Otolith phenotypic analysis for the endemic Anatolian fish species, Caucasian bleak Alburnus escherichii Steindachner, 1897 (Teleostei, Leuciscidae), from Selevir Reservoir, Akarçay Basin, Turkey

Alburnus escherichii samples were collected from the Selevir Reservoir in the Akarçay Basin (Turkey) by local fishermen (Fig. 1). First, Alburnus samples from the Selevir Reservoir were identified as Alburnus nasreddini, an endemic fish species from the Akarçay Basin (Bektaş et al. 2020). However, recent molecular (Bektaş et al. 2020) and morphometric (Bayçelebi et al. 2020) studies revealed that Alburnus nasreddini is a synonym of Alburnus escherichii. A total of 154 Caucasian bleak specimens were examined for total length (L_t; ± 0.1 cm) and weight (W; ± 0.01 g). Their sex was determined by macroscopic examination of the gonads.

Figure 1

Utricular (lapillus) and lagenar (asteriscus) otoliths were removed, distinguishing between left and right ones. Otoliths were removed through an incision in the cranium, then cleaned with ethanol and stored in a dry place. All otoliths were photographed on the distal (asterisci) and dorsal (lapilli) surface using a Leica DF295 digital camera. The asteriscus length (AL), asteriscus height (AH), asteriscus perimeter (AP), asteriscus area (AA), lapillus length (LL) and lapillus width (LW), lapillus perimeter (LP) and lapillus area (LA) were measured to the nearest 0.001 mm using Leica Application Suit ver. 3.8 Imaging Software (Fig. 2) and weighted (AOW for asterisci and LOW for lapilli; 0.00001 g). The length of asteriscus and lapillus otoliths was defined as the greatest distance between their anterior and posterior margins. The height of asteriscus otoliths was defined as the greatest distance between their dorsal and ventral edges. The width of lapillus otoliths was determined as the greatest distance from their lateral to medial margins (Yilmaz et al. 2015). Lapilli and asterisci pairs were also photographed on their distal and proximal sides using a Hitachi SU 1510 Scanning Electron Microscope (SEM) for their phenotypic description (Fig. 3).

Figure 2

All variables were tested for normality and homogeneity of variance using Kolmogorov–Smirnov, Shapiro and Levene's tests, respectively. Different tests (paired t-test, Wilcoxon test, Independent Two Sample t test, Mann–Whitney U test) were used depending on whether the data were normally distributed or not. Differences between sexes were also examined by the independent t-test. A standardized model was used to remove the size effect on otolith measurements according to the following equation (Elliot et al. 1995; Lleonart et al. 2000): Zij=Yij[X0Xj]bj {Z_{ij}} = {Y_{ij}}{\left[ {{{{X_0}} \over {{X_j}}}} \right]^{bj}} where Y_ij is each original measurement of individual j, X_j is the total length of individual j, X₀ is the reference length, b_j is the allometric parameter relating the dependent variable Y_i to the independent variable X, Z_ij is the value of the standardized measurement.

Linear and nonlinear (power) models (y = a + bx, y = ax^b, where y is the otolith measurement and x is the fish length) were applied to determine the relationships between the otolith morphometrics and L_t. Parameters a and b were estimated through linear regression analysis based on logarithms, log Y = log a + b log X (Zar 1999).

The asteriscus and lapillus of A. escherichii were described morphologically according to the terminology used by Assis (2003) and Schulz-Mirbach & Reichenbacher (2006). Otolith shape indices such as the Form Factor (FF), Circularity (C), Roundness (RO), Rectangularity (RE), Aspect Ratio (AR) and Ellipticity (E) were used to analyze the shape of otoliths. They were calculated using the following formulas: Roundness = (4 × OA) / (π × OL²); Circularity = OP² / OA; Form Factor = (4 × π × OA) / OP²; Ellipticity = (OL − OB) / (OL + OB); Rectangularity = OA / (OL × OB) and Aspect Ratio = (OL / OB) (Tuset et al. 2003; Ponton 2006).

Two length classes (Class I: 6.7–10.9 cm L_t; Class II: 11.0–15.0 cm L_t) were generated for otolith shape analysis. Sexual maturity status, as determined by macroscopic examination of the gonads, was considered when establishing these classes. The population of A. escherichii from the Selevir Reservoir begins spawning in June when water temperature reaches 20°C and continues through mid-July (Gülle et al. 2017). The development of gonads in the Caucasian bleak was described using five stages following Brown-Peterson et al. (2011), which was also accepted by other researchers (Dopeikar et al. 2015; Keivany et al. 2017): stage I – testes and ovaries were very often clear and threadlike, blood vessels indistinct, and the sex could not be determined; stage II – ovaries and testes like an opaque reddish tube occupying almost half the body cavity; stage III – gonads filled the body cavity, ovaries were orange and testes were white; stage IV – ovaries were flaccid and blood vessels prominent, testes were small and flaccid and did not release milt under pressure; stage V – ovaries and testes were empty, loose and red.

In addition, multivariate analyses (PCA and DFA) were performed to detect morphometric differences in the shape of otoliths between mature and immature samples. Discriminant function analysis (DFA) was performed to detect differences in otolith shape variation between the sampling sites. Wilks’ lambda (λ) was employed to assess the performance of DFA.

SPSS 20, Past 3.0, Minitab 15.0 and Excel were used for statistical analysis.

The sex ratio was 1:2.02 based on the total sample size of females (n = 51; 33.12%) and males (n = 103; 66.88%). The maximum and minimum total length and weight of individuals ranged from 6.70 to 15.00 cm and from 2.58 to 33.48 g, respectively.

There are differences between females and males in terms of the total length and weight (Mann–Whitney U Test, p < 0.05). Therefore, a standardized model was used to remove the effect of size on otolith measurements. When asteriscus and lapillus otoliths from females and males were compared, there were no differences in AL, AH, AP, AA, AOW, LL, LW, LP, LA and LOW (p > 0.05). When comparing left and right otoliths, statistically significant differences were found in AH, AA, LL, LP and LA (p < 0.001). Descriptive statistics of otolith morphometrics are presented in Table 1.

The relationships between L_t and otolith dimensions were determined and the best fit for L_t and area for asterisci (r² > 0.610) and length for lapilli (r² > 0.569) was obtained (Table 2).

The medial face of asteriscus otoliths features a furrow. Although other authors used the term ‘sulcus’ to describe this furrow, the asteriscus does not look like a groove. For this reason, it is proposed that the name ‘fossa acustica’, used by Berinkey (1956) to name any furrow-like depression that exteriorly surrounds the fossa in cyprinids, is reintroduced. Asteriscus otoliths in the Caucasian bleak have serrated edges all around their periphery in both length classes. According to Assis (2003), the proximal surface has an acoustic pit (fossa acustica). The fossa acustica surrounded by a lobe (lobus major) can be easily detected in the asteriscus. However, the general shape of the lapillus is elongate. The antero- and posteromedial edges are small and rounded. In most lapilli, the antero- and posterolateral edges are well defined. In general, the posterolateral edge forms the pointed posterior end of the lapillus. The medial and posterior margins are slightly concave or rounded (Fig. 2). A bump (cranial umbo) is observed at its anterior end. The ventral region of the lapillus is convex and bumpy.

FF, C, RO, RE, AR and E were calculated for left and right asteriscus and lapillus otolith pairs separately. There were no differences in values of the shape indices for the left and right asteriscus and lapillus otolith pairs (p > 0.05) except Ellipticity for lapilli (p < 0.001). Otolith shape indices were calculated for asteriscus and lapillus otolith pairs (Table 3). The shape of otoliths was compared between the length classes. The discriminant function included otolith shape characteristics explaining the variation between mature and immature samples. According to the discriminant function analysis, 79.9% of A. escherichii samples were correctly classified, with Wilks’ lambda (λ) score of 0.661 (Table 4).

Morphological characteristics of fish otoliths vary with species and fish size (Popper et al. 2005; Romero et al. 2020). The relationships between fish length and otolith dimensions are used as a basis for research in fish biology and fisheries. Linear and nonlinear functions are preferred to describe the relationships between otolith dimensions and fish size. In general, the nonlinear function is used to study the relationships between otolith morphometrics and total length of fish (Jawad et al. 2017; Yılmaz et al. 2019; Saygın et al. 2020). According to Lleonart et al. (2000), using a linear model to explain the relationship between otolith morphometrics and fish length is not useful because the independent coefficient “a” is not relevant in morphometry. There are no studies for A. escherichii on the relationship between the fish length and otolith dimensions. However, there are many studies involving other Alburnus species that investigated the relationships between otolith dimensions and fish length. Bostanci & Polat (2011) studied the relationships between otolith dimensions and fork length in A. tarichi from Lake Van and found strong correlations. Basusta et al. (2013) investigated the relationships between otolith morphometrics and total length of fish in the population of A. heckeli. Saygın et al. (2017) investigated the relationships between fork length and otolith dimensions in A. tarichi from different lakes across Van Basin. In present study, the best fit was obtained between L_t and area for asterisci (r² > 0.610), and length for lapilli (r² > 0.569).

Relationships between fish size and otolith morphometrics are a baseline for studies of prey–predator interactions. Analysis of otoliths retrieved from the stomachs or feces of piscivorous predators can provide information on the type, size, mass and energy content of their fish prey (Więcaszek et al. 2020). Results of research on the relationships between fish length and otolith biometrics can be used to determine the size distribution of fish consumed by predators and stock discrimination (Jawad et al. 2017; Park et al. 2018). In the Selevir Reservoir, A. escherichii (in previous studies described as A. nasreddini) feeds on zooplankton, amphipods, dipteran larvae and pupae, and rarely on plant material (Gülle et al. 2017). In this habitat, Sander lucioperca is the top predator species. The presented study will also provide a reference for dietary and ecological studies on S. lucioperca.

This study provides data on biometric relationships between total fish length and otolith measurements for A. escherichii and presents the first quantitative analysis of total fish length and utricular and lagenar otolith size parameters. It may be useful for sustainable use and management of fisheries resources in this region.

Asteriscus otoliths in the Caucasian bleak have serrated edges around their periphery. The proximal surface features an acoustic pit. Mesial and lateral surfaces are convex. Otherwise, the general shape of the lapillus (Fig. 3) is consistent with the description provided by Schulz-Mirbach & Reichenbacher (2006). Bostanci et al. (2015) described the shape of asteriscus otoliths in four Alburnus species. According to these authors, the shape of otoliths in A. escherichii is also oval, the mesial and lateral surfaces are convex, and their dorsal margins are rounded. The shape of otoliths is an important feature in the identification of fish species. The characteristics of otoliths are a useful tool for identifying intra- and interspecific relationships. Otolith shape analysis has been used in many studies (Mapp et al. 2017; Ozpicak 2020; Bano & Serujiddin 2021).

Figure 3

The otolith shape varies from round in fish larvae to a specific shape in adults (Lagardère et al. 1995). Therefore, in many studies, including this one, data are standardized to eliminate the length and size factor (Zhao et al. 2017; Tuset et al. 2018; Saygın et al. 2020). In the present study, two length classes were established according to the stages of gonadal development and the classification score was calculated as 79.9%. In recent years, many studies have been conducted using different length classes according to morphology, ontogenetic effects and sexual variation (Capoccioni et al. 2011; Cerna et al. 2019; Carvalho et al. 2020; Teimori et al. 2020; Więcaszek et al. 2020). Waessle et al. (2003) identified a strong modification in the morphology of Micropogonias furnieri otoliths, which was associated with sexual maturation. Capoccioni et al. (2011) indicated ontogenetic and environmental effects on the Anguilla anguilla otolith shape variation. Souza et al. (2020) investigated ontogenetic and interpopulation differences in the otolith shape for Perca fluviatilis and found that otoliths tend to be elongated when fish are young and old, while more rounded in middle age. Bostanci et al. (2015) used the shape of otoliths as a tool to distinguish between different Alburnus species. They established four length classes and the overall canonical discriminant analysis (CDA) classification score was 93.8% for different Alburnus species. However, there are still ambiguities in the taxonomical status of Alburnus species. Phenotypic variation is a dynamic and flexible concept that affects population structure within a short period of time. As a phenotypic approach, research on otoliths can support the taxonomy of this species because they are species specific.

Unfortunately, changes in climatic conditions, which affect the water regime, and human activities are the main threats to the basin and lakes. There are no doubts that all these negative phenomena will have a negative impact on the distribution of the species and the current status of its population. For all these reasons, it is necessary to develop a conservation strategy for A. escherichii.