Distribution, Abundance, and Habitat Characteristics of
Fundulus jenkinsi (Evermann) (Saltmarsh Topminnow)
in Coastal Mississippi Watersheds, with Comments on
Range-wide Occurrences Based on Non-vouchered and
Museum Records
Mark S. Peterson, William T. Slack, and Erik T. Lang
Southeastern Naturalist, Volume 15, Issue 3 (2016): 415–430
Full-text pdf (Accessible only to subscribers.To subscribe click here.)
Southeastern Naturalist
415
M.S. Peterson, W.T. Slack, and E.T. Lang
22001166 SOUTHEASTERN NATURALIST 1V5o(3l.) :1451,5 N–4o3. 03
Distribution, Abundance, and Habitat Characteristics of
Fundulus jenkinsi (Evermann) (Saltmarsh Topminnow)
in Coastal Mississippi Watersheds, with Comments on
Range-wide Occurrences Based on Non-vouchered and
Museum Records
Mark S. Peterson1,*, William T. Slack2, 3, and Erik T. Lang4
Abstract - Fundulus jenkinsi (Saltmarsh Topminnow) is listed as “at risk” by the USFWS
and as a Tier 2 conservation priority in Mississippi, in part, because of marsh-habitat loss
due to storms, urbanization, and its specialized habitat requirements and limited geographic
distribution. To provide additional quantitative data for conservation planning, our objectives
were to (1) determine the distribution and abundance of Saltmarsh Topminnow within
coastal Mississippi, (2) characterize its habitat requirements, and (3) organize and present
all Saltmarsh Topminnow data records (non-vouchered and museum records and those from
this study) for use in the development of management/conservation plans. We collected
497 fish and associated habitat data from 27 February to 1 August 2009. PCA produced 3
meaningful components: (1) a landscape-position axis (32.40% of the total variance), (2) a
seasonal/spatial axis of species occurrence (18.99%), and (3) a geomorphic bank-slope
axis (18.78%). Ninety-six percent of all fish (representing 78.8% of collection effort) were
captured in water with salinity less than 13 psu. We compiled 831 geo-referenced occurrences with
collection dates ranging from 1891 to 2015. To better quantify and conserve the closelylinked
habitat requirements of this species within a reduced salinity range, additional
sampling should be focused in undersampled areas between Lake Borgne, LA, to west of
Galveston Bay, TX.
Introduction
Fundulus jenkinsi (Evermann) (Saltmarsh Topminnow) occurs sporadically
from Galveston Bay, TX to Escambia Bay, FL, although Peterson et al. (2003) and
Lopez et al. (2011) suggested that this species might be more widely distributed
and abundant than previously believed. Two recent studies confirmed that initial
hypothesis. First, Martin et al. (2012) reported new records of Saltmarsh Topminnow
in Texas that documented a western range extension (the Tres Palacios River).
Second, Guillen et al. (2015) reported new localities in the Galveston-Trinity Bay
system as well as Sabine Lake, TX, and also documented Saltmarsh Topminnow
1Department of Coastal Sciences, The University of Southern Mississippi, 703 East Beach
Drive, Ocean Springs, MS 39564. 2Mississippi Department of Wildlife, Fisheries and Parks,
Mississippi Museum of Natural Science, 2148 Riverside Drive, Jackson, MS 39202-1353.
3Current address - US Army Engineer Research and Development Center, Waterways Experiment
Station EE-A, 3909 Halls Ferry Road, Vicksburg, MS 39180-6199. 4Fisheries
Management, Louisiana Department of Wildlife and Fisheries, 2000 Quail Drive, Baton
Rouge, LA 70817. *Corresponding author - ecofishconsulting@gmail.com.
Manuscript Editor: Carol Johnston
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
416
in the type locality watershed; no specimens had been found at the type locality
(Dickinson Bayou; Evermann 1892) in Galveston-Trinity Bay since 1951 despite
considerable collecting efforts
Historically, 2 environmental factors have been suggested as influencing this
species’ distribution and abundance. Saltmarsh Topminnow uses Spartina alterniflora
Loiseleur (Smooth Cordgrass) marsh (Peterson and Turner 1994, Suttkus
et al. 1999, Thompson 1980), and has been found mainly where salinity ranges
between 1 and 4 psu (Bailey et al. 1954, Boschung and Mayden 2004, Gilbert and
Relyea 1992, Thompson 1980). The most comprehensive study (Lopez et al. 2011)
collected 661 Saltmarsh Topminnows from the Barataria-Terrebonne basin, in LA,
to as far east as Escambia Bay, FL, and used principal component analysis (PCA)
to ordinate physical–chemical data into a geomorphic axis (water depth, bank
slope, and plant-stem density) and a seasonal/spatial axis of species occurrence
(water temperature, salinity, and turbidity). PCA illustrated a higher mean CPUE
in habitats comprised of low to moderate stem density (<25 stems/0.25 m2), depth
(<25 cm), bank slope (<15°), turbidity (<30 NTU), and salinity (<16 psu) coupled
with spring and early summer water temperatures (>15 °C). Saltmarsh Topminnow
CPUE was significantly higher in Spartina cynosuroides (L.) Roth (Big Cordgrass)
marsh edge compared to 5 other habitat types, even though it was one of the leastsampled
habitats.
Given its apparent low relative abundance and patchy distribution, there is a
real need to obtain distribution-wide data on habitat characteristics for Saltmarsh
Topminnow to better manage and conserve this species and its habitat (Guillen et
al. 2015, Lopez et al. 2011). In fact, little is known about the distribution and habitat
characteristics of Saltmarsh Topminnow throughout its entire range in coastal
Mississippi (MMNS 2015). This situation is particularly important for 2 reasons.
First, the extensive development of the dock-side gaming industry and associated
activities in coastal Mississippi has added further concern about the status of Saltmarsh
Topminnow in coastal wetlands which are being urbanized at a faster rate
than estimated in the past (Meyer-Arendt et al. 1998, MMNS 2015) following shifts
in general population demographics (e.g., Crossett et al. 2004, European Environmental
Agency 2006). Second, the “at risk” designation of Saltmarsh Topminnow
on the USFWS website (http://ecos.fws.gov/speciesProfile/profile/speciesProfile.
action?spcode=E0BO) indicates considerable coastal-marsh habitat loss in the
north-central Gulf of Mexico (GOM) attributable, in part, to multiple hurricanes
(George [1998], Ivan [2004], Dennis, Katrina, and Rita [2005]), and continued
coastal urbanization (Bulleri and Chapman 2010; Chapman and Underwood 2011;
Lowe and Peterson 2014, 2015; Peterson and Lowe 2009). The Mississippi State
Wildlife Plan (MMNS 2015) considers this species as a species of greatest conservation
need and ranks it as a Tier 2 conservation-priority because of “… specialized
habitat needs or habitat vulnerability.” Furthermore, the 2010 Deepwater Horizon
oil spill (Alford et al. 2014) and projected sea-level rise (Fulford et al. 2014) threaten
further potentially deleterious consequences for these communities. In addition
to these impacts, dredging of waterways will have cumulative effects on hydrology,
Southeastern Naturalist
417
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
water quality, and habitat. Specifically, dredging allows salt-water intrusion further
up bayous and tidal rivers, and shifts optimal salinity to locations with different
habitat structure (Peterson 2003, Peterson et al. 2007, Guillen et al. 2015), potentially
changing distribution patterns and reducing estuarine production. Saltmarsh
Topminnow is also listed as a species of concern in Mississippi (Ross 2001), threatened
in Florida (Gilbert and Relyea 1992), endangered in Alabama (Boschung and
Mayden 2004), a species of concern in Louisiana (R. DeMay, Barataria-Terrebonne
National Estuary Program, Thibodaux, LA; pers. comm.), and a species of greatest
conservation need for the Gulf Coast Prairies and Marshes ecoregion of Texas
(TPWD 2005, 2011, 2012).
The specific objectives of this study were to (1) determine the distribution and
abundance of Saltmarsh Topminnow within coastal Mississippi, (2) characterize its
habitat requirements, and (3) organize and present all data records (non-vouchered
and museum records and those from this study) of Saltmarsh Topminnow for use in
the development of management/conservation plans.
Methods
We set and fished Breder traps that faced inshore at 8 general sites occurring
within identified watersheds of the Pascagoula River, Old Ft. Bayou, Biloxi River,
Tchoutacabouffa River, Bernard Bayou, Wolf River, Jourdan River, and Pearl
River, which are all major waterways within the coastal Mississippi landscape.
Based on previous research (Lang et al. 2012, Lopez et al. 2011, Peterson et al. 2003),
we selected sites according to salinity regimes and vegetation present along the
marsh edge within each watershed. We focused on sampling pure Juncus roemarianus
Scheele (Black Needlerush), Smooth Cordgrass bordered by Black Needlerush,
and pure Smooth Cordgrass habitats; other habitat types were sampled as encountered.
Initially, we sampled at 3 sites using 5 Breder traps (Breeder 1960, Lang et al.
2012, Lopez et al. 2011, Peterson et al. 2003) and 5 Gee minnow traps on 27 February
2009 and again on 2 March 2009 (30 Breder traps and 30 Gee traps) in order to
compare catch-per-unit-effort (CPUE; fishing from high to low tide) with time each
trap fished and CPUE by gear type in order to choose gear type. There was no correlation
between time fished and CPUE pooled by gear type (r = 0.180, P = 0.504),
Breder trap samples only (r = 0.210, P = 0.369) or Gee minnow traps only (r = 0.230,
P = 0.279). Also, there was no difference between Breder trap and Gee minnow trap
CPUE (ANOVA: F1,17 = 0.320; P = 0.580); thus, we only used the Breder trap geartype
collections for the entire study because that collection method is consistent with
the one we used in earlier research (Lopez et al. 2010, 2011).
We fished the falling tide from 27 February to 1 August 2009 (21 dates) in each
habitat type with up to 30 Breder traps (Fig. 1) at a time (6 traps per habitat type). At
each site, we placed the traps 1 m apart at 3–5 locations within a watershed across
the entire study area. Each Breder traps was constructed out of 0.635-cm-thick
Plexiglass to make a box that was 30 cm x 15 cm with two 15 cm x 15 cm wings
protruding from each side. The wings formed a “V” with a 12-mm opening to catch
anything that was moving with the outgoing tide. At each site, we measured water
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
418
temperature (°C), turbidity (NTU, nephlometric turbidity units), and salinity (psu)
once for each 6-trap set. For each individual trap, we measured depth (cm) and bank
slope (in degrees), and we used a 0.25-m2-PVC quadrat to determine plant-stem
density (number of stems/0.25 m2) immediately in front of each trap. We used a
Garmin GPS 76 unit (Garmin International, Inc., Olathe, KS) to geo-reference each
trap location. We also measured dissolved oxygen (DO, mg/L) at each site, but did
not include it in our analyses because DO was negatively correlated with water
temperature (see Lopez et al. 2011).
To examine the relationship between physical–chemical variables and Saltmarsh
Topminnow CPUE, we used a 2-step multivariate procedure (Lopez et al.
2011, Peterson and Vanderkooy 1997). First, we ordinated the 7 physical–chemical
variables (only 6 variables used in final analysis; see below) using PCA of the
correlation matrix (Field 2013) with varimax rotation to maximally resolve loadings,
and considered any variable that loaded on a component at an absolute value
≥0.50 as making a significant contribution to interpreting that component (Hair et
al. 1984, Lopez et al. 2011). All variables used in the PCA were either the mean
value of the 6 traps (water depth, bank slope and stem density) or the single value
for the set of 6 traps as noted earlier. Thus, descriptive summary statistics are grand
mean (± SEM) values for water depth, bank slope, and stem density, and standard
mean (± SEM) values for the other variables. We plotted standardized factor-scores
Figure 1. Image of a Breder trap (Breder 1960) used in this study.
Southeastern Naturalist
419
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
of the most meaningful components for each sample with mean CPUE as a third
axis. We employed the Kaiser-Meyer-Olin (KMO) measure of sampling adequacy
and Bartlett’s test of sphericity to test the adequacy of our sample size for the PCA
analysis. The KMO statistic ranges from 0 to 1, with 0 indicating diffusion in the
pattern of correlations (hence, PCA is likely not appropriate for the data set) and
values close to 1 indicating the pattern of correlation is compact and thus PCA is
appropriate (Field 2013). We considered values 0.5–0.7 as mediocore, 0.7–0.8 as
good, 0.8–0.9 as great, and >0.9 as superb. KMO values less than 0.5 indicate that data are
not appropriate to use in PCA (Field 2013). Bartlett’s measure tests the null hypothesis
that the original correlation matrix is an identity matrix; for PCA to work
correctly, the measure must be significantly (P < 0.05) different from the null because
the variables must have some degree of correlation. Finally, we used stepwise
multiple-regression analysis to examine the relationship between the standardized
factor-scores from the important components of the PCA analysis and mean CPUE
of Saltmarsh Topminnow as a support tool for the results of the PCA analysis. We
also conducted a one-way ANOVA to compare Saltmarsh Topminnow CPUE across
habitat types and among locations sampled. When F-values were significant, we
used either a Sidak (equal variances) or a Games-Howell (G-H) (heterogeneous
variances) post-hoc test to separate mean responses. All statistical analyses were
conducted in SPSS software (ver. 15.0; SPSS, Inc., Chicago, IL) and we set significance
at P ≤ 0.05 (Field 2013).
We used ArcMap 10 to plot occurrence records of Saltmarsh Topminnow based
on our collections, non-vouchered records, and compiled museum records to depict
the currently recognized range of the species and to identify potential regions for
future studies (see Supplemental File 1, available online at http://www.eaglehill.
us/SENAonline/suppl-files/s15-3-S2249-Peterson-s1, and, for BioOne subscribers,
at http://dx.doi.org/10.1656/S2249.s1). We obtained data directly from partnering
institutions (GCRL, Inland Fishes of Mississippi database [Ross, 2001], and
MMNS), online data sources (FishNet2 [http://www.fishnet2.net/], FoTX [https://
sites.cns.utexas.edu/hendricksonlab/FoTX_Sandbox], Tulane Museum of Natural
History, New Orleans, LA), and through personal communication (J. Knight,
Florida FWCC, Holt, FL; G. Guillen, University of Houston-Clear Lake, TX). We
obtained occurrence records from ANSP, AUM, CAS, CUMV, the current study,
FMNH, FoTX, FSBC, FWC, GCRL, INHS, KU, LSUMZ, MCZ, MMNS, TCWC,
TNHC, TU, UAIC, UF, UHCL, UMMZ, UNO, USM, USNM, UT and YPM. Source
abbreviations follow Sabaj Pérez (2010) except for FoTX (Fishes of Texas database
[Hendrickson and Cohen, 2015]), FWC (Florida Fish and Wildlife Conservation
Commission) and UHCL (University of Houston - Clear Lake). We accessed Tulane
Museum of Natural History collection records on 17 December 2009 through the
GBIF data portal, Tulane University Museum of Natural History (http://data.gbif.
org/datasets/resource/8077), and records from the Fishnet2 Portal (www.fishnet2.
net) on 16 July 2015. We did not examine museum voucher material except where
noted (current study, GCRL, and MMNS).
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
420
Results
Between February and August 2009, we sampled 674 Breder traps and collected
497 Saltmarsh Topminnow at 8 sites across the state of Mississippi from the Pearl
to the Pascagoula watersheds (Fig. 2), areas not previously systematically sampled.
Table 1 shows CPUE, water quality and vegetation summary-statistics (mean ± 1
standard error of the mean [SEM]) for the 8 sites for all Breder trap sets. Collections
from the Tchoutacoubouffa watershed had the highest CPUE (1.44 ± 0.49),
followed by the Pascagoula (1.27 ± 0.44), and then the Pearl (0.77 ± 0.19); other
watersheds had lower CPUE values (Table 1). However, Saltmarsh Topminnow
CPUE (log10) did not differ significantly (ANOVA: F7,113 = 1.823, P = 0.090) among
the 8 sites or among the 12 replicated vegetation types (within watersheds) at our
sites (range = 2–18 replicate samples each) (ANOVA: F11,74 = 0.939, P = 0.510).
Saltmarsh Topminnow CPUE (log10) also did not differ among the 3 main habitat
types pooled across sites (Smooth Cordgrass, Black Needlerush, mixed Smooth
Cordgrass–Black Needlerush); F1,25 = 0.255, P = 0.777). Finally, we collected
97.2% of all Saltmarsh Topminnow (representing 84.9% of our collection effort) at
salinity levels of less than 16 psu and 95.8% (78.8% of our collection effort) when salinity
was less than 13 psu. The highest salinity at any site where we collected Saltmarsh Topminnow
was 19.8 psu.
The KMO measure of sampling adequacy of 0.610 suggests a mediocre factorsolution
of the data set (Field 2013) and that the PCA analysis produced distinct
Figure 2. Overall project area in coastal Mississippi where sites were sampled with Breder
traps during the 2008–2009 Saltmarsh Topminnow project (top left panel). Sites sampled
west of Highway 49 (Panel A) and sites sampled east of Highway 49 (Panel B) during the
project period noting samples where Saltmarsh Topminnow were present (solid circle) or
absent (gray square). What may appear to be darker or black squares are actually overlapping
gray squares due to close spatial positioning of some of the Breder traps.
Southeastern Naturalist
421
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
Table 1. Summary statistics for Saltmarsh Topminnow CPUE, water quality, and plant-stem density by watershed for all Breder traps set. Number of traps
set is found parenthetically under each watershed name. Temperature, salinity, dissolved oxygen (DO) and turbidity values are presented as mean ± 1
SEM (standard error of the mean); bank slope, depth, and total stem-density are presented as grand mean ± 1 SEM. Stem density is expressed as number
of stems/0.25m2.
Watershed CPUE Temp. (°C) DO (mg/L) Salinity (psu) Turbidity (NTU) Depth (cm)
Pascagoula (n = 120) 1.27 ± 0.44 25.01 ± 0.24 7.49 ± 0.30 1.31 ± 0.20 13.17 ± 0.99 22.97 ± 0.51
Ft. Bayou (n = 90) 0.03 ± 0.02 29.63 ± 1.68 6.67 ± 0.23 13.13 ± 2.59 8.00 ± 1.39 22.38 ± 0.97
Biloxi (n = 60) 0.58 ± 0.19 30.70 ± 0.30 5.77 ± 0.38 8.32 ± 0.63 2.03 ± 0.31 24.27 ± 0.65
Tchoutacabouffa (n = 90) 1.44 ± 0.49 26.67 ± 1.49 5.99 ± 0.40 5.89 ± 1.10 10.85 ± 2.65 19.68 ± 0.60
Bernard Bayou (n = 12) 0.33 ± 0.19 30.00 ± 1.20 5.35 ± 2.10 10.00 ± 0.10 2.12 ± 0.47 19.00 ± 1.57
Wolf (n = 120) 0.44 ± 0.12 29.65 ± 0.37 5.41 ± 0.47 7.15 ± 1.66 11.62 ± 2.72 16.50 ± 0.75
Jourdan (Diamondhead) (n = 90) 0.49 ± 0.12 28.36 ± 0.80 6.52 ± 0.37 11.20 ± 2.11 14.76 ± 2.96 20.64 ± 0.57
Pearl (n = 90) 0.77 ± 0.19 30.57 ± 1.97 7.41 ± 0.41 7.43 ± 1.89 19.95 ± 3.30 20.40 ± 0.46
Juncus roemerianus Spartina alterniflora Spartina cynosuroides
Watershed Bank slope (°) density density density Total stem density
Pascagoula (n = 120) 6.95 ± 0.44 4.60 ± 0.98 3.50 ± 0.63 1.05 ± 0.61 6.32 ± 0.81
Ft. Bayou (n = 90) 5.43 ± 0.49 2.69 ± 0.79 1.54 ± 0.56 - 3.73 ± 0.73
Biloxi (n = 60) 9.56 ± 0.75 5.87 ± 1.93 4.00 ± 1.06 - 4.98 ± 1.07
Tchoutacabouffa (n = 90) 5.70 ± 0.49 3.48 ± 1.19 - - 4.37 ± 0.97
Bernard Bayou (n = 12) 4.42 ± 0.89 - - - -
Wolf (n = 120) 6.56 ± 0.41 2.98 ± 0.77 4.11 ±0.75 1.00 ± 0.72 7.27 ± 0.80
Jourdan (Diamondhead) (n = 90) 7.82 ± 0.63 1.93 ± 0.65 2.54 ± 0.57 0.17 ± 0.17 3.68 ± 0.61
Pearl (n = 90) 8.21 ± 0.67 0.83 ± 0.83 1.67 ± 1.00 0.42 ± 0.29 1.11 ± 0.29
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
422
components based on sampling adequacy. The Bartlett’s test of sphericity was also
significant (P < 0.001), supporting the KMO results. Results of the PCA indicated
that the 6 original variables were reduced to 3 meaningful components (eigenvalues
> 1.00) that explained 70.18% of the variation (Table 2). Component I explained
32.40% of the total variance and was composed of positive correlations with salinity
and mean water-depth, and a negative correlation with mean turbidity, which
we interpret as a landscape position axis. The standardized factor-scores were
scattered across the axis suggesting that we sampled a range of conditions. Component
II explained an additional 18.99% of the total variance and was composed of
water temperature and mean total plant-stem density (both positive correlations);
however, the standardized factor-scores were somewhat narrowly distributed. This
result suggests that we made our collections across a limited range of both water
temperature and plant-stem density. We interpreted this component as a seasonal/
spatial axis of species occurrence (Table 2). Component III explained 18.78% of
the total variance and was composed of a positive correlation with mean bankslope.
The standardized factor-scores were distributed across a range of conditions
(Table 2), and we interpreted Component III as a geomorphic bank- slope axis.
Plots of combinations of Components I, II, and III arrayed the mean Saltmarsh
Topminnow CPUE and suggested that these axes provide important information
about the habitat characteristics of Saltmarsh Topminnow across all systems
studied (Fig. 3). The stepwise linear regression supported Component I and II as
significant but weak predictors of mean CPUE of Saltmarsh Topminnow across all
systems (adjusted R2 = 0.063, n = 107, P = 0.013).
We used ArcMap 10 to plot all occurrence records of Saltmarsh Topminnow based
on our collections, non-vouchered specimens, and museum records (Fig. 4). We compiled
874 total occurrence records, which included 10 non-vouchered observations
and 33 records lacking sufficient data for geo-referencing (see Supplemental File 1,
available online at https://www.eaglehill.us/SENAonline/suppl-files/s15-3-S2249-
Peterson-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/S2249.s1).
The range map includes 831 geo-referenced occurrences representing 27 institutions
(museums, universities and agencies) and 5 states (FL: 79, AL: 172, MS: 355, LA:
216 and TX: 52). The occurrences within Alabama, Mississippi and Louisiana comprise
nearly 85% of the total records compiled. Dates for collection range from 1891
through 2015, with 91% of the records dating since 1980.
Table 2. Varimax-rotated component matrix of the PCA. * indicates loadings used to identify the
components.
Components (% explained = 70.18)
Variables I (32.40%) II (18.99%) III (18.78%)
Water temperature (log10) 0.377 0.618* -0.309
Turbidity (log10) -0.777* -0.090 -0.012
Salinity (log10) 0.831* 0.013 -0.121
Mean plant-stem density (log10) -0.150 0.861* 0.199
Mean bank slope (log10) 0.061 0.050 0.920*
Mean water depth (log10) 0.694* -0.070 0.362
Southeastern Naturalist
423
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
Figure 3. 3-D plots of (A) PC I vs.
PC II, (B) PC I vs. PC III, and (C) PC
II vs. Pc III versus the mean CPUE
of Saltmarsh Topminnow (n = 490)
based on collections with 674 Breder
traps fished across coastal Mississippi
environments. H = high loadings, L
= low loadings, WD = mean water
depth, BS = mean bank slope, Sal
= salinity, Turb= turbidity, Temp =
water temperature, and plant stem =
mean plant-stem density.
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
424
Discussion
Results of this study suggested a link between the landscape/geomorphic and
seasonal/spatial environmental conditions (PCA axes) that characterized the habitat
types with the highest Saltmarsh Topminnow CPUE, and also suggested that
multiple environmental factors influence habitat use by this species. Water depth,
bank slope and plant-stem density are inter-related within marsh environments and
have been shown to influence marsh access to a number of nekton species (Johnston
and Sheaves 2007, Lang et al. 2012, McIvor and Rozas 1996, McIvor et al.
1989). For example, a number of other studies have illustrated that water depth
and marsh-edge type (i.e., depositional or erosional) influence access and use of
low, intermediate, and high marsh by fishes (Ennis and Peterson 2015, McIvor and
Odum 1988, McIvor and Rozas 1996, McIvor et al. 1989, Meyer and Posey 2009).
Peterson et al. (2003) reported that Saltmarsh Topminnow abundance was highest
at water depths of ~50 cm (depositional marsh edges) in main-channel marsh-edge
habitats but mainly in low-salinity areas (≤12 psu). Our data are similar to those
presented in Lopez et al. (2011) in that it showed Saltmarsh Topminnow caught in
Breder traps were very common in depths less than 25 cm, but this may be because of how
the traps are placed within the shallow marsh. However, we also recognize that
Saltmarsh Topminnow are collected with seines at low tide when Breder traps are
not as efficient (Fulling et al. 1999, Peterson et al 2013). Finally, coupled with the
Figure 4. Upper-left panel illustrates watersheds (8 digit HUC) where Saltmarsh Topminnow
has been documented. Lower panel illustrates the regional distribution of Saltmarsh
Topminnow based on results of the current project and compiled non-vouchered and
museum records (see Supplemental File 1, available online at https://www.eaglehill.us/
SENAonline/suppl-files/s15-3-S2249-Peterson-s1, and, for BioOne subscribers, at http://
dx.doi.org/10.1656/S2249.s1).
Southeastern Naturalist
425
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
geomorphology relative to marsh access, habitats with high vegetation-biomass or
stem-density provide greater food-availability (food attachment) and refuges from
predation (Dibble et al. 2006, Warfe and Barmuta 2004) when compared to lessdense
or less-complex habitats.
Although we did not collect specifically at different marsh elevations, we hypothesize
that Saltmarsh Topminnow use the intermediate and high marsh when
they are flooded, and our samples were collected as these fish moved into the low
marsh at low tide (sensu Ennis and Peterson 2015, Lopez et al. 2011). In Louisiana
marshes (Rozas and Reed 1993), the euryhaline fundulids Fundulus grandis Baird
and Girard (Gulf Killifish) and Adinia xenica (Jordan and Gilbert) (Diamond Killifish)
were abundant in high-, intermediate-, and low-marsh elevations, but the
Saltmarsh Topminnow was found in only high and intermediate marsh. Peterson
and Turner (1994) found that Saltmarsh Topminnow were more abundant near the
marsh-edge habitat (>3 m from the creek) in Louisiana; however, Ennis and Peterson
(2015) found Saltmarsh Topminnows used micro-topography and especially
rivulets to access shallow interior-marsh habitat in Mississippi. These data support
our hypothesis that the small-bodied Saltmarsh Topminnow uses intermediate to
high marsh and may favor shallow waters for refuge or foraging as do other fundulids,
including Fundulus luciae (Baird) (Spotfin Killifish; Kneib 1984, Yozzo
and Ottman 2003, Yozzo and Smith 1998). Shields and Mayes (1983) in North
Carolina collected Spotfin Killifish most often in high-marsh habitat dominated by
either Black Needlerush or Spartina patens (Ait.) Muhl. (Salt Meadow Cordgrass),
whereas Able et al. (1983) collected Spotfin Killifish in New Jersey’s high marsh
mainly from areas dominated by the short form of Smooth Cordgrass or Salt Meadow
Cordgrass. In fact, Peterson et al. (2003), who only sampled in main-channel
marsh-edge habitat, collected fewer Saltmarsh Topminnow than Lopez et al. (2011)
and the present study, where the focus was collecting in small dendritic creeks off
of main channels. Finally, Lang et al. (2012) noted that the oocyte composition
of ovaries of Saltmarsh Topminnow suggested spawns occur over multiple days
around the time of spring tides both within a population and on the individual level.
Our findings contribute to a better understanding of the importance of linkages of
intertidal-saltmarsh habitat to Saltmarsh Topminnow because spawning intensity
appears to increase with tidal height and marsh inundation (Lang et al. 2012). Our
work also suggests that small creeks are important vectors for marsh access by
Saltmarsh Topminnow and supports the value of the dendritic nature of salt marshes
to marsh residents (Ennis and Peterson 2015; Kneib 2000, 2003; Lopez et al. 2010,
2011; Meyer and Posey 2009).
It has been shown that abiotic factors such as water temperature, salinity,
and turbidity can initiate movements, drive distribution and abundance patterns,
and affect the foraging ecology of a species (Fulford et al. 2011, Peterson et
al. 2004). The environmental conditions where Saltmarsh Topminnow was
most abundant represent a subset of available conditions across the geographic
range we sampled. These findings support earlier work from other geographically
smaller-scale Saltmarsh Topminnow collections in Mississippi and Alabama
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
426
(Ennis and Peterson 2015, Fulling et al. 1999, Peterson et al. 2003), other general
sources across its range (Boschung and Mayden 2004, Gilbert and Relyea 1992,
Guillen et al. 2015, Peterson and Ross 1991, Ross 2001) and the spatially wideranging
study of Lopez et al. (2011). This more-limited range of environmental
conditions described herein appears to be preferred by Saltmarsh Topminnow
within systems as compared to other resident fundulids. Thus, these habitats must
be protected if the Saltmarsh Topminnow is to remain extant. These saltmarshes
are in coastal regions where increasing urbanization, land subsidence, and projected
sea-level rise threaten this species and its restricted habitat.
Our study documented the distribution and abundance of Saltmarsh Topminnow
across the coastal systems of Mississippi in most systems sampled. Landscape/geomorphic
and seasonal/spatial axes of physical–chemical variables showed a narrow
distribution within these systems for Saltmarsh Topminnow as compared to other
resident fundulids (sensu Lopez et al. 2011). We collected more individuals of Saltmarsh
Topminnow during the spring and summer months with increased numbers
of juveniles; it has been suggested that seasonal abiotic cues such as water temperature,
salinity, and turbidity may be influencing life-history traits like reproduction
and spawning (Lang et al. 2012, Lopez et al. 2011). Our results suggest other factors
such as water depth, bank slope, and stem density (landscape/geomorphic characters)
influence CPUE and distribution when nested within the seasonal/spatial axis.
Our data present the current knowledge of the distribution of the rare Saltmarsh
Topminnow across most of its known range. The species has a fairly contiguous
distribution from eastern localities in Florida, through Alabama and Mississippi.
However, Saltmarsh Topminnow is patchily distributed from Lake Pontchartrain to
Galveston Bay. Four potential regions for future studies which should include systematic
sampling are: (1) Lake Borgne to the Mississippi River Delta (Louisiana),
(2) Atchafalaya/Vermillion bays to Sabine Lake (Louisiana), (3) Sabine Lake to
Galveston Bay (Louisiana and Texas), and (4) west of Galveston Bay (Texas). Additional
sampling should be focused in regions that have been minimally sampled
within coastal habitats based on the Saltmarsh Topminnow’s closely linked habitat
requirements within a reduced salinity range. Findings from these studies will also
support state and federal conservation initiatives.
Acknowledgments
We thank J.D. Lopez, P. Grammer, M. Lowe, J.-M. Havrylkoff, M. Andres,
and S. Manning for all their help on this project. Sara LeCroy provided records
from the GCRL Museum, and S.T. Ross provided access to the compiled database
from the Inland Fishes of Mississippi. George Guillen and John Knight
provided non-vouchered collection records from their regions. This Project was
funded by the US Fish and Wildlife Service through State Wildlife Grant T-7-1
funds administered through the Mississippi Department of Wildlife, Fisheries,
and Parks. This research was approved and performed under the University of
Southern Mississippi IACUC protocol # 10040802.
Southeastern Naturalist
427
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
Literature Cited
Able, K.W., C.W. Talbot, and J.K. Shisler. 1983. The Spotfin Killifish, Fundulus luciae, is
common in New Jersey salt marshes. Bulletin of the New Jersey Academy of Sciences
28:7–10.
Alford, J.B., M.S. Peterson, and C. Green (Eds.). 2014. Impacts of Oil Spill Disasters on
Marine Habitats and Fisheries in North America. CRC Press, Boca Raton, FL. 320 pp.
Bailey, R.M., W.E. Winn, and C.L. Smith. 1954. Fishes from the Escambia River, Alabama
and Florida, with ecologic and taxonomic notes. Proceedings of the Natural Sciences of
Philadelphia 106:109–164.
Boschung, H.T., Jr., and R.L. Mayden. 2004. Fishes of Alabama. Smithsonian
Books,Washington, DC. 736 pp.
Breder, C.M., Jr. 1960. Design a fry trap. Zoologica 45:155–160.
Bulleri, F., and M.G. Chapman. 2010. The introduction of coastal infrastructure as a driver
of change in marine environments. Journal of Applied Ecology 47:26–35.
Chapman, M.G., and A.J. Underwood. 2011. Evaluation of ecological engineering of “armoured”
shorelines to improve their value as habitat. Journal of Experimental Marine
Biology and Ecology 400:301–313.
Crossett, K.M, T.J. Culliton, P.C. Wiley, and T.R. Goodspeed. 2004. Population trends
along the coastal United States: 1980–2008. NOAA, Coastal Trends Report Series,
Washington, DC. 47 pp.
Dibble, E.D., S.M. Thomaz, and A.A. Padial. 2006. Spatial complexity measured at a multiscale
in three aquatic plant species. Journal of Freshwater Ecology 21:239–247.
Ennis, B., and M.S. Peterson. 2015. Nekton and macro-crustacean habitat use of Mississippi
micro-tidal saltmarsh landscapes. Estuaries and Coasts 38 :1399–1413.
European Environmental Agency. 2006. The changing face of Europe’s coastal areas.
Publications Office of the European Union, European Environmental Agency Report
6/20006. Luxembourg City, Luxembourg.
Evermann, B.W. 1892. A Report upon investigations made in Texas in 1891. Bulletin of the
US Fisheries Commission 11:61–90.
Field, A. 2013. Discovering Statistics Using IBM SPSS Statistics. 4th Edition. Sage Publications,
Los Angeles, CA. 915 pp.
Fulford, R.S. M.S. Peterson, and P.O. Grammer. 2011. An ecological model of the habitat
mosaic in estuarine nursery areas: Part I—Interaction of dispersal theory and habitat variability
in describing juvenile fish distributions. Ecological Modelling 222:3203–3215.
Fulford, R.S., M.S. Peterson, W. Wu, and P.O. Grammer. 2014. An ecological model of the
habitat mosaic in estuarine nursery areas: Part II. Projecting effects of sea-level rise on
fish production. Ecological Modeling 273:96–108.
Fulling, G.L., M.S. Peterson, and G.J. Crego. 1999. Comparison of Breder traps and seines
used to sample marsh nekton. Estuaries 22:224–230.
Gilbert, C.R., and K. Relyea 1992. Saltmarsh Topminnow, Fundulus jenkinsi. Pp. 68–72, In
C.R. Gilbert (Ed.). Rare and Endangered Biota of Florida, Volume II, Fishes. University
Press of Florida, Gainesville, FL. 247 pp.
Guillen, G., J. Roberston, J. Oakley, and S. Curtis. 2015. Distribution, abundance, and
habitat use of the Saltmarsh Topminnow (Fundulus jenkinsi). EIH Final Report 15-002
submitted to the Texas Parks and Wildlife, Dickinson, TX. 61 pp.
Hair, J.F., Jr., R.E. Anderson, R.L. Tatham, and B.J. Grablowsky. 1984. Multivariate Data
Analysis with Readings. Macmillan Publishing Company, New York, NY. 360 pp.
Hendrickson, D.A., and A.E. Cohen. 2015. Fishes of Texas Project Database (Version 2.0.
doi:10.17603/C3WC70. Accessed 17 February 2016.
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
428
Johnston, R., and M. Sheaves. 2007. Small fish and crustaceans demonstrate a preference
for particular small-scale habitats when mangrove forests are not accessible. Journal of
Experimental Marine Biology and Ecology 353:164–179.
Kneib, R.T. 1984. Patterns in the utilization of intertidal salt marsh by larvae and juveniles
of Fundulus heteroclitus (L.) and Fundulus luciae (Baird). Journal of Experimental Marine
Biology and Ecology 83:41–51.
Kneib, R.T. 2000. Saltmarsh ecoscapes and production transfers by estuarine nekton in the
southeastern US. Pp. 267–292, In M.P. Weinstein and D.A. Kreeger (Eds.). Concepts and
Controversies in Tidal Marsh Ecology. Kluwer Academic, Dordrecht, The Netherlands.
875 pp.
Kneib, R.T. 2003. Bioenergetics and landscape considerations for scaling expectations
of nekton production from intertidal marshes. Marine Ecology Progress Series
264:279–296.
Lang, E.T., N.J. Brown-Peterson, M.S. Peterson, and W.T. Slack. 2012. Seasonal and tidally
driven reproductive patterns in the Saltmarsh Topminnow, Fundulus jenkinsi (Evermann
1892). Copeia 2012(3):451–459.
Lopez, J.D., M.S. Peterson, E.T. Lang, and A.M. Charbonnet. 2010. Linking habitat and life
history for conservation of the rare Saltmarsh Topminnow (Fundulus jenkinsi): Morphometrics,
reproduction, and trophic ecology. Endangered Species Research 12:141–155.
Lopez, J.D., M.S. Peterson, J. Walker, G.L. Grammer, and M.S. Woodrey. 2011. Distribution,
abundance, and habitat characterization of the Saltmarsh Topminnow, Fundulus
jenkinsi (Evermann 1892). Estuaries and Coasts 34:148–158.
Lowe, M.R., and M.S. Peterson. 2014. Effects of coastal urbanization on salt marsh faunal
assemblages in the northern Gulf of Mexico. Marine and Coastal Fisheries: Dynamics,
Management, and Ecosystem Science 6:89–107.
Lowe, M.R., and M.S. Peterson. 2015. Relative condition and foraging patterns of nekton
from salt-marsh habitats arrayed along a gradient of urbanization. Estuaries and Coasts
38:800–812.
Martin, F.D., A.E. Cohen, and D.A. Hendrickson. 2012. Using the fishes of Texas project
database and recent collections to detect range expansions by four fish species on the
lower coastal plain of Texas. Gulf and Caribbean Research 24:63–72.
Meyer, D.L., and M.H. Posey. 2009. Effects of life-history strategy on fish distribution
and use of estuarine salt-marsh and shallow-water flat habitats. Estuaries and Coasts
32:797–812.
McIvor, C.C., and W.E. Odum. 1988. Food, predation risk, and microhabitat selection in a
marsh-fish assemblage. Ecology 69:1341–1351.
McIvor, C.C., and L.P. Rozas. 1996. Direct nekton use of intertidal saltmarsh habitat and
linkages with adjacent habitats: A review from the southeastern United States. Pp.
311–334, In K.F. Nordstrom and C.T. Roman (Eds.). Estuarine Shores: Evolution, Environments
and Human Alterations. John Wiley and Sons, Ltd., New York, NY. 486 pp.
McIvor, C.C., L.P. Rozas, and W.E. Odum. 1989. Use of the marsh surface by fishes in
tidal freshwater wetlands. Pp. 311–334, In R.R. Sharitz and J.W. Gibbons (Eds.). Proceedings
of a Symposium on Freshwater Wetlands and Wildlife, CONF-8603101, DOE
Symposium Series No. 61, USDOE Office of Scientific and Technical Information, Oak
Ridge, TN.
Meyer-Arendt, K., S.M. Oivanki, and B. Yassin. 1998. Wetland changes in coastal Mississippi,
1950–1992. Pp. 377–401, In L.A. Klein, M. Landry, and J.E. Seward (Eds.).
Marine Resources and History of the Mississippi Gulf Coast, Volume II. Mississippi
Department of Marine Resources, Biloxi, MS. 502 pp.
Southeastern Naturalist
429
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
Mississippi Museum of Natural Science (MMNS). 2015. Mississippi’s comprehensive
wildlife conservation strategy. Mississippi Department of Wildlife, Fisheries, and Parks,
Jackson, MS. Available online at https://www.mdwfp.com/seek-study/state-wildlifeaction-
plan.aspx. Accessed 29 March 2016.
Peterson, G.W., and R.G. Turner. 1994. The value of salt-marsh edge vs. interior as a habitat
for fish and decapod crustaceans in a Louisiana tidal marsh. Est uaries 17:235–262.
Peterson, M.S. 2003. A conceptual view of environment–habitat production linkages in
tidal river estuaries. Reviews in Fisheries Science 11:291–313.
Peterson, M.S., and M.R. Lowe. 2009. Implications of cumulative impacts to estuarine and
marine-habitat quality for fish and invertebrate resources. Reviews in Fisheries Science
17:505–523.
Peterson, M.S., and S.T. Ross. 1991. Dynamics of littoral fishes and decapods along a
coastal river–estuarine gradient. Estuarine, Coastal, and Shelf Science 33:467–483.
Peterson, M.S., and S.J. VanderKooy. 1997. Distribution, habitat characterization, and aspects
of reproduction of a peripheral population of Bluespotted Sunfish, Enneacanthus
gloriosus (Holbrook). Journal of Freshwater Ecology 12:151–161
Peterson, M.S., G.L. Fulling, and C.M. Woodley. 2003. Status and habitat characteristics
of the Saltmarsh Topminnow, Fundulus jenkinsi (Evermann), in eastern Mississippi and
western Alabama coastal bayous. Gulf and Caribbean Research 15:51–59.
Peterson, M.S., M.R. Weber, M.L. Partyka, and S.T. Ross. 2007. Integrating in situ quantitative
geographic-information tools and size-specific laboratory-based growth zones in
a dynamic river-mouth estuary. Aquatic Conservation: Marine and Freshwater Ecosystems
17:602–618.
Peterson, M.S., B.H. Comyns, C.F. Rakocinski, and G.L. Fulling. 2004. Defining the fundamental
physiological niche of young estuarine fishes and its relationship to understanding
distribution, vital metrics, and optimal nursery conditions. Environmental Biology
of Fishes 71:143–149.
Ross, S.T. 2001. Inland Fishes of Mississippi. University of Mississippi Press, Jackson,
MS. 624 pp.
Rozas, L.P., and D.J. Reed. 1993. Nekton use of marsh-surface habitats in Louisiana
(USA) deltaic salt marshes undergoing submergence. Marine Ecology Progress Series
96:147–157.
Sabaj Pérez, M.H. (Ed.). 2010. Standard symbolic codes for institutional resource collections
in herpetology and ichthyology: An online reference. Version 1.5. American Society
of Ichthyologists and Herpetologists, Washington, DC. Available online at http://
www.asih.org/. Accessed 15 May 2016.
Shields, M.A., and C.H. Mayes. 1983. Occurrence and habitat preference of Fundulus
luciae (Baird) (Pisces: Cyprinodontidae) of a southeastern North Carolina salt marsh.
Brimleyana 9:141–144.
Suttkus, R.D., D.A. White, and M.F. Mettee. 1999. Atlas of fishes of the St. Louis Bay
drainage in southern Mississippi. Gulf and Caribbean Research 11(Supplement
1):S1–S91.
Texas Parks and Wildlife Department (TPWD). 2005. Texas Comprehensive Wildlife Conservation
Strategy. Austin, TX. Available online at http://tpwd.texas.gov/publications/
pwdpubs/pwd_pl_w7000_1187a/. Accessed 31 March 2016.
TPWD. 2011. Texas conservation action plan: Key to conservation status and listing ranks.
Austin, TX. Available online at http://www.tpwd.state.tx.us/landwater/land/tcap/documents/
species_key_tcap_2011.pdf. AAccessed 31 March 2016.
Southeastern Naturalist
M.S. Peterson, W.T. Slack, and E.T. Lang
2016 Vol. 15, No. 3
430
TPWD. 2012. Texas Conservation Action Plan 2012–2016: Gulf Coast Prairies and Marshes
Handbook. Austin, TX. Available online at https://tpwd.texas.gov/landwater/land/tcap/
documents/gcpm_tcap_2012.pdf. Accessed on 31 March 2016.
Thompson, B.A. 1980. Fundulus jenkinsi (Evermann), Saltmarsh Topminnow. P. 518, In
D.S. Lee, C.R. Gilbert, C.H. Hocutt, R.E. Jenkins, D.E. McAllister, and J.R. Stauffer Jr.
(Eds.). Atlas of North American Freshwater Fishes. NC State Museum of Natural History,
Raleigh, NC. 854 pp.
Warfe, D.M., and L.A. Barmuta. 2004. Habitat structural complexity mediates the foraging
success of multiple predator species. Oecologia 141:171–178.
Yozzo, D.J., and F. Ottman. 2003. New distribution records for the Spotfin Killifish, Fundulus
luciae (Baird), in the lower Hudson River estuary and adjacent waters. Northeastern
Naturalist 10:399–408.
Yozzo, D.J., and D.E. Smith. 1998. Composition and abundance of resident marsh-surface\
nekton: Comparison between tidal freshwater and salt marshes in Virginia, USA. Hydrobiologia
362:9–19.