Assessment of Population Parameters for Fisheries Management in a Pressured Ecosystem: A Case Study of Sapanca Lake (Türkiye)

Abstract

Sapanca Lake is a tectonic freshwater lake ecosystem whose water balance and ecological integrity are increasingly threatened by urbanization, pollution, climate change, and declining water levels. To assess recent changes in the fish community, length–weight relationships (LWRs), condition factors (CFs), sex ratios (F/M), and standing stock biomass estimates were determined for eight fish species (Blicca bjoerkna, Scardinius erythrophthalmus, Carassius gibelio, Esox lucius, Silurus glanis, Perca fluviatilis, Abramis brama and Cyprinus carpio) sampled between December 2024 and April 2025. According to standing stock biomass assessments, Cyprinus carpio showed the highest biomass (297.67 tons), while Abramis brama had the lowest (22.72 tons). Most species showed positive allometric growth (b > 3), suggesting generally favorable feeding conditions. In contrast, the main predatory species, E. lucius (b = 2.89) and Silurus glanis (b = 0.21), exhibited negative allometric growth, likely due to limitations in food availability and prey abundance. The CF values were generally >1, indicating good physiological status; however, lower CF values in A. brama (1.18) and B. bjoerkna (1.19) suggest species-specific ecological limitations. Sex ratio analysis revealed pronounced female dominance across species, ranging from complete female dominance (F/M = 1/0) in Carassius gibelio to a female-biased ratio of 1/0.04 in Scardinius erythrophthalmus, likely driven by seasonal sampling effects, sex-specific behavior, and species-specific reproductive strategies. Overall, the results indicate increasing trophic imbalance and ecological stress in Sapanca Lake, emphasizing the need for standing stock biomass assessments and ecosystem-focused fisheries management in tectonic lakes under hydrological pressure.

1. Introduction

Sapanca Lake is a tectonic freshwater lake located in the eastern part of the Marmara Region in Türkiye, with reed beds along its shoreline [1]. The lake is fed by surface precipitation, streams and groundwater from its bottom due to its faulted geological structure. The Çark River, which originates from the northeastern end of the lake, contributes to its constantly renewed hydrological regime by emptying into the Sakarya River [2]. Sapanca Lake is used for fishing and recreation, as well as for drinking water [3]. However, the lake and its surroundings are under significant pressure due to illegal construction, rapid urbanization, and pollution, which disrupts the lake’s water balance and increases the pressure on it [4]. Although Sapanca is the smallest district in the Sakarya province, its population has grown by 26% over the past 23 years, reaching 46,080 inhabitants [5]. Furthermore, the population can triple during the summer months due to local and foreign tourists [6].

Monitoring studies have been initiated to limit such threats to the lake basin ecosystem [7]. Sustainability is being damaged by increasing pollution and activities that are excessive with regard to the number of fish that can be caught, resulting in a decrease in the age and size of fish and the number that can be caught. As long as there is sufficient food and a suitable habitat, sustainability is ensured through the replacement of individuals removed due to hunting and natural causes [8]. Length and weight data are two important components of fish species biology at both the individual and population levels. This is particularly important for the effective management and development of fish populations [9]. Therefore, determining fish stocks and analyzing parameters such as length, weight, and condition factors will contribute significantly to the sustainable conduct of these monitoring studies. In this respect, stock estimation studies are important. Declines in the populations of some species due to overfishing and the destruction of breeding areas can be evidenced by changes in fish abundance and size [10].

In fisheries biology, LWRs are useful for converting length–weight equations into growth–weight equations for use in stock assessment models, as well as for estimating stock biomass from limited sample sizes [11]. They are also used to estimate growth type (isometric or allometric) from fish length and weight [12], to assess fish condition, and to compare the historical characteristics and living conditions of fish species in different regions [13]. These values are important for yield per fry analysis in stock assessment studies [14]. They can also provide important details about climatic and environmental changes, as well as adjustments to people’s subsistence practices [15]. Alongside LWRs, the condition factor (CF) is used to estimate the interaction between biotic and abiotic factors affecting the physiological condition of a fish. The CF enables the collection of information on stock composition, life history, growth, maturity, and production [16]. It also facilitates comparisons between species and populations, as well as the estimation of fish population health [17].

In this respect, LWRs and CF are effective tools in fisheries biology for informing sustainable fisheries management guidelines for natural water resources [16]. In this study, our aim was to estimate the LWRs of fish species obtained from Sapanca Lake, evaluate the CF, determine the suitability and quantity of the fish population stocks in Sapanca Lake, and determine the fishing techniques and conditions necessary for protecting these stocks. This information will contribute to the management and protection of this water resource, allowing comparisons to be made between the current and future states of the natural populations of these economically important fish species. It will also help us to determine the necessary fishing techniques and conditions in Sapanca Lake, which has a dynamic ecosystem.

2. Materials and Methods

2.1. Sampling Area

Sapanca Lake is located within the borders of Sakarya Province (40°43′ N 30°15′ E/40.717° N 30.250° E), and its catchment area covers 251 km². The lake has a surface area of 46 km², a length of 16 km from east to west and 5 km from north to south, a maximum depth of 55 m, and an average depth of 25.6 m. It is located in the eastern part of the Marmara Region of Türkiye (Figure 1). It is fed by small streams descending from the mountains, including Keçi Creek, Istanbul Creek, Mahmudiye Creek, Kuruçay Creek, Karaçay Creek, Balıkhane Creek, Maden Creek and Arifiye Creek, as well as by springs on the lake bed. The lake water flows into the Sakarya River via Çark Creek from the eastern end [18]. It is located 12 km west of Sakarya, in the eastern half of a long depression connecting the Adapazarı Plain to the Gulf of Izmit [19,20].

2.2. Sample Collection

The study was carried out using plain galsama gillnets with mesh sizes of 32, 40, 50, 80, and 100 mm. Galsama gillnets are fish-catching devices that trap a fish by forcing its head through the mesh, but not its entire body. When it attempts to escape, it becomes caught by its gills. To catch a fish, the mesh size must be larger than the fish’s head and smaller than its body. Furthermore, while galsama gillnets optimally catch individuals of a certain size, they catch proportionally fewer smaller and larger individuals; when the fish exceed this size, the gillnets’ effectiveness approaches zero [21,22]. Therefore, although the nets exhibited size selectivity, this factor was not considered in the calculations in the present study. Instead, the stock estimates were based on the biomass of the size classes to which the gillnets are most sensitive. As a result, the values reported in this study reflect the biomass of the size classes to which the nets are sensitive during a specific time period (December 2024 and January, February, March, and April 2025), rather than the total stock biomass of the lake. Samplings were conducted at four stations, selected to represent the entire lake area, during December 2024 and January, February, March, and April 2025 (Figure 1). Five sets of single-layer gillnets made of monofilament material (0.20 mm polyamide) were used in the study. The Donam factor (E) was 0.50 for both the cork and lead collars. The total length of the nets used in the study was approximately 250 m, with one size of net measuring approximately 50 m.

Table 1. Total area of used gillnets.

Gillnet Mesh Size (mm)	Surface Area of All the Gillnets (ha)	Gillnet Length (m)	Number of Meshes per Gillnet Depth	Gillnet Depths (m)	Surface Area of Gillnets (ha)
32	0.131	50	50	3.20	0.016
40		50	50	4.00	0.020
50		50	50	5.00	0.025
80		50	50	6.00	0.030
100		50	50	8.00	0.040

shows the total area of the nets used in the fishery to determine the catchable stock of fish, based on estimates of the biomass present at the time of sampling. In addition, the water temperature, oxygen, salinity, electrical conductivity (EC), and pH values were measured from the sampling areas of the lake using the PCE-PHD Data Logger throughout the sampling period, and average values were taken (

Table 2. Water parameters (temperature, salinity, EC, DO, pH) for Sapanca Lake.

Sampling Periods	Water Parameters
	Temperature ± SD (°C)	Dissolved Oxygen ± SD (mg/L)	pH ± SD	Salinity (ppt)	EC (µS/cm)
December	9.5 ± 1.45	6.84 ± 2.00	7.36 ± 0.62	0.13 ± 0.01	299.57 ± 32.94
January	8.6 ± 1.35	7.84 ± 2.00	7.46 ± 0.62	0.12 ± 0.02	219.67 ± 20.98
February	9.6 ± 1.40	8.94 ± 2.00	8.16 ± 0.62	0.10 ± 0.0	219.67 ± 22.95
March	12.6 ± 0.30	9.94 ± 3.12	7.16 ± 0.52	0.11 ± 0.02	259.67 ± 21.95
April	14.6 ± 0.30	9.84 ± 3.12	9.16 ± 0.57	0.11 ± 0.02	249.67 ± 21.95

).

Gillnets were laid in the water close to sunset and left for approximately 12 h before being collected in the early morning (between 06:00 and 07:00 a.m). The areas where fishing took place were sandy, muddy and clayey, with meadows, and the depths at which the nets were dropped varied between 5 and 6 m, depending on the study area. The fish samples caught in the gillnets were counted and grouped according to mesh size. The number of species caught in each net was recorded, and the samples were stored in an ice box in a cold environment for the necessary measurements. In this study, eight fish species (Blicca bjoerkna Linnaeus, 1758; Scardinius erythrophthalmus Linnaeus, 1758; Carassius gibelio Bloch, 1782; Esox lucius Linnaeus, 1758; Silurus glanis Linnaeus, 1758; Perca fluviatilis Linnaeus, 1758; Cyprinus carpio Linnaeus, 1758; and Abramis brama Linnaeus, 1758) were collected from commercial fishermen in Sapanca Lake following the researchers’ standardized sampling protocol, including sampling time, procedure, gillnet type, and gillnet depth. Among these species, P. fluviatilis, Silurus glanis, and E. lucius were identified as the most commercially important [23,24].

2.3. Biometrics, Stock Estimation, and Analytical Procedures

The biometrics of fish captured by the gillnets were compiled separately for each net, and the total standing stock biomass was estimated [25]. The total length (TL) was measured to the nearest millimeter (mm) using a fish measuring board, and the total weight (TW) was determined using a digital balance (Weightlab WL-3002L, Shanghai, China) with an accuracy of 0.1 grams (g).

The surface area covered by a single gillnet (A_i, m²) was calculated as the product of net width and net length:

$$A_i=w_i l_i,$$

(1)

where w_i is the net width (m), and l_i is the net length (m).

The total area covered by all the gillnets was obtained as follows:

$$A_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}={\sum }_{i=1}^n A_i$$

(2)

Assuming that all fish $A_{\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}$ within the gillnet-covered area were susceptible to capture, the unit-area yield for sub-region $j$ was calculated as follows:

$$Y_j={\displaystyle \frac{C_j}{A_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}}} ({\displaystyle \frac{\mathrm{k}\mathrm{g}}{\mathrm{h}\mathrm{a}}}),$$

(3)

where $C_j$ is the average catch (kg) obtained in sub-region $j$.

The biomass for sub-region $j$ was then estimated as follows:

$$B_j=Y_j A_j,$$

(4)

where $A_j$ is the total area of sub-region $j$ (m²) and $B_j$ is expressed in tons.

2.4. Length–Weight Relationships (LWRs), Condition Factor (CF), and Sex Ratio

The relationship between total length (TL, cm) and total weight (TW, g) was analyzed using linear regression, and the length–weight relationships (LWRs) were expressed by the model TW = aTL^b [26]:

$$\mathrm{T}\mathrm{W}=a \mathrm{T}{\mathrm{L}}^{ b},$$

(5)

where TW is total weight (g), TL is total length (cm), $a$ is the intercept, and $b$ is the allometric growth exponent. The degree of association between variables was evaluated using the coefficient of determination (R²) [27].

Fulton’s condition factor (CF) was calculated to assess the individual condition of the fish, assuming that heavier fish at a given length are in better condition [28,29]:

$$\mathrm{C}\mathrm{F}=100 {\displaystyle \frac{\mathrm{T}\mathrm{W}}{\mathrm{T}{\mathrm{L}}^3}}$$

(6)

where TW is the total weight (g), and TL is the total length (cm).

Sex was determined by macroscopic examination of the gonads after dissection. Individuals with paired, elongated, whitish testes were classified as males, whereas those with larger, lobulated, yellow to orange ovaries containing visible oocytes were classified as females. The gonadal morphology and coloration were evaluated following standard ichthyological procedures described in the literature [26,30].

The sex ratio (female/male) was calculated as follows:

$$\mathrm{S}\mathrm{e}\mathrm{x} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}={\displaystyle \frac{F}{M}},$$

(7)

and the female frequency (F %) was calculated as follows:

$$F \%=100 {\displaystyle \frac{F}{F+M}},$$

(8)

where F denotes the number of females, and M denotes the number of males.

2.5. Statistical Analysis

Associations among variables were evaluated using Spearman’s rank correlation. Group differences were tested with independent-samples Student’s $t$-tests. A chi-square (${\chi }^2$) test assessed whether the female-to-male ratio deviated from expected proportions. Differences in CF among species were evaluated by one-way ANOVA ($p<0.05$). All analyses were conducted in MINITAB 21.

The departure of $b$ from isometry ($b=3$) was tested as follows [31]:

$$t_s={\displaystyle \frac{ b-3 }{S_b}},$$

(9)

where $t_s$ is the test statistic and $S_b$ is the standard error of $b$. Growth was interpreted as isometric if $b=3.0$, negative allometric if $b<3.0$, and positive allometric if $b>3.0$. The statistical significance of departures from isometry was determined using $t_s$ and the 95% confidence interval (CI) of $b$ [32].

3. Results

3.1. Estimation of Total Biomass of the Fish Species

Among the studied species, Cyprinus carpio was the most abundant species in terms of the number of individuals (n = 157) and also exhibited the highest estimated biomass per unit area and total biomass (66.15 kg·ha⁻¹; 297.67 tons). This was followed by E. lucius (n = 55), which showed a high biomass per unit area (53.92 kg·ha⁻¹) and a total biomass of 242.66 tons. In contrast, Abramis brama was the least abundant species (n = 12) and displayed the lowest biomass values both per unit area (5.05 kg·ha⁻¹) and in total biomass (22.72 tons). Scardinius erythrophthalmus ranked seventh among the species in terms of biomass, with 28 individuals recorded, a biomass of 7.83 kg·ha⁻¹ per unit area, and an estimated total biomass of 35.25 tons during the sampling period (

Table 3. Total biomass of fish species in Sapanca Lake during the sampling period.

Fish Species	n	N	Total Fish Weight (kg)	Average Fish Weight (kg [N − 1]−1)	Biomass of the Unit Area (kg ha−1)	Total Biomass (tons)
Carassius gibelio	91	12	67.21	6.11	46.64	209.87
Esox lucius	55	12	77.70	7.06	53.92	242.66
Bilicca bjoerkna	46	12	30.31	2.78	21.03	94.65
Abramis brama	12	12	7.27	0.66	5.05	22.72
Perca fluviatilis	90	12	21.23	1.93	14.73	66.29
Cyprinus carpio	157	12	95.32	8.67	66.15	297.67
Scardinus erythrophthalmus	28	12	11.29	1.03	7.83	35.25
Silurus glanis	15	12	37.99	3.45	26.37	118.64
TOTAL	494		348.32	31.69	241.72	1087.75

n = Sample size, N = Number of gillnet sets (deployments and retrievals).

). Overall, species differed markedly in both numerical abundance and biomass contribution, indicating that dominant species in terms of biomass were not always the most numerous.

3.2. Length–Weight Relationships (LWRs) and Condition Factor (CF)

A total of 494 fish samples were collected from Sapanca Lake during the study period. The average total length (TL), total weight (TW), length–weight relationship (LWR) parameters (a and b), coefficient of determination (R²), growth type, condition factor (CF), and sex rate values for each species are given in

Table 4. Mean LWR parameters of the fish species in Sapanca Lake (p < 0.05).

Species	n	Mean TL ± SD	Mean TW ± SD	a	b	R2	B (SE)	95% Cl	Student t Test of b	Mean CF ± SD	LWR Equations	Growth Type	p	F/M
Carassius gibelio	91	31.80 ± 6.85	737.42 ± 403.77	0.012	3.15	0.98	3.68	43.09–57.71	12.88	1.99 ± 0.20	TW = 0.012L3.15	A(+)	0.001 *	1/0
Esox lucius	55	44.81 ± 11.76	1412.79 ± 824.22	0.020	2.89	0.93	202.00	−422–386	−0.11	1.34 ± 0.35	TW = 0.020L2.8	A(−)	0.92	1/0.28
Bilicca. bjoerkna	46	36.32 ± 7.57	657.19 ± 408.43	0.004	3.33	0.93	42.00	−95.70–73.5	−0.34	1.19 ± 0.28	TW = 0.004L3.33	A(+)	0.74	1/0.48
Abramis brama	12	35.73 ± 6.11	606.15 ± 351.36	0.001	3.60	0.95	5.01	13.88–35.95	4.37	1.18 ± 0.210	TW = 0.001L3.60	A(+)	0.01 *	1/0.33
Perca fluviatilis	90	23.39 ± 3.83	216.70 ± 172.148	0.001	3.73	0.90	2.83	6.00–17.26	3.04	1.45 ± 0.49	TW = 0.001L3.73	A(+)	0.003 *	1/0.2
Cyprinus carpio	157	33.55 ± 5.15	607.13 ± 372.19	0.007	3.23	0.89	0.89	16.33–19.83	17.04	1.47 ± 0.32	TW = 0.007L3.23	A(+)	0.001 *	1/0.06
Scardinus erythrophthalmus	28	30.53 ± 3.32	403.13 ± 149.46	0.005	3.30	0.96	0.65	11.53–14.19	15.21	1.35 ± 0.11	TW = 0.005L3.30	A(+)	0.001 *	1/0.04
Silurus glanis	15	57.81 ± 4.43	2532.75 ± 264.23	7066	0.82	0.21	229.79	−224.00–761.00	1.16	1.35 ± 0.29	TW = 7066L0.21	A(−)	0.267	1/0.07

n = Sample size, TL = Total length (cm), TW = total weight (g), a = intercept, b = slope, Cl = confidence intervals, R² = Coefficient of determination, CF = Fulton’s condition factor, GT = growth type, A(+):positive allometric growth, and A(−): negative allometric growth, * = <0.05, Regression analysis results are significant, F/M = Sex rate-Female/Male.

. P. fluviatilis had the lowest mean total length and weight (23 cm and 217 g, respectively), while Silurus glanis had the highest (58 cm and 1640 g, respectively). The b value ranged from 0.82 to 3.73 (Silurus glanis to P. fluviatilis, respectively). Silurus glanis and E. lucius showed negative allometric growth (b < 3), whereas the other six species showed positive allometric growth (b > 3) (Table 4). Carassius gibelio ranked in first place with 1.99, and Cyprinus carpio was in second place with 1.47, among the fish with the best CF values in Sapanca Lake. Statistically, a significant difference was found among the fish species in terms of CF values. The most significant differences in CF were between Carassius gibelio and the other species. On the other hand, A. brama, E. lucius, Silurus glanis, Scardinius erythrophthalmus, P. fluviatilis, and Cyprinus carpio showed similar CF characteristics. The difference in their CF values was not statistically significant (p < 0.05). The difference in their CF values was not statistically significant (p < 0.05). Additionally, the lowest R² values were in Silurus glanis (0.82), Cyprinus carpio (0.89), and P. fluviatilis (0.90), whereas the highest R² values were in Carassius gibelio (0.98), Scardinius erythrophtalmus (0.96), and A. brama (0.95). E lucius and B. bjoerkna had the same R² value, 0.93 (Table 4).

4. Discussion

People have fished to meet their nutritional needs as the population has grown since the first years of humanity. This has led to an increase in fishing, which in turn has caused important problems such as water pollution, unregulated fishing, and damage to fish stocks. Controlled and conscious fishing is essential for conserving fish populations and ensuring their long-term survival [33]. It has been well recognized that Sapanca Lake has been experiencing severe water loss in recent years due to increasing drought and climate change pressures. Shortly after the completion of the present study, which was conducted between December 2024 and April 2025, the lake water level declined to approximately 28.5 m by December 2025, representing one of the lowest levels ever recorded. This critical reduction led to shoreline retreats of up to 15–20 m in certain areas, rendering many piers and coastal structures nonfunctional. These pronounced hydromorphological changes clearly indicate that Lake Sapanca has undergone one of the driest periods of the past 60–65 years, highlighting serious risks to the lake’s sustainability both as a drinking water resource and as a natural freshwater ecosystem [34,35].

Previous studies conducted in Sapanca Lake have demonstrated that patterns of species dominance vary depending on whether assessments are based on commercial landings, numerical abundance, or biomass. Several studies have reported that E. lucius, one of the most economically important fish species in the lake, dominates commercial fisheries, accounting for up to 31.25% of total catches, followed by Cyprinus carpio (25%) and Scardinius erythrophthalmus (18.75%), whereas other species contribute relatively little to total landings [7,36]. In contrast, studies focusing on numerical abundance have revealed a different structure of the fish community. For example, Scardinius erythrophthalmus was identified as the most numerically abundant species, comprising 24.4% of the catch, followed by Blicca bjoerkna (22.98%), while E. lucius contributed only 7.37% to total abundance; Cyprinus carpio and Silurus glanis accounted for 1.10% and 0.52%, respectively [37]. Similarly, Rutilus rutilus, B. bjoerkna, and Coregonus albula have been consistently reported as the most numerically abundant species in the lake, despite their relatively low contribution to commercial catches [7]. In the present study, however, biomass-based estimates revealed a markedly different pattern. The highest total biomass was recorded for Cyprinus carpio (297.67 tons; 27.47% of total biomass; n = 157), reflecting the combined effects of high numerical abundance and moderate-to-high individual body mass. This was followed by E. lucius (242.66 tons; 22.39% of total biomass; n = 55), which contributed disproportionately to total biomass despite its relatively low abundance, owing to its large individual body size. In contrast, A. brama (22.72 tons; 2.10% of total biomass; n = 12) and Scardinius erythrophthalmus (35.25 tons; 3.25% of total biomass; n = 28) exhibited the lowest biomass values, consistent with their low abundance and smaller average body weights. From a biological perspective, these differences reflect species-specific life-history strategies and trophic roles: large predatory species such as E. lucius tend to achieve biomass dominance primarily through somatic growth, whereas omnivorous species such as Cyprinus carpio attain high biomass through population density. In addition, the relatively low biomass of Scardinius erythrophthalmus compared to earlier reports may be associated with increased fishing pressure in recent years, as this species has been intensively exploited and increasingly marketed for human consumption in tourist areas surrounding the lake. Methodological factors may also contribute to these patterns, as gillnet selectivity favors the capture of larger-bodied individuals, potentially underrepresenting small-bodied but numerically abundant species. Overall, these results highlight the importance of integrating both biomass- and abundance-based metrics in fisheries assessments to more accurately capture population structure and exploitation dynamics. In this context, numerical abundance and biomass represent different dimensions of community structure, demonstrating that biomass-based patterns may be dominated by a small number of large-bodied species, even when these species are not numerically abundant, whereas numerically dominant small-bodied species may contribute relatively little to total biomass [38]. This distinction is particularly relevant for interpreting biomass estimates of large predatory fishes such as E. lucius, which may exhibit high biomass values despite relatively low numerical abundance.

The LWRs of fish species may vary temporally or spatially with length range; reproductive activities; or environmental conditions such as temperature and water quality, food quality, food availability, disease, and competition (39). In this study, the LWR parameters (a, b), coefficient of determination (R²), growth type and CF values were determined separately for each fish species. The “a” value in the LWR equation of fish species indicates the average condition of the individuals, and the “b” value indicates the growth type of the fish according to the conditions [39]. In the LWR equations of fish species, if the b value is equal to 3, it indicates isometric growth (I); if it is greater than 3, it indicates positive allometric growth A(+); and if it is less than 3, it indicates negative allometric growth A(−) [40]. In the present study, the predominance of b > 3 in most species (Carassius gibelio, B. bjoerkna, A. brama, P. fluviatilis, Cyprinus carpio, and Scardinius erythrophthalmus) indicates that positive allometric growth is dominant, with weight increasing faster than length in individuals. This suggests that food availability in Sapanca Lake is generally sufficient and that the energy intake can be effectively allocated to somatic growth and energy storage. In contrast, the b < 3 values observed in E. lucius and Silurus glanis indicate that growth in predatory species is strongly influenced by prey availability, trophic structure, and species-specific feeding strategies. Variations in the a coefficient among species reflect differences in body shape and condition, while the high R² values calculated for most species demonstrate the reliability and consistency of the length–weight relationship models. The CF values further indicate that the overall physiological status of the populations is relatively good, although condition appears to be sensitive to species-specific ecological traits and environmental pressures.

A previous study [41] conducted in the same research area (Sapanca Lake) between January and May 2003 reported that E. lucius showed positive allometric growth. The present study was conducted on fish samples taken between December 2024 and April 2025. The sampling periods were almost identical, and it is possible that this change in the pike, one of the most valuable species of freshwater fish in Türkiye, may also be due to changes in the lake’s conditions over time, as many studies report that E. lucius shows negative allometric growth. Furthermore, aquatic keystone predators (piscivorous predators) at the top of the food web, such as E. lucius, can strongly impact the structure, function, and biodiversity of aquatic habitats, such as freshwater and coastal ecosystems [42]. This makes them good predators, and over time, this trait can cause them to grow taller, slimmer, and heavier. Positive growth occurs when anabolism exceeds catabolism, causing fish to become larger or deeper-bodied as they increase in length. Negative growth occurs when catabolism exceeds anabolism, causing fish to become thinner as they increase in weight [43]. Silurus glanis is an invasive species in oligotrophic deep lakes [44] such as Sapanca Lake. A review of previous studies has shown that Silurus glanis exhibits both negative and positive allometric growth. In this study in Sapanca Lake, although it exhibits negative allometric growth, it has a very low b value (0.21). These findings can be attributed to several factors. Firstly, sampling was conducted at the beginning of the breeding season, when newly recruited individuals had not yet reached sexual maturity, and when full gonadal development and the reproductive potential of Silurus glanis peak later in April [45]. In addition, gillnets with a mesh size of 100 mm were used in this study. Adult Silurus glanis individuals generally exceeding 1 m have been reported [46]. Moreover, Silurus glanis is characterized by a relatively large head morphology, which likely reduces its susceptibility to the sampling gear employed. Consequently, larger adult individuals were probably underrepresented in the collected samples, as the representation of large and small fish varies depending on the sampling methods employed. Sampling gear selectivity can bias length-based representation because different gears tend to sample specific size classes more effectively than others [47]. This represents a methodological limitation of the study. Consequently, the measured values in Silurus glanis predominantly represent the LWR parameters of smaller, pre-adult individuals.

As shown in Table 4, the growth form of the species used in this study varies from that in the other studies. Similarly to this study, many studies have emphasized that B. bjoerkna showed positive allometric growth [23,41,48,49]. There have also been studies indicating that B. bjoerkna showed negative allometric growth [24,50,51]. It was emphasized that the b value of B. bjoerkna was lower in eutrophic and relatively shallow lakes, and this species was found in shallow parts of warm lakes with vegetation [51]. It is known that changes in b values in the same species or in different species can be due to various factors. The growth patterns of species can vary among different populations of the same species or within the same population in different years, depending on food availability [12]; water quality [52]; biological, temporal, and sampling factors [53]; fish condition; season [54,55]; diet; sex; health; habitat; gonadal maturity; preservation techniques; stomach fullness; and region [56,57,58]. Therefore, differences may occur in the same species across different regions and even in different years of the same period.

The LWR and b values obtained in this study, when compared with those reported from different habitats in the literature (

Table 5. Mean length–weight relationship parameters of the fish species in different locations reported by different authors (p < 0.05).

Species	n	L (cm) (Min–Max)	W (g) (Min–Max)	Length–Weight Parameters			Growth Type	Location	CF	References
				a	b	R2
Carassius gibelio	363	5.20–30.20	-	0.008	3.25	0.99	A(+)	İznik Lake	-	[34]
	95	11.30–35.50	-	0.022	2.88	0.90	A(−)	Anzali Wetlands	1.55	[60]
	395	9.90–34.50	16.17–774.40	0.012	3.11	0.99	A(+)	Büyükçekmece Lake	1.72	[61]
	46	1.90–36.50	113.00–984.00	0.019	3.04	0.88	A(+)	Aşartepe Dam Lake	2.03	[62]
	179	9.30–32.40	13.76–592.75	0.026	2.87	0.97	A(−)	Sakarya River	-	[23]
	3987	6.90–38.20	3.70–1266.00	0.011	3.17	0.98	A(+)	Eğirdir Lake	-	[63]
	88	10.2–29.8	19.47–408.59	0.013	3.08	0.96	A(+)	Asi River	1.76	[64]
	64	14.20–40.20	58.00–1420.00	0.010	3.23	0.99	A(+)	Çaltıcak Lake	0.21–2.57	[24]
	730	3–35.70	-	0.009	3.18	0.99	A(+)	Ömerli Dam Lake	-	[41]
	363	5.20–30.20	-	0.008	3.25	0.99	A(+)	İznik dam lake	-	[41]
	2213	6.80–27.50	4.90–372.20	0.0173	2.97	0.98	A(−)	Marmara Lake	-	[65]
	159	7.6–25.5	7.60–246.43	0.0130	3.060	0.97	A(+)	Euphrates River	-	[66]
Esox lucius	13	26.3–57.6	-	0.003	3.21	0.97	A(+)	Sapanca Lake	-	[41]
	48	40.20–76.30	689.40–3421.50	0.066	2.48	0.94	A(−)	Sakarya River	-	[67]
	311	22.50–33.39	101.50–319.81	0.023	2.72	0.95	A(−)	Çapalı Lake	0.88	[68]
	313	22.80–66.00	92.90–3342.00	0.003	3.21	0.98	A(+)	Işıklı Dam Lake	0.87	[69]
	100	16.5–53.4	260.00–1870.00	0.036	2.69	0.998	A(−)	Kesikköprü Dam Lake	0.86–1.04	[70]
	44	27.20–259.80	153.16–1353.12	0.0097	2.91	0.93	A(−)	Sakarya River	-	[23]
	39	31.10–73.00	180.00–2480.00	0.002	3.32	0.96	A(+)	Çaltıcak Lake	0.54–0.87	[24]
	204	25.50–70.50	115.00–3174.00	0.0035	3.21	0.98	A(+)	Ladik lake	0.78	[71]
	652	5.30–82.50	1.00–3851	0.006	3.001	0.99	I	Elbe River	-	[72]
Bilicca bjoerkna	196	12.00–21.20	-	0.007	3.18	0.90	A(+)	Sapanca Lake	-	[41]
	434	13.20–27.80	22.80–259.00	0.004	3.36	0.97	A(+)	Ladik Lake	1.59	[73]
	434	13.87–18.72	42.90–112.30	0.007	3.32	0.97	A(+)	Ladik Lake	1.49–1.78	[48]
	1250	8.00–30.00	-	0.033	2.93	0.97	A(−)	Kyiv Reservoir	2.68	[51]
	350	6.60–24.30	-	0.004	3.39	0.97	A(+)	Sapanca Lake	1.12	[74]
	183	7.40–18.50	6.78–124.16	0.115	2.58	0.81	A(−)	Uluabat Lake	1.27–2.87	[50]
	547	6.20–30.40	3.15–311.15	0.015	3.12	0.96	A(+)	Sakarya River	-	[23]
	392	13.70–27.80	26.00–289.00	-	3.44	1 × 10−6	A(+)	Aras Dam Lake	-	[49]
	708	10.80–52.00	10.00–1650.00	0.023	2.77	0.93	A(−)	Çaltıcak lake	0.23–2.11	[24]
	2871	3.30–39.00	0.20–660.00	0.006	3.23	0.98	A(+)	Elbe River	-	[72]
Abramis brama	21	20.9–39.70	-	0.005	3.35	0.99	A(+)	Terkos Dam Lake	-	[41]
	184	21.50–45.00	228.00–2044.00	0.028	2.91	0.97	A(−)	Volga River	2.01–2.15	[75]
	143	14.30–53.70	33.35–1977.48	0.007	3.12	0.97	A(+)	Sakarya River	-	[27]
	1420	14.00–56.00	50.00–5600.00	0.010	3.21	0.98	A(+)	Middle Dnieper	2.21	[76]
	722	8.10–44.60	8.00–1790.00	0.009	3.18	0.99	A(+)	Ladik Lake	1.59	[48]
	19	26.60–46.50	172.00–1156.00	0.013	2.94	0.91	A(−)	Çaltıcak Lake	0.89–1.59	[24]
	141	19.50–38.00	-	0.184	2.06	0.69	A(−)	Beni- Haroun Dam	0.31	[77]
	4490	3.80–56.50	0.5–2282	0.01	2.97	0.98	A(−)	Elbe River	-	[72]
Perca fluviatilis	11	7.20–21.20	-	0.008	3.20	0.98	A(+)	Büyükçekmece Lake	-	[41]
	689	5.90–29.30	1.37–449.00	0.006	3.26	0.99	A(+)	Büyükçekmece Lake	1.24	[63]
	858	8.20–27.50	7.16–365.20	0.005	3.36	0.98	A(+)	Ladik Lake	1.28	[78]
	107	11.40–28.70	20.45–370.51	0.015	2.94	0.93	A(−)	Sakarya River	-	[23]
	128	10.50–26.20	20.00–615.00	0.0096	3.24	0.989	A(+)	Volga River	1.72–2.25	[75]
	3438	3.30–43.50	0.30–1438	0.004	3.34	0.987	A(+)	Elbe River	-	[72]
	18	14.60–26.70	55.30–230.00	0.640	1.73	0.88	A(−)	Çaltıcak Lake	0.91–2.37	[24]
Cyprinus carpio	142	21.10–77.60	152.00–7045	0.026	2.83	0.99	A(−)	Altınkaya Dam Lake	-	[79]
	36	24.80–44.3	225.00–1315.00	0.020	2.90	0.98	I	Karaboğaz Lake	-	[79]
	302	12.20–42.40	34.95–1177.04	0.001	3.14	0.93	A(+)	Kızılırmak River	1.52	[80]
	342	9.0–68.90	12.00–5050.00	0.033	2.78	0.96	A(−)	Ganga River	1.48–1.61	[9]
	92	9.30–26.30	24.00–385.00	0.068	2.68	0.97	A(−)	Heviz Lake	1.65–3.59	[81]
	190	11.2–55.5	30.3–2860.5	0.053	2.72	0.97	A(+)	Tercan Dam Lake	2.15	[82]
	120	37.00–43.00	765.00–1415.00	6.982	0.25	0.66	A(−)	Mechraa Sfa Dam	2.05	[83]
	402	24.90–46.50	226.70–1369.50	0.016	2.94	0.94	I	Yeşilırmak	1.3	[84]
Scardinus erythrophthalmus	409	13.00–34.90	20.00–6199	0.004	3.73	-	A(+)	Sapanca Lake	1.24	[85]
	19	7.80–22.90	-	0.008	3.21	0.99	A(+)	Sapanca Lake	-	[41]
	141	6.40–17.70	-	0.006	3.36	0.96	A(+)	Anzali Wetlands	1.40	[60]
	305	6.90–27.00	-	0.008	3.17	0.98	A(+)	Büyükçekmece Lake	1.38	[62]
	43	10.20–30.20	13.46–364.67	0.009	3.15	0.99	A(+)	Sakarya River	-	[23]
	270	9.40–17.90	11.98–98.50	0.006	3.29	0.96	A(+)	Anzali Lagoon	1.58–2.30	[86]
	144	4.80–28.00	1.00–259	0.007	3.17	0.995	A(+)	Elbe River	-	[72]
	159	16.10–30.00	42.00–326.40	0.013	2.99	0.90	I	Çaltıcak Lake	0.12–1.59	[24]
Silurus glanis	64	22.50–86.70	66.10–5987.60	0.003	3.22	0.99	A(+)	Sakarya River	-	[67]
	21	20.5–250.0	-	0.032	2.57	0.96	A(−)	Seyhan Dam Lake	-	[87]
	257	92.7–101.8	6578.2–9041.10	0.010	2.01	0.97	A(−)	Menzelet Reservoir	-	[88]
	66	48.50–68.32	704.00–6560.00	0.0004	3.06	0.96	A(+)	Çelik Lake	0.58	[89]
	108	24.80–67.90	92.40–2066.50	0.005	3.02	0.99	A(+)	İznik Lake	0.60	[69]
	128	33.80–103.00	165.00–7600.00	0.007	2.99	0.99	I	Altınkaya Dam Lake	0.63	[90]
	35	40.70–83.00	422.00–3142.00	0.013	2.81	0.97	A(−)	Çaltıcak Lake	0.62	[24]

), indicate that fish growth patterns are strongly shaped by environmental conditions. The broad range of b values reported for the same species across habitats (≈2.0–3.7) demonstrates that growth can shift between isometric, positive, and negative allometry depending on local ecological conditions. In contrast, the predominance of b values greater than 3 in the populations examined here (e.g., Carassius gibelio, B. bjoerkna, A. brama, P. fluviatilis, Cyprinus carpio, and Scardinius erythrophthalmus) suggests that weight gain exceeds length increase, indicating dominant positive allometric growth. This pattern likely reflects favorable habitat conditions, including adequate food availability, efficient energy allocation to somatic growth and lipid storage, and metabolically suitable environmental factors such as temperature, water quality, and trophic structure. Consistent with our findings, previous studies have associated elevated b values in freshwater ecosystems with good feeding conditions and high habitat quality [12,59]. Conversely, the relatively low b values (<3) observed for predatory species such as E. lucius and Silurus glanis in some lentic and lotic systems support the view that growth in piscivorous fishes is closely linked to prey availability, trophic dynamics, and ontogenetic dietary shifts [12,40,59]. Notably, lower b values reported in Table 5 are frequently associated with ecosystems experiencing environmental stress, increased competition, or reduced water quality, reinforcing the interpretation of LWR parameters as indirect indicators of ecosystem health. Overall, the high R² values and relatively elevated b estimates obtained in this study suggest homogeneous size structure and relatively stable environmental conditions, emphasizing that variations in LWRs should be interpreted in conjunction with habitat characteristics, food-web structure, reproductive period, and prevailing environmental pressures, in addition to species-specific life-history traits [12,59].

In Sapanca Lake, there were fish in better condition, such as Carassius gibelio, which ranks first with 1.99, and Cyprinus carpio, which ranks second with 1.47. The introduction of a species into a new habitat can be devastating. Carassius gibelio is known as a successful invasive species due to its high reproductive capacity and tolerance to environmental changes. Owing to these characteristics, it can become the dominant species in the new habitat in a very short time [91]. The pressure of Carassius gibelio probably affects the fitness of the other species. Carassius gibelio shows positive allometric growth (b = 3.15) and has the highest R² value of 0.98, indicating a high correlation between length and weight. Cyprinus carpio, which had the second-highest CF, has a wide CF range (1.3–3.59) when compared to the values reported by some studies (Table 5). Sapanca Lake’s Cyprinus carpio has an average value of almost 1.47 (1.12–3.14). The reason for this wide CF range is, of course, that it is one of the most common freshwater fish in the world [92] and that it is a hardy fish, able to withstand large fluctuations in temperature. Furthermore, it has a wide adaptability to changing climate and environmental conditions, as well as different feed availability. It is omnivorous and consumes available nutrients. It can tolerate changing oxygen levels in the water [93]. The lowest mean CF values in Lake Sapanca were found in A. brama (1.18; 0.89–1.62) and B. bjoerkna (1.19; 0.86–2.16). The values closest to those in the present study were documented by Aydın et al. [24] in Çaltıcak Lake in the same region (Table 5). Similar to the b values, the observed differences in CF values can be attributed to factors such as the fish’s feeding regime, seasonal timing, heavier female ovaries, the organisms consumed, feeding behavior, biological characteristics, and environmental disturbances [74]. Differences in CF may be due to a combination of one or more of these factors [58]. The CF values of these species during the sampling period indicate that Sapanca Lake provided rich food resources and suitable environmental conditions for the fish species examined in this study, because the CF value is greater than 1.

A high coefficient of determination (R²) value indicated that the model used for the analysis fits the data and confirms the model’s suitability [86]. The regression analysis showed that the length of Carassius gibelio, A. brama, P. fluviatilis, Cyprinus carpio, and Scardinius erythrophthalmus correlated more strongly with weight than did the length of E. lucius and B. bjoerkna (p < 0.05) in Sapanca Lake (Table 4).

This study also evaluated the sex ratios of fish species in Sapanca Lake during the sampling season. A sex population estimate is the abundance of any sex under natural conditions at a given time. Knowing the sex ratio of fish is important for ensuring proportional catches of the two sexes and provides essential information for assessing the reproductive potential of a population [94]. Additionally, sex ratio is an important characteristic of fish populations because it depends on the reproduction, growth, or stagnation of specific species [95]. Under normal conditions, generations generally contain equal numbers of individuals of different sexes, resulting in a 1:1 sex ratio. This ratio can vary during different periods of the year due to different environmental factors and spawning [96]. Deviations from this ratio are commonly observed during winter and early spring, when reproductive migration, gonadal development, and sex-specific behavior can influence catch composition [97]. Females often become more vulnerable to capture during these periods due to increased feeding activity and physiological changes associated with oocyte development, whereas males may exhibit different habitat use or reduced catchability [98,99]. This can also be explained by the characteristic female dominance and gynogenesis observed in European Carassius gibelio populations. The European Carassius gibelio population (Poland) was bisexual, with a female/male ratio of 3.1:1. Triploid and diploid populations were in a 1:1 ratio. The sex ratio among diploids was strikingly even; most males (78.8%) were diploid, and the remainder, surprisingly, were triploid. The sex and ploidy ratios suggest that Carassius gibelio exhibits both gynogenesis and bisexual reproduction [100]. Similarly, in the Galabovo region of Bulgaria, the sex composition of Carassius gibelio populations was determined to be predominantly female (99.30%), witha very low proportion of males (0.70%) [101]. We think that for such reasons, the expected 1:1 ratio was not achieved in the present study, and the combination of winter–spring sampling and gynogenesis likely contributed to the extreme deviation from the expected sex ratio in this species. When female frequency was examined, the proportion of females (%) was 100% in Carassius gibelio, 78% in E. lucius, 67% in B. bjoerkna, 67% in A. brama, 83% in P. fluviatilis, 94% in Cyprinus carpio, 96% in Scardinius erythrophthalmus, and 93% in Silurus glanis. This showed that the females were more abundant than the males in the sampling periods. This suggested that the factors listed above, the breeding season and especially gynogenesis in Carassius gibelio are more influential. In addition, only female individuals of Carassius gibelio were observed during the sampling period. Consequently, this species was not included in the chi-square test, as the absence of males makes the sex ratio data unsuitable for the test and ensures more reliable results. The chi-square (χ²) test showed a significant deviation of sex ratios from the expected distribution across seven species (χ² = 46.79, df = 6, p = 0.0001, p < 0.001). The χ² test indicated an important significant difference in sex ratio (the proportion of females to males) in all the fish species (χ² = 12.28, p = 0.001). The consistent female bias observed across groups suggests non-random structuring of the population, potentially influenced by biological traits or sampling selectivity. Furthermore, the restriction of sampling to the December–April period represents a clear seasonal limitation. Such seasonality may influence sex-specific patterns in length–weight relationships, condition factor, and sex ratios, as the observed female dominance could be associated with spawning aggregations, seasonal growth dynamics, or sex-dependent habitat use reported for many freshwater fish species [30,40].

5. Conclusions

This study provides an updated, biomass-based assessment of the fish community in Lake Sapanca, revealing clear shifts in stock structure, growth patterns, and population condition under increasing environmental and exploitation pressures. The highest standing stock biomass was estimated for Cyprinus carpio and E. lucius, whereas A. brama and Scardinius erythrophthalmus exhibited markedly low stock levels. These findings demonstrate that biomass-based rankings may diverge substantially from numerical abundance patterns, particularly for large-bodied species.

Positive allometric growth (b > 3) predominated in most species, indicating generally favorable feeding conditions and energy allocation to somatic growth. In contrast, the key predatory species E. lucius and Silurus glanis exhibited negative allometric growth, suggesting heightened sensitivity to prey availability, trophic restructuring, and cumulative environmental stress. Despite its high biomass contribution, the low numerical abundance and reduced growth performance of E. lucius indicate a potentially vulnerable population state.

Condition factor analyses showed that invasive or opportunistic species, particularly Carassius gibelio and Cyprinus carpio, were in comparatively good physiological condition, highlighting a concerning combination of strong invasive performance and weakened top-predator growth. This pattern suggests an increasing trophic imbalance within the lake ecosystem.

Overall, the results indicate that Sapanca Lake is experiencing a convergence of declining predator performance, elevated invasive species success, and hydrological stress. Effective management should therefore prioritize the protection of large predatory species, the regulation of invasive and opportunistic stocks, and the integrated use of biomass- and abundance-based metrics in stock assessments to support long-term ecosystem stability.