Trophic niche overlap between native and non‐native fishes in central Anatolian lakes, Türkiye

Abstract

The Central Anatolian region of Türkiye exhibits exceptionally high freshwater fish diversity and endemism and is increasingly exposed to the introduction of non‐native taxa. We examined the diet composition, prey selectivity and interspecific dietary overlap among eight native and five non‐native fish species across seven lakes in Central Anatolia. Fish were sampled using multimesh gill nets and fyke nets, and their stomach contents were identified. In addition, zooplankton and benthic macroinvertebrate communities were quantified to evaluate prey selectivity patterns. Aquatic insects were the main food sources (averaging 60%), followed by cladocerans (25%), for both native and non‐native fish species. Horn's index indicated a high dietary overlap (>0.60) between several native and non‐native species, suggesting substantial overlap in resource use and the potential for exploitative interactions. Native species generally exhibited more specialized diets, whereas non‐native species tended to be generalists. Prey selectivity analysis revealed that both native and non‐native species preferentially consumed aquatic insects over other available prey types, indicating shared reliance on benthic food resources. These findings suggest that non‐native fishes may influence trophic structure and resource use patterns in freshwater ecosystems with high endemism. We emphasize the importance of region‐specific monitoring and management strategies to mitigate the ecological impacts of non‐native fishes and support the conservation of Türkiye's freshwater biodiversity.

1. INTRODUCTION

Freshwater ecosystems cover approximately 2.3% of the Earth's surface (Reid et al., 2019), and they are particularly sensitive to environmental change (Döll & Zhang, 2010; Nel et al., 2009). Despite their relatively small spatial extent, rivers and lakes support a rich biodiversity and supply vital ecosystem services (Dawson, 2012; Dudgeon et al., 2006), but they are highly vulnerable to environmental disturbances such as climate change, habitat degradation, pollution and the introduction of non‐native species (Döll & Zhang, 2010; Dudgeon & Strayer, 2024; Nel et al., 2009). The introduction of non‐native fish species poses a major ecological threat, as they disrupt trophic networks, outcompete native taxa and undermine community integrity (Costantini et al., 2018; Gallardo et al., 2016; Kalogianni et al., 2014). Evidence from freshwater systems worldwide shows that non‐native fishes can restructure food webs, reduce the abundance of native species and alter key ecosystem processes such as nutrient cycling and energy transfer (Bernery et al., 2022; Britton, 2023; Cucherousset & Olden, 2011; Haubrock et al., 2026). For instance, Cyprinus carpio Linnaeus, 1758 has been shown to trigger shifts toward turbid states, decrease water quality and reduce native biodiversity through benthivory (Thresher et al., 2018). In Europe, non‐native cyprinids and catfish Silurus glanis Linnaeus, 1758 have been linked to long‐term changes in fish community composition, while sport fish, such as Micropterus salmoides (Lacepède, 1802), listed among the world's most disruptive non‐native species, exert intense predation and competition pressures that diminish native populations (Canonico et al., 2005; Haubrock et al., 2021; Van der Veer & Nentwig, 2015). Such ecological disruptions also erode ecosystem services, including fisheries yields, water provisioning and recreational value (Dudgeon et al., 2006; Gallardo et al., 2016), which are particularly concerning in biodiversity hotspots.

Türkiye is a freshwater biodiversity hotspot situated at the intersection of several major freshwater ecoregions, including Central, Northern, Southern and Western Anatolia, as well as the Upper Tigris–Euphrates system (Abell et al., 2008). It hosts more than 420 freshwater fish species, of which 215 are endemic (Çiçek et al., 2018, 2023). This high level of endemism is primarily shaped by the country's complex geology, diverse climatic conditions and a great variety of aquatic habitats (Şekercioğlu et al., 2011). Similar to other freshwater biodiversity hotspots worldwide, many endemic species in Türkiye are highly specialized and vulnerable to environmental changes (Costa et al., 2024). The Central Anatolia region alone is home to 43 endemic species, many of which have a restricted geographic distribution and face increasing threats from habitat degradation and the introduction of non‐native species (Çolak et al., 2022; Korkmaz, Mangıt, Dumlupınar, et al., 2023; Korkmaz, Mangıt, & Yerli, 2023; Yılmaz et al., 2021).

Fish play a crucial role in aquatic food webs, functioning as both predators and prey, thereby regulating population dynamics, community structure and overall ecosystem functioning (Baker et al., 2014). However, native fish communities can be severely impacted by non‐native fish species. In Mediterranean regions, such as the Iberian Peninsula and the Balkans, invasive fish species have caused significant declines in native assemblages through competition, predation and habitat alteration (Gallardo et al., 2016; Maceda‐Veiga, 2013). These outcomes are consistent with ecological theory, which proposes that resource partitioning facilitates coexistence among native species, whereas non‐native fishes with broader ecological tolerances may overlap in resource use with native specialists, potentially increasing the likelihood of exploitative interactions and competitive displacement (Bøhn et al., 2008; Chase & Leibold, 2003). Empirical evidence clearly illustrates this: for instance, introductions of largemouth bass (M. salmoides) and other centrarchids in Iberian rivers have led to sharp reductions in endemic cyprinids, while in the Balkans, non‐native trout and cyprinids have displaced native species and reshaped community structure (Crăciun et al., 2025; Ribeiro & Leunda, 2012). In Central Anatolia, a striking case is Lake Eğirdir, where the introduction of pikeperch Sander lucioperca (Linnaeus, 1758) in the 1950s triggered the collapse of several endemic fish populations, and subsequent introductions of gibel carp Carassius gibelio (Bloch, 1782) further altered ecosystem dynamics (Tarkan et al., 2012; Yalçın‐Özdilek et al., 2019). Together, these examples illustrate how non‐native fish species may alter trophic networks and suggest that freshwater biodiversity hotspots with high endemism can be particularly sensitive to biological invasion.

The trophic plasticity and competitive advantage of non‐native over native species (Cicala et al., 2024) may further lead to niche compression, forcing native taxa into suboptimal conditions and increasing their risk of decline. Such patterns have been documented across Mediterranean and semi‐arid freshwater systems (Costantini et al., 2018; Kalogianni et al., 2014; Tesfaye et al., 2021), potentially generating ecological feedbacks that facilitate the further spread of non‐native taxa, thereby contributing to long‐term biodiversity loss (Cicala et al., 2024; Gallardo et al., 2016). In some cases, competition extends beyond fish, impacting other aquatic organisms. For example, studies have shown that C. carpio may compete with globally endangered diving duck species, such as the common pochard, Aythya ferina (Linnaeus, 1758), and the white‐headed duck, Oxyura leucocephala (Scopoli, 1769), for benthic macroinvertebrates (García, 2001; Maceda‐Veiga et al., 2017).

Understanding fish diet preferences and feeding habits is essential for characterizing the trophic structure of aquatic communities, clarifying predator–prey relationships and informing conservation and management strategies. A widely used approach to investigate these dynamics is stomach content analysis, which provides detailed insights into the dietary composition, feeding strategies and ecological roles of fish species (Nyunja et al., 2002). Such knowledge can also shed light on broader ecological theories, such as niche differentiation and resource partitioning, by revealing how species coexist (or fail to) based on their trophic behaviour. Recent studies in Türkiye and the wider Mediterranean region have begun to characterize the trophic interactions between native and non‐native fishes using stomach content and stable isotope approaches. For example, stable isotope analysis has been used to anticipate the predatory and competitive impacts of a potential Perca fluviatilis Linnaeus, 1758 introduction into a biodiversity‐rich Anatolian lake (Tarkan et al., 2023), while studies on invasive pumpkinseed Lepomis gibbosus (Linnaeus, 1758) in a southwestern Anatolian stream revealed trophic overlap with an endemic chub but only minimal consequences for its somatic growth and condition (Karakuş et al., 2023). At the community level, non‐native species and artificial water bodies have been shown to reshape the trophic and functional niche structure of Mediterranean freshwater fish assemblages (Toutain et al. 2024), and isotopic niche overlap between co‐occurring percids indicates a potential for trophic competition when resources become limiting (Toutain et al. 2026). Despite these advances, the trophic dynamics and resource‐use interactions between native and non‐native fishes remain poorly understood, particularly in the lakes of Central Anatolia, where high endemism coincides with widespread non‐native fish introductions.

This study aims to fill this knowledge gap by assessing dietary overlap and its ecological implications through stomach content analysis to evaluate food preferences, trophic dynamics and interspecific interactions, while also examining how species distributions relate to environmental variables. By contextualizing these findings within a broader ecological and biogeographical framework, our research contributes to a better understanding of how non‐native fishes may influence trophic structure and resource use in freshwater biodiversity hotspots. We hypothesized that (1) non‐native fishes exhibit substantial dietary overlap with native taxa, particularly through shared reliance on aquatic insects, and that (2) native species exhibit more specialized diets, whereas non‐native species function primarily as generalists.

2. MATERIALS AND METHODS

2.1. Study area

The study area comprised seven natural lakes located in Central Anatolia, Türkiye: Gökgöl, Uyuz, Kaya, Bakı, Narlı, Burdur and Salda (Table 1, Figure 1). These lakes are distributed within two major endorheic basins, the Konya Closed Basin and the Burdur Closed Basin, and encompass a broad range of environmental conditions, including variation in depth, salinity, nutrient concentrations and trophic status (Beklioğlu et al., 2020; Yılmaz et al., 2021). The systems include shallow eutrophic lakes such as Gökgöl and Uyuz, as well as deep oligo‐mesotrophic lakes such as Salda and Burdur, thereby providing a heterogeneous environmental gradient for evaluating fish trophic interactions and resource use patterns.

TABLE 1.

General water quality characteristics of the seven investigated lakes in Central Anatolia, Türkiye, grouped according to the Konya Closed Basin and Burdur Closed Basin.

Basin	Lakes	Maximum depth (m)	Water temperature (°C)	DO (mg/L)	Salinity (‰)	SPC (mS/cm)	pH	NO3 mg/L	SRP mg/L	TN mg/L	TP μg/L	Chl‐a mg/L
Konya Closed Basin	Gökgöl	1.4	24.3	10.6	1.00	1.75	8.56	39.5	121	2597	451	153
	Uyuz	1.5	20.3	13.8	1.64	2.81	9.58	39.2	NA	3546	609	3.7
	Kaya	2.5	25.9	13.5	0.26	0.48	7.33	111	21	575	132	2.9
	Bakı	4.5	22.6	11.8	0.25	0.45	6.61	1885	35	2578	175	6.6
	Narlı	19.5	24.7	9.2	2.43	4.03	8.33	467	13	883	28	2.0
Burdur Closed Basin	Burdur	110	27.2	9.3	22.43	32.04	9.07	429	4	1173	113	22
	Salda	184	23.9	9.2	0.91	1.69	9.35	108	4	425	15	0.9

FIGURE 1.

Location of the seven lakes in Central Anatolia, Türkiye, where fish sampling was conducted. The lakes are grouped within the Konya Closed Basin (Lakes Gökgöl, Narlı, Uyuz, Kaya and Bakı) and the Burdur Closed Basin (Lakes Burdur and Salda).

The lakes also differ in fish community composition, ranging from systems dominated by endemic native fishes to lakes dominated by non‐native species. Species were classified as native or non‐native using the terminology of Soto et al. (2024) and the most recent national assessment of established non‐native species in Türkiye (Tarkan et al., 2026). Under this framework, native species occur within their natural biogeographic range, whereas non‐native species are transported beyond their native range by human activity; established non‐native species are those that have formed self‐sustaining wild populations. In addition, several lakes are exposed to substantial anthropogenic pressures, including eutrophication, water abstraction, habitat degradation and biological invasions, all of which may influence prey availability and trophic dynamics within these ecosystems (Çolak et al., 2022; Korkmaz, Mangıt, Dumlupınar, et al., 2023; Korkmaz, Mangıt, & Yerli, 2023). In the Konya Closed Basin in particular, recent water‐quality assessments have identified organic and nutrient pollution alongside salinization as pervasive stressors that intensify under semi‐arid, low‐flow conditions, further threatening the basin's highly endemic biota (Mangıt, 2026).

2.2. Lake water sampling and laboratory analyses

Lakes were sampled in July 2021 following a snapshot methodology, an approach commonly used in shallow‐lake studies in which all systems are sampled within a short and standardized time window to characterize ecological conditions at a specific moment and facilitate broad spatial comparisons among lakes (Beklioğlu et al., 2020; Kosten et al., 2012). This approach is particularly useful for comparing trophic structure, environmental conditions, and community composition across multiple lakes while minimizing temporal variability due to seasonal changes. Physicochemical variables, including water temperature, salinity, specific conductivity (SPC), dissolved oxygen (DO) and pH, were measured using a YSI 650 MDS multiparameter probe at the deepest location in the lake. Water samples (~40 L) were collected using a Ruttner water sampler, which enables targeted sampling at specific depths, ensuring comprehensive coverage of the entire water column. The collected samples were then subsampled for subsequent analyses of water chemistry, Chl‐a and zooplankton. Water chemistry samples were stored frozen until analysis for nitrate (NO₃‐), soluble reactive phosphorus (SRP), total nitrogen (TN) and total phosphorus (TP). TP and SRP concentrations were determined using the ammonium molybdate and ascorbic acid reduction method (Mackereth et al., 1978). NO₃‐ and TN concentrations were assessed using an automated Skalar N autoanalyser, with TN samples digested at 90°C in an alkaline persulfate solution (Kroon, 1993). Water samples were immediately filtered through Whatman GF/C filters for chlorophyll a (Chl‐a) determination. Chl‐a was extracted in cold absolute ethanol, and its absorbance was subsequently measured by spectrophotometry at 663 nm, corrected for turbidity at 750 nm (Jespersen & Christoffersen, 1987).

2.3. Fish sampling and stomach content analysis

Fish sampling was conducted using benthic multimesh gill nets in accordance with the European standard methodology EN 14757:2015 (CEN, 2015). The nets, 30 m long and 1.5 m deep, had 12 different mesh sizes ranging from 5 to 55 mm. Gill nets and fyke nets were deployed during the evening and retrieved the following morning after approximately 10 h of exposure. In addition, fyke nets were used to capture larger fish that might evade the multi‐mesh gill nets. For each species, fish were measured for morphometric characteristics, including total length (TL), fork length (FL) and standard length (SL). TL measurements above 15 cm were made using a ruler, whereas a digital calliper was used for smaller individuals. TL measurements of analysed individuals are provided in Table S1. Most analysed specimens represented relatively similar size classes within each species. Fish were dissected shortly after retrieval, and stomachs were preserved in 4% formaldehyde and transported to the Limnology Laboratory in the Department of Biological Sciences at Middle East Technical University. Although prolonged net exposure may affect stomach fullness due to digestion during capture, all lakes were sampled using the same standardized protocol, thereby minimizing methodological variation among systems. Stomach contents from 337 specimens were thoroughly examined under a dissecting microscope at 10× magnification. Prey items were identified to the lowest possible taxonomic level, counted, and weighed before analysis. Insect identification followed the taxonomic keys of Merritt et al. (2019), while non‐insect items were identified using the keys of Thorp and Covich (2010).

2.4. Sampling for natural prey availability

To ensure a robust assessment of prey selectivity, we conducted extensive and systematic sampling of macroinvertebrates in sediment and the water column. The obtained abundance and composition data were used in the food selection analysis, allowing us to compare prey availability in the habitat with their actual consumption by fish.

We collected benthic macroinvertebrate samples using a Kajak Sediment Corer (KC Denmark A/S). A total of ten cores were taken from randomly selected points at varying depths within each lake to account for the spatial variability of macroinvertebrate assemblages. In the field, the upper 10 cm of the cores was sampled, sieved through two mesh size filters (250–500 μm) and preserved in 70% ethanol. In the laboratory, the macroinvertebrates were carefully sorted, counted and identified to species or genus level using relevant taxonomic keys (Merritt et al., 2019; Thorp & Covich, 2010).

Zooplankton were sampled by filtering 10–20 L of lake water sample through a 20‐μm mesh filter and preserving the samples with 4% Lugol's solution. Species identification and counting were performed using a stereo microscope (Leica DM80), and approximately 25–30 individuals of each species were measured for body size, when possible. Biomass calculations were based on length–weight relationships available in the literature (Bottrell et al., 1976; Dumont et al., 1975). Identification of zooplankton species, including cladocerans and copepods, was based on relevant taxonomic keys (Dussart & Defaye, 2001; Flössner, 2000; Kiefer, 1978; Smirnov, 1996).

2.5. Data analysis

To determine whether the number of stomachs analysed was sufficient to accurately describe the diet of each species, we established a prey accumulation curve following the criteria of Ferry and Cailliet (1996). Prey dominance was assessed using the percentage of the prey‐specific index of relative importance (%PSIRI), using the following formula (Equation 1, Brown et al., 2012):

$$\%\text{PSIRI}=\%{\mathrm{FO}}_i\times \left(\%{\mathrm{PN}}_i+\%{\mathrm{PW}}_i\right)/2$$ (1)

Here, $\%{\mathrm{FO}}_i$ represents the percentage frequency of occurrence, calculated as the number of stomachs containing prey divided by the total number of stomachs. %PN _i and %PW _i represent the numerical and weight percentages of prey items, respectively.

To calculate the diet breadth of the organism, we used the standardized Levin's index:

$$B_i=1/\sum {p_j}^2$$ (2)

where B _i is diet breadth, and p _j is the proportion of individuals consuming resource j. The index ranges from 0 to 1, with 0 indicating a narrow diet and 1 a broad diet (Equation 2, Krebs, 1989).

To gain information on the position that organisms occupied in the food web, we estimated the trophic level of each taxon using the Trophlab software developed by Pauly et al. (2000). This software calculates trophic levels based on the weighted mean of prey trophic levels obtained from diet composition data, integrating the proportional contribution of each prey item in the consumer's diet to produce an overall trophic level value for each taxon. Trophic levels for prey categories were assigned using the default values in the TrophLab database (Pauly et al., 2000). The resulting trophic levels ranged from 1 to 5, where values between 0 and 1 correspond to decomposers, 1 and 2 to primary producers, 2 and 3 to primary consumers, 3 and 4 to secondary consumers and 4 and 5 to tertiary consumers.

We used Horn's niche overlap index (Equation 3) to estimate dietary overlap among fish species within each lake:

$$R_{\mathrm{o}}=\sum \left(P_{\mathit{ij}}+P_{\mathit{ik}}\right)\log \left(P_{\mathit{ij}}+P_{\mathit{ik}}\right)-\sum P_{\mathit{ij}}\log P_{\mathit{ij}}-\sum P_{\mathit{ik}}\log P_{\mathit{ik}}/2\log 2$$ (3)

Here, R _o is the Horn's index of overlap for species j and k. P _ij represents the proportion of resources utilized by species j, and P _ik is the proportion of the same resource used by species k. Index values range from 0 to 1, with values indicating low (0–0.29), medium (0.30–0.65), or high (0.66–1) overlap based on Langton's (1982) classification scale.

We assessed fish prey preference using the PSIRI index and calculated Jacob's electivity index (D) using the formula D = (r − p)/([r + p] − 2rp), where r represents the percentage (%) of a particular food item in the stomach content (PSIRI), and p represents its percentage in the lake environment (Jacobs, 1974). For the D‐value calculation, we categorized the zooplankton into Cladocera and Copepoda and combined macroinvertebrate data categorized into insects and gastropods from each site to assess the impact of habitat on the species' food selection within each lake. The relative abundance of each prey type was calculated as a proportion of the total available prey in the environment. These proportions were then compared to the percentages of prey items found in the stomach contents to assess selectivity. The D values range from −1 to 1, with negative values indicating avoidance of specific food items and positive values indicating preference.

To choose the appropriate constrained ordination method, a detrended correspondence analysis (DCA) was first performed on the species composition data. The DCA indicated unimodal species responses along the environmental gradients, and canonical correspondence analysis (CCA) was therefore selected. Prior to the CCA, correlations among environmental variables were examined using a Spearman correlogram. Among the correlated variables, those with the highest variance inflation factor (VIF > 10) were stepwise removed to reduce multicollinearity. The final CCA model included TN, depth, salinity, zooplankton abundance and macroinvertebrate abundance. The statistical significance of the CCA model and selected variables was tested using Monte Carlo permutation analysis with unrestricted iterations. Analyses were conducted using the CCA package (v. 1.2.2; González & Déjean, 2023) in R (R Core Team, 2021).

2.6. Ethics statement

All procedures involving fish collection and examination in this study conformed to the national regulations of the Republic of Türkiye for the scientific sampling of aquatic organisms. Fieldwork was conducted under official research permits issued by the Ministry of Agriculture and Forestry, General Directorate of Agricultural Research and Policies (Permit No: E‐92190712‐605.99‐9608765), and the General Directorate of Nature Conservation and National Parks (Permit No: E‐21264211‐288.04‐9521803). All sampling and handling procedures adhered to the conditions specified in these permits, and efforts were made to minimize stress and handling time for all specimens.

3. RESULTS

3.1. Physicochemical and biological characteristics of the study lakes

The physicochemical variables showed large variations among the lakes (Table 1). Lake depth ranged from 1.4 m (Lake Gökgöl) to 184 m (Lake Salda). Salinity ranged from 0.25‰ (Lake Bakı) to 22.4‰ (Lake Burdur). Nitrate (NO₃ ⁻) ranged from 39.2 μg/L (Lake Uyuz) to 1885 μg/L (Lake Bakı), while SRP ranged from 4 μg/L (Lakes Burdur and Salda) to 121 μg/L (Lake Gökgöl). Chl‐a reached its maximum in Lake Gökgöl (153 μg/L). TP (609 μg/L) and TN (3546 μg/L) peaked in Lake Uyuz (Table 1).

The zooplankton assemblages varied among lakes (Table S2). Cladocera were generally scarce, ranging from 0.2% in Lake Burdur to 33.3% in Lake Salda, except in Lake Kaya, where they constituted 51% of the zooplankton community. Copepods also varied markedly among lakes, from 3.2% in Lake Gökgöl to 72.3% in Lake Bakı, where they were the dominant group. Rotifera exhibited the broadest variation, ranging from 7.2% in Lake Uyuz to 96% in Lake Gökgöl, and dominated the zooplankton assemblage in several lakes, including Burdur (84.3%) and Narlı and Kaya (43%–44%).

Macroinvertebrate assemblages also differed across the lakes (Table S3). Chironomidae were the dominant taxon overall, representing up to 46% of total macroinvertebrate abundance and reaching 85.7% in Lake Narlı and 75% in Lake Uyuz. In contrast, Glossiphoniidae dominated in Lake Bakı (83%) and Lake Salda (78.7%). Lake Kaya exhibited a more balanced assemblage, composed mainly of Chironomidae (48%), Glossiphoniidae (36%) and Corixidae (16%). In Lake Gökgöl, gastropods were particularly abundant, with Valvatidae and Physidae together representing 61% of the assemblage.

3.2. Fish community, stomach content and food selection

Thirteen fish species were recorded in the seven lakes—eight native and five non‐native. A total of 1735 individuals were collected, from which a subset of 337 specimens representing various size groups was selected for stomach content and food selection analyses. However, for the native Anatolichthys sureyanus (Neu, 1937), the only species sampled in Lake Burdur and endemic to this lake, all sampled individuals had empty stomachs. Consequently, although seven lakes were sampled, stomach content analyses were only conducted for six lakes and twelve fish species.

Diet items identified in the stomach contents included Cladocera, Plumatelidae, algae, Copepoda, unidentified organic matter, plant remains, aquatic insect remains, ostracods, fish or fish remains and gastropods, for which PSIRI values were calculated (Table 2). C. carpio consumed all these food items (Table 2). The PSIRI analysis revealed that aquatic insect remains were the most relevant food item for all fish species, 60.3% on average, ranging from 1 to 100% (Table 2). The Levin index indicated that the native species were specialists, while the non‐native species were generalists (Bi >0.5), except for Pseudorasbora parva (Temminck & Schlegel, 1846) and Tinca tinca (Linnaeus, 1758) (Table 3). Based on trophic levels, most species were primary consumers; however, Cobitis battalgili Băcescu, 1962, P. parva, Seminemacheilus ekmekciae Yoğurtçuoğlu, Kaya, Geiger & Freyhof 2020 (in Lake Narlı) and Pseudophoxinus burduricus Küçük, Gülle, Güçlü, Çiftçi & Erdoğan, 2013 were secondary consumers (Table 3).

TABLE 2.

Diet composition of fish species from lakes within the Konya Closed Basin and Burdur Closed Basin in Central Anatolia, Türkiye.

Basin	Lakes	Species	n	UOM	PR	IR	Alga	Cope	Clad	Ostr	Plum	Gast	Fish
Konya closed basin	Gökgöl	A. anatoliae a	15			35.28			32.36	32.36
		C. gibelio	16	4.74		66.02			15.02	14.22
		C. battalgili a	21			93.10							6.90
		C. carpio	19	12.77	10.00	47.83	0.85	3.48	15.68	3.10	1.16	2.71	2.42
		P. crassus a	29	18.43	0.58	77.91	0.58	1.45	0.19	0.64		0.19
		P. parva	19			94.05		5.95
		S. ekmekciae a	18			99.16	0.10	0.64	0.10
	Uyuz	C. gibelio	16	26.76	3.49	61.66		0.87	3.20	2.62
		T. tinca	18		5.85	93.05		1.10
	Kaya	C. gibelio	18	14.51	0.60	4.34	3.63		76.91
		C. carpio	26	24.16	6.47	3.37	1.15	2.26	60.23	0.32		0.05
	Bakı	C. carpio	16		47.11	1.08		3.77	10.77	37.15		0.54
	Narlı	C. carpio	18	34.93	11.02	15.82	7.62	0.81	28.13	0.05
		S. ekmekciae a	16			100
Burdur closed basin	Salda	A. saldae a	25	7.7	0.18	90.01		0.18					1.85
		A. boyeri	15	52.50		47.5
		C. phrygica a	16	15.31		79.59	5.10
		P. burduricus a	16	4.46		76.21							19.33

Note: Values for each dietary item represent percentages of the prey‐specific index of relative importance (PSIRI%).

Abbreviations: Alga, algae; Clad, Cladocera; Cope, Copepoda; Fish, fish and fish remains; Gast, gastropods; IR, aquatic insect remains; N, number of stomachs analysed; Ostr, ostracods; Plum, Plumatelidae; PR, plant remains; UOM, unidentified organic matter.

^a Indicates native species.

TABLE 3.

Levin's niche breadth index (Bi) and trophic level estimates for fish species from lakes within the Konya Closed Basin and Burdur Closed Basin in Central Anatolia, Türkiye.

Basin	Lakes	Species	Levin's index (Bi)	Trophic level
Konya closed basin	Gökgöl	A. anatoliae a	0.32	2.6 ± 0.41
		C. gibelio	0.50	2.9 ± 0.41
		C. battalgili a	0.35	3.8 ± 0.61
		C. carpio	0.88	2.8 ± 0.32
		P. crassus a	0.68	2.6 ± 0.28
		P. parva	0.36	3.1 ± 0.26
		S. ekmekciae a	0.13	3.1 ± 0.21
	Uyuz	C. gibelio	0.75	2.7 ± 0.28
		T. tinca	0.37	2.7 ± 0.21
	Kaya	C. gibelio	0.68	2.4 ± 0.22
		C. carpio	0.70	2.7 ± 0.31
	Bakı	C. carpio	0.68	2.9 ± 0.41
	Narlı	C. carpio	0.79	2.7 ± 0.27
		S. ekmekciae a	0.06	3.2 ± 0.40
Burdur closed basin	Salda	A. saldae a	0.15	2.9 ± 0.33
		A. boyeri	0.52	2.8 ± 0.31
		C. phrygica a	0.16	2.4 ± 0.22
		P. burduricus a	0.43	3.4 ± 0.53

Note: Lower Bi values indicate narrower and more specialized dietary niches. Trophic levels were estimated using TrophLab software based on the weighted mean trophic levels of prey items derived from diet composition data. Values are presented as mean ± standard error.

^a Indicates native species.

Fish food selectivity varied between lakes and among species (Figure 2). Preference for copepods was higher in Lake Narlı for C. carpio and in Lake Uyuz for T. tinca (R ² = 0.72, F = 1.85, p < 0.001). The analysis of insect D‐values revealed significant positive selection by all fish species in Lake Gökgöl (R ² = 0.64, F = 1.72, p < 0.001). In contrast, preference for Cladocera was lower than for insects in all study lakes (R ² = 0.54, F = 1.93, p < 0.001). Overall, insects emerged as the most preferred dietary items for almost all species in all lakes (R ² = 0.84, F = 1.94, p < 0.001), while gastropods were selectively consumed only in Lake Gökgöl (Figure 2).

FIGURE 2.

Preference index (D) values for Copepoda, Cladocera and macroinvertebrates (insects and gastropods in fish stomachs) from the six study lakes in Central Anatolia, Türkiye. *Indicates native species. Error bars represent standard deviations.

3.3. Dietary niche overlap

In Lake Gökgöl, the Horn's niche overlap index indicated high dietary overlap between the native species Anatolichthys anatoliae (Leidenfrost, 1912) and the non‐native C. gibelio and C. carpio, as well as between the native species S. ekmekciae, Pseudophoxinus crassus (Ladiges, 1960) and C. battalgili and the non‐native C. gibelio, C. carpio and P. parva. In Lake Salda, high dietary overlap was also observed between the native species Anatolichthys saldae (Akşiray, 1955), C. battalgili and P. burduricus and the non‐native Atherina boyeri Risso, 1810, as well as between the native Cobitis phrygica Battalgazi, 1944 and P. burduricus with the non‐native A. boyeri. In Lake Narlı, the native S. ekmekciae exhibited moderate dietary overlap with non‐native C. carpio (Table 4). In lakes Uyuz, Kaya and Bakı, dietary overlap occurred exclusively among non‐native species.

TABLE 4.

Horn's niche overlap index values across fish species co‐occurring within lakes of the Konya Closed Basin and Burdur Closed Basin in Central Anatolia, Türkiye.

Basin	Lake	Species	Aa a	Cg	Cb a	Cc	Pc a	Pp	Se a
Konya closed basin	Gökgöl	Aa a	1	0.89	0.54	0.71	0.54	0.55	0.56
		Cg		1	0.78	0.85	0.83	0.78	0.81
		Cb a			1	0.66	0.85	0.94	0.96
		Cc				1	0.83	0.70	0.70
		Pc a					1	0.88	0.89
		Pp						1	0.98
		Se a							1

Basin	Lake	Species	As a	Ab	Cp a	Pb a
Burdur closed basin	Salda	As a	1	0.81	0.96	0.93
		Ab		1	0.87	0.49
		Cp a			1	0.85
		Pb a				1

Basin	Lake	Species	Cg	Cc
Konya closed basin	Kaya	Cg	1	0.97
		Cc		1

Basin	Lake	Species	Cg	Tt
Konya closed basin	Uyuz	Cg	1	0.81
		Tt		1

Basin	Lake	Species	Cc	Se a
Konya closed basin	Narlı	Cc	1	0.35
		Se a		1

Abbreviations: Aa ^a , A. anatoliae; As ^a , A. saldae; Ab, A. boyeri; Cg, C. gibelio; Cb ^a , C. battalgili; Cp^a, C. phrygica; Cc, C. carpio; Pb^a, P. burduricus; Pc^a, P. crassus; Pp, P. parva; Se^a, S. ekmekciae; Tt, T. tinca.

^a Indicates native species.

The CCA indicated that TN, depth, salinity, zooplankton and macroinvertebrates explained 97.3% of the total compositional variance, with the first two canonical axes accounting for 61.3% of the constrained variation (CCA1 = 30.7%, CCA2 = 30.6%), with an overall significance of F = 7.13, p = 0.003. Permutation tests showed that both CCA1 and all selected environmental variables were significant in explaining the fish composition in the lakes (CCA1: p = 0.004; TN p = 0.009; depth p = 0.003; salinity p = 0.002; zooplankton p = 0.006; and macroinvertebrates p = 0.0018). Lake Burdur was closely associated with A. sureyanus, a native species endemic to the lake, which explains why it appears isolated in the CCA plot.

Variables such as depth, pH and macroinvertebrates were aligned with species such as C. gibelio and S. ekmekciae, suggesting that these prefer deeper, more alkaline lakes with richer benthic communities. Nutrient (SRP, TP, TN) and DO concentrations were associated with P. parva, A. anatoliae and C. carpio, indicating that these species thrive in more eutrophic lakes. Lakes with higher macroinvertebrate abundance tended to host species like C. gibelio, while zooplankton abundance correlated with species such as T. tinca, A. anatoliae and C. carpio, indicating a potential dietary or habitat preference linked to plankton‐rich conditions.

The lakes were widely distributed across the plot, suggesting a large ecological differentiation among them. For instance, Lake Salda, located in the upper‐central portion of the plot, was associated with endemic species such as A. saldae, C. phrygica and P. burduricus. These species are indicative of oligotrophic conditions with high water clarity and low productivity. In contrast, Lake Kaya, Lake Gökgöl and Lake Bakı clustered in the central‐right area of the plot, corresponding to higher nutrient levels and diverse fish assemblages (Figure 3).

FIGURE 3.

Canonical Correspondence Analysis (CCA) of seven Central Anatolian lakes, showing fish species, macroinvertebrate and zooplankton abundances, as well as environmental variables. *Indicates native species.

4. DISCUSSION

Our results support the hypothesis that there is a substantial dietary overlap between non‐native and native fish species, primarily due to their shared reliance on aquatic macroinvertebrates and zooplankton. Native species generally exhibited more specialized feeding strategies, whereas non‐native species tended to be generalists (Table 3). These patterns suggest potential resource overlap and exploitative interactions between native and non‐native fish species in the Central Anatolian lakes where they co‐occur.

Due to the widespread introduction of non‐native fish, many lakes in Central Anatolia are currently characterized by the predominance of non‐native fish species (Giannetto & Innal, 2021; Korkmaz, Mangıt, Dumlupınar, et al., 2023), reflecting a broader global trend in freshwater ecosystems (Carpio et al., 2019; Lyach, 2022; Toussaint et al., 2016). Non‐native fishes may overlap with native species in resource use and potentially influence trophic dynamics within invaded systems. In our study, native and non‐native species showed similar dietary patterns, with aquatic insects as the predominant prey category. Larger‐bodied zooplankton, such as cladocerans and copepods, were more frequently selected than smaller rotifers (Table S2), suggesting a size‐selective feeding behaviour. PSIRI and D‐value analyses further indicated that macroinvertebrates, particularly aquatic insects, constituted the most frequently selected prey. Similar feeding patterns have been reported for visually feeding omnivorous fishes, which commonly prefer macroinvertebrates and larger zooplankton over smaller prey such as rotifers (e.g., Pacheco et al., 2021).

Many of the non‐native fish species in the study lakes were historically introduced for commercial and recreational purposes, a pattern documented for Turkish and Central Anatolian freshwaters where common carp and gibel carp have been widely stocked or spread through fishery‐enhancement activities (Aksu et al., 2025; Gaygusuz et al., 2015), consistent with broader global evidence on the ecological impacts of non‐native freshwater fishes (Cucherousset & Olden, 2011). These introductions have often involved species with broad feeding ecologies, as generalist fishes are typically more resilient and easier to maintain than specialist endemics. In our study, native species generally exhibited more specialized diets, whereas non‐native species displayed broader feeding strategies, enabling the use of a wide range of prey resources (Garcia et al., 2018). Horn's niche overlap index further revealed substantial dietary overlap between several native and non‐native species. For example, in Lake Gökgöl, the non‐native C. gibelio and C. carpio exhibit dietary overlaps with native species, such as C. battalgili. Similar patterns have been reported in other Turkish freshwater systems. Yalçın‐Özdilek et al. (2019) found that the non‐native C. gibelio displayed the widest isotopic niche among coexisting fishes in the Karamenderes River (Western Anatolia, Türkiye), with substantial isotopic overlap with Mugil cephalus Linnaeus, 1758 at the river mouth. Likewise, studies in the Karacaören Reservoir have documented isotopic niche overlap between native species such as Capoeta antalyensis (Battalgil, 1943) and the non‐native C. gibelio and C. carpio (Kaymak et al., 2023). Together, these studies suggest that broad dietary and isotopic niches may facilitate the persistence of non‐native fishes and increase the potential for resource overlap with native assemblages, particularly under environmentally variable or resource‐limited conditions. Similarly, a nationwide stable‐isotope assessment of natural and artificial Turkish water bodies found that the non‐native common carp and gibel carp exhibited highly similar growth patterns, body condition and trophic ecology, resulting in substantial trophic niche overlap and a strong potential for interspecific competition where they co‐occur (Aksu et al., 2025). This aligns specifically with our finding that, in lakes where only non‐native species were present (Uyuz, Kaya and Bakı), dietary overlap occurred exclusively among these introduced cyprinids. Importantly, trophic overlap does not always translate into measurable fitness consequences; for instance, Karakuş et al. (2023) found that although invasive pumpkinseed overlapped trophically with the endemic Anatolian chub, this overlap had only minimal effects on the growth and condition of the native species. This reinforces the interpretation that the dietary overlap documented here reflects the potential for, rather than direct evidence of, competitive interactions.

We found that non‐native fish species, such as C. gibelio and C. carpio, exhibited broad trophic niches, exploiting both pelagic and benthic prey resources. The species C. gibelio has been reported to occupy multiple ecological niches and display high ecological flexibility in invaded freshwater systems (Dominguez‐Almela et al., 2021), whereas C. carpio, although generally considered omnivorous (Alaş et al., 2010; Ölmez & Aydın, 2023), may shift between benthivory and zooplanktivory depending on the environmental conditions (Pothoven, 2020; Zhang et al., 2008). In Lake Narlı, C. carpio consumed both zooplankton and macroinvertebrates, while in Lake Kaya and Lake Bakı, it fed primarily on cladocerans, reflecting dietary flexibility across systems. Similarly, C. gibelio exhibited broad feeding preferences and substantial dietary overlap with native species such as C. battalgili and P. crassus. Dietary generalism in cyprinids has been frequently reported in Mediterranean freshwater ecosystems and may contribute to their persistence in invaded habitats (García et al., 2018; Bissattini et al., 2021; Neves et al., 2024; Cicala et al., 2024).

Our findings revealed consistent omnivory–insectivory among several native and non‐native fish species, particularly in lakes with higher macroinvertebrate abundance, such as Lake Gökgöl, as previously reported for Central Anatolian reservoirs (Yalçın‐Özdilek et al., 2013). However, dietary shifts across lakes, particularly among C. carpio and C. gibelio, indicate a pronounced trophic flexibility for these introduced cyprinids. These species exploited both zooplankton and benthic invertebrates, consistent with their classification as omnivorous generalists (Neves et al., 2024; Ölmez & Aydın, 2023). Such feeding plasticity aligns with the omnivory theory (Gutgesell et al., 2022), which proposes that generalist consumers may persist successfully in fluctuating or disturbed environments by shifting among available trophic resources. Accordingly, the dietary adaptability observed in C. gibelio and C. carpio may contribute to their widespread occurrence and invasive potential in Central Anatolian lakes.

Fish dietary preferences varied among lakes and appeared closely associated with differences in zooplankton composition and prey availability (Table S2). Variability in the dominance of cladocerans, rotifers and copepods among lakes likely influenced prey selectivity patterns observed in several fish species. In particular, positive selection for copepods by species such as C. carpio and T. tinca suggests that even less energetically profitable prey may serve as important trophic resources under specific environmental conditions (Alaş et al., 2010; Dadebo et al., 2015).

In Lake Gökgöl, all fish species displayed a clear preference for copepods, despite their low representation (~3%) within the zooplankton community. The particularly high D‐values recorded for native species such as A. anatoliae indicate preferential selection of this relatively scarce prey resource. This pattern suggests that copepods, although not numerically dominant, may represent an important food resource when available, possibly due to higher encounter rates or accessibility compared to other planktonic taxa. In contrast, cladocerans, despite being relatively abundant in Lake Kaya and Lake Salda, were consumed less frequently, as reflected by consistently low D‐values across lakes, which may be related to their smaller size and lower energetic contribution. Macroinvertebrates consistently represented the dominant dietary component across lakes, likely reflecting their larger body size and higher nutritional value relative to zooplankton (Vesterinen et al., 2021). This widespread reliance on macroinvertebrates highlights their ecological importance in sustaining fish diets, as reported for other freshwater ecosystems (Guo et al., 2018).

The multivariate analyses indicated that environmental gradients contributed to the variation in fish assemblage structure across lakes. The spatial segregation of endemic species, such as A. sureyanus and C. battalgili, may reflect sensitivity to local physicochemical conditions and limited tolerance to eutrophication. In contrast, non‐native generalists, such as C. gibelio, P. parva and C. carpio, were primarily associated with nutrient‐rich lakes, highlighting their ecological flexibility under degraded environmental conditions. These findings partially align with the species‐sorting framework (Chase & Leibold, 2003), suggesting that habitat filtering associated with eutrophication may contribute to shaping fish assemblages (Figure 3). However, because systematic historical faunal data are unavailable, the relative contributions of environmental filtering, invasion history and past native declines to present‐day community structure cannot be fully disentangled.

In Lake Salda, omnivorous fish dominated the community, with A. boyeri consuming large proportions of unidentified organic matter. Multivariate analysis also identified Lake Salda as ecologically distinct, which is consistent with its low invertebrate richness. Only three zooplankton taxa were recorded: Bosmina longirostris (O.F. Müller, 1776) (33.3%), copepods (50.0%) and Keratella quadrata (Müller, 1786) (16.7%) (Table S2). Likewise, the macroinvertebrate assemblage was restricted to only two groups, Chironomidae (21.3%) and Glossiphoniidae (78.7%) (Table S3). This reduced richness in both zooplankton and macroinvertebrates is consistent with previous studies attributing the impoverished biological communities of Lake Salda to elevated magnesium and chlorine concentrations (Kazancı et al., 2004).

Lake Burdur, a thalassic, hyposaline and deep lake characterized by low macroinvertebrate diversity (Girgin et al., 2004; Green et al., 1996), has undergone a marked decline in water levels, driven by uncontrolled agricultural water extraction and dam construction (Çolak et al., 2022; Küçük et al., 2016). Consequently, the lake has experienced severe littoral habitat loss, with approximately 27% of its surface area disappearing over 27 years (Giannetto & Innal, 2021; Saç et al., 2025). Consistent with this, both Lake Burdur and Lake Salda have been identified as freshwater Key Biodiversity Areas whose endemic fish faunas are threatened by a combination of water abstraction, salinization, habitat degradation and non‐native species (Mangıt et al., 2025). Currently, only A. sureyanus, an endemic species, has been recorded in Lake Burdur, and all sampled individuals were found to have empty stomachs. This species, which feeds on phytoplankton, zooplankton and benthic organisms (Güçlü et al., 2007), is currently classified as endangered on the IUCN Red List, primarily due to extensive habitat loss (Freyhof, 2014). The absence of food items in the stomachs of A. sureyanus may reflect the extremely low prey diversity and availability observed in Lake Burdur, where zooplankton were dominated by rotifers (84.3%), particularly Brachionidae (83% of the rotifer community). In contrast, cladocerans and copepods occurred at much lower abundances (0.2% and 15.5%, respectively). Macroinvertebrates were represented exclusively by Chironomidae (100%). However, methodological factors may also have contributed to the absence of stomach contents, as prolonged net exposure and capture‐related stress can accelerate digestion in small‐bodied fishes such as A. sureyanus. Seasonal feeding inactivity or stress‐induced physiological responses cannot be ruled out, and additional targeted sampling is necessary to distinguish among these potential explanations.

Most fish species recorded in the studied lakes functioned as primary consumers, whereas only C. battalgili, P. parva, S. ekmekciae (in Lake Narlı) and P. burduricus were classified as secondary consumers. The broad feeding strategies exhibited by several non‐native fishes may facilitate persistence in the environmentally variable and often disturbed habitats characteristic of Anatolian lakes (Milardi et al., 2019). Similar patterns have been reported in other Turkish freshwater ecosystems, where non‐native species frequently dominate altered fish assemblages (Mangıt et al., 2018; Mollot et al., 2017; Tarkan et al., 2015; Tarkan et al., 2017). In our study, only non‐native fishes were recorded in lakes Uyuz, Kaya and Bakı, including C. carpio, C. gibelio, T. tinca, P. parva and A. boyeri. The predominance of generalist non‐native fishes in these systems may influence prey communities and trophic structure, particularly in shallow eutrophic lakes, where benthic invertebrates constitute key food resources (Giannetto & Innal, 2021; Mangıt et al., 2017; Tarkan et al., 2015). In lakes Uyuz and Gökgöl, where the endangered common pochard (Aythya ferina) and white‐headed duck (Oxyura leucocephala) breed (Yılmaz et al., 2021), C. carpio and C. gibelio consumed substantial proportions of aquatic insects. Because aquatic insects are also important dietary resources for these bird species (Cramp, 1977; Salvador et al., 2020), this pattern may indicate potential trophic overlap within these lake ecosystems. However, additional studies would be necessary to evaluate the ecological significance of these interactions.

Non‐native fish species continue to be stocked and translocated in freshwater systems worldwide, including Türkiye, where introductions of C. carpio and C. gibelio are common in reservoirs (Gaygusuz et al., 2015; Savini et al., 2010). In our study, C. gibelio was recorded in lakes Gökgöl, Uyuz and Kaya, while P. parva occurred in Lake Gökgöl, where both species exhibited dietary overlap with several native fishes. These overlap patterns indicate shared use of trophic resources and suggest the potential for exploitative interactions between native and non‐native species in systems where they co‐occur. However, present‐day fish assemblages likely also reflect historical introductions, species translocations and possible local declines or extirpations of native fishes. Importantly, none of these lakes is naturally fishless; however, systematic historical ichthyofaunal surveys are lacking for lakes Uyuz, Kaya and Bakı, which prevents a direct assessment of whether non‐native fishes have displaced formerly present native species or have instead occupied assemblages already depauperate through habitat degradation. This limitation is not unique to our study systems: across the Konya Closed Basin, long‐term monitoring of better‐documented wetlands has shown that habitat loss and water abstraction have driven marked declines of endemic fishes concurrent with the expansion of non‐native taxa (Yılmaz et al., 2021). By analogy, the current dominance of non‐native generalists in lakes Uyuz, Kaya and Bakı is consistent with, but cannot by itself confirm, a history of native decline. Distinguishing displacement from the colonization of already degraded systems would require historical faunal baselines that are currently unavailable for these lakes. The broad dietary niche and ecological flexibility of non‐native species such as C. gibelio may facilitate their persistence under changing environmental conditions, highlighting the importance of continued monitoring and management efforts to prevent further introductions and conserve endemic fish diversity in Central Anatolian lakes.

Although we observed substantial dietary overlap between native and non‐native species, our study focused primarily on feeding habits and therefore does not provide direct evidence of competitive interactions. The dietary patterns documented here represent a temporal snapshot of resource use under summer conditions (July 2021), and seasonal fluctuations in prey availability and feeding behaviour may influence trophic overlap over time. Consequently, our findings should be interpreted as reflecting summer trophic dynamics rather than year‐round species interactions. Future research incorporating seasonal sampling, long‐term monitoring and complementary approaches such as stable isotope and gut DNA analyses could provide a more comprehensive understanding of temporal variation in resource use and trophic overlap under different environmental conditions. Nevertheless, a direct evaluation of competition would require additional experimental, behavioural, or population‐level approaches. Recognizing these limitations is essential for interpreting our findings and guiding future studies on the ecological effects of non‐native fishes in freshwater ecosystems.

5. CONCLUSIONS

Our results revealed substantial dietary overlap between native and non‐native fish species in Central Anatolian lakes, primarily associated with a shared reliance on aquatic macroinvertebrates and zooplankton. Native species generally exhibited more specialized feeding behaviours, whereas non‐native fishes displayed broader dietary preferences characteristic of generalist foraging strategies. Such trophic flexibility may facilitate the exploitation of diverse prey resources under variable environmental conditions and increase the potential for resource overlap in lakes where native and non‐native species co‐occur. The continued stocking and translocation of non‐native cyprinids may further alter trophic structure and resource‐use patterns in these freshwater ecosystems. Given the high endemism and unique freshwater biodiversity of Central Anatolia, the increasing prevalence of non‐native fishes represents an important conservation and management concern. Accordingly, site‐specific management strategies remain essential to limit further introductions and support the conservation of native fish assemblages. Based on our findings, effective management actions should include (1) monitoring and control measures to reduce the spread of non‐native species, and (2) conservation and restoration of critical habitats, particularly littoral zones that sustain diverse prey communities. These actions may help maintain ecosystem functioning and support the long‐term persistence of native fish populations in these ecologically sensitive lake systems.

AUTHOR CONTRIBUTIONS

Conceptualization: ARG, MK, MB, KÖ, EJ. Developing methods: ARG, MK, JPP, EJ. Data analysis: ARG, MK, JP, ÜNT, EJ. Preparation of figures and tables: ARG and MK. Conducting the research, data interpretation and writing: ARG, MK, JJP, ÜNT, MVPZ, VA, CA, MM, CD, GY, SE, CAA, MAÇ, İKÖ, ZA, KÖ, MB, EJ.

CONFLICT OF INTEREST STATEMENT

The authors declare that there are no conflicts of interest that could have influenced the outcomes of this study. All funding sources and affiliations have been disclosed, and there are no personal, financial or professional relationships that could be perceived as influencing the research process or findings.

Supporting information

Table S1. Total length (TL, mm) of fish specimens used in stomach content analyses from lakes within the Konya Closed Basin and Burdur Closed Basin in Central Anatolia, Türkiye. Values are presented as a minimum–maximum range and mean ± standard deviation (SD).

Table S2. Percentage abundances of zooplankton species in the study lakes.

Table S3. Percentage abundance of invertebrate species in the study lakes.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.