Spatial, Temporal, and Interspecific Differences in Composition of Stable Isotopes in Fishes in Maryland Coastal Bays

Abstract

Carbon (δ¹³C) and nitrogen (δ¹⁵N) isotopes were used to evaluate spatial, temporal, and interspecific differences in trophic relationships of four fish species (Paralichthys dentatus, Anchoa mitchilli, Leiostomus xanthurus, and Bairdiella chrysoura) in Maryland’s coastal bays. The δ¹³C values for all species were more enriched in 2017 than in 2018, a year of higher-than-average rainfall that likely caused higher amounts of terrestrial carbon to enter the estuary. There were significant differences among species in the δ¹³C values, with L. xanthurus being the least depleted (−17.2‰ in 2017; −18.8‰ in 2018). Spatially, the δ¹³C values of the species, particularly P. dentatus and B. chrysoura, were more depleted in the northern bays, which have a higher nutrient content and receive more freshwater inflow directly from tributaries, than the southern bays. The observed δ¹³C values (−19.5 ± 0.2‰ to –17.2 ± 0.3‰), however, indicate that marine phytoplankton was the primary carbon source of the fishes. Overall, A. mitchilli was the most enriched in δ¹⁵N (13.0‰), and L. xanthurus was the most depleted (10.2‰). δ¹⁵N was more enriched in fish from the more human-impacted northern bays than in fish from the southern bays, though this might also have stemmed from the differences in the diet composition of the species in the northern and southern bays. A. mitchilli had the highest trophic level, while L. xanthurus and P. dentatus had the lowest trophic levels. Niche breadth was widest in L. xanthurus compared to the other fish species, suggesting a higher variability in diets among L. xanthurus individuals, leading to specialized diets. There was a high niche overlap between B. chrysoura, A. mitchilli, and L. xanthurus, which indicates they fed on similar prey resources.

1. Introduction

Stable isotopes are useful for analyzing food web structure because they provide time and space integration into trophic relationships among organisms [1]. Stable isotope analysis can be used to determine both source and trophic level information. In particular, carbon and sulfur isotopes are used to determine source information, and nitrogen isotopes are used to determine an organism’s trophic position [2,3]. In a predator, the heavier isotope, ¹⁵N, is enriched by 3–5‰ compared to its diet [2,3,4]. The ¹⁵N enrichment in animals compared to their diet mainly occurs from fractionation during deamination and transamination that transfers nitrogen from amino acids to urea for excretion of the isotopically lighter nitrogen in the urine [2,4,5,6]. Due to nitrogen successively increasing with trophic level, δ¹⁵N is useful for determining an organism’s trophic position. Stable isotopes of C and N have been widely used to examine trophic relationships in fishes and show that the diet composition of a species may vary spatially between and within estuaries, ontogenetically, and between years [7,8]. However, little information exists on the trophic relationships of fishes in Maryland coastal bays (MCBs), and the few studies [9,10] were based on traditional gut content analysis.

MCBs provide food resources and spawning areas for many fish species, as well as nursery habitats for young-of-year fish [11,12,13]. The four fish species (Paralichthys dentatus, Anchoa mitchilli, Leiostomus xanthurus, and Bairdiella chrysoura) used in this study are among the most abundant species in MCBs. In 2018, A. mitchilli, B. chrysoura, L. xanthurus, and P. dentatus ranked 3rd, 4th, 5th, and 18th out of 74 species in overall finfish abundances [14]. These fish species are economically and ecologically important.

A. mitchilli are pelagic and zooplanktivorous. The larvae generally prey on copepod nauplii and rotifers; the juveniles prey on copepods, crab zoea, and other zooplankton; whereas the adult diet includes crustaceans such as mysids and ostracods in addition to small fish and small benthic mollusks [15].

Juvenile B. chrysoura tend to be pelagic, feeding mainly on zooplankton [16]. B. chrysoura show ontogenetic diet shifts; as they increase in size, they consume larger prey that are more energetically valuable. Their primary diet consists of calanoid copepods and mysid shrimp, with mysids being the predominant prey of choice as B. chrysoura increase in size [17].

L. xanthurus larvae act as selective plankton feeders that consume ciliates, invertebrate eggs, and copepod nauplii. L. xanthurus juveniles and adults are benthic feeders that consume ostracods, amphipods, mysids, gastropods, nematodes, and polychaetes [18].

Due to P. dentatus searching, stalking, active eye motion, and visual fixation on prey, they are primarily considered visual feeders on prey found on the bottom and in the water column [19]. Studies have also shown that P. dentatus are opportunistic feeders [20]. At the juvenile stage, P. dentatus tend to feed on mysid shrimp and small fish, and when they reach adulthood, they shift their diet to decapod crustaceans and larger fish [20,21].

In this study, we used δ¹³C and δ¹⁵N stable isotope values to compare the feeding ecology of A. mitchilli, B. chrysoura, L. xanthurus, and P. dentatus between the northern and southern MCBs and between 2018, a year of higher-than-normal precipitation, and 2017, a year of relatively lower precipitation [22]. We hypothesized that the δ¹³C of the fish species would be more depleted in the more eutrophic northern part of the MCBs, which receives freshwater from St. Martin River, than in the southern bays and more depleted in 2018 than in 2017.

2. Materials and Methods

2.1. Study Area

MCBs occupy an area of 453.2 km² that includes Berlin, Ocean City, Snow Hill, and Pocomoke (Figure 1). They consist of more than 117,000 acres of land, 71,000 acres of water, and 450.6 km of shoreline [12]. The system has a uniform depth < 10 ft [23] and is divided into the northern coastal bays that include Isle of Wight Bay, Assawoman Bay, and St. Martin River and the southern bays including Sinepuxent Bay, Newport Bay, and Chincoteague Bay. The bays connect to the Atlantic Ocean through the Ocean City inlet in the north and the Chincoteague inlet in the south [23].

Groundwater is the main pathway through which freshwater and nutrients enter the system, while winds and tidal exchange at the Ocean City and Chincoteague inlets regulate the system [23]. Flushing times of the bays are slow, causing the system to accumulate contaminants and nutrients [24].

Land use for the coastal bays includes agricultural land or cropland, forestry, pasture, and urban area [25]. The primary economic operations include intensive poultry production, crop production (corn, soybeans, and barley), fishing, and tourism [25]. Land use in the northern bays is predominantly residential and urban development, leading to more anthropogenic eutrophication from wastewater [13]. The southern bays have more agricultural nitrogen inputs [13]. Freshwater inflow directly into the MCBs is higher in the northern bays because of the location of the St. Martin River. Phytoplankton biomass, macroalgae abundance and the associated amphipods, as well as fish abundance, are higher in the northern than southern bays [13,26,27].

2.2. Sample Collection

Samples of P. dentatus, A. mitchilli, L. xanthurus, and B. chrysoura were collected by a beam trawl at 13 sites in MCBs (Figure 1) from April to October in 2017 and 2018. The fish specimens were kept on ice until arrival in the laboratory, where they were stored in a −20 °C freezer until analyzed. The length (cm) and weight (g) measurements for each individual fish were obtained before muscle tissues were removed from the side of the fish between the dorsal and ventral fin. All the fish sampled were young-of-year (YoY) fish except A. mitchilli, which consisted of YoY and age-2 fish. The fish were classified as YoY by the following: P. dentatus, 1.3–16 cm TL [28]; L. xanthurus, <17.0 cm TL [29]; and B. chrysoura <9.6 cm TL [30]. A. mitchilli YoY were <6.5 cm, and age-2 were 6.5–7.9 cm TL [31].

2.3. Stable Isotope Analyses

Freeze-dried and homogenized muscle tissues (1.2–1.3 mg) were placed in tin capsules and sent to the Stable Isotope Facility, University of California Davis, for analysis of carbon and nitrogen isotopes. Samples were analyzed with a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ Europa 20–20 isotope ratio mass spectrometer (Sercon Ltd., Cheshire, UK). The isotopic composition (¹³C and ¹⁵N) was expressed in parts per thousand deviations from an internationally accepted standard:

$$\delta \left(‰\right)=\left(\frac{R_{sample}}{R_{standard}}-1\right)\times 1000$$

where R is the ratio of heavy-to-light isotope, $R_{sample}$ is the ratio in the sample, and $R_{standard}$ is the ratio in the standard. A positive delta (δ) signifies that the sample has more of the heavy isotope than the standard, while a negative δ signifies the sample has less of the heavy isotope compared to the standard [32].

2.4. Data Analyses

The δ¹³C and δ¹⁵N values were reported as averages for each year, season, and sampling location. The δ¹⁵N values for each fish species were used to calculate trophic levels (TL) according to the following formula [3]:

$$TL=\lambda +\left({\delta }^{15}{N}_{consumer}-{\delta }^{15}{N}_{base}\right)/\mathsf{\Delta }n$$

where λ is the trophic position of the organism used to estimate δ¹⁵$N_{base}$ (1 was used as the trophic position of suspended particulate organic matter (SPOM), the base organism in this study), δ¹⁵$N_{consumer}$ is the mean of δ¹⁵N values for each fish species, and δ¹⁵$N_{base}$ is the mean of δ¹⁵N values (~6.0‰) for SPOM collected in summer and fall 2016 from the bays; no significant differences were observed in the mean values from the northern and southern bays [33]. Finally, $\mathsf{\Delta }n$ is the trophic fractionation per trophic level, and 3.4‰ is the typical average fractionation per trophic level [3,34,35]. No lipid correction was applied to the samples, as the mean values of the C:N ratios from the results of the mass spectrometry generally did not exceed 3.5 [36].

The trophic niche sizes of the four fish species based on their δ¹³C and δ¹⁵N values were determined using SIBER within the stable isotope package analysis in R (SIAR; v. 4.2). The standard ellipse area (SEA) and corrected standard ellipse area (SEAc) for small sample sizes were used for comparison of niche size. The percentage of ellipse overlap was also calculated by the ratio between the area of overlap between paired ellipses of species and each individual ellipse area for a species [37]. When the overlap percentage between two species was >60%, it was considered significant [38,39].

Multivariate statistical analyses were conducted using Sigma Plot. Kruskal–Wallis analysis was used to determine if there were intraspecific and interspecific differences in stable isotope values, and Dunn’s pairwise test was used to determine which groups were different. The Mann–Whitney U test was used to determine if there were differences between years and between the northern and southern MCBs in stable isotope values, as well as to compare fish lengths between years and between northern and southern MCBs.

3. Results

A summary of data on the fish used in the diet study is presented in

Table 1. A summary of the number (n) and total length (mean, minimum, maximum) of each fish species used in the stable isotope analyses of diet of fishes in Maryland coastal bays (a dash indicates there were either no length data or not enough data for that year and/or location).

	2017 (April–October)	2018 (June–October)
	Mean ± SE (n); Range (cm)	Mean ± SE (n); Range (cm)
P. dentatus	11.9 ± 0.5 (141); 5.5–39.5	11.7 ± 0.4 (90); 7.2–33.0
A. mitchilli	7.3 ± 0.3 (16); 6.1–12.1	6.2 ± 0.2 (44); 4.0–9.9
L. xanthurus	11.4 ± 0.4 (41); 7.1–16.5	13.7 ± 0.5 (21); 8.5–18.3
B. chrysoura	8.4 ± 0.2 (53); 5.5–11.2	8.7 ± 0.2 (50); 4.5–15.0
	2017 (North MCBs)	2017 (South MCBs)
	Mean ± SE (n); Range (cm)	Mean ± SE (n); Range (cm)
P. dentatus	12.0 ± 0.8 (51); 7.7–39.5	11.9 ± 0.6 (90); 5.5–35.0
A. mitchilli	-	7.3 ± 0.3 (16); 6.1–12.1
L. xanthurus	13.6 ± 0.4 (13); 12.0–16.5	10.4 ± 0.4 (28); 7.1–16.3
B. chrysoura	8.5 ± 0.3 (24); 5.9–11.2	8.3 ± 0.2 (29); 5.5–10.5
	2018 (North MCBs)	2018 (South MCBs)
	Mean ± SE (n); Range (cm)	Mean ± SE (n); Range (cm)
P. dentatus	12.0 ± 1.2 (21); 7.6–33.0	11.5 ± 0.4 (69); 7.2–32.4
A. mitchilli	6.5 ± 0.2 (20); 4.0–8.3	5.9 ± 0.3 (24); 4.0–9.9
L. xanthurus	13.4 ± 0.7 (11); 11.5–18.3	13.9 ± 0.7 (10); 8.5–15.7
B. chrysoura	8.6 ± 0.3 (41); 4.5–15.0	9.3 ± 0.8 (9); 6.0–13.9

for each species, including the mean total length, minimum and maximum length, and number of fish analyzed each season and year. A. mitchilli and L. xanthurus lengths (cm) were significantly different between 2017 and 2018 (Mann–Whitney, p < 0.05). A. mitchilli were larger in 2017 (7.3 ± 0.3SE) than in 2018 (6.2 ± 0.2SE), while L. xanthurus were larger in 2018 (13.7 ± 0.5SE) than in 2017 (11.4 ± 0.4SE). No significant differences were observed between 2017 and 2018 in the mean lengths of P. dentatus and B. chrysoura (p > 0.05). There was also a significant difference between the northern and southern bays in the mean length of L. xanthurus collected in 2017 (Mann–Whitney, p < 0.001), with fish collected in the northern MCBs being larger, but no differences were observed in 2018. For the other species, no significant differences were observed between the northern and southern bays in the mean length of the fish (p > 0.05).

3.1. Interspecific Differences in Isotopic Compositions of the Four Fish Species in MCBs

Mean values of δ¹³C and δ¹⁵N in the years 2017 and 2018 for P. dentatus, A. mitchilli, L. xanthurus, and B. chrysoura are shown in

Table 2. Mean values ± standard error of δ¹³C and δ¹⁵N values in fish species from MCBs collected in 2017 and 2018. δ¹³C and δ¹⁵N isotopic values of fish species with similar letters within each year are not significantly different (p > 0.05).

	2017	2017	2018	2018
	δ13C	δ15N	δ13C	δ15N
P. dentatus	−18.7 ± 0.1 a	10.9 ± 0.1 c	−19.1 ± 0.1 a	10.7 ± 0.1 b
A. mitchilli	−18.9 ± 0.3 a	13.0 ± 0.2 a	−19.5 ± 0.2 c	12.5 ± 0.2 a
L. xanthurus	−17.2 ± 0.3 b	10.2 ± 0.2 d	−18.8 ± 0.5 c	12.4 ± 0.2 a
B. chrysoura	−18.0 ± 0.2 a	11.5 ± 0.1 b	−19.4 ± 0.2 b	12.0 ± 0.1 a
p-value	<0.001	<0.001	0.013	<0.001

. There were significant differences among fish species in δ¹³C values in 2017 (Kruskal–Wallis, p < 0.001). Pairwise comparisons revealed that L. xanthurus had higher δ¹³C values than A. mitchilli, B. chrysoura, and P. dentatus (Dunn’s, p < 0.050). In 2018, there were also significant differences (Kruskal–Wallis, p = 0.013) between fish δ¹³C values; pairwise comparison (Dunn’s, p = 0.026) showed that P. dentatus and B. chrysoura were different. The δ¹⁵N values were also significantly different among species (Kruskal–Wallis, p < 0.001) in both years. In 2017, δ¹⁵N values for A. mitchilli were significantly higher than for L. xanthurus, P. dentatus, and B. chrysoura; B. chrysoura were significantly higher than for P. dentatus and L. xanthurus; and P. dentatus and L. xanthurus were significantly different. In 2018, the pairwise test (Dunn’s, p < 0.001) revealed that δ¹⁵N values in L. xanthurus, A. mitchilli, and B. chrysoura were not significantly different but were higher than in P. dentatus.

3.2. Interannual Differences in Isotopic Composition of Each Fish Species in MCBs

The δ¹³C values were significantly higher in 2017 than in 2018 for all four fish species (Mann–Whitney, p < 0.05) (

Table 3. Mean values ± standard error of δ¹³C and δ¹⁵N in fish species from MCBs collected in 2017 and 2018. δ¹³C and δ¹⁵N isotopic values of fish species with similar letters between years are not significantly different (p > 0.05) as in Table 2.

	2017	2018		2017	2018
	δ13C	δ13C	p-Value	δ15N	δ15N	p-Value
P. dentatus	−18.7 ± 0.1 a	−19.1 ± 0.1 b	<0.05	10.9 ± 0.1 a	10.7 ± 0.1 a	>0.05
A. mitchilli	−18.9 ± 0.3 a	−19.5 ± 0.2 b	<0.05	13.0 ± 0.2 a	12.5 ± 0.2 b	<0.05
L. xanthurus	−17.2 ± 0.3 a	−18.8 ± 0.5 b	<0.05	10.2 ± 0.2 a	12.4 ± 0.2 b	<0.05
B. chrysoura	−18.0 ± 0.2 a	−19.5 ± 0.2 b	<0.05	11.5 ± 0.1 a	12.0 ± 0.1 b	<0.05

). A. mitchilli, L. xanthurus, and B. chrysoura also showed annual intraspecific differences in δ¹⁵N values between 2017 and 2018 (Mann–Whitney, p < 0.05), but no difference was observed for P. dentatus (p > 0.05).

Isotope biplots (Figure 2a,b) indicated that in 2017, A. mitchilli was the most enriched in δ¹⁵N (13.0‰) and most depleted in δ¹³C (−18.9‰) compared to the other fish species. In contrast, L. xanthurus was the most depleted in δ¹⁵N (10.2‰) but most enriched in δ¹³C (−17.2‰) compared to the other fish species. P. dentatus and B. chrysoura had intermediate levels of δ¹⁵N and δ¹³C relative to the other species. The biplots in 2018 revealed that A. mitchilli (12.5‰), B. chrysoura (12.0‰), and L. xanthurus (12.4‰) had enriched δ¹⁵N values and relatively depleted δ¹³C values (−19.5 to −18.8‰), whereas P. dentatus had depleted δ¹⁵N (10.7‰) but an intermediate δ¹³C value (−19.1‰).

3.3. Seasonal Differences in Isotopic Composition of Each Fish Species in MCBs

Seasonal comparisons were made only for P. dentatus and B. chrysoura in 2017 and for all species in 2018 because of insufficient sample sizes in spring and summer months in 2017. In 2017, P. dentatus showed no significant difference between spring (−18.4 ± 0.2), summer (−18.6 ± 0.1), and fall (−19.0 ± 0.2) for δ¹³C, while significant differences (Kruskal–Wallis, p = 0.001) were noted for δ¹⁵N in spring (10.8 ± 0.4), summer (11.2 ± 0.1), and fall (10.4 ± 0.2). Pairwise testing revealed that δ¹⁵N values of P. dentatus caught in summer and fall were significantly different (Dunn’s method, p = 0.003). In 2017, B. chrysoura showed significant differences (Kruskal–Wallis, p = 0.001) between summer (−17.8 ± 0.2) and fall (−20.0 ± 0.1) for δ¹³C, as well as for δ¹⁵N in summer (11.4 ± 0.1) and fall (12.7 ± 0.3).

In 2018, no significant differences were observed between summer and fall in δ¹³C content of the fish species examined, except for B. chrysoura, which showed significant differences for δ¹⁵N (Kruskal–Wallis, p = 0.001).

Seasonal biplots (Figure 3a–c) indicated that in summer 2017, B. chrysoura was enriched in δ¹⁵N and had intermediate levels in δ¹³C compared to the other fish species. P. dentatus was enriched in both isotopes, and L. xanthurus was depleted in both isotopes compared to the other fish. In summer 2018, L. xanthurus was enriched in both δ¹⁵N and δ¹³C compared to the other fish species, while A. mitchilli and B. chrysoura were enriched in δ¹⁵N and depleted in δ¹³C. P. dentatus was fairly depleted in both isotopes compared to the other species. Comparing the fish species in fall 2018, A. mitchilli was enriched in δ¹⁵N and had intermediate levels of δ¹³C, L. xanthurus was enriched in δ¹⁵N and depleted in δ¹³C, B. chrysoura had intermediate levels of both isotopes, and P. dentatus had depleted levels of both isotopes.

3.4. Spatial Variation in Stable Isotopes of the Four MCB Fish Species

Mean values of δ¹³C and/or δ¹⁵N in 2017 and 2018 for P. dentatus, A. mitchilli, L. xanthurus, and B. chrysoura differed between the northern and southern bays (

Table 4. Mean ± standard error of δ¹³C and δ¹⁵N values of fish species collected in the northern and southern MCBs in 2017 and 2018 (a dash indicates there were either no isotopic data or not enough data for that location). For each species and year, values in the north and south bays with similar letters were not significantly different.

	2017			2018
	δ13C			δ13C
	North	South	p-Value	North	South	p-Value
P. dentatus	−19.1 ± 0.1 a	−18.4 ± 0.1 b	<0.05	−19.6 ± 0.1 a	−18.9 ± 0.1 b	<0.001
A. mitchilli	-	−18.9 ± 0.3	-	−19.4 ± 0.4 a	−19.6 ± 0.2 a	>0.05
L. xanthurus	−17.4 ± 0.6 a	−17.0 ± 0.3 a	>0.05	−19.3 ± 0.8 a	−18.3 ± 0.7 a	>0.05
B. chrysoura	−18.5 ± 0.3 a	−17.6 ± 0.2 b	<0.05	−19.8 ± 0.2 a	−17.8 ± 0.3 b	<0.001
	δ15N			δ15N
	North	South	p-value	North	South	p-value
P. dentatus	11.4 ± 0.2 a	10.6 ± 0.1 b	<0.05	11.9 ± 0.2 a	10.4 ± 0.1 b	<0.05
A. mitchilli	-	13.0 ± 0.2		12.7 ± 0.3 a	12.3 ± 0.2 a	>0.05
L. xanthurus	11.3 ± 0.2 a	9.7 ± 0.3 b	<0.05	12.9 ± 0.2 a	11.9 ± 0.3 a	>0.05
B. chrysoura	11.8 ± 0.1 a	11.2 ± 0.2 b	<0.05	12.1 ± 0.1 a	11.7 ± 0.1 a	>0.05

). In 2017, the δ¹³C contents of P. dentatus and B. chrysoura were higher in the southern than northern parts of MCBs, but no spatial differences were observed for L. xanthurus. For P. dentatus, B. chrysoura, and L. xanthurus, the δ¹⁵N contents were higher in the northern bays than southern bays (Mann–Whitney, p < 0.05). In 2018, P. dentatus and B. chrysoura δ¹³C contents were higher in fish collected from the southern than northern MCBs, while δ¹⁵N contents for P. dentatus were significantly higher in the northern than southern bays (Mann–Whitney, p < 0.001), but no differences were observed between areas for the other species (Table 4).

Isotopic biplots (Figure 4a–d) indicated that in 2017, the fish had similar enrichment/depletion trends for both the northern and southern MCBs. A. mitchilli was enriched in δ¹⁵N and most depleted in δ¹³C. B. chrysoura had intermediate δ¹³C but the most enriched δ¹⁵N levels. P. dentatus had an intermediate δ¹⁵N level and the most depleted δ¹³C, and L. xanthurus was the least depleted in δ¹⁵N but most enriched in δ¹³C. In 2018, L. xanthurus and A. mitchilli from the northern MCBs were enriched in both δ¹⁵N and δ¹³C compared to the other species, while B. chrysoura and P. dentatus were relatively depleted in both δ¹⁵N and δ¹³C. In the southern MCBs, L. xanthurus and B. chrysoura were enriched in both isotopes, P. dentatus was depleted in both isotopes, and A. mitchilli was enriched in δ¹⁵N but depleted in δ¹³C.

The trophic levels of A. mitchilli, B. chrysoura, L. xanthurus, and P. dentatus were determined to be 3.04, 2.77, 2.55, and 2.51, respectively, based on their δ¹⁵N values. The standard ellipse areas estimated for the species are presented in

Table 5. Standard ellipse area (SEA) and standard ellipse corrected area (SEAc) for the four MCB fish species.

Species	SEA (δ2)	SEAc (δ2)
P. dentatus	4.2	4.3
A. mitchilli	3.8	3.8
L. xanthurus	10.6	10.8
B. chrysoura	4	4

and Figure 5. The SEA and SEAc values for the species are similar. L. xanthurus had the largest SEA/SEAc values compared to the other fish species, while A. mitchilli had the smallest SEA/SEAc values.

The trophic niche overlap between species based on δ¹³C and nitrogen δ¹⁵N ranged from 20.1 to 57.2% (P. dentatus), 22.2 to 76.4% (A. mitchilli), 22.7 to 33.6% (L. xanthurus), and 46.1 to 89.1% (B. chrysoura) (

Table 6. Percentage area overlap between the SEA of the four MCB fish species. (The table should be read horizontally. For example, the ellipse for P. dentatus overlaps the ellipse for A. mitchilli by 20.1%, whereas the ellipse of A. mitchilli overlaps the ellipse of P. dentatus by 22.2%.)

	P. dentatus	A. mitchilli	L. xanthurus	B. chrysoura
P. dentatus		20.1	57.2	49.8
A. mitchilli	22.2		76.4	48.5
L. xanthurus	22.7	27.4		33.6
B. chrysoura	52.3	46.1	89.1

). The percentage of the B. chrysoura ellipse that overlapped L. xanthurus was the highest (89.1%), while the ellipse of P. dentatus that overlapped A. mitchilli was the lowest percentage (20.1%).

4. Discussion

4.1. Interspecific Differences in Stable Isotope Composition of the Four MCB Fish Species

There were significant differences between species in the δ¹³C and δ¹⁵N mean values of the muscle tissue. In 2017, the fish were most enriched to most depleted in δ¹³C as follows: L. xanthurus, B. chrysoura, P. dentatus, and then A. mitchilli, which was similar to the pattern observed in 2018. The higher enrichment of δ¹³C in L. xanthurus than in the pelagic species such as A. mitchilli is probably related to a more benthic feeding that might have caused the addition of a benthic microalgal carbon source, which has δ¹³C values around −17.0‰ [40,41]. Nevertheless, the δ¹³C values overlapped between fish species and indicate that the fish species were dependent on a similar organic matter source [42]. The mean ± SE value of δ¹³C ranged from −18.9 ± 0.5‰ to −17.2 ± 0.3‰ in 2017 and −19.5 ± 0.2‰ to −18.8 ± 0.5‰ in 2018 and suggest that the carbon source for the fish were mainly of marine phytoplankton, which ranges from −24‰ to −18‰ [5,40].

In 2017, the fish were most enriched to most depleted in δ¹⁵N as follows: A. mitchilli, B. chrysoura, P. dentatus, and then L. xanthurus, which is slightly different from what was observed in 2018, when the most enriched to most depleted δ¹⁵N values were A. mitchilli, L. xanthurus, B. chrysoura, and then P. dentatus. A. mitchilli had the highest δ¹⁵N values (13.0 ± 0.2‰ in 2017 and 12.5 ± 0.2‰ in 2018) in both years compared to the other species. In another study, Winemiller et al. [8], found that A. mitchilli had the lowest δ¹⁵N value (12.8 ± 0.8‰) compared to L. xanthurus (13.3 ± 2.0‰) and B. chrysoura (14.7 ± 2.0‰), though the δ¹⁵N value they observed in A mitchilli was comparable to the value we observed in our study in the MCBs. The A. mitchilli used for stable isotope analysis were approximately age-2 fish based on the length values, while the other fish species were age-0 [28,29,30,31]. It is known that δ¹⁵N increases with the size and age of a species [5,42,43].

It was found that in 2017, A. mitchilli was the most enriched in δ¹⁵N and most depleted in δ¹³C compared to the other fish species; L. xanthurus was the most depleted in δ¹⁵N and the most enriched in δ¹³C, while P. dentatus and B. chrysoura had intermediate levels of δ¹⁵N and δ¹³C compared to the other two fish species. In 2018, A. mitchilli, B. chrysoura, and L. xanthurus had enriched δ¹⁵N values and depleted δ¹³C values, and P. dentatus had depleted δ¹⁵N and an intermediate δ¹³C level. When fish have relatively enriched δ¹³C values and depleted δ¹⁵N values, this indicates that they were feeding on a more benthic-based prey; depleted δ¹³C values and enriched δ¹⁵N values indicate that the fish were feeding on more of a planktonic- or pelagic-based prey [6,41]. This suggests that L. xanthurus in 2017 and P. dentatus in 2018 had a more benthic influence in their diet, while A. mitchilli in 2017 and 2018 and B. chrysoura in 2018 had a more planktonic influence on their diet, which is consistent with what is known about their feeding habits [10].

4.2. Temporal Variations in Stable Isotope Composition of the Four MCB Fish Species

The δ¹³C values of the fish were more enriched in 2017 than in 2018. In 2017, L. xanthurus mean δ¹³C value was slightly more enriched and might have had a benthic microalgal carbon source, which has δ¹³C values around −17.0‰ [40,41]. MacKenzie et al. [44] attributed annual variations in δ¹³C to the feeding areas of salmon (Salmo salar) being subjected to interannual variability in primary production. It has also been shown that δ¹³C can be more depleted in wet years compared to dry years because more terrestrially derived carbon is carried by rivers into an estuary in the wet years [45]. There was more precipitation in the year 2018 compared to 2017 in Maryland. The average annual rainfall values in 2017 and 2018 were 104.7cm and 164.1cm, respectively [22]. Since 2018 had more depleted δ¹³C, especially in the northern bays, it is possible that the higher rainfall increased freshwater inputs into the MCBs and affected the carbon signatures of organisms at the base of the food web [33,46] which, in turn, resulted in variations in the fish species’ δ¹³C [32,33,47,48,49,50].

4.3. Spatial Variations in Stable Isotopes of the Four MCB Fish Species

The δ¹³C values in P. dentatus and B. chrysoura collected from the southern bays in both 2017 and 2018 were enriched compared to the individuals collected from the northern bays. In contrast, δ¹⁵N values in P. dentatus, B. chrysoura, and L. xanthurus collected from the northern bays in 2017 were more enriched compared to the individuals collected from the southern bays, whereas only P. dentatus had significantly higher δ¹⁵N in the northern than southern bays in 2018. Thus, the fish that displayed significant differences were more depleted in δ¹³C and enriched in δ¹⁵N when collected from the northern MCBs but enriched in δ¹³C and depleted in δ¹⁵N when collected from the southern MCBs. This spatial difference reflects the differences in the amount of freshwater inflow from MCB tributaries in the northern and southern bays and the extent of anthropogenic impact. St. Martin River, the main tributary that transports water directly into the MCBs, is located in the northern bays and thus carries organic materials from land and river into the bays, thereby influencing the δ¹³C values in the fish. It has been determined that increased anthropogenic nutrients influences productivity, causing increases in isotopic values [51]. The northern MCBs have more anthropogenic eutrophication from residential and urban development compared to the southern MCBs [13,52]. It is also plausible that the observed difference between the northern and southern MCBs was due to different prey assemblages that influenced the diet of the species in the areas. Estrada et al. [53] found isotopic variations in bluefin tuna (Thunnus thynnus), suggesting that the diets of individual bluefin tuna differed and explained by the fact that the bluefin tuna came from areas with different prey assemblages.

In a study on stable isotopes of juvenile fish in MCBs, O’Brien [54] found that A. mitchilli collected in the northern MCBs had δ¹³C values around −20‰, indicating the presence of a more pelagic carbon source. The δ¹³C values (−19.8 ± 0.2‰ to −17.4 ± 0.6‰) for the fish collected from the northern and southern bays in this study are similar to the values reported by O’Brien [54].

4.4. Trophic Levels and Niche Overlap of the Four MCB Fish Species

The trophic levels estimated for A. mitchilli, B. chrysoura, L. xanthurus, and P. dentatus were 3.04, 2.77, 2.55, and 2.51, respectively. The higher trophic level of A. mitchilli relative to the other species may be because they were older and likely included more fish in their diets. Fish size/age has a positive relationship with trophic level [55,56].

The isotopic niche breadth differed between the species, with L. xanthurus having a higher SEA/SEAc than the other fish species, indicating a larger niche, and A. mitchilli having the lowest SEA/SEAc, indicating a smaller niche. The broader niche width of L. xanthurus reveals a wider use of MCBs’ habitat and/or prey sources [57]. The variation in niche widths could be related to species with smaller niche widths having a trophic behavior that is more flexible compared to species with larger niche widths, which include individuals with specialized trophic behavior [58]. Generalist species are believed to have a narrower isotopic niche width than specialist species due to generalists having similar isotopic signatures because they are sampling the same prey items in similar proportions and thus integrate their diet [59,60]. The niche width data from this study suggest a higher variability in the diets among L. xanthurus individuals, resulting in specialized diets [61], compared to individuals belonging to other fish species that were thus, relatively, generalists.

The percentage of the B. chrysoura ellipse that overlapped L. xanthurus, as well as the A. mitchilli ellipse that overlapped L. xanthurus, was >60%, meaning they had significant niche overlaps [38,39]. Greater niche overlaps suggest that the species are feeding on similar prey resources [62,63]. This high dietary overlap between the species can increase resource competition [64].

5. Conclusions

The δ¹³C values of the fish species in both years indicate marine phytoplankton as the dominant source of carbon and that interannual variations in δ¹³C in some species were likely associated with interannual variability in the amount of precipitation and terrestrially derived organic carbon transported via freshwater discharge into the bays. A. mitchilli had the highest δ¹⁵N in both years, probably because the individuals used in the study were ages-1 to 2 fish, while the other fish species were juveniles. Overall, the species were more δ¹⁵N enriched when collected from the northern MCBs than when collected from the southern MCBs, especially in 2017. This was probably due to a higher anthropogenic influence in the northern MCBs and different prey assemblages in the two locations. L. xanthurus had a broader niche breadth than the other fish species, whereas B. chrysoura and A. mitchilli had the highest niche overlap with L. xanthurus; thus, there is a potential for competition between the two species and L. xanthurus. Our results are useful for interpreting food web dynamics in estuaries and underscore the need to take into account spatial and temporal variability of δ¹³C and δ¹⁵N when examining trophic relationships of fishes.