**2. Materials and methods**

#### **2.1. The study area**

Lake Eğirdir is the second largest freshwater reservoir in Turkey with a total of 457 km2 (48 km x 16 km) surface area [39, 40] and located in the lakes region, southwestern part of Turkey. The maximum depth of lake is 13 m. The water income of the lake is supported from underground water source, surface springs, runoff water, rain and small streams. Evaporation and water flow into Lake Kovada through a channel are main outflows of the lake [41]. The lake is an important source of drinking water as well as tourism and agricultural irrigation. The previous fauna and flora studies carried out in the lake yielded a rich biodiversity. According to the QB/T [42], the Rotifera index showed that the lake has mesotrophic features in terms of zooplankton. Carlson's trophic state index also supports that the lake shows both mesotrophic and eutrophic characteristics. Annual mean concentration of chlorophyll-a (3.0 ± 0.2 mg/ m<sup>3</sup> ) also supported the proposed trophic status of the lake [43]. Zooplanktonic organisms, which is significant part of the lakes, consisted of Rotifera (40 species), Cladocera (22 species) and Copepoda (3 species). Rotifers, *Poyarthra dolichoptera* and *Keratella cochlearis* known as indicator of mesotrophic conditions, were reported to occur predominantly in the lake [43].

A total of 129 algal taxa belonging to six groups Ochrophyta (65 species), Chlorophyta (30 species), Charophyta (13 species), Cyanophyta (12 species), Euglenophyta (6 species) and Myzozoa (3 species) were determined [44]. The average abundance of 24 zoobenthic species was recorded recently as 4.195 individuals/m2 . Dominant species were Oligochaeta with 53.4% relative density. The proportions of Insecta, Bryozoa and Malacostraca were reported as 17.6%, 11.7% and 10.6%, respectively [45].

In the first and most comprehensive study on lake, it was reported that the lake fish fauna consisted of 10 different (*Cyprinus carpio, Schizothorax prophylax, Varicorhinus pestai, Acanthorutilus handlirschi, Vimba vimba, Thylognathus klatti, Aphanius chantrei, Cobitis taenia, Nemachilus angorae, Pararhodeus niger*) species [46]. In fact nine species occurs in the lake since *S. prophylax* and *V. pestai* are synonyms of each other. Perch (*Sander lucioperca*) was the first fish introduced into the lake in 1955. A total of nine different non-native species were reported at different times over a period of about 70 years. Today, there are totally 14 fish species (2 native, 6 endemic, 7 non-native) belonging to 8 families in the lake. These species are listed as *Cyprinus carpio, Vimba vimba* (native), *Oxynoemacheilus mediterraneus*, *Seminemacheilus ispartensis, Cobitis turcica, Capoeta pestai, Pseudophoxinus egridiri, Aphanius anatoliae* (endemic), *Atherina boyeri, Carassius gibelio, Pseudorasbora parva, Knipowitschia caucasica, Gambusia holbrooki* and *Sander lucioperca* (non-native) [41, 47–49].

#### **2.2. Specimen sampling and data analysis**

**1. Introduction**

358 Selected Studies in Biodiversity

Reservoir in Ukraine.

Eğirdir, Turkey.

**2.1. The study area**

**2. Materials and methods**

and could decrease fish biodiversity in Europe [38].

The topmouth gudgeon, *Pseudorasbora parva* (Temminck and Schlegel, 1846) (sub fam. Gobioninae) is a small cyprinid fish distributed in Japan, China, Korea, Hungary, Germany, Serbia, Austria, Greece, Poland, North Africa, Romania, Czech Republic, United Kingdom, Azerbaijan and Ukraine [1–10]. It usually occupies a range of lotic and lentic habitats, including rivers, reservoirs, canals, ponds, shallow lakes and oxbows [11–13]. This species is one of the most effective invasive species to have been introduced into inland waters in Turkey for the past 30 years [14–23]. Generally, the topmouth gudgeon is considered as an important predator on crustaceans, zooplankton, ostracods, molluscs, chironomid larvae, rotifers and benthic organisms. It also feeds on phytoplankton (diatoms and other algae), zooplankton (cladocerans, copepods), the larvae and eggs of native fish species, insects and detritus [7, 24–26]. Although there are some data available on its age, growth, reproduction and habitat

Feeding habits and feeding ecology of topmouth gudgeon were studied by Wolfram-Wais et al. [4] in Neusiedler See (Austria) and Xie et al. [34] in the Biandangtang Lake of China. Hliwa et al. [35] studied the diet of the species in the Kis-Balaton Reservoir, whereas Nikolova et al. [36] investigated seasonal variation in the diet of topmouth gudgeon from shallow eutrophic lakes along River Vit in Bulgaria. Yalçin-Özdilek et al. [37] carried out research on the feeding ecology of the species from Gelingüllü Reservoir and Karakuş [22] studied dietary interactions between non-native species topmouth gudgeon and some native fish species in Sarıçay Stream in Turkey. Didenko and Kruzhylina [10] investigated trophic interaction between topmouth gudgeon and the co-occurring species during summer in the Dniprodzerzhynsk

Asian cyprinid, *Pseudorasbora parva*, causes increased mortality and totally inhibiting spawning of endangered native fish, the European cyprinid *Leucaspius delineatus*. This threat is caused by an infectious pathogen, a rosette-like intracellular eukaryotic parasite that is a deadly, non-specific agent. It is probably carried a vector of an emergent infectious disease

The topmouth gudgeon is successfully inhabited invasive fish in Lake Eğirdir. However, its feeding properties have not been sufficiently studied yet. The aim of the present study was to determine the diet composition of *Pseudorasbora parva* and its prey selectivity in the Lake

Lake Eğirdir is the second largest freshwater reservoir in Turkey with a total of 457 km2 (48 km x 16 km) surface area [39, 40] and located in the lakes region, southwestern part of Turkey. The maximum depth of lake is 13 m. The water income of the lake is supported from

[27–33], published information on the diet of topmouth gudgeon is still scarce.

Fish samples were collected between April and August in the years of 2010 and 2011. All fish caught by the gill nets and purse seine were evaluated in the diet study. Fishing nets with mesh size 10, 16, 45 and 0.9 mm were used. Sampling was performed at two different sites (**Figure 1**), one in the southern of the lake (St 1, 5–7 m of depth) and the other in the southeast (St 2, with 2–5 m depth). Fish specimens were measured to the nearest 0.1 mm fork length (FL) and weighed to the nearest 0.001 g. The contents of the stomach were removed and the empty stomach was reweighed to the nearest 0.001 g. A total of 88 topmouth gudgeon were analyzed. Each prey item was determined to the lowest possible taxonomic level and counted. Proportion of full and empty stomachs was also determined. Volume calculation was used in Malacostraca, Annelida and Insecta. In addition, average volume was estimated [50] for *Disparalona rostrata, Chydorus sphaericus* and *Nitocra hibernica, Mesocyclops leuckarti* (1.0 × 107 μ<sup>3</sup> ), *Graptoleberis testudinaria, Alona* 

non-normally distributed. Therefore, those non-normally distributed data were compared using Wilcoxon signed-rank test, followed by Tukey-Kramer honestly significant difference test (HSD). Wilcoxon test displayed a Wilcoxon rank sums test if there were two groups and a Kruskal-Wallis nonparametric one-way analysis of variance if there were more than two groups. To estimate prey selectivity of topmouth gudgeon, Pearre's selection index (V) [57]

Prey Selection of *Pseudorasbora parva* (Temminck and Schlegel, 1846) in a Freshwater Ecosystem (Lake Eğirdir/Turkey)

\_\_\_\_\_\_\_\_\_\_\_\_\_

is the relative abundance of species a in the environment and bd is the rela-

. Diet similarity among size classes and months was estimated using the

is the relative abundance of all other species in the

, d = a<sup>d</sup> + bd, e = a<sup>e</sup> + b<sup>e</sup>

http://dx.doi.org/10.5772/intechopen.70471

.

361

) where,

, b = b<sup>d</sup> + b<sup>e</sup>

\_\_\_\_\_\_\_\_\_\_ (*a*<sup>∗</sup> *b*<sup>∗</sup> *d*<sup>∗</sup> *e*)

√

The selection index (Va) is statistically tested using the chi-squared test. (X2 = n\*V2

If the C value is bigger than 0.80, it means that the diet of the two groups is similar.

Where Va is Pearre's index for topmouth gudgeon selectivity of species a, ad is relative

Schoener Overlap Index (C) [58]. Cry = 1–0.5 ∑ |pxi-pyi|; where pxi and pyi are the proportions by number of prey type i in the diets of groups (length or season) x and y, respectively.

In this study, topmouth gudgeon ranged from 6.1 to 11.1 cm in fork length (FL) with an average value of 7.71 ± 0.18 cm and their total weight ranged from 3.52 g to 25.49 g, with an average value of 8.13 ± 0.78 g. The number of species, minimum and maximum fork length and minimum and maximum weights from different months in the lake are presented in

**Sampling date The number of fish Min–Max Mean value ±SD Min–Max Mean value ± SD** April 9 6.2–10.0 7.30 ± 0.48 3.85–20.17 7.65 ± 2.15 May 46 6.1–11 7.71 ± 0.18 3.52–25.20 8.13 ± 0.78 June 6 7.4–11.1 9 ± 0.58 7.06–25.49 11.65 ± 2.80 July 10 6.8–9.6 7.88 ± 0.29 4.74–13.05 7.18 ± 0.82 August 17 6.5–10 8.15 ± 0.25 3.7–16.2 8.48 ± 0.89

**Table 1.** Number of the fish caught during the study, their minimum, maximum and average fork length and minimum,

**Fork length (FL, cm) Weight (W, g)**

was calculated.

environment, a<sup>e</sup>

n = a<sup>d</sup> + a<sup>e</sup> + b<sup>d</sup> + b<sup>e</sup>

**3. Results**

**Table 1**.

maximum and average weight.

Va <sup>=</sup> (ad<sup>∗</sup> be) <sup>−</sup> (ae<sup>∗</sup> bd)

tive abundance of all other species in the diet. a = a<sup>d</sup> + a<sup>e</sup>

**3.1. The size and weight ranges of topmouth gudgeon**

abundance of species a in the diet, b<sup>e</sup>

**Figure 1.** Location of the Lake Eğirdir and sampling sites.

*quadrangularis, Alona guttata, Coronatella rectangula, Pleuroxus aduncus* and Nauplii (5.0 × 107 μ<sup>3</sup> ), *Bosmina longirostris* (4.0 × 107 μ<sup>3</sup> ), *Daphnia cucullata* (1.0 × 108 μ<sup>3</sup> ), *Gomphonema* sp., (6.0x104 μ<sup>3</sup> ) and *Pediastrum* sp. (8.0 × 103 μ<sup>3</sup> ). All topmouth gudgeon caught were divided into six size classes according to fork length (FL) measuring 6.0–6.9 cm, 7.0–7.9 cm, 8.0–8.9 cm, 9.0–9.9 cm, 10.0–10.9 cm and 11.0–11.9 cm. Fish weight were classified into four groups: ≤5.0 g, 5.1–9.9 g, 10.0–14.9 g and ≥15.0 g.

Feeding intensity (stomach fullness) was estimated by I<sup>F</sup> = (WSC/W<sup>F</sup> )\*10,000 [51]. Where, IF is the fullness index, WSC is the weight of the stomach contents and W<sup>F</sup> is the weight of the fish. Percentage and frequency of occurrence were used to estimate the dietary importance of each prey category [52, 53]. The percentage of the relative importance index [54] and three-dimensional graphical representations [55] were used to express prey importance. IRI<sup>i</sup> = (N<sup>i</sup> % + W<sup>i</sup> %) \* O<sup>i</sup> %, where W<sup>i</sup> and N<sup>i</sup> are the total net weight and number of prey and Oi is the number of stomachs containing prey i. Shannon-Weaver (H′ ) were used to evaluate the variety of foods in stomach. This index provides a general indication of changes in species diversity [56]. In the first step of statistical analysis, the normality of data was tested for each parameter using the Shapiro-Wilk test, and it was shown that dataset was non-normally distributed. Therefore, those non-normally distributed data were compared using Wilcoxon signed-rank test, followed by Tukey-Kramer honestly significant difference test (HSD). Wilcoxon test displayed a Wilcoxon rank sums test if there were two groups and a Kruskal-Wallis nonparametric one-way analysis of variance if there were more than two groups. To estimate prey selectivity of topmouth gudgeon, Pearre's selection index (V) [57] was calculated.

 $\text{Vara cuncuneter:}$  
$$\text{Va = \frac{(\text{ad}^\* \text{be}) - (\text{ae}^\* \text{bd})}{\sqrt{(a^\* \text{ b}^\* \text{d}^\* \text{e})}}$$

Where Va is Pearre's index for topmouth gudgeon selectivity of species a, ad is relative abundance of species a in the diet, b<sup>e</sup> is the relative abundance of all other species in the environment, a<sup>e</sup> is the relative abundance of species a in the environment and bd is the relative abundance of all other species in the diet. a = a<sup>d</sup> + a<sup>e</sup> , b = b<sup>d</sup> + b<sup>e</sup> , d = a<sup>d</sup> + bd, e = a<sup>e</sup> + b<sup>e</sup> . The selection index (Va) is statistically tested using the chi-squared test. (X2 = n\*V2 ) where, n = a<sup>d</sup> + a<sup>e</sup> + b<sup>d</sup> + b<sup>e</sup> . Diet similarity among size classes and months was estimated using the Schoener Overlap Index (C) [58]. Cry = 1–0.5 ∑ |pxi-pyi|; where pxi and pyi are the proportions by number of prey type i in the diets of groups (length or season) x and y, respectively. If the C value is bigger than 0.80, it means that the diet of the two groups is similar.
