6 7 6 7 6 6 7 4
AR 5.19 5.89 5.65 5.91 5.17 5.96 5.6 3.98
HE 0.577 0.565 0.613 0.579 0.573 0.627 0.562 0.557
PHW 0.31 0.966 0.252 0.982 0.049 0.012 0.514 0.824
FIS −0.025 0.042 0.046 −0.041 0.118 0.049 0.028 0.103

Lca107 n 93 84 98 98 101 48 97 46
#A 9 8 9 10 10 9 10 10
AR 8.46 7.98 8.41 9.15 9.22 8.93 9.36 9.96
HE 0.776 0.801 0.781 0.814 0.781 0.775 0.8 0.767
PHW 0.145 0.714 0.034 0.912 0.272 0.119 0.495 0.027
FIS 0.099 0.019 0.059 0.035 0.023 −0.021 0.021 0.065

Prs55 n 93 84 101 98 101 48 97 46

#A 7 5 6 4 7 4 6 4
AR 5.5 4.06 5.27 3.45 5.19 3.93 4.6 3.98
HE 0.271 0.265 0.253 0.256 0.297 0.193 0.251 0.255
PHW 0.1 0.272 0.753 0.426 0.942 1 0.867 0.611
FIS 0.166 −0.032 −0.057 0.082 −0.034 −0.077 −0.11 0.062

Prs137 n 93 85 101 98 99 47 97 46
#A 13 12 11 11 12 12 11 8
AR 10.24 9.78 9.09 9.34 9.63 11.74 8.83 7.96
HE 0.732 0.68 0.721 0.676 0.72 0.694 0.682 0.676
PHW 0.013 0.613 0.13 0.513 0.032 0.714 0.054 0.102
FIS 0.105 0.049 0.066 0.019 0.031 0.049 0.063 0.131

Prs221 n 93 84 101 98 101 48 97 46

#A 12 12 12 13 15 13 16 11
AR 10.67 10.23 10.78 10.93 11.93 12.68 12.54 10.91
HE 0.766 0.772 0.8 0.767 0.803 0.807 0.797 0.771
PHW 0.366 0.36 0.522 0.185 0.532 0.625 0.44 0.388
FIS 0.032 0.06 0.022 0.082 0.075 −0.007 0.056 0.069

Prs240 n 92 81 101 98 100 47 97 46

#A 18 18 18 19 18 18 20 18
AR 17.03 16.79 15.9 16.61 16.88 17.83 17.13 17.87
HE 0.902 0.901 0.889 0.899 0.892 0.902 0.907 0.875
PHW 0.269 0.298 0.983 0.681 0.85 0.926 0.022 0.042
FIS −0.037 0.054 −0.002 −0.022 −0.009 −0.038 0.034 0.056

Prs248 n 93 84 101 98 101 48 97 46

#A 18 17 20 23 18 14 19 16
AR 15.3 14.29 15.5 16.6 14.64 13.8 16.05 15.89
HE 0.872 0.868 0.888 0.865 0.877 0.88 0.869 0.879
PHW 0.644 0.337 0.319 0.952 0.407 0.403 0.906 0.551
FIS 0.001 −0.056 0.052 −0.05 −0.005 0.029 0.015 0.035

Prs260 n 93 85 101 98 101 48 97 46

#A 5 3 5 5 3 4 3 3
AR 3.93 3 4.52 4.17 3 3.94 3 3
HE 0.305 0.397 0.398 0.439 0.394 0.379 0.357 0.393
PHW 0.064 0.218 0.309 0.017 0.814 1 1 0.4
FIS 0.118 −0.187 −0.019 −0.047 −0.029 −0.044 −0.011 −0.05

Prs275 n 93 85 101 98 101 48 97 45

#A 6 8 6 7 10 5 7 5
AR 5.34 6.64 5.05 5.75 7.33 4.94 5.36 5
HE 0.609 0.579 0.564 0.61 0.632 0.603 0.595 0.608
PHW 0.837 0.511 0.392 0.272 0.93 0.849 0.539 0.446
FIS −0.095 −0.057 0.016 −0.121 −0.05 −0.002 0.099 0.05

Prs282 n 93 85 101 98 101 48 97 46

#A 12 12 11 11 11 11 11 13
AR 9.561 10.688 9.188 9.787 9.677 10.681 10.176 12.912
HE 0.599 0.66 0.636 0.64 0.684 0.565 0.655 0.693
PHW 0.27 0.443 0.399 0.895 0.261 0.63 0.594 0.006
FIS 0.049 0.091 0.082 −0.004 0.073 0.079 −0.022 0.027

Prs328 n 92 84 101 98 101 48 97 46

#A 3 4 4 4 6 3 4 3
AR 3 3.98 3.45 3.63 4.58 3 3.83 3
HE 0.547 0.567 0.568 0.531 0.559 0.516 0.546 0.563
PHW 0.243 0.487 0.106 0.856 0.93 0.592 0.702 0.842
FIS −0.173 −0.008 0.128 0.039 -0.01 0.153 −0.075 −0.042

Prs333 n 93 85 101 98 101 46 97 46

#A 5 5 7 5 7 5 6 6
AR 4.85 4.09 5.08 3.76 5.53 4.98 5.01 5.96
HE 0.382 0.277 0.336 0.361 0.395 0.401 0.33 0.414
PHW 0.104 0.7 0.925 1 0.361 0.424 0.604 0.305
FIS −0.04 −0.148 −0.032 −0.019 0.073 0.078 0 0.212
24 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25

Sample NC SC GA DA ML SA MG PC

Ra6 n 92 85 101 98 101 48 97 46

#A 7 7 7 6 7 7 7 7
AR 6.44 5.95 5.72 5.41 6.02 6.88 6.03 6.96
HE 0.392 0.401 0.384 0.461 0.355 0.506 0.384 0.409
PHW 0.202 0.134 0.325 0.994 0.467 0.049 0.126 0.958
FIS 0.085 0.12 −0.032 −0.018 −0.003 0.218 0.14 −0.169

mtDNA
ND4 n 20 20 20 20 20 20 20 20
#H 8 11 7 8 11 10 7 6
h 0.758 0.884 0.784 0.805 0.905 0.842 0.711 0.658
 0.091 0.128 0.106 0.066 0.185 0.095 0.05 0.051

tations, and FIS = inbreeding coefficient. For mtDNA: n = number of Arèvalo, E., Davis, S.K., Sites, J.W., 1994. Mitochondrial DNA sequence divergence
individuals sampled, #H = number of unique haplotypes observed, and phylogenetic relationships among eight chromosome races of the
Sceloporus grammicus complex (Phrynosomatidae) in central Mexico. Syst. Biol.
h = haplotype diversity,  = nucleotide diversity. 43, 387–418.
Bagley, M., Geller, J., 1998. Characterization of microsatellite loci in the vermilion
snapper Rhomboplites aurorubens (Percoidei: Lutjanidae). Mol. Ecol. 7,
Appendix B 1089–1090.
Beaumont, M.A., 1999. Detecting population expansion and decline using
microsatellites. Genetics 153, 2013–2029.
Spatial distribution of mitochondrial (ND4) haplotypes. Beerli, P., 2006. Comparison of Bayesian and maximum-likelihood inference of
Haplotype NC SC GA DA ML SA MG PC population genetic parameters. Bioinformatics 22, 341–345.
Beerli, P., Felsenstein, J., 2001. Maximum likelihood estimation of a migration
#1 1 2 matrix and effective population sizes in n subpopulations by using a coalescent
#2 9 6 8 7 4 6 10 11 approach. Proc. Natl. Acad. Sci. 98, 4563–4568.
#3 1 Beerli, P., Palczewski, M., 2010. Unified framework to evaluate panmixia and
#4 5 4 5 6 5 6 5 5 migration direction among multiple sampling locations. Genetics 185,
#5 1 313–326, http://dx.doi.org/10.1534/genetics.109.112532
#6 1 1 1 1 Begg, G.A., Friedland, K.D., Pearce, J.B., 1999. Stock identification and its role in
#7 1 1 1 1 stock assessment and fisheries management: an overview. Fish. Res. 43, 1–8.
Bell, J.D., Bartley, D.M., Lorenzen, K., Loneragan, N.R., 2006. Restocking and stock
#8 1 1
enhancement of coastal fisheries: potential problems and progress. Fish. Res.
#9 1
80, 1–8.
#10 1
Bielawski, J.P., Gold, J.R., 2002. Mutation patterns of mitochondrial H-and L-strand
#11 2 DNA in closely related cyprinid fishes. Genetics 161, 1589–1597.
#12 1 Blaber, S.J.M., Dichmont, C.M., Buckworth, R.C., Badrudin, Sumiono, B., Nurhakim,
#13 1 1 1 S., Iskandar, B., Fegan, B., Ramm, D.C., Salini, J.P., 2005. Shared stocks of
#14 1 1 snappers (Lutjanidae) in Australia and Indonesia: Integrating biology,
#15 1 population dynamics and socio-economics to examine management scenarios.
#16 1 1 Rev. Fish Biol. Fish 15, 111–127.
#17 3 2 Botsford, L.W., Micheli, F., Hastings, A., 2003. Principles for the design of marine
#18 1 reserves. Ecol. Appl. 13, S25–S31.
#19 1 Brown-Peterson, N., Burns, K., Overstreet, R., 2009. Regional differences in Florida
#20 1 1 red snapper reproduction. Proc. Gulf Caribb. Fish. Inst. 61, 149–155.
#21 2 Burns, K.M., Brown-Peterson, N.J., Overstreet, R.M., Gannon, J., Simmons, P.,
Sprinkle, J., Weaver, C., 2008. Evaluation of the efficacy of the current
#22 1
minimum size regulation for selected reef fish based on release mortality and
#23 1
fish physiology. Mote Tech. Rep., 1176.
#24 1 Burns, K.M., Parnell, N.F., Wilson, R.R., 2004. Partitioning release mortality in the
#25 1 undersized red snapper bycatch: comparison of depth vs. hooking effects.
#26 1 Mote Tech. Rep., 932.
#27 1 Camper, J.D., Barber, R.C., Richardson, L.R., Gold, J.R., 1993. Mitochondrial DNA
#28 1 variation among red snapper (Lutjanus campechanus) from the Gulf of Mexico.
#29 1 Mol. Mar. Biol. Biotechnol. 2, 154–161.
#30 1 Carson, E., Saillant, E., 2011. Population structure, long-term connectivity, and
#31 1 effective size of mutton snapper (Lutjanus analis) in the Caribbean Sea and
#32 1 Florida Keys. Fish Bull. 109, 416–428.
#33 1 1 Carvalho, G., Hauser, L., 1994. Molecular genetics and the stock concept in
#34 1 fisheries. Rev. Fish Biol. Fish. 4, 326–350.
Cavalli-Sforza, L., Edwards, A., 1967. Phylogenetic analysis: models and estimation
#35 1
procedures. Am. J. Hum. Genet. 19, 233–257.
#36 1
Christman, M., 1997. Peer Review of Red Snapper (Lutjanus campechanus) Research
#37 1
and Management in the Gulf of Mexico: Statistics Review, Office of Science and
#38 1 Technology, NOAA. Silver Springs, MD.
#39 1 Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. jModelTest 2: more models,
new heuristics and parallel computing. Nat. Methods 9, 772.
Davey, J.W., Hohenlohe, P.A., Etter, P.D., Boone, J.Q., Catchen, J.M., Blaxter, M.L.,
References 2011. Genome-wide genetic marker discovery and genotyping using
next-generation sequencing. Nat. Rev. Genet. 12, 499–510.
Estoup, A., Largiader, C.R., Perrot, E., Chourrout, D., 1996. Rapid one-tube DNA
Akçakaya, H.R., Mills, G., Doncaster, C.P., 2007. The role of metapopulations in extraction for reliable PCR detection of fish polymorphic markers and
conservation. In: Macdonald, D.W., Service, K. (Eds.), Key Topics in transgenes. Mol. Mar. Biol. Biotechnol. 5, 295–298.
Conservation Biology. Blackwell Publishing, pp. 64–84. Excoffier, L., Lischer, H., 2010. Arlequin suite ver 3. 5: a new series of programs to
Allen, G., 1985. FAO Species Catalogue Snappers of the World: An Annotated and perform population genetics analyses under Linux and Windows. Mol. Ecol.
Illustrated Catalogue of Lutjanid Species Known to Date, vol. 6. Rome, Food and Resour. 10, 564–567.
Agriculture Organization of the United Nations. Fischer, A.J., Baker, M.S., Wilson, C.A., 2004. Red snapper (Lutjanus campechanus)
Allendorf, F.W., Hohenlohe, P.A., Luikart, G., 2010. Genomics and the future of demographic structure in the northern Gulf of Mexico based on spatial
conservation genetics. Nat. Rev. Genet. 11, 697–709. patterns in growth rates and morphometrics. Fish. Bull. 102, 593–603.
Allendorf, F.W., Waples, R.S., 1996. Conservation and genetics of salmonid fishes. Fitzhugh, G.R., Duncan, M.S., Collins, L.A., Walling, W.T., Oliver, D.W., 2004.
In: Avise, J.C., Hamrick, J.L. (Eds.), Conservation Genetics: Case Histories from Characterization of red snapper (Lutjanus campechanus) reproduction: for the
Nature. Chapman & Hall, pp. 238–280.
 C.M. Hollenbeck et al. / Fisheries Research 172 (2015) 17–25 25

2004 Gulf of Mexico SEDAR. NMFS, SEFSC, Panama City, Florida. Contrib. 1–4. Raymond, M., Rousset, F., 1995. GENEPOP (version 1.2): population genetics
Available at: http://www.sefsc.noaa.gov/sedar/download/SEDAR7 DW35. software for exact tests and ec.
pdf?id=DOCUMENT Renshaw, M.A., Saillant, E., Bradfield, S.C., Gold, J.R., 2006. Microsatellite multiplex
Frankham, R., 1995. Effective population size/adult population size ratios in panels for genetic studies of three species of marine fishes: red drum
wildlife: a review. Genet. Res. 66, 95–107. (Sciaenops ocellatus), red snapper (Lutjanus campechanus), and cobia
Garber, A.F., Tringali, M.D., Stuck, K.C., 2004. Population structure and variation in (Rachycentron canadum). Aquaculture 253, 731–735.
red snapper (Lutjanus campechanus) from the Gulf of Mexico and Atlantic coast Rice, W.R., 1989. Analyzing tables of statistical tests. Evolution 43, 223–225.
of Florida as determined from mitochondrial DNA control region sequence. Rousset, F., 2008. genepop’007: a complete re-implementation of the genepop
Mar. Biotechnol. 6, 175–185. software for Windows and Linux. Mol. Ecol. Resour. 8, 103–106.
Gold, J.R., Pak, E., Richardson, L.R., 2001. Microsatellite variation among red snapper Russ, G., Alcala, A., 1989. Effects of intense fishing pressure on an assemblage of
(Lutjanus campechanus) from the Gulf of Mexico. Mar. Biotechnol. 3, 293–304. coral reef fishes. Mar. Ecol. Prog. Ser. 56, 13–27.
Gold, J.R., Saillant, E., Ebelt, N.D., Lem, S., 2009. Conservation genetics of gray Ruzzante, D.E., Taggart, C.T., Cook, D., 1999. A review of the evidence for genetic
snapper (Lutjanus griseus) in U. S. waters of the Northern Gulf of Mexico and structure of cod (Gadus morhua) populations in the NW Atlantic and
Western Atlantic Ocean. Copeia, 277–286. population affinities of larval cod off Newfoundland and the Gulf of St.
Gold, J.R., Sun, F., Richardson, L.R., 1997. Population structure of red snapper from Lawrence. Fish. Res. 43, 79–97.
the Gulf of Mexico as inferred from analysis of Mitochondrial DNA. Trans. Am. Saillant, E., Bradfield, S.C., Gold, J.R., 2010. Genetic variation and spatial
Fish. Soc. 126, 386–396. autocorrelation among young-of-the-year red snapper (Lutjanus campechanus)
Goudet, J., 2001. FSTAT, a program to estimate and test gene diversities and in the northern Gulf of Mexico. ICES J. Mar. Sci. 67, 1240–1250.
fixation indices (version 2.9.3). Available at: http://www2.unil.ch/popgen/ Saillant, E., Gold, J.R., 2006. Population structure and variance effective size of red
softwares/fstat.htm snapper (Lutjanus campechanus) in the northern Gulf of Mexico. Fish. Bull. 104,
Grimm, V., Reise, K., Strasser, M., 2003. Marine metapopulations: A useful concept? 136–148.
Helgoland Marine Research, pp. 222–228. Saillant, E.A., Renshaw, M.A., Cummings, N.J., Gold, J.R., 2012. Conservation
Guindon, S., Gascuel, O., 2003. A simple, fast, and accurate algorithm to estimate genetics and management of yellowtail snapper, Ocyurus chrysurus, in the US
large phylogenies by maximum likelihood. Syst. Biol. 52, 696–704. Caribbean and South Florida. Fish. Manag. Ecol. 19, 301–312.
GMFMC, 2013. Red Snapper 2013. Quota Increase and Supplemental Recreational Schirripa, M.J., Legault, C.M., 1999. Status of red snapper in the US waters of the
Season Including Environmental Assessment, Regulatory Impact Review, and Gulf of Mexico: updated through 1998, National Marine Fisheries Service,
Regulatory Flexibility Act Analysis. Gulf of Mexico Fishery Management Southeast Fisheries Science Center. Sustainable Fisheries Division contribution
Council. Tampa, FL. SFD-99/00–75. Miami Laboratory, Miami, Florida. Miami, Florida.
Hardy, O.J., Vekemans, X., 1999. Isolation by distance in a continuous population: Schroepfer, R.L., Szedlmayer, S.T., 2006. Estimates of residence and site fidelity for
Reconciliation between spatial autocorrelation analysis and population red snapper Lutjanus campechanus on artificial reefs in the northeastern Gulf of
genetics models. Heredity 83, 145–154. Mexico. Bull. Mar. Sci. 78, 93–101.
Hastings, A., 1993. Complex interactions between dispersal and dynamics: lessons SEDAR, 2009. Southeast Data, Assessment, and Review: SEDAR 15 Stock
from coupled logistic equations. Ecology 74, 1362–1372. Assessment Report 1 (SAR 1) South Atlantic Red Snapper. North Charleston, SC.
Hauser, L., Carvalho, G.R., 2008. Paradigm shifts in marine fisheries genetics: ugly SEDAR, 2010. Southeast Data, Assessment, and Review: SEDAR 24 South Atlantic
hypotheses slain by beautiful facts. Fish Fish. 9, 333–362. Red Snapper. North Charleston, SC.
Heist, E.J., Gold, J.R., 2000. DNA microsatellite loci and genetic structure of red SEDAR, 2013. Southeast Data, Assessment, and Review: SEDAR 31 Gulf of Mexico
snapper in the Gulf of Mexico. Trans. Am. Fish. Soc. 129, 469–475. Red Snapper Stock Assessment Report. North Charleston, SC.
Higgins, K., Lynch, M., 2001. Metapopulation extinction caused by mutation Seutin, G., White, B.N., Boag, P.T., 1991. Preservation of avian blood and tissue
accumulation. Proc. Natl. Acad. Sci. 98, 2928–2933. samples for DNA analyses. Can. J. Zool. 69, 82–90.
Hilborn, R., Quinn, T.P., Schindler, D.E., Rogers, D.E., 2003. Biocomplexity and Smith, B.J., 2005. Bayesian output analysis program (BOA) for MCMC. Available at:
fisheries sustainability. Proc. Natl. Acad. Sci. U. S. A. 100, 6564–6568. http://phase.hpcc.jp/mirrors/stat/R/CRAN/doc/packages/boa.pdf
Hood, P., Strelcheck, A., 2007. A history of red snapper management in the Gulf of Storz, J.F., Beaumont, M.A., 2002. Testing for genetic evidence of population
Mexico. In: Patterson III, W.F., Cowan Jr., J.H., Fitzhugh, G.R., Nieland, D.L. expansion and contraction: an empirical analysis of microsatellite DNA
(Eds.), Red Snapper Ecology and Fisheries in the U.S. Gulf of Mexico. American variation using a hierarchical Bayesian model. Evolution 56, 154–166.
Fisheries Society, Bethesda, Maryland, pp. 241–256. Strelcheck, A.J., Hood, P.B., Patterson III, W.F., Cowan Jr., J.H., Fitzhugh, G.R.,
Hudson, R.R., 2000. A new statistic for detecting genetic differentiation. Genetics Nieland, D.L., 2007. Rebuilding red snapper: Recent management activities and
155, 2011–2014. future management challenges. Am. Fish. Soc. Symp. 60, 353–363.
Jackson, M.W., Cowan Jr., J.H., Nieland, D.L., Patterson, W., Cowan, Jhj., Fitzhugh, Szedlmayer, S.T., 1997. Ultrasonic telemetry of red snapper, Lutjanus campechanus,
G.R., 2007. Demographic differences in northern Gulf of Mexico red snapper at artificial reef sites in the northeast Gulf of Mexico. Copeia 1997, 846–850.
reproductive maturation: implications for the unit stock hypothesis. Am. Fish. Tamura, K., Nei, M., 1993. Estimation of the number of nucleotide substitutions in
Soc. Symp. 60, 197–206. the control region of mitochondrial DNA in humans and chimpanzees. Mol.
Kough, A.S., Paris, C.B., Butler, M.J., 2013. Larval connectivity and the international Biol. Evol. 10, 512–526.
management of fisheries. PLoS One 8, e64970. Tringali, M., Higham, M., 2007. Isolation-by-Distance Gene Flow Among Vermilion
Kritzer, J., Sale, P., 2006. Marine Metapopulations. Academic Press. Snapper (Rhomboplites aurorubens Cuvier, 1829) From the Gulf of Mexico and
Mills, L.S., Allendorf, F.W., 1996. The one-migrant-per-generation rule in Southeastern United. Gulf Mex. Sci. 1, 2–14.
conservation and management. Conserv. Biol. 10, 1509–1518. Van Oosterhout, C., Hutchinson, W.F., Wills, D.P.M., Shipley, P., 2004.
Moe, M.A., Jr., 1963. A survey of offshore fishing in Florida. Professional Papers Micro-checker: software for identifying and correcting genotyping errors in
Series, Number Four. Florida State Board of Conservation. Available at: http:// microsatellite data. Mol. Ecol. Notes 4, 535–538.
www.sefsc.noaa.gov/sedar/download/SEDAR24-RD08 Moe1963. Waples, R.S., 2005. Genetic estimates of contemporary effective population size: to
pdf?id=DOCUMENT what time periods do the estimates apply. Mol. Ecol. 14, 3335–3352.
Nielsen, E.E., Hemmer-Hansen, J., Larsen, P.F., Bekkevold, D., 2009. Population Waples, R.S., 2006. A bias correction for estimates of effective population size based
genomics of marine fishes: identifying adaptive variation in space and time. on linkage disequilibrium at unlinked gene loci. Conserv. Genet. 7, 167–184.
Mol. Ecol. 18, 3128–3150. Waples, R.S., 2010. Spatial-temporal stratifications in natural populations and how
NOAA, 2012. Status of stocks: Report on the status of U.S. fisheries for they affect understanding and estimation of effective population size. Mol.
2011—Report to Congress. Available at: http://www.nmfs.noaa.gov/stories/ Ecol. Resour. 10, 785–796.
2012/05/docs/status of stocks 2011 report.pdf Waples, R.S., Antao, T., Luikart, G., 2014. Effects of overlapping generations on
Orozco-Terwengel, P., Corander, J., SchlÖtterer, C., 2011. Genealogical lineage linkage disequilibrium estimates of effective population size. Genetics 197,
sorting leads to significant, but incorrect Bayesian multilocus inference of 769–780.
population structure. Mol. Ecol. 20, 1108–1121. Waples, R.S., Do, C., 2008. LDNE: a program for estimating effective population size
Patterson III, W.F., Watterson, J.C., Shipp, R.L., Cowan Jr, J.H., 2001. Movement of from data on linkage disequilibrium. Mol. Ecol. Resour. 8, 753–756.
tagged red snapper in the northern Gulf of Mexico. Trans. Am. Fish Soc. 130, Waples, R.S., Do, C., 2010. Linkage disequilibrium estimates of contemporary Ne
533–545. using highly variable genetic markers: a largely untapped resource for applied
Patterson, W.F., Cowan, J.H., 2003. Site fidelity and dispersion of red snapper conservation and evolution. Evol. Appl. 3, 244–262.
associated with artificial reefs in the northern Gulf of Mexico. Am. Fish. Soc. Watterson, J.C., Patterson, W.F., Shipp, R.L., Cowan, J.H., 1998. Movement of red
Symp. 36, 181–193. snapper, Lutjanus campechanus, in the north central Gulf of Mexico: Potential
Pruett, C.L., Saillant, E., Gold, J.R., 2005. Historical population demography of red effects of hurricanes. Gulf Mex. Sci. 16, 92–104.
snapper (Lutjanus campechanus) from the northern Gulf of Mexico based on Weir, B.S., Cockerham, C.C., 1984. Estimating F-statistics for the analysis of
analysis of sequences of mitochondrial DNA. Mar. Biol. 147, 593–602. population structure. Evolution, 1358–1370.
R Core Team, 2013. R: A Language and Environment for Statistical Computing. R Wilson, C.A., Nieland, D.L., 2001. Age and growth of red snapper Lutjanus
Foundation for Statistical Computing. Vienna, Austria. Available at: http:// campechanus from the northern Gulf of Mexico off Louisiana. Fish. Bull. 99,
www.r-project.org/ 653–664.