Linking non-marine Ostracoda (Crustacea) assemblages to environmental drivers using species archetype models and indicator value analysis

Abstract

We investigated ostracod assemblages from 120 aquatic sites across Adana Province, southern Türkiye, to evaluate community–environment relationships and identify diagnostic taxa along strong physicochemical gradients across a range of habitat types. A total of 35 species were recorded from 11 different aquatic habitats. Species Archetype Modelling (SAM), Indicator Species Analysis (IndVal), and multivariate ordination were applied to characterize distributional patterns across sites. SAM resolved four archetypes, including a widespread assemblage dominated by Heterocypris incongruens and three less common assemblages associated with distinct environmental conditions. Conductivity was the primary variable associated with variation among archetypes, with pH and water temperature contributing additional variation. Canonical Correspondence Analysis accounted for 63.6% of the variation in species–environment relationships and indicated patterns of separation among assemblages along conductivity and thermal–chemical gradients, although these relationships should be interpreted cautiously. Cluster analysis further identified groups of sites sharing similar ecological and physicochemical characteristics. IndVal analysis supported the SAM-defined groupings, highlighting taxa associated with particular habitat types, including H. salina and Prionocypris zenkeri in saline or restricted environments and Cypridopsis vidua in freshwater sites. Species with broad ecological tolerances were widely distributed across sites, whereas less frequent taxa showed more restricted occurrences. The combined use of SAM and IndVal provides a robust framework for identifying patterns of ecological specialization and non-random assemblage structure in inland waters. These results highlight the association of non-marine ostracods with conductivity-related gradients and support their use as indicators in biomonitoring and conservation studies in Mediterranean freshwater ecosystems.

Introduction

Non-marine ostracods (Crustacea, Ostracoda) are small bivalved crustaceans inhabiting freshwater, brackish, and ephemeral aquatic habitats in almost all continents. Being found in variety of aquatic habitats of lotic and lentic ecosystems, ostracods are probably the most widespread aquatic invertebrates. Despite their small size (ca. 1 mm of length), ostracods as part of the food web play important ecological and biological roles in the aquatic ecosystems (Kantorek, 1992; Dole-Olivier et al., 2000; Külköylüoğlu, 2005; Çapraz et al., 2022). They consist of a pair of calcium-carbonated carapaces which make them found as fossils from Ordovician (c. 450 myr) (Maddocks, 1982) or the Cambrian age (Sars, 1928; Williams et al., 2008; Oakley et al., 2013; Wolfe et al., 2016). Contemporary studies revealed that majority of the species tend to show species-specific habitat requirements in different environmental conditions where ecological tolerances and optimum estimates of the species can be very broad for the type of variables (Anadón et al., 1986; Yılmaz and Külköylüoğlu, 2006; Külköylüoğlu et al., 2020). Many (if not all) non-marine ostracod species are also highly sensitive to environmental gradients such as water temperature, salinity, dissolved oxygen, pH, and habitat permanence, making them valuable bioindicators for freshwater ecosystems (Wise, 1961; Delorme, 1991; Rosati et al., 2014; Schmitz et al., 2025).

Adana Province (hereafter Adana), located in southern Türkiye, encompasses a wide variety of aquatic habitats, ranging from temporary ponds and permanent lakes to brackish wetlands. This environmental heterogeneity provides an excellent setting to study the factors shaping ostracod community structure. However, the ecological mechanisms driving community assembly, particularly the roles of habitat permanence and environmental variables such as temperature, pH and salinity remain poorly understood. Although limited knowledge about autecology of a few individual ostracod species is known (see, e.g., Yavuzatmaca et al., 2024), we do not have better understanding of species’ occurrence patterns. This means that correlational driving patterns between groups of species, which inhabit and/or prefer similar environmental conditions, and ecosystems need more elusively detailed research and supportive data than is already present.

The idea of Species Archetype Models (SAM) is a model-based approach for clustering species (e.g., common and widely distributed species) into groups that exhibit similar environmental responses (Silva and Souza, 2018). In this context, a similar perspective, albeit with a different hypothesis, was proposed by Külköylüoğlu et al. (2020), who suggested that similar occurrence patterns of two or more species (referring to archetypes) can reflect shared ecological characteristics. Accordingly, these species groups can serve as indicators of that of particular habitat(s). When SAM reduces complex community data into ecologically meaningful assemblages, IndVal identifies species that characterize each assemblage, linking community patterns to environmental gradients (Dolédec et al., 2000; Dunstan et al., 2011). Applying this combined approach to ostracod communities allows detection of general assembly patterns and identification of species that reliably indicate environmental conditions. This study represents the first application of a SAM-IndVal framework to ostracod communities in Adana, providing a baseline for understanding community assembly, environmental filtering, and bioindicator potential in a variety of freshwater ecosystems. Based on these considerations, this study tested two main hypotheses. First, ostracod community composition in Adana is associated with environmental gradients such as water temperature, salinity, pH and habitat permanence. The rationality of this hypothesis is that distinct ecological assemblages are expected to correspond to different habitat types, with generalists (referring to cosmopolitan species) occupying variable habitats, freshwater specialists occurring in permanent waters, salinity-tolerant species inhabiting brackish sites, and opportunistic species thriving in ephemeral habitats. Second, we hypothesize that selected ostracod species (e.g., Heterocypris incongruens, H. salina, Ilyocypris gibba, Prionocypris zenkeri, Cypridopsis vidua) are expected to show significant associations with specific environmental conditions and assemblages, indicating their potential as ecological indicators. The rationality here is that these species are expected to reflect distinct habitat characteristics (e.g., salinity, pH, water temperature, and habitat permanence). This hypothesis is based on the idea that ostracod species distributions are largely governed by their ecological tolerances and habitat preferences. Testing these relationships provides insight into the predictability of ostracod–environment associations in Mediterranean-climate basins and evaluates the potential of these taxa as effective bioindicators for assessing ecological integrity and habitat heterogeneity in inland waters of southern Türkiye. Based on these two hypotheses, the aims of this study were to (1) conduct the first extensive field survey of non-marine ostracod species in Adana, including the collection of ecological and geographical data in situ, and (2) apply both SAM and IndVal analyses to distinguish archetypes with the indicator species recorded in the present study.

Materials and methods

Site description

Adana Province, bordering the Mediterranean Sea to the south, is characterized by a typical Mediterranean climate, with hot, dry summers and mild, rainy winters. Precipitation in the region generally results from orographic rainfall and the interaction of moving air masses, with an average annual total of 625 mm where the last about 40 years mean annual temperature is 8.7 °C (Çevre Şehircilik ve İklim Değişikliği Bakanlığı, 2024). The Seyhan (506 km) and Ceyhan (509 km) Rivers, which are the two largest rivers of the Mediterranean Region, flow through the territory of Adana and play a crucial role in the region’s hydrology and irrigation systems.

Sampling methodology

Between 29 October and 4 November 2024, a total of 120 aquatic sites were randomly visited across Adana (13,844 km² surface area) to assess non-marine ostracod species composition (Figure 1). Prior to sediment sampling, site-specific geographical information (e.g., coordinates and elevation) was recorded in the field using a Garmin GPS device (GARMIN eTrex 30). Each site was also photographed for future reference. To avoid disturbance of the sampling site, water samples and measurements were taken prior to sediment sampling. It is noted that our sampling represents a temporal snapshot of the system and does not account for seasonal variability. At each site, 100 mL of water was taken in plastic bottles for major ion analysis, immediately placed in ice boxes, and later analyzed according to standard protocols provided in APHA (2012) using a Dionex ICS-1100 ion chromatography system at the Technical University of Eskişehir (Türkiye). A multiparameter probe (Hanna HI-9829) was employed on site to record various physicochemical variables, including pH, water temperature (Tw, °C), salinity (psu), dissolved oxygen (DO, mg/L), percent dissolved oxygen (%DO), electrical conductivity (EC, µS/cm), total dissolved solids (TDS, mg/L), water transparency or turbidity (NTU) and atmospheric pressure (mmHg). Air temperature (Ta, °C) was determined using a Kestrel-3000 anemometer.

FIGURE 1

Approximately 200 mg of sediment samples were collected from about a depth of 30 cm of shallow waters with a phytoplankton hand net (150 mm mesh size) in each site and kept in 70% ethanol. These samples were filtered under three standard sieves (1.0; 0.5; 0.25 mm mesh sizes) in the laboratory. Specimens were separated from the sediment with the aids of fine needles and kept in 70% ethanol. Adult individuals representing each taxonomic group were dissected in the lactophenol solution for taxonomic descriptions. Species were identified using a light microscope (Olympus BX-51) following the taxonomic keys of Meisch (2000) and Karanovic (2012). For more detailed examination, the morphological characteristics of carapaces and valves were investigated using a Zeiss EVO50 scanning electron microscope (SEM) at the Department of Geological Engineering, Hacettepe University (Türkiye). All specimens and prepared slides were systematically labeled, catalogued, and deposited in the Laboratory of Limnology, Department of Biology, Bolu Abant İzzet Baysal University (Türkiye).

Statistical analyses

For the statistical analyses, 58 sites containing at least one recorded species were selected, encompassing 35 species in total. We only used adult individuals with complete carapaces and soft body parts. Ecological archetypes were identified using the Species Archetype Model (SAM), which groups species with similar occurrence patterns across sites, thereby highlighting distinct assemblage structure (Cutler and Breiman, 1994; Dunstan et al., 2011). Hence, species archetype refers to the species with similar responses to environmental gradients (Dunstan et al., 2011). Following archetype identification, Indicator Species Analysis (IndVal) was conducted using the multipatt function from the R package indicspecies to detect species strongly associated with specific site groups (De Cáceres and Legendre, 2009). IndVal calculates the specificity and fidelity of species to site groups, with statistical significance assessed via permutation tests (n = 999). The integration of SAM, IndVal, and environmental data allowed for a robust evaluation of species assemblages, identification of indicator species, and ecological interpretation of patterns across the sampled freshwater sites. All analyses were performed in R (version 4.3.0) (R Core Team, 2023) within RStudio (RStudio Team, 2023), using the packages vegan (Oksanen et al., 2023), reshape2 (Wickham, 2007), ggpubr (Kassambara, 2023), RColorBrewer (Neuwirth, 2022), and indicspecies (De Cáceres and Legendre, 2009).

Species assemblages were compared along with environmental gradients to test the effects of environmental variables and habitat permanence. Hence, we used Unweighted Pair Group Method with Arithmetic Mean (UPGMA) Bray-Curtis Hierarchical clustering analyses to illustrate similarities among the sampling sites. Besides, to elaborate (i) relationship among the species and variables measured in situ and (ii) display the most effective variables on species occurrences, we used multivariate analyses such as non-metric multidimensional scaling (NMDS) and Canonical Correspondence Analyses (CCA). Before CCA, Detrended Correspondence Analysis (DCA) was run to confirm the suitability of our data for CCA (ter Braak and Šmilauer, 1998). Accordingly, the lengths of gradient (6.299) supported suitability of the data for CCA. Then after, CCA along with Monte Carlo permutation test (499 permutations) was used for those species that occurred in two or more sites. To avoid multicollinearity, inflation factors (VIF) were compared for the variables. During the CCA process, rare species were excluded, and data was log-transformed. Rare species (occurring in fewer than two sites) were excluded to reduce noise and improve the stability and interpretability of multivariate analyses. CANOCO 4.5 program was used (ter Braak and Verdonschot, 1995) for these analyses. Ternary plots in the PAST program were provided to exhibit chemical (and ionic) characteristics of the habitats (Hammer et al., 2001). Ecological tolerance and optimum values for each species were calculated using a weighted averaging regression transfer function in the C2 program (Juggins, 2003).

Results

The physicochemical characteristics of the 120 water bodies in Adana exhibited a wide ecological gradient, ranging from fresh to highly saline conditions (Figure 2). Water pH values (7.4–10.7, mean = 8.51 ± 0.56) indicated generally alkaline environments typical of carbonate-rich basins, reflecting strong buffering capacity and intense photosynthetic activity in some shallow systems. Dissolved oxygen concentrations (1.3–12.7 mg L⁻¹) and oxygen saturation values (14%–132%) varied considerably among sites, from hypoxic conditions in warm or stagnant waters to supersaturation in productive habitats, suggesting high temporal and spatial heterogeneity in metabolic processes. Conductivity values (33–8654 μS cm⁻¹) spanned several orders of magnitude, revealing the coexistence of oligohaline to hypersaline ecosystems within the study area. Water temperatures ranged between 7.9 °C and 24.8 °C (mean = 14.6 °C), while oxidation–reduction potentials (mean = 187 mV) were generally positive, indicating predominantly oxidizing conditions. Ionic composition was dominated by calcium (mean = 50.5 mg L⁻¹) and magnesium (41.9 mg L⁻¹), characteristic of limestone geology, although sodium (19.3 mg L⁻¹) and strontium (0.45 mg L⁻¹) were also notable in saline and evaporative basins. Trace metals such as chromium (0.004 mg L⁻¹) and manganese (0.003 mg L⁻¹) remained at background concentrations. Turbidity varied widely (0–317 NTU), reflecting differences in substrate type and organic matter content. Overall, these gradients in alkalinity, ionic strength, and redox conditions underscore the strong environmental heterogeneity of Adana’s inland aquatic habitats, which is consistent with the pronounced species–environment structuring observed in ostracod assemblages.

FIGURE 2

We found 53 ostracod taxa (35 species and 18 subfossils (empty carapace or valves)) from 11 different aquatic habitats (lake, reservoir, dam, pond, ditch, waterfall, river, creek, spring, trough, canal). SEM photographs of the species are presented in Supplementary Appendix 1. Ostracod communities across 120 aquatic sites in Adana (southern Türkiye) exhibited distinct ecological structuring captured by ordination and species archetype modeling. The NMDS ordination based on species composition revealed a structured distribution of sampling sites in two-dimensional space (k = 2; stress = 0.188; Figure 3). The arrangement of sites indicates the presence of four main ecological assembled patterns associated with differences in dominant ostracod taxa. Sites characterized by the dominance of Heterocypris incongruens were broadly distributed across the ordination space, indicating a wide ecological tolerance. In contrast, sites dominated by Cypridopsis vidua were positioned toward one side of the ordination, suggesting a more constrained distribution. Similarly, H. salina showed a relatively distinct placement, while sites associated with Prionocypris zenkeri occupied intermediate to partially separated positions. Among cations, calcium was dominant ion (Ca > Mg > Na), while water temperature, pH, and dissolved oxygen were among the most influential environmental variables (Figure 2). Environmental boxplots showed consistent differences in conductivity, temperature, and pH among archetypes, particularly high conductivity in Saline sites and low values in Freshwater ones, highlighting the association of physicochemical gradients with differences among assemblages (Supplementary Appendix 2).

FIGURE 3

The Species Archetype Model (SAM) indicated an optimal solution of four archetypes (k = 4) based on the residual-sum-of-squares screeplot, which showed a rapid decline up to k = 4 followed by a plateau (Figure 4; Table 1). Although higher solutions (k = 5–6) revealed additional sub-structuring, they contributed only minor improvements and were therefore not retained as the primary model. Archetype assignments based on species occurrences were uneven, with 53 sites grouped in Archetype 1 and the remaining archetypes representing environmentally unique or restricted conditions. Integrated with Indicator Species Analysis (IndVal), these archetypes defined coherent ecological units (Table 1). Archetype 1 (Generalist Assemblage) was dominated by H. incongruens, a tolerant species occurring across varied habitats. Archetype 2 (Freshwater/Slightly Saline Specialists) was characterized by Ilyocypris gibba (IndVal = 0.80, p = 0.02), H. salina (0.65, p = 0.03), and C. vidua (0.86, p = 0.02), all showing statistically significant associations. Archetype 3 (Transitional Group) exhibited intermediate conditions and weak species associations, suggesting ecological overlap with Archetype 1. Archetype 4 (Sensitive/Restricted Assemblage) represented stable, slightly alkaline, and well-oxygenated waters where most taxa were scarce, while Archetypes 5–6 (Stress-Tolerant and Secondary Mixed Groups) were defined by low species richness and weak indicator values, likely reflecting resource-poor or environmentally stressed habitats. Mean salinity ranged from 1.5 psu (Archetype 3) to 7.1 psu (Archetype 2), confirming pronounced environmental segregation. Heatmaps of standardized abundances supported these trends: H. incongruens dominated generalized conditions, H. salina and P. zenkeri prevailed in saline environments, and I. gibba appeared sporadically, consistent with its narrower niche (Figure 5). IndVal results corroborated the diagnostic role of these species in distinguishing ecological states.

FIGURE 4

FIGURE 5

Gradient analyses further reinforce these patterns. DCA indicated a gradient length of 6.299, signifying high species turnover and justifying a unimodal model. CCA identified pH, dissolved oxygen, conductivity, temperature, and elevation as the five major factors associated with variation in species composition, explaining 63.6% of cumulative variance (Figure 6; Table 2). The UPGMA dendrogram grouped sites with similar environmental and taxonomic characteristics, supporting ordination-based clustering (Figure 7). Overall, species with cosmoecious characteristics exhibited relatively wide ecological tolerances but lower optimum values to different variables (Supplementary Appendix 3).

FIGURE 6

FIGURE 7

Collectively, these results indicate that ostracod assemblages in Adana’s aquatic systems vary along conductivity and pH gradients. The combined SAM–IndVal–CCA framework highlights both the diagnostic potential of indicator species and consistent patterns of community organization, providing a solid ecological baseline for biomonitoring and conservation of freshwater and brackish habitats in the region.

Discussion

There are so far five nonmarine living ostracods (I. gibba, H. incongruens, E. virens, Neglecandona neglecta, E. hamadanensis) reported from Adana (Gülen, 1988). Except E. hamadanensis, first four species were already found in the present study. Thus, total numbers of nonmarine living ostracods increased to 36 species (Supplementary Appendix 1). Comparing to the earlier studies (Malatya (12313 km², 25 spp.) (Batmaz et al., 2020), Eskişehir (13960 km², 36 spp.) (Batmaz et al., 2025), Kastamonu (13136 km², 20 spp.) (Külköylüoğlu et al., 2024), Mersin (15853 km², 36 spp.) (Dalgakıran et al., 2020), Muğla (12974 km², 24 spp.) (Akdemir et al., 2020)), this number portrays relatively high species richness and diversity of the province. In contrast, fossil records do not show such richness for the nonmarine fauna. Indeed, there are so far five nonmarine species (Cyprideis torosa, Eucypris (syn: Arctocypris) dulcifons, Loxoconcha rhomboidei, I. gibba) (Avşar et al., 2006) and D. stevensoni (Nazik et al., 1992; 1999) known from the fossil records. Of which, C. torosa, I. gibba and D. stevensoni were also found during the present study. Two possible reasons for scarcity of nonmarine fossil forms can be considered as (1) the paleoenvironmental changes when sea waters influence (i.e., The Messinian salinity crisis, a geological phenomenon that occurred between 5.97 and 5.33 mya) (Garcia-Castellanos and Villasenor, 2011) were dominant on the formation of freshwater habitats, and (2) lack of studies within the region. Indeed, except those stated here, there are no extensive paleontological studies and contemporary studies in the area. Accordingly, our knowledge about fossil forms is limited. The ecological gradients structuring modern ostracod assemblages in Adana partially mirror those documented in fossil records, implying possibility of a long-term continuity of adaptive strategies (see Nazik et al., 1992; Darbaş et al., 2008). Late Miocene–Early Pliocene salinity oscillations in Paratethyan basins promoted diversification of euryhaline lineages such as Cyprideis and Candona (Rausch et al., 2020), while non-marine faunas from the central Apennines (Spadi et al., 2019), Japan (Ozawa et al., 2008), and the North Atlantic (Wood, 2009) reveal comparable ecological partitioning. These fossil parallels suggest that generalist tolerance, salinity specialization, and resilience to ionic stress were already established by the Pliocene, when fluctuating hydrology selected for broad physiological versatility. Accordingly, the archetypes identified in modern Adana waters represent present-day expressions of long-standing evolutionary pathways shaped by Pliocene environmental instability.

Ostracod communities across Adana seem to display clear ecological structuring associated with conductivity, pH, and temperature gradients. These variables have long been identified as major determinants of non-marine ostracod distribution, as they are closely related to osmoregulation, shell calcification, and reproductive success. Comparable environmental control has been reported from Mediterranean aquatic bodies (Mezquita et al., 1999a; b; Rossetti et al., 2004; Pieri et al., 2009), semi-arid steppe systems (Frenzel and Boomer, 2005), and Anatolian water bodies such as lakes and springs (Çapraz et al., 2022). The predominance of calcium among major ions reported in the present study parallels ionic hierarchies observed in the waters of arctic polygon ponds in north-east Siberia (Schneider et al., 2016), in the highland lakes in east-central Mexico (Pérez et al., 2015), in central Anatolia that Ca–HCO₃-dominated chemistry stabilizes carapace formation of the species (Wansard and Mezquita, 2001; McCormack et al., 2019) and supports broad ecological tolerances of the species (e.g., cosmoecious species) (Külköylüoğlu, 2013). Such gradients are associated with the partitioning of communities into distinct hydrochemical domains, producing the major assemblage types revealed by ordination and archetype modeling. The NMDS ordination revealed clear differences in species composition among sampling sites, with assemblages associated with different dominant taxa. These patterns indicate that ostracod communities are not strictly separated into discrete groups. Instead, the ordination suggests gradual transitions among assemblages, reflecting underlying environmental gradients that structure species distributions across sites. Hence, as indicated above, the SAM identified four major ecological archetypes that summarize community–environment relationships among the Adana’s aquatic systems. The largest and most generalized group, dominated by H. incongruens, represents a tolerant assemblage occupying varieties of lentic habitats with wide ranges of conductivity and temperature. Hence, finding an increasing trend toward archetype 1 is consistent with our second hypothesis (Figures 3, 4; Table 1).

These species (i.e., cosmopolitan or cosmoecious) exhibited wider tolerance values to various ecological variables than the species with rare occurrences (Supplementary Appendix 3). Such high tolerances may provide an advantage for species survival. The ubiquity of this species in both ephemeral and permanent water bodies across Europe and western Asia (Meisch, 2000; Zhai and Zhao, 2014; Akita et al., 2016) is consistent with its physiological adaptability and reproductive versatility. A second group, characterized by I. gibba and H. salina, corresponds to freshwater to slightly saline conditions, commonly found in shallow temporary pools. Similar species associations have been reported from the Aegean and Paratethyan regions, where fluctuating ionic regimes favor moderately halotolerant taxa (Nazik et al., 2008; Rausch et al., 2020; Hoehle and Wrozyna, 2022). Transitional and restricted communities, which include C. vidua and P. zenkeri, appear confined to microhabitats with distinctive oxygen or organic profiles (Külköylüoğlu et al., 2012; Escrivà et al., 2014). By contrast, the least diverse archetypes likely represent stressed environments, such as saline depressions or nutrient-poor ponds, where only a few resistant species persist. These findings are in accordance with the studies of Schornikov et al. (2014) and Hong et al. (2017), who reported a significant decrease in the number of ostracod species with declining water quality, where only a few opportunistic (e.g., cosmopolitan or cosmoecious) species persisted.

Indicator Species Analysis (IndVal) confirmed that several species can serve as diagnostic markers for environmental states within the region. Ilyocypris gibba exhibited the highest fidelity to low-salinity sites, while H. salina was associated with saline and mesohaline conditions (Rosenfeld and Ortal, 1983; Frenzel and Boomer, 2005), consistent with its broad osmotic tolerance. As indicated above, C. vidua showed a good interpretable association with moderately enriched habitats, a trend also noted in northern Anatolian assemblages (Külköylüoğlu, 2005). The species, like those mentioned above, is considered a cosmoecious species with a wide geographical distribution and relatively high ecological tolerances to different environmental variables (Külköylüoğlu, 2013). It is very common in lentic habitats (e.g., lakes, reservoirs, ponds) and is known to be resistant to pesticides, heavy metals, and organic pollution in European waters as well (Kantorek, 1992; Rossi et al., 2003; Martínez-García et al., 2015). These results indicate that even ostracods with cosmopolitan characteristics maintain ecological specificity when examined within a regional environmental framework. The combined SAM–IndVal approach therefore provides complementary perspectives: SAM identifies groups of species with shared responses, while IndVal statistically validates their habitat fidelity.

This perspective of using the combination of these two approaches is supported with the multivariate analyses. For example, as indicated above, ordination analyses further underscored the strong association of environmental gradients with species distributions. Although not significant (p > 0.05), CCA was able to explain 63.6% of the cumulative variance through pH, conductivity, dissolved oxygen, temperature, and elevation (Figure 6; Table 2). Hence, our first hypothesis is consistent with these results, where species located closer to the center of the CCA diagram are mainly species with cosmopolitan characteristics, suggesting relatively high tolerance levels to those variables used in the CCA. The ordination analyses support the conclusion that the archetypes are not only compositionally distinct but also environmentally segregated. Sites within each archetype cluster together (Figure 7), indicating similarity in environmental conditions, whereas distances between archetypes suggest clear ecological separation. Similar supportive results have already been reported in several contemporary field works. For example, Akita et al. (2016) reported 11 ostracods from a deep brackish lake, Tangra Yumco (ca. 4595 m a.s.l.) and water bodies located on the southern Tibetan Plateau. These authors showed seven key indicator species assemblages from the (1) Ca-depleted brackish waters with ubiquitous distribution, (2) fresh to brackish waters, (3) mesohaline to polyhaline waters, (4) exclusively brackish-lacustrine deep waters, (5) freshwaters, (6) shallow temporary waters, and (7) ponds. Their results also revealed that distribution and abundance of species were significantly associated with physico-chemical variables such as conductivity and habitat types. Similarly, Külköylüoğlu et al. (2018), based on common characteristics (e.g., habitat requirements, ecological preferences), indicated three clustering groups of 23 species collected from 11 different water bodies located at 460–1380 m a.s.l. Additional evidence was provided by Külköylüoğlu et al. (2024), who recognized five commonly known diagnostic species (N. neglecta, C. vidua, H. incongruens, I. bradyi, P. olivaceus) for specific habitats based on their habitat selection. Similarly, all species were encountered during the present study, where two species (H. incongruens, C. vidua) were listed with indicator values (Table 1).

The variables have also been highlighted as key predictors in recent multi-regional datasets (Li et al., 2021; Hoehle and Wrozyna, 2022). The dominance of conductivity and pH supports the view that ionic strength and carbonate equilibria are primary axes of ostracod community organization, whereas temperature and oxygen modulate diversity through their effects on metabolism and microhabitat stability. The relatively high proportion of explained variance suggests that environmental gradients play an important role in structuring the assemblages, although stochastic or dispersal-related processes cannot be excluded in the Adana system. However, this interpretation should be considered with caution, as the present study does not include biotic interactions, habitat structure, or temporal variability, which may also contribute to community assembly. Accordingly, these factors may also contribute to species-specific habitat preferences. Indeed, our results revealed that certain habitats (e.g., natural springs or related habitats) and saline depressions (e.g., shallow water bodies) within the study area host environmentally distinct communities. These sites contain stress-tolerant or restricted species with limited distribution elsewhere, indicating a high degree of ecological singularity. Similar patterns have been reported in other studies, although such comparisons should be interpreted cautiously (Plotnikov et al., 2021). Because these habitats support highly specialized faunas adapted to extreme chemical (e.g., ionic concentration) and thermal conditions (e.g., elevated water temperature, reduced dissolved oxygen), they warrant high conservation priority. Their degradation or loss through processes such as groundwater extraction, agricultural runoff, habitat modification (e.g., residential development, road construction), or desiccation would likely lead to disproportionate declines in regional biodiversity and the loss of unique evolutionary lineages (Hutchins et al., 2025).

Integrating SAM, IndVal, and multivariate analyses (e.g., CCA, NMDS) offers a comprehensive framework for interpreting ostracod ecology across temporal and spatial scales. By quantifying co-response patterns, validating indicator taxa, and linking them explicitly to environmental gradients, this approach advances beyond descriptive faunal lists toward a framework for describing and interpreting ecological patterns relevant to conservation. The patterns identified here reinforce the dual message emerging from current knowledge of species’ habitat preferences: ostracod assemblages are highly sensitive to hydrochemical change, yet remarkably consistent in their ecological organization through time, although this requires further confirmation. These insights provide an essential baseline for monitoring the effects of ongoing salinization and climatic variability in aquatic ecosystems of southern Türkiye. However, this approach requires additional data on the ecology, biology, and taxonomic characteristics of living ostracods to better reconstruct past environmental conditions. Overall, the SAM–IndVal approach provides a robust framework for detecting ecological patterns in ostracod communities. By linking species composition to environmental gradients, the analysis highlights how both generalist and specialist species shape community structure, and how environmental variables such as salinity and temperature are associated with differentiation among habitats. This framework has practical applications for monitoring ecological health, detecting environmental change, and understanding community assembly in freshwater and brackish systems.

Conclusion

Ostracod assemblages across Adana’s inland aquatic systems show clear ecological organization associated with conductivity, pH, and temperature gradients. The combined use of SAM, IndVal, and multivariate ordination techniques successfully identified distinct ecological archetypes representing generalist, freshwater, saline, and restricted assemblages. Calcium-dominated water chemistry and varying environmental conditions were associated with variation in species distributions and community structure. The results indicate that the observed patterns are consistent with environmental structuring in explaining species coexistence and habitat differentiation, although stochastic processes cannot be excluded. Overall, the integration of community modeling and environmental analyses provides a useful framework for understanding species–environment relationships in non-marine ostracods and illustrates how hydrochemical variability is associated with ecological diversification within Mediterranean-type inland waters. Furthermore, the analyses are based on in situ environmental variables, and broader landscape or hydrological factors were not explicitly incorporated. The present study is therefore based on pattern-based analyses, and no formal tests of community assembly processes (e.g., null models or variance partitioning) were conducted. Interpretations related to environmental filtering should be considered as inferential and not as direct evidence of underlying ecological mechanisms. It should also be noted that habitats can be influenced by a variety of abiotic and biotic factors. This study does not include biotic factors and is limited to the variables measured in situ. Therefore, the conclusions presented here are primarily based on environmental associations and may not fully reflect the combined influence of other ecological processes. Future studies should address this gap.

References

• 1

  AkdemirD.KülköylüoğluO.YavuzatmacaM.TanyeriM.GürerU.AlperA.et al (2020). Ecological characteristics and habitat preferences of Ostracoda (crustacea) with a new bisexual population record (Muğla, Turkey). Appl. Ecol. Environ. Res.18, 1471–1487. 10.15666/aeer/1801_14711487

  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  AkitaL. G.FrenzelP.WangJ.BornerN.PengP. (2016). Spatial distribution and ecology of the recent Ostracoda from Tangra Yumco and adjacent waters on the southern Tibetan Plateau: a key to palaeoenvironmental reconstruction. Limnologica59, 21–43. 10.1016/j.limno.2016.03.005

  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  AnadónP.De DeckkerP.JuliàR. (1986). The Pleistocene lake deposits of the NE Baza Basin (Spain): salinity variations and ostracod succession. Hydrobiologia143, 199–208. 10.1007/BF00026662

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  APHA (2012). Standard Methods for the Examination of Water and Wastewater. 22nd Edn.Washington, DC: American Public Health Association, American Water Works Association, Water Environment Federation.

  • Google Scholar
  • View reference in article

• 5

  AvşarN.NazikA.DinçerF.DarbaşG. (2006). Adana Havzası Kuzgun formasyonunun mikrofosiller ile ortamsal yorumu. Hacet. Univ. J. Earth Sci. Appl. Res. Cent.27, 1–21.

  • Google Scholar
  • View reference in article

• 6

  BatmazF.KülköylüoğluO.AkdemirD.YavuzatmacaM. (2020). Effective roles of ecological factors on non-marine Ostracoda (Crustacea) in shallow waters of Malatya (Turkey). Ecol. Res.35, 1–13. 10.1111/1440-1703.12120

  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  BatmazF.KülköylüoğluO.YavuzatmacaM. (2025). Distribution, co-occurrence, and habitat characteristics of non-marine ostracods (Crustacea) in lentic and lotic waters of Eskişehir Province, Türkiye. Anim. Taxon. Ecol.71, 652–669. 10.1556/1777.2025.00074

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  ÇaprazÇ.KülköylüoğluO.YavuzatmacaM. (2022). Species archetype models: linking species distributions to environmental gradients. Int. J. Limnol58, 3. 10.1051/limn/2022002

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  Çevre, Şehircilik ve İklim Değişikliği Bakanlığı (2024). Adana İl Çevre, Şehircilik ve İklim Müdürlüğü: iletişim bilgileri – Adana Valiliği.

  • Google Scholar
  • View reference in article

• 10

  CutlerA.BreimanL. (1994). Archetypal analysis. Technometrics36, 338–347. 10.1080/00401706.1994.10485840

  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  DalgakıranE.KülköylüoğluO.YavuzatmacaM.AkdemirD. (2020). Correlational analyses of the relationships between altitude and carapace size of Ostracoda (Crustacea). Int. J. Lim.56, 2. 10.1051/limn/2019025

  • CrossRef
  • Google Scholar

• 12

  DarbaşG.NazikA.TemelA.GürbüzK. (2008). A paleoenvironmental test of the Messinian Salinity Crisis using Miocene–Pliocene clays in the Adana Basin, southern Turkey. Appl. Clay Sci.40, 108–118. 10.1016/j.clay.2007.09.007

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  De CáceresM.LegendreP. (2009). Associations between species and groups of sites: indices and statistical inference. Ecology90, 3566–3574. 10.1890/08-1823.1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  DelormeL. D. (1991). “Ostracoda,” in Ecology and Classification of North American Freshwater Invertebrates. Editors ThorpJ. H.CovichA. P. (San Diego, CA: Academic Press), 695–726.

  • Google Scholar
  • View reference in article

• 15

  Dole-OlivierM. J.GalassiD. M. P.MarmonierP.Creuzé des ChatelliersM. (2000). The biology and ecology of lotic microcrustaceans. Freshw. Biol.44, 63–91. 10.1046/j.1365-2427.2000.00590.x

  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  DolédecS.ChesselD.Gimaret-CarpentierC. (2000). Niche separation in community analysis: a new method. Ecology81, 2914‐2927.

  • Google Scholar
  • View reference in article

• 17

  DunstanP. K.FosterS. D.DarnellR. (2011). Model-based grouping of species across environmental gradients. Ecol. Model.222, 955–963. 10.1016/j.ecolmodel.2010.11.030

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  EscrivàA.RuedaJ.ArmengolX.Mesquita-JoanesF. (2014). Artificial dam lakes as suitable habitats for exotic invertebrates: Ostracoda ecology and distribution in reservoirs of the eastern Iberian Peninsula. Knowl. Manag. Aquat. Ecosyst.412, 9. 10.1051/kmae/2014008

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  FrenzelP.BoomerI. (2005). The use of ostracods from marginal marine, brackish waters as bioindicators of modern and Quaternary environmental change. Palaeogeogr. Palaeoclimatol. Palaeoecol.225, 68–92. 10.1016/j.palaeo.2004.02.051

  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  Garcia-CastellanosD.VillaseñorA. (2011). Messinian Salinity Crisis regulated by competing tectonics and erosion at the Gibraltar Arc. Nature480, 359–363. 10.1038/nature10651

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  GülenD. (1988). Contribution to the taxonomy of the freshwater ostracod fauna of Turkey. Istanb. Univ. Su Urun. Derg.1, 199–203.

  • Google Scholar
  • View reference in article

• 22

  HammerØ.HarperD. A. T.RyanP. D. (2001). PAST: paleontological statistics software package for education and data analysis. Palaeontol. Electron.4, 9. Available online at: http://palaeo-electronica.org/2001_1/past/issue1_01.htm.(Accessed February 10, 2025).

  • Google Scholar
  • View reference in article

• 23

  HoehleM.WrozynaC. (2022). Spatio-temporal distribution of ostracod species in saline inland lakes (Mansfeld Lake area, central Germany). PeerJ10, e13668. 10.7717/peerj.13668

  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  HongY. Y.YasuharaM.IwataniH.SetoK.YokoyamaY.YoshiokaK.et al (2017). Freshwater reservoir construction by damming a marine inlet in Hong Kong: paleoecological evidence of local community change. Mar. Micropaleontol.132, 53–59. 10.1016/j.marmicro.2017.04.003

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  HutchinsB. T.SchwartzB. F.KülköylüoğluO. (2025). The conservation status of Texas groundwater Ostracoda. Tex. J. Sci.77, 1–17. 10.32011/txjsci_77_1_Article2

  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  JugginsS. (2003). Software for Ecological and Palaeoecological Data Analysis and Visualization: C2 User Guide. Newcastle upon Tyne: University of Newcastle.

  • Google Scholar

• 27

  KantorekJ. (1992). Ekologické Hodnocení Cenóz Lasturnatek (Ostracoda) a Možnosti Jeho Využití Pro Bioindikaci Kvality Vody. Bratislava: Přírodovědecká fakulta Univerzity Komenského. PhD thesis.

  • Google Scholar

• 28

  KaranovicI. (2012). Recent Freshwater Ostracods of the World: Crustacea, Ostracoda, Podocopida. Berlin: Springer-Verlag. 10.1007/978-3-642-21810-1

  • CrossRef
  • Google Scholar

• 29

  KassambaraA. (2023). ggpubr: “Ggplot2” Based publication-ready Plots (Version 0.6.0). Available online at: https://CRAN.R-project.org/package=ggpubr. (Accessed February 10, 2025).

  • Google Scholar

• 30

  KülköylüoğluO. (2005). Factors affecting the occurrence of Ostracoda (Crustacea) in the Yumrukaya reedbeds (Bolu, Turkey). Wetlands25, 224–227. 10.1672/0277-5212(2005)025[0224:fatooo]2.0.co;2

  • CrossRef
  • Google Scholar

• 31

  KülköylüoğluO. (2013). Diversity, distribution and ecology of non-marine Ostracoda (Crustacea) in Turkey: application of pseudorichness and cosmoecious species concepts. Recent Res. Dev. Ecol.4, 1–18.

  • Google Scholar

• 32

  KülköylüoğluO.YavuzatmacaM.AkdemirD.SarıN. (2012). Distribution and local species diversity of freshwater Ostracoda in relation to habitat in the Kahramanmaraş Province of Turkey. Int. Rev. Hydrobiol.97, 247–261. 10.1002/iroh.201111490

  • CrossRef
  • Google Scholar

• 33

  KülköylüoğluO.YavuzatmacaM.AkdemirD.ÇelenE.DalkıranN. (2018). Ecological classification of freshwater Ostracoda (Crustacea) based on physicochemical properties of waters and habitat preferences. Ann. Limnol. - Int. J. Limnol.54, 26. 10.1051/limn/2018017

  • CrossRef
  • Google Scholar

• 34

  KülköylüoğluO.AkdemirD.YavuzatmacaM. (2020). Non-marine Ostracoda (Crustacea) as indicator species of habitat types. Aquat. Ecol.54, 519–533. 10.1007/s10452-020-09757-x

  • CrossRef
  • Google Scholar

• 35

  KülköylüoğluO.YavuzatmacaM.TunoğluC.TunçkolB. (2024). Determining diagnostic characteristics of non-marine Ostracoda (Crustacea) and their habitat selection with unique functional traits in the Kastamonu Province (Türkiye). Limnologica108, 126199. 10.1016/j.limno.2024.126199

  • CrossRef
  • Google Scholar

• 36

  LiX.ZhaiD.WangQ.WenR.JiM. (2021). Depth distribution of ostracods in a large freshwater lake on the Qinghai–Tibet Plateau and its ecological and palaeolimnological significance. Ecol. Indic.129, 108019. 10.1016/j.ecolind.2021.108019

  • CrossRef
  • Google Scholar

• 37

  MaddocksR. (1982). Ostracoda, in The Biology of Crustacea, Vol. 1: Systematics, the Fossil Record and Biogeography. Editor AbeleL. G. (New York: Academic Press), 221–239.

  • Google Scholar

• 38

  Martínez-GarcíaB.Suarez-HernandoO.MendicoaJ.MurelagaX. (2015). Living ostracod species from permanent and semi-permanent ponds of Bardenas Reales de Navarra (northern Spain) with remarks on their ecological requirements. Ameghiniana52, 598–612. 10.5710/amgh.17.07.2015.2895

  • CrossRef
  • Google Scholar

• 39

  McCormackJ.ViehbergF.AkdemirD.ImmenhauserA.KwiecienO. (2019). Ostracods as ecological and isotopic indicators of lake hydrology and hydrochemistry. the Lake Van example. Biogeosciences16, 2095–2111. 10.5194/bg-16-2095-2019

  • CrossRef
  • Google Scholar

• 40

  MeischC. (2000). “Freshwater Ostracoda of Western and central Europe,″ in Süßwasserfauna von Mitteleuropa 8/3. Editor Chwoerbel, J. P. Z. (Heidelberg, Berlin:Spektrum Akademischer Verlag), 1‐522.

  • Google Scholar

• 41

  MezquitaF.TapiaG.RocaJ. R. (1999a). Ostracoda from springs on the eastern Iberian Peninsula: ecology, biogeography and palaeolimnological implications. Palaeogeogr. Palaeoclimatol. Palaeoecol.148, 65–85.

  • Google Scholar

• 42

  MezquitaF.GriffithsH. I.SanzaS.SoriaJ. M.PiñónA. (1999b). Ecology and distribution of ostracods associated with flowing waters in the eastern Iberian Peninsula. J. Crustac. Biol.19, 344–354. 10.2307/1549241

  • CrossRef
  • Google Scholar

• 43

  NazikA.ŞafakÜ.ŞenolM. (1992). Micropaleontological investigation of the Pliocene sequence of the Tufanbeyli (Adana) areaGeosound, Spec. Issue. 56–72.

  • Google Scholar

• 44

  NazikA.EvansG.GürbüzK. (1999). Sedimentology and paleontology with special reference to the Ostracoda fauna of Akyatan Lagoon (Adana, SE Turkey). Geosound35, 127–147.

  • Google Scholar

• 45

  NazikA.Türkmenİ.KoçC.AksoyE.AvşarN.YayıkH. (2008). Fresh and brackish water ostracods of Upper Miocene deposits, Arguvan/Malatya (Eastern Anatolia). Turk. J. Earth Sci.17, 481–495.

  • Google Scholar

• 46

  NeuwirthE. (2022). RColorBrewer: Colorbrewer Palettes (Version 1.1–3). Available online at: https://CRAN.R-project.org/package=RColorBrewer. (Accessed February 10, 2025).

  • Google Scholar

• 47

  OakleyT. H.WolfeJ. M.LindgrenA. R.ZaharoffA. K. (2013). Phylotranscriptomics to bring the understudied into the fold: monophyletic Ostracoda, fossil placement, and pancrustacean phylogeny. Mol. Biol. Evol.30, 215–233. 10.1093/molbev/mss216

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  OksanenJ.SimpsonG. L.BlanchetF. G.KindtR.LegendreP.MinchinP. R.et al (2023). Vegan: Community Ecology Package (Version 2.6–4). Available online at: https://CRAN.R-project.org/package=vegan. (Accessed February 10, 2025).

  • Google Scholar

• 49

  OzawaH.NagamoriH.TanabeK. (2008). Pliocene ostracods (Crustacea) from the Togakushi area, central Japan: palaeobiogeography of trans-Arctic taxa and Japan Sea endemic species. J. Micropalaeontol.27, 161–175. 10.1144/jm.27.2.161

  • CrossRef
  • Google Scholar

• 50

  PérezL.Lozano-GarcíaS.CaballeroM. (2015). Non-marine ostracodes from highland lakes in east-central Mexico. Rev. Biol. Trop.63, 401–425. 10.15517/rbt.v63i2.15240

  • CrossRef
  • Google Scholar

• 51

  PieriV.MartensK.StochF.RossettiG. (2009). Distribution and ecology of non-marine ostracods (Crustacea, Ostracoda) from Friuli Venezia Giulia (NE Italy). J. Limnol.68, 1–15. 10.4081/jlimnol.2009.1

  • CrossRef
  • Google Scholar

• 52

  PlotnikovI.SmurovA.AladinN. (2021). Large saline lakes of Central Asia. J. Arid. Land Stud.31, 29–44. 10.14976/jals.31.2_29

  • CrossRef
  • Google Scholar

• 53

  R Core Team (2023). R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. Available online at: https://www.R-project.org/ (Accessed February 10, 2025).

  • Google Scholar

• 54

  RauschL.StoicaM.LazarevS. (2020). A Late Miocene–Early Pliocene Paratethyan-type ostracod fauna from the Denizli Basin (SW Anatolia) and its palaeogeographic implications. Acta Palaeontol. Romaniae16, 3–56. 10.35463/j.apr.2020.02.01

  • CrossRef
  • Google Scholar

• 55

  RosatiM.CantonatiM.PrimicerioR.RossettiG. (2014). Biogeography and relevant ecological drivers in spring habitats: a review on ostracods of the Western Palearctic. Int. Rev. Hydrobiol.99, 1–22. 10.1002/iroh.201301726

  • CrossRef
  • Google Scholar

• 56

  RosenfeldA.OrtalR. (1983). “Ostracodes as indicators of water pollution in Nahal Harod, northern Israel,” in Applications of Ostracoda. Editor MaddocksR. F. (Houston, TX: University of Houston Geoscience), 229–237.

  • Google Scholar

• 57

  RossettiG.BartoliM.MartensK. (2004). Limnological characteristics and recent ostracods (Crustacea, Ostracoda) of freshwater wetlands in the Parco Oglio Sud (northern Italy). Ann. Limnol. - Int. J. Limnol.40, 329–341. 10.1051/limn/2004030

  • CrossRef
  • Google Scholar

• 58

  RossiV.BenassiG.VeneriM.BellavereC.MenozziP.MoroniA.et al (2003). Ostracoda of the Italian ricefields thirty years on: new synthesis and hypotheses. J. Limnol.62, 1–8. 10.4081/jlimnol.2003.1

  • CrossRef
  • Google Scholar

• 59

  RStudio Team (2023). RStudio: Integrated Development Environment for R. Boston, MA: Posit Software, PBC. Available online at: https://posit.co/. (Accessed February 10, 2025)

  • Google Scholar

• 60

  SarsG. O. (1928). An account of Crustacea of Norway: Ostracoda. Bergen:The Bergen Museum9, 1–277.

  • Google Scholar

• 61

  SchmitzO.AliverniniM.YasuharaM.FrenzelP. (2025). Ostracoda (Crustacea) as indicators of anthropogenic impacts: a review. Earth-Sci. Rev.263, 105049. 10.1016/j.earscirev.2025.105049

  • CrossRef
  • Google Scholar

• 62

  SchneiderA.WetterichS.SchirrmeisterL.HerzschuhU.MeyerH.PestryakovaL. A. (2016). Freshwater ostracods (Crustacea) and environmental variability of polygon ponds in the tundra of the Indigirka Lowland, north-east Siberia. Polar Res.35, 25225. 10.3402/polar.v35.25225

  • CrossRef
  • Google Scholar

• 63

  SchornikovE. I.ZeninaM. A.IvanovaE. V. (2014). Ostracods as indicators of aquatic environmental conditions on the northeastern Black Sea shelf over the past 70 years. Russ. J. Mar. Biol.40, 455–464. 10.1134/s1063074014060200

  • CrossRef
  • Google Scholar

• 64

  SilvaK. J. P.SouzaA. F. (2018). Common species distribution and environmental determinants in South American coastal plains. Ecosphere9, e02224. 10.1002/ecs2.2224

  • CrossRef
  • Google Scholar

• 65

  SpadiM.GliozziE.MediciM. C. (2019). Piacenzian–Gelasian non-marine ostracods from the Dunarobba Fossil Forest (Tiberino Basin, Umbria, central Italy). Pap. Palaeontol.5, 307–334. 10.1002/spp2.1240

  • CrossRef
  • Google Scholar

• 66

  ter BraakC. J. F.ŠmilauerP. (1998). CANOCO Reference Manual and User’s Guide to Canoco for Windows: Software for Canonical Community Ordination (Version 4). Ithaca, NY: Microcomputer Power.

  • Google Scholar

• 67

  ter BraakC. J. F.VerdonschotP. F. M. (1995). Canonical correspondence analysis and related multivariate methods in aquatic ecology. Aquat. Sci.57, 255–289. 10.1007/bf00877430

  • CrossRef
  • Google Scholar

• 68

  WansardG.MezquitaF. (2001). The response of ostracod shell chemistry to seasonal change in a Mediterranean freshwater spring environment. J. Paleolimnol.25, 9–16. 10.1023/a:1008121029324

  • CrossRef
  • Google Scholar

• 69

  WickhamH. (2007). Reshaping data with the reshape package. J. Stat. Softw.21, 1–20. 10.18637/jss.v021.i12

  • CrossRef
  • Google Scholar

• 70

  WilliamsM.SiveterD.SalasM.VannierJ.PopovL.Ghobadi PourM. (2008). The earliest ostracods: the geological evidence. Palaeobiodivers. Palaeoenviron.88, 11–21. 10.1007/BF03043974

  • CrossRef
  • Google Scholar

• 71

  WiseC. D. (1961). Taxonomy and Ecology of Freshwater Ostracods of south-central Texas (Phd Thesis). Albuquerque, NM: University of New Mexico.

  • Google Scholar

• 72

  WolfeJ. M.DaleyA. C.LeggD. A.EdgecombeG. D. (2016). Fossil calibrations for the arthropod tree of life. Earth-Sci. Rev.160, 43–110. 10.1016/j.earscirev.2016.06.008

  • CrossRef
  • Google Scholar

• 73

  WoodA. M. (2009). The phylogeny and palaeozoogeography of cold-water species of ostracod Ostracod (Crustacea) from the pre-ludhamian stage (late pliocene-early pleistocene), red crag formation, East Anglia, England; with reference to the earliest arrival of pacific species. Paleontol. Res.13, 345–356. 10.2517/1342-8144-13.4.345

  • CrossRef
  • Google Scholar

• 74

  YavuzatmacaM.KülköylüoğluO.AtamanA.AyteğinA.Yılmazİ. Ö. (2024). Assessment of rarity and ecological preferences of the non-marine ostracods (Crustacea) in Çanakkale Province (Türkiye). Aquat. Sci.86, 74. 10.1007/s00027-024-01088-5

  • CrossRef
  • Google Scholar

• 75

  YılmazF.KülköylüoğluO. (2006). Tolerance, optimum ranges, and ecological requirements of freshwater Ostracoda (Crustacea) in Lake Aladağ (Bolu, Turkey). Ecol. Res.21, 165–173. 10.1007/s11284-005-0121-2

  • CrossRef
  • Google Scholar

• 76

  ZhaiD.ZhaoW. (2014). On some recent non-marine ostracods from northern China, with description of one new species. Crustaceana87, 985–1026. 10.1163/15685403-00003339

  • CrossRef
  • Google Scholar