Spatiotemporal Variations in Trophic Diversity of Fish Communities in a Marine Bay Ecosystem Based on Stable Isotope Analysis

Abstract

Climate change has led to significant fluctuations in marine ecosystems. As a component of the food web, the trophic diversity and spatiotemporal changes of fish communities are crucial for understanding ecosystems. In recent years, stable isotope analysis has been increasingly used as a comprehensive tool for quantitative assessment of trophic diversity to explore spatiotemporal variations in fish community diversity. This study is based on carbon ($\delta$ ¹³C) and nitrogen ($\delta$ ¹⁵N) stable isotope analysis using different biomass-weighted isotope diversity indices, including isotopic divergence index (IDiv), isotopic dispersion index (IDis), isotopic evenness index (IEve), and isotopic uniqueness index (IUni). The overall results indicate that IDis, IEve, and IUni values of fish communities were relatively low, while IDiv was relatively high in the Haizhou Bay ecosystem. IDiv, IDis, IEve, and IUni were lower in autumn than in spring; IDiv and IDis were relatively higher in offshore waters, while IEve and IUni were relatively higher in inshore waters. The changes in species composition and intensive pelagic–benthic coupling in Haizhou Bay may lead to significant spatiotemporal variations in the trophic diversity of fish communities in the area. These findings highlight the importance of incorporating trophic relationships into ecosystem models, which will help to enhance our understanding of the complexity of the trophic structure of fish communities.

1. Introduction

Trophic interactions between species in the food web reveal energy flow and matter conversion in ecosystems [1], which play a crucial role in maintaining the trophic structure and function of marine ecosystems [2]. As an important part of marine ecosystems, fish communities have attracted much attention through trophic cascades in the energy flow and material circulation of food webs [3,4,5]. However, factors such as human activities, climate change and environmental pollution have significant impacts on ecosystem processes, community structure and stability [1,6,7]. Analysis of the trophic diversity of fish communities can not only provide important information for management and protection of marine fishery resources [8], but also help to mitigate the adverse effects of environmental degradation and human activities on marine ecosystems and improve the resilience and stability of ecosystems [9,10,11].

Trophic diversity has been used to reflect the differences and diversity of trophic sources and trophic niches of species within a community [12]. The application of trophic diversity can be realized through the use of diversity indices, which can provide important information for monitoring the impact of human disturbance, environmental change or trophic downgrading on fishery resources [13,14]. Although methods developed by functional ecologists to quantify functional diversity could be adapted for stable isotope ecology, assessment of multiple aspects of trophic diversity is lacking [8,15]. Therefore, the proposal of a complementary set of diversity indicators, taking into account different aspects of isotopic diversity, is widely accepted [8,12,16,17,18].

Stable isotope analysis is powerful for quantifying the interactions between organisms and the fluxes of energy in marine ecosystems [19]. It can reflect the ecological information of organisms’ feeding over a relatively long period of time [20], which effectively makes up for the shortcomings of traditional stomach content analysis [21,22]. The similarity of the carbon stable isotope ratio ($\delta$ ¹³C) in the consumer’s body to that of the food organism can provide long-term feeding information and is often used to study the feeding ecology of fish [23]. The nitrogen stable isotope ratio ($\delta$ ¹⁵N) has a 3.00–4.00‰ enrichment effect between consumers and food organisms, which can be used to estimate the trophic position of consumers [24]. In this process, the selection of appropriate baseline organisms is crucial [25]. Post [26] suggested that primary consumers with stable diets should be selected as baseline organisms. Long-lived stationary consumers integrate spatial and temporal variation in stable isotope ratios to reflect a time-averaged isotopic signal, which can provide a good isotopic baseline for estimating the trophic position of higher trophic level consumers in aquatic ecosystems [26]. Filter-feeders, as sedentary consumers, are generally used to track stable isotope baseline variations. This method has been widely used by ecologists as an integrative tool in food web ecology [27,28] to quantify the trophic implications of ecological processes [19,29]. In recent years, it has been used in the analysis of trophic diversity of fish communities [30,31].

Haizhou Bay is an open bay ecosystem in the southern Yellow Sea, which is one of the important fishing grounds and spawning habitats in China Seas [32]. Since the 1980s, the ecological environment of Haizhou Bay has undergone large changes due to the impacts of fishing pressure, habitat degradation, and climate change [33], with fish diversity and fishery resources declining precipitously in this ecosystem [34,35]. In order to quantify the trophic relationships among species, the isotopic diversity of fish communities in Haizhou Bay was evaluated through the analysis of carbon and nitrogen stable isotopes. Four biomass-weighted isotope diversity indices were used to quantify different aspects of isotopic diversity in fish communities, including isotopic divergence index (IDiv), isotopic dispersion index (IDis), isotopic evenness index (IEve), and isotopic uniqueness index (IUni). Based on this, the spatiotemporal variations of trophic structure and trophic diversity of fish communities in Haizhou Bay were analyzed. This study will help to enhance the understanding of the spatiotemporal variations in the trophic diversity of fish communities and advance the understanding of the complexity of the trophic structure of this sea area, which will provide essential information for Ecosystem-based Fisheries Management (EBFM).

2. Materials and Methods

2.1. Study Area

The survey area was in Haizhou Bay in China, which ranges from 34.4° to 35.6° N and 119.4° to 121.1° E (Figure 1). The bottom trawl surveys were conducted in 2018, with stratified random sampling surveying of 18 stations in spring (April) and autumn (September). A 220 kW otter trawler, with a sampling net of 12 m open width and mesh size of the codend at 17 mm, was used in bottom trawl surveys, which towed for about 1 h at a speed of 2–3 knots at each station. The surveys were only conducted diurnally. Catch data were standardized at a tow speed of 2 knots and trawl of 1 h. A detailed description of the survey design is available in Xu et al. [36]. In addition, according to the K-means clustering analysis results of Li et al. [35] on the fishery communities in this area, the stations were roughly divided into inshore (<20 m) and offshore (>20 m) areas according to the 20 m isobath.

2.2. Sample Treatment

The biomass of fish species analyzed in this study accounts for more than 90% of the estimated total biomass of the fish community, providing a representative sample of their trophic diversity in Haizhou Bay. The white muscle near the first dorsal fin of fish samples was used for stable isotope analysis [37,38]. Tissue samples were rinsed with distilled water to remove exogenous materials (e.g., remaining scales or bones), and dried in an oven at 60 °C for 48 h. The tissue sample was ground into a fine powder (with an agate mortar and pestle) and packed into a clean glass scintillation vial. The lipids and inorganic carbon in the sample were extracted by methanol–chloroform–water (volume ratio 10:5:4) and hydrochloride (1 mol L⁻¹), respectively [39,40,41]. Eventually, the samples were frozen and stored in glass vials for further processing. Samples were sent to the Institute of Hydrobiology of the Chinese Academy of Sciences for stable isotope analysis. The ratio of stable isotopes of carbon and nitrogen of species were analyzed by isotope ratio mass spectrometers (Isoprime100 IRMS).

2.3. Stable Isotope Analysis

Stable isotopic values were reported in international standard $\delta$ notation using the following equations:

$$\delta {}^{13}\mathrm{C}\mathrm{or}\delta {}^{15}\mathrm{N}=1000(R_{\mathrm{sample}}-R_{\mathrm{standard}})/R_{\mathrm{standard}}$$

(1)

Results for carbon and nitrogen are reported as $\delta$ ¹³C and $\delta$ ¹⁵N values, respectively, and are derived from the relation between the isotopic value for the sample and a known standard of $\delta$ ¹³C or $\delta$ ¹⁵N, where R corresponds to ratio $\frac{{13}_{\mathrm{C}}}{{12}_{\mathrm{C}}}$ or ratio $\frac{{15}_{\mathrm{N}}}{{14}_{\mathrm{N}}}$. Standards used for $\delta$ ¹³C and $\delta$ ¹⁵N included Vienna Pee Dee Belemnite (VPDB) and atmospheric nitrogen with international standard R values of 11.24‰ and 3.68‰, respectively. In the process of analysis, one working standard is added after five samples to ensure the stability of the instrument and accuracy of the results. The accuracy of $\delta$ ¹³C and $\delta$ ¹⁵N is the ±0.15 and ±0.18, respectively.

In this study, scallops (Azumapecten farreri) collected in Haizhou Bay during the same period were chosen as a realistic trophic baseline in this study, which is widely distributed in the sea area and effectively reflects the stable isotope characteristics of primary producers [42,43]. Based on the baseline organism, the formula for calculating the trophic level using stable isotope data is as follows [44]:

$$T{L}_{consumers}=\frac{(\delta {}^{15}{\mathrm{N}}_{consumers}-\delta {}^{15}{\mathrm{N}}_{baseline})}{TEF}+2$$

(2)

where $T{L}_{consumers}$ and $\delta {}^{15}{\mathrm{N}}_{consumers}$ represent the trophic level and stable nitrogen isotope ratios of fish, respectively. $\delta {}^{15}{\mathrm{N}}_{baseline}$ represents the stable nitrogen isotope ratio of the baseline organisms. In this study, scallop (A. farreri) was selected as the baseline organism, with a mean $\delta$ ¹⁵N of 4.50‰ [45]. This study used a trophic enrichment factor ($TEF$) of 3.40‰ [45,46,47]. The trophic level of the baseline organism is 2.

2.4. Trophic Diversity Indices

Ecologists have developed metrics to quantify the multiple aspects of functional diversity in an n-dimensional space based on functional traits [12]. Methods developed to quantify functional diversity could be adapted for stable isotope ecology [8,15]. In this study, we repurposed some functional diversity metrics based on functional traits into a set of isotopic diversity metrics based on stable isotope values [12]. Four trophic diversity indices were selected to quantify different aspects of isotope diversity in the fish community of Haizhou Bay [8,17,18], including isotopic divergence (IDiv), isotopic dispersion (IDis), isotopic evenness (IEve), and isotopic uniqueness (IUni). These indices are mathematically independent of the number of organisms. In this study, four trophic diversity indices were used to describe the trophic diversity of fish communities in Haizhou Bay by constructing the weights of coordinate points relative to biomass assigned by two-dimensional isotope space [12]. The four trophic diversity indices analyzed in this study include:

(1) Isotopic divergence (IDiv) is derived from the functional divergence index indicating the degree to which a species’ distribution in isotopic space maximizes trophic divergence within the food web [8]. The formula for IDiv index is as follows:

$$IDiv=\frac{\Delta d+\overline{dG}}{\Delta \left|d\right|+\overline{dG}}$$

(3)

where $\Delta d$ is the dispersion weighted by species richness. $\overline{dG}$ is the average distance between species and the center of gravity. IDiv is maximal (i.e., tends to 1) if all the species (or most of their weight) are located on the edges of the convex hull.

(2) Isotopic dispersion (IDis) is calculated by dividing the weighted deviation to the average position of points in the stable isotope space by the maximal distance to the center of gravity [17]. The formula for IDis index is as follows:

$$d{O}_i=\sqrt{{\displaystyle {\sum }_{k=1}^{SI}(\delta k_i-\frac{1}{N}{\displaystyle \sum _{i=1}^N\delta k_i{)}^2}}}$$

(4)

$$IDis={\displaystyle \sum _{i=1}^N{w}_i\times d{O}_i}/\max (d{O}_1,\dots ,d{O}_i,\dots ,d{O}_N)$$

(5)

where $\delta k_i$ is the value for stable isotope k [1, SI] of each species i [1, N] and $w_i$ is the weight of each species i [1, N]. IDis is maximal when most of the points (or their weight) are far from the center of gravity of the group of points.

(3) Isotopic evenness (IEve) is derived from FEve, indicating the evenness of abundance distribution in a multi-dimensional stable isotope space [8]. The formula for IEve index is as follows:

$$IEve=\frac{{\displaystyle {\sum }_{l=1}^{N-1}\min }(\frac{E{W}_l}{{\displaystyle {\sum }_{l=1}^{S-1}E{W}_l}},\frac{1}{N-1})-\frac{1}{N-1}}{1-\frac{1}{N-1}}$$

(6)

where EW is weighted evenness, l is a branch of the minimum spanning tree linking species in the stable isotopic space. IEve increases when abundance is more evenly distributed among species or species are evenly distributed in the stable isotope space.

(4) Isotopic uniqueness (IUni) is defined as the inverse of the average isotopic redundancy which appears in the average closeness of species in the stable isotope space. The formula for IUni index is as follows:

$$NN{D}_i=\underset{i\ne j}{{\displaystyle \min }}(\sqrt{{\displaystyle {\sum }_{k=1}^{SI}(\delta k_i-\delta k_j}{)}^2})$$

(7)

$$IUni={\displaystyle \sum _{i=1}^N{w}_i\times NN{D}_i}/\max (NN{D}_1,\dots ,NN{D}_i,\dots ,NN{D}_N)$$

(8)

where $NN{D}_i$ is the minimal distance of a species to its nearest neighbors. IUni tends to 1 when most of the species (or species with the highest abundance) are isolated in the stable isotope space.

These trophic diversity indices were calculated using the R package’s geometry version 0.4.5 [48], rcdd version 1.5 [49] and ape version 5.6 [50]. In addition, the relative biomass was assigned to a coordinate point by constructing the isotope space, and the structure and trophic diversity of fish communities in Haizhou Bay were analyzed by combining four trophic diversity indices (IDiv, IDis, IEve, and IUni). The t-test was used to determine whether there was significant difference in trophic diversity of the community in different seasons and areas.

3. Results

3.1. Temporal Variation in Trophic Diversity of Fish Communities

The trophic diversity of fish communities in Haizhou Bay was significantly different between seasons (p < 0.05), with the four isotopic diversity indices being higher in spring than those in autumn (Figure 2). In spring, the fish with large biomass mostly concentrated at points near the boundary of the convex hull (irregular polygon), while in autumn they were close to the center of gravity of the convex hull. In addition, the mean range of $\delta$ ¹³C and $\delta$ ¹⁵N values in spring was (−21.4‰, −16.8‰) and (8.5‰, 13.8‰), respectively, while the mean range of $\delta$ ¹³C and $\delta$ ¹⁵N values in autumn was (−21.7‰, −15.6‰) and (8.9‰, 12.3‰), respectively (Table S1). Notably, the fish with the maximum and minimum mean $\delta$ ¹⁵N values were Gadus macrocephalus and Coilia mystus in spring, and Thryssa kammalensis and Syngnathus acus in autumn. The fish with the maximum and minimum mean $\delta$ ¹³C values were Hippocampus trimaculatus and Ammodytes personatus in spring, and M. cinereus and Enedrias nebulosus in autumn.

3.2. Spatial Variation in Trophic Diversity of Fish Communities

The trophic diversity of fish communities in Haizhou Bay was different between inshore and offshore waters (p < 0.05). The results of the four isotope diversity indices showed that IDiv and IDis were higher in offshore waters (>20 m), while IEve and IUni were higher in inshore waters (<20 m). Meanwhile, the four isotope diversity indices showed that inshore waters’ (<20 m) fish communities had higher $\delta$ ¹³C, while offshore waters’ (>20 m) fish communities had higher $\delta$ ¹⁵N (Figure 3).

3.3. Composition and Trophic Diversity of Fish Communities

A total of 54 fish species in Haizhou Bay were captured in 2018 and analyzed for their stable isotope values. The $\delta$ ¹³C values vary more widely than the $\delta$ ¹⁵N values, with the mean $\delta$ ¹³C value range being (−21.7‰, −15.6‰) and the mean $\delta$ ¹⁵N value range being (8.5‰, 13.8‰) (Table S2). The maximum and minimum $\delta$ ¹³C vales were found in Enedrias nebulosus and Muraenesox cinereus, respectively. The maximum and minimum $\delta$ ¹⁵N values were found in Gadus macrocephalus and Coilia mystus, respectively. Detailed $\delta$ ¹³C and $\delta$ ¹⁵N values for other fish species are shown in Table S2. Trophic levels of these fish ranged from 3.19 to 4.73 (Table S2). In addition, the values of four trophic diversity indices of fish communities in Haizhou Bay were IDiv 0.60, IDis 0.33, IEve 0.40, and IUni 0.18, respectively (Figure 4).

4. Discussion

Trophic diversity and trophic redundancy are often linked to stability and resilience to perturbations, because of their roles in promoting the retention of functional roles in ecosystems [51,52]. The trophic structure of fish communities in Haizhou Bay has multiple trophic levels, with trophic positions ranging from 3.19 to 4.73. Both IEve and IUni of fish communities in Haizhou Bay were low, indicating high dietary overlap and trophic redundancy among these fish species. The plasticity in resource use among the dominant species makes it possible for species with extensive overlap in resource use to coexist with other species [53]. Therefore, in the Haizhou Bay ecosystem, fish species with different feeding habits may increase productivity by maximizing resource utilization (e.g., through niche complementarity). Liu [54] suggested that fish with high overlapped trophic niches could reduce food competition through habitat differentiation and prey selection. A previous study found that species could reduce their interspecific competition by changing their food composition and spatial distributions [55]. However, low IDis indicates a low overall average level of trophic diversity in the food web of Haizhou Bay, which may be due to a significant decrease in the abundance of top predator fish and an increase in the abundance of small-sized pelagic and demersal fish due to intense fishing pressure in Haizhou Bay [56,57]. Although isotope diversity is an important reference for studying ecosystem function, IDiv, IEve, and IUni provide more valuable information about the degree of habitat coupling affected by predators in complex food webs [57].

Variation in species’ compositions are usually the main reason for the seasonal variation in trophic diversity of fish communities [38]. The discrepancy can be attributed to the fact that offshore migratory fish (such as Eupleurogrammus muticus, M. cinereus, Conger myriaster, Chelidonichthys kumu, and Saurida elongata) in autumn have higher trophic levels and wider trophic diversity, which may replace some fish with similar isotopes in spring and expand the trophic niche of fish communities. The IDiv, IDis and IEve indices of fish communities in Haizhou Bay were higher in spring than those in autumn, indicating that dominant fish species (such as C. kumu and Collichthys lucidus) in Haizhou Bay were closer to the center of isotopic gravity in autumn, with high overlap of trophic niches and intense competition for resources among species. The study on feeding habits of C. kumu and C. lucidus in Haizhou Bay suggested that Leptochela gracilis were the main prey organisms for the two species, and they will compete for food resources in this area [55,58]. Meanwhile, dominant species (such as Lophius litulon, Enedrias fangi, and T. kammalensis) in spring were at the edge of the polygonal convex hull in isotopic space, indicating a high degree of niche differentiation and extensive resource utilization.

This study also showed that there were obvious spatial variations in isotopic trophic diversity of fish communities in Haizhou Bay. The IDiv and IDis of fish communities in inshore waters (<20 m) were lower, indicating that the interspecific competition between fish was higher in inshore waters than that in offshore waters. This is consistent with the findings of Zhang et al. [56], who found greater functional divergence in offshore waters (>20 m), revealing weak interspecific competition in this area. In contrast, lower IEve and IUni indices in offshore waters (>20 m) suggested most fishes in this area had similar isotope ratios, and greater trophic redundancy can buffer external disturbances. In inshore waters (<20 m), fish communities can obtain different carbon sources through both the pelagic and benthic pathways, and with the decrease in water depth, the vertical migration of zooplankton, benthic organisms and pelagic fishes will promote the pelagic–benthic coupling [59]. The pelagic–benthic coupling intensity of different water depths may be one of the main reasons for the spatial variation of isotopic diversity indices of fish communities. Compared with inshore waters, the effects of pelagic–benthic coupling were weaker in offshore waters, and the carbon source of fish communities in this area may be mainly obtained from pelagic waters [59]. Therefore, low food resource diversity in this area contributes greatly to the low IEve value of fish communities.

As one of the important components of ecosystem structure and function, trophic relationships between species are the main aspects of interspecific relationships in fish communities, which are intricately linked to fluctuations in population dynamics and community structure. This study showed that the trophic structure of fish communities in the Haizhou Bay ecosystem had a high degree of trophic redundancy, especially at intermediate trophic levels, which will increase the resilience of this ecosystem to disturbances. The spatial and seasonal variation shown here in the trophic diversity of a fish community highlights the importance of including trophic patterns into ecosystem models, which will provide essential information for ecosystem-based fishery management. In the future, detailed and long-term monitoring of trophic diversity, biological and abiotic factors of fish in the ecosystem is needed to combine biological characteristics with environmental factors, which will contribute to a comprehensive understanding of the mechanism affecting the change in trophic diversity of fish communities and its relationship with ecosystem functions.

Different fish species may react variously to environmental variability at the community level because they have high diversity of life-history traits, feeding strategies and associated morphologies. However, few studies have analyzed the feedback of trophic diversity of fish communities to ecosystem function and changes in food web structure. In the future, a more comprehensive understanding of mechanisms influencing changes in the trophic diversity of fish communities and their relationship to ecosystem function should be considered. In addition, further studies on the variations in trophic diversity of fish communities at larger temporal and spatial scales should also be carried out to assess the stability and vulnerability of food webs under global changes, which will help guide Ecosystem-based Fisheries Management (EBFM).