This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Introduction
Pipefish are representatives of ichthyofauna from vegetated habitats in coastal and estuarine areas (Howard & Koehn 1985; Kuiter 2000). These are ideal habitats for feeding, reproduction, protection against predators and wintering (Teixeira & Musick 1995). Decaying algae and dead seagrass are known to serve as a shelter and refuge, and even as means of transportation to shallower or deeper waters (Teixeira & Musick 1995). Syngnathus abaster (Risso 1827) is a euryhaline pipefish species adapted to freshwater bodies with sandy, muddy and detritus-containing habitats, such as lagoons and lakes (Cakić et al. 2002; Didenko et al. 2018). Syngnathus acus (Linnaeus 1758) occurs mostly in densely vegetated regions of sandy and muddy estuarine waters (Dawson 1986).
Lake Bafa is an alluvial lake located at the southeastern end of the Aegean region and connected to the Greater Meander delta (Turgutcan 1957). The lake is fed by the Meander River and therefore harbors a wide variety of fish species, thus representing a large ground for research on ichthyofauna. Changes in the populations of diadromous and potamodromous fish species are closely associated with the connection between lakes and rivers (Sarı et al. 1999). Research on fish species in Lake Bafa dates back 30 years and so far 20 species have been identified in the lake (Kasparek 1988; Balık & Ustaoglu 1989; Kuru et al. 2001). In addition to economic species, species from the Syngnathus genus occur in the lake (Sarı et al. 1999; Kuru et al. 2001; Gurkan & Innal 2018). In Europe, research on feeding strategies, especially of S. abaster, is common in certain freshwater habitats and estuarine systems, but there are no studies on S. acus. Pipefish are passive predators (Howard & Koehn 1985). They select their prey depending on the habitat in which they live and generally feed on small crustaceans (Ryer & Orth 1987; Teixeira & Musick 1995). Moreover, the mouth structure and size of prey also affect their choice of prey groups in microhabitats in which they hunt (De Lussanet & Muller 2007). Prey groups of S. abaster include phytal organisms, copepods and small crustaceans (Franzoi et al. 1993; Oliviera et al. 2007; Didenko et al. 2018), while prey groups of S. acus include planktic crustaceans and zooplankton (Moreira et al. 1992; Taskavak et al. 2010).
The study carried out by Franzoi et al. (1993) involved feeding habits of two Syngnathus species in an estuarine system of the Mediterranean Sea (Po River delta). Research by Vizzini & Mazzola (2004) on the trophic structure of pipefish species from the western Mediterranean coast is important, because the authors addressed the issue in detail. The most recent study, according to our review of the relevant literature, determined the feeding model of S. abaster in a freshwater habitat in Ukraine (Didenko et al. 2018). The first detailed study on feeding habits of S. acus on the European coasts (Aegean coast) was carried out by Taskavak et al. (2010). A subsequent study determined feeding habits of various species, including S. acus and S. abaster, from the Black Sea coast (Gürkan & Uncumusaoglu 2016). Both S. abaster and S. acus are listed in the Least Concern (LC) category of the International Union for Conservation of Nature (IUCN) Red List of Threatened Species (Smith-Vaniz 2015; Pollom 2016). Thus, the lack of knowledge about the populations of the two species is somewhat remedied, but there is a need for research on how these two species, adapted to estuarine systems, feed and share food in alluvial lake habitats.
This is an important study, because it is the first time the trophic ecology of two congeneric pipefish species occurring in alluvial Lake Bafa (Aegean region, Turkey) has been determined. The study investigates changes in prey group choices of both Syngnathus species with respect to two seasons (autumn and spring), length groups, sex and morphological differences in the mouth apparatus.
Materials and methods
Sample collection
Lake Bafa is a natural barrier lake located in the southeastern region of the Greater Meander River delta and has an indented shoreline. Pipefish samples were collected seasonally, from November 2014 to March 2016, by seining in the coastal area of the Kapıkırı region (between 37°30′N–27°31′E and 37°29′N–27°31′E), the maximum depth of which is 1–1.5 m (Fig. 1). The potential effect of the mesh opening was ignored. Ethics committee regulations were followed when fish species were caught. A total of 205 fish individuals were caught (S. abaster – 76 ind., S. acus – 129 ind.). The collected fish samples were brought to the laboratory in sealed vessels containing 70% ethanol. In the laboratory, the total length of each fish specimen (TL, mm) was measured using a measuring board, the total weight (W, g) was determined using an analytical balance with a sensitivity of 0.01 g and the mouth width (MW) was measured using a digital caliper with an accuracy of 0.001 mm. Fish samples were assessed according to three length groups (Length Group I: < 60 mm, Length Group II: 60–100 mm and Length Group III: >100 mm). The sex of the collected specimens was determined macroscopically based on the pouch in males (Vincent et al. 1995). Females were dissected on their ventral side up to the anus and gonads were analyzed macroscopically.
Figure 1
Fish samples
The tubular gastrointestinal system of pipefish is not fully developed. For the purpose of examining digested food particles, the anterior of the gastrointestinal tube was defined as the gut. The guts were dissected and removed. The gut content analysis was performed on 149 samples with full guts. The dissected guts were kept in 4% formalin solution. Empty and full guts were examined under a binocular microscope. Prey items found in full guts were identified according to different groups. Food particles were counted and weighed with a sensitivity of 0.0001 g. Dominant prey groups were determined in the gut contents of both fish species.
Data analysis
Fish with empty and full guts were determined. The feeding activity was described using the fullness index FI (Hureau 1969). The fullness index was calculated using the following formula: FI = 100*(Gut W) *(W)−1, where Gut W is the gut content weight and W is the fish weight.
Frequency of gut contents (F%), numerical occurrence (N%), weight percentage (W%), the index of relative importance (IRI) and the percent index of relative importance (IRI %) with respect to length groups and two seasons were separately calculated in accordance with Liao & Larscheid (2001). The t-test was used both to determine the differences in the gut contents with respect to length groups and seasons as well as the correlation and regression relationships between the total length and mouth width (Sokal, Rohlf 1969). One-way ANOVA was performed using the Statistica software.
Results
Syngnathus abaster
A total of 76 samples (44 females, 25 males, 7 immatures) were collected in the study period, i.e. in spring and autumn. The total length of the samples varied from 55 mm to 130 mm, with Length Group II containing the largest number of samples (75%). The gut fullness ratio in the samples was 68.42% and it was higher in female samples (75%) than in male samples.
Eight major prey groups were identified in S. abaster depending on its feeding activity. Amphipods were the most dominant prey group (F%, 68.09%) in the food composition and C. aquaedulcis was the most important prey group (IRI%; Table 1). The highest gut fullness index was determined in spring and was 6.70, whereas the lowest fullness index was 0.36 and was determined in autumn in Length Group II (Table 2; Fig. 2). Table 2 shows the food composition of S. abaster samples in terms of the length groups.
Relationship between the total length and the stomach fullness index in S. abaster
According to Table 2, guts in the samples of Length Group I were empty, while the samples in Length Group II contained eight prey groups. In these medium-size samples, amphipods and C. aquaedulcis (56.10%, F%) were the most dominant prey groups, while amphipods were the most important prey group (IRI%, 82.12%). On the other hand, the same fish samples contained also cyclopoid copepod, ostracod, and amphipod particles, chironomid larvae, and unidentified eggs. Moreover, the same group preferred pipefish larvae as food. Length Group III preyed on four prey groups. Amphipods were both the most dominant and important prey group in these large fish samples, while C. aquaedulcis was the second most important prey group. Syngnathus larvae and unidentified eggs were preferred so frequently that the question of cannibalism was raised. The results of the t-test revealed no significant differences between the prey groups in the length groups (p > 0.05). The food composition in S. abaster and S. acus samples was also compared for the same length group. The comparison revealed significant differences between the food composition in the length groups of both species (Group I: t-value = 16.68, p = 0.00001; Group II: t-value = 27.315, p = 0.0005; Group III: t-value = 27.314, p = 0.0001, p < 0.05). Table 3 shows prey groups consumed by S. abaster specimens in autumn and spring.
Food composition in S.abaster stomachs in two seasons
Prey items
Autumn
Spring
F%
IRI%
F%
IRI%
Copepods
Calanipeda aquaedulcis
0.00
0.00
96.00
97.55
Other cyclopoids
6.67
3.22
0.00
0.00
Ostracods
16.67
3.63
0.00
0.00
Amphipods
53.33
76.81
64.00
2.17
Amphipod pieces
43.33
0.00
0.00
0.00
Diptera
Chironomidae larvae
6.67
16.14
0.00
0.00
Fish larvae (Syngnathus sp.)
3.33
0.18
4.00
0.002
Unknown eggs
0.00
0.00
8.00
0.28
Only six prey groups were favored by individuals caught in autumn. Amphipods were the most dominant and important prey group in autumn, followed by chironomid larvae, which are aquatic insects. Fish larvae were also preferred, albeit at a low level. Four prey groups were consumed in spring and C. aquaedulcis and amphipods were both the most dominant and important prey groups (Table 3). However, there were no statistically significant differences between the food groups in the S. abaster samples across seasons (t-value = 1.285, p = 0.219, p > 0.05).
Syngnathus acus
A total of 129 samples (69 females, 30 males, 30 immatures) were collected during the study period. The total length values of the samples ranged from 55 mm to 125 mm and Length Group II contained the largest number of samples (57.5%). The gut fullness ratio for the species was 75.1% and it was higher in female samples (70%) than in male samples (63.76%). Of the 16 prey groups identified based on the feeding activity of S. acus, 15 prey groups were most frequently identified. According to the food composition, amphipods were the most dominant and important prey group (66.15%, 54.04%; Table 4).
Stomach contents in Syngnathus acus
Prey items
F%
N%
W%
IRI %
Algae
Diatom (Surirella sp.)
1.54
0.15
0.00
0.004
Brown algae pieces
1.53
0.00
0.00
0.00
Copepods
Calanipeda aquaedulcis
44.62
51.35
1.58
42.37
Other Calanoids
1.54
0.30
0.002
0.009
Oithona sp.
1.54
0.15
0.009
0.004
Other cyclopoids
10.77
3.61
0.64
0.82
Harpacticoids
1.54
0.15
0.0002
0.004
Ostracods
4.61
0.45
0.24
0.06
Mysids
1.54
0.15
0.09
0.007
Amphipods
66.15
42.62
2.90
54.04
Amphipod pieces
36.92
0.00
0.00
0.00
Gastropods
3.08
0.15
0.01
0.009
Diptera
Chironomidae larvae
1.54
0.15
46.96
1.29
Other Diptera larvae
1.54
0.15
46.69
1.29
Fish larvae (Syngnathus sp.)
3.08
0.30
4.35
0.08
Fish scales
1.54
0.15
0.00
0.04
Total prey number
664
Number of full stomachs
65
The highest value of the gut fullness index was 17.9 and was determined in Length Group II in spring, while the lowest value was 0.22 and was determined in Length Group III (Table 2, Fig. 3).
Figure 3
Relationship between the total length and the stomach fullness index in S. acus
According to Table 5, samples in Length Group I contained only one type of prey group (C. aquaedulcis). Samples in Length Group II contained an increased number of prey groups, totaling 13 prey groups. The dominant prey groups in these medium-size fish samples were amphipods and C. aquaedulcis, with C. aquaedulcis being the most important prey group. However, chironomid larvae, unidentified eggs and algae pieces were not included in the food composition. Nine prey groups were identified in Length Group III. Amphipods and amphipod pieces were both the most dominant and most important prey groups in these large-size fish samples. There was an increase in the consumption of larvae (Chironomidae and Syngnathus sp.) in the group. The results of the t-test revealed no statistically significant differences between the prey groups in the length groups (p > 0.05).
Food composition in S. acus size groups
Prey Items
Length groups
< 60 mm
60–100 mm
100 mm >
F%
IRI%
F%
IRI%
F%
IRI%
Algae
Diatom (Surirella sp.)
0.00
0.00
2.13
0.004
4.00
0.00
Brown algae pieces
0.00
0.00
0.00
0.00
4.00
0.00
Copepods
Calanipeda aquaedulcis
100.00
100.00
51.06
76.57
16.00
3.23
Other Calanoids
0.00
0.00
2.13
0.01
0.00
0.00
Oithona sp.
0.00
0.00
2.13
0.004
0.00
0.00
Other Cyclopoids
0.00
0.00
14.89
0.72
4.00
0.02
Harpacticoids
0.00
0.00
2.13
0.004
0.00
0.00
Ostracods
0.00
0.00
4.26
0.02
4.00
0.02
Mysids
0.00
0.00
0.00
0.00
4.00
0.02
Amphipods
0.00
0.00
59.57
22.47
60.00
96.65
Amphipod pieces
0.00
0.00
31.91
0.00
40.00
0.00
Gastropods
0.00
0.00
2.13
0.04
0.00
0.00
Diptera
Chironomidae larvae
0.00
0.00
0.00
0.00
4.00
0.03
Other Diptera larvae
0.00
0.00
2.13
0.18
0.00
0.00
Fish larvae (Syngnathus sp.)
0.00
0.00
2.13
0.01
4.00
0.02
Fish scales
0.00
0.00
2.13
0.004
0.00
0.00
Table 6 shows the prey groups identified in the S. acus samples collected in autumn and spring.
Prey groups consumed by S. acus during two seasons
Prey items
Autumn
Spring
F%
IRI%
F%
IRI%
Algae
Diatom (Surirella sp.)
3.13
0.22
0.00
0.00
Brown algae pieces
0.00
0.00
1.35
0.00
Copepods
Calanipeda aquaedulcis
3.13
0.22
37.84
93.19
Other calanoids
3.13
0.44
0.00
0.00
Oithona sp.
3.13
0.23
0.00
0.00
Other cyclopoids
3.13
0.23
8.11
0.0002
Harpacticoids
0.00
0.00
1.35
0.0002
Ostracods
9.38
1.98
0.00
0.00
Mysids
3.13
0.22
0.00
0.00
Amphipods
50.00
85.18
36.49
6.75
Amphipod pieces
43.75
0.00
13.52
0.00
Gastropods
3.13
0.23
0.00
0.00
Diptera
Chironomidae larvae
3.13
9.18
0.00
0.00
Other Diptera larvae
0.00
0.00
1.35
0.06
Fish larvae (Syngnathus sp.)
6.25
1.42
0.00
0.00
Fish scales
3.13
0.22
0.00
0.00
Thirteen prey groups dominated in the samples in autumn. Amphipods and amphipod pieces were the most dominant and important prey group in autumn. In addition, Ostracods, Chironomidae larvae and fish larvae were preferred, albeit at a low level. The collected fish individuals consumed seven prey groups in spring, and C. aquaedulcis and amphipods were both the most dominant and most important prey groups. On the other hand, there were no statistically significant differences between the prey groups identified in the S. acus samples with respect to seasons (t-value = 1.406, p = 0.169, p > 0.05). Figure 4 shows the relationship between the total length and the mouth width. The food composition of the S. abaster and S. acus samples was also compared for the same season. The comparison revealed no statistically significant differences between the food composition of the two species in either autumn (t-value = 1.341, p = 0.194, p > 0.05) or spring (t-value = 0.355, p = 0.727, p > 0.05).
Figure 4
Relationships between the total length and the mouth width in pipefish species
The mouth width (MW) of S. abaster was greater than that of S. acus (Fig. 4). There was a significant difference between the two species in terms of mouth width (t-value = 2.457, p = 0.015, p < 0.05). According to the regression relationship between the total length and the mouth width, the latter in the S. abaster samples gradually increased until the total length value reached 95 mm (MW = 0.044, TL − 2.4463, r2 = 0.48). On the other hand, feeding continued in sampled S. acus, although there was no significant change in the width of mouths as their total length increased (MW = 0.0193, TL − 0.7534, r2 = 0.449). Figure 5 shows the relationship between the total length and the weight of digested food.
Figure 5
Relationship between the total length and the total weight of ingested food in pipefish species
Figure 5 shows the regression relationship between the total length (TL) and the food weight for both species. The figure reveals that feeding in the captured individuals of S. acus, which showed a wide distribution considering their increasing total length values, was more intense than in the S. abaster individuals. However, the results of the t-test revealed that the difference between the species in terms of food was not significant [t-value = −0.058, p = 0.95, p >0.05; y(abaster) = 0.00009, r2 = 0.124; y(acus) = 0.0003TL − 0.0157, r2 = 0.456].
Discussion
Pipefish species are bony fish with high trophic specialization, distinguished from other demersal fish species by distinct mouth morphology and feeding habits (Kendrick & Hyndes 2005). The main food items of pipefish are usually groups of small crustaceans (Howard & Koehn 1985; Tipton & Bell 1988; Franzoi et al. 1993; Lyons & Dunne 2004; Taskavak et al. 2010). The mouth apparatus of pipefish has a structural effect on prey capture (De Lussanet & Muller 2007). In this study, the length distribution of S. abaster individuals was greater than that of their relatives living in a coastal lagoon (Spain) and similar to the length distribution of S. abaster samples collected in an estuary in Greece. Similarities and differences between the length and weight values depend on season, habitat, hunting strategy and time, changes in environmental factors (water temperature, waves, etc.), gonad development, gut fullness, and sex (Tesh 1971; Bagenal & Tesch 1978).
The food of S. abaster consists of zooplankton that include copepods (Didenko et al. 2018), benthic organisms such as amphipods (Vizzini & Mazzola 2004), phytal harpacticoids (Franzoi et al. 1993), Cirripedia and Harpacticoida (Gürkan & Uncumusaoğlu 2016). The food of S. acus consists of Harpacticoida and amphipods (Taskavak et al. 2010), planktic crustaceans, zooplankton (Moreira et al. 1992) and Euterpina acutifrons (Gürkan, Uncumusaoğlu 2016). Benthic organisms from six classes (Bivalvia, Crustacea, Gastropoda, Polychaeta, Insecta, and Isopoda) occur in Lake Bafa (Tuna 2015). However, there were no significant differences between the two species in terms of food composition (t-value = −0.058, f-ratio = 1.045, p > 0.05). It has been reported that some congeneric species can feed in different microhabitats without competing (Franzoi et al. 1993; Vizzini & Mazzola 2004). This is consistent with the results of this study, which revealed that the species feed on different habitats in the lake without causing any difference in prey composition (t-value = 2.457, p = 0.015, p < 0.05). Campolmi et al. (1996) emphasized the mouth effect on food sharing in the environment. The difference is essentially attributable to the differences in the mouth aperture of the two species that are both passive predators, as well as the differences in the predatory strategy, which involves filtering from the medium or foraging (Howard & Koehn 1985; Franzoi et al. 1993). The mouth apparatus of S. abaster is morphologically short and cylindrical, while the mouth apparatus of S. acus is relatively longer. Franzoi et al. (1993) and Vizzini & Mazzola (2004) pointed out that morphological adaptations in congeneric species can be important factors in determining differences in ecological niches. This indicates that S. abaster with its short snout filters food from rich habitats, while another Syngnathus species with longer mouths capture their prey that floats above the water column by aiming (Franzoi et al. 1993). Thus, given the prey types identified in the samples, it is clear that S. abaster and S. acus maintain a similar niche in the lake habitat.
In this study, the captured individuals of S. abaster fed on 17 prey groups and amphipods were the dominant prey group in their food composition, while C. aquaedulcis was the most important prey group. Among 36 prey groups of the same species in the Stugna River, copepods were the most important prey (Didenko et al. 2018). Copepods were also the most important prey group on the western Mediterranean (Stagnon di Marsela) coast (Vizzini, Mazzola 2004). Copepods and Cirripede nauplii were the most dominant and important prey group on the coast of the Black Sea (Gürkan & Uncumusaoğlu 2016). Copepod species are irreplaceable for S. abaster both in coastal areas and other estuary regions. A total of 17 prey groups were identified depending on the feeding activity of S. acus, with 15 prey groups being the most frequently found. Amphipods were both the most dominant and important prey group in the food composition. There are two studies investigating the food composition of S. acus in Turkey. According to the study by Taskavak et al. (2010), the Aegean populations of S. acus preferred harpacticoid copepods and amphipods, while according to the study by Gürkan & Uncumusaoğlu (2016), the populations of the species from the Black Sea fed mainly on Euterpina acutifrons. Seasonal and vertical distribution of prey groups in their microhabitats and habitat structure (sandy, vegetated, etc.) are important factors in the feeding of Syngnathidae species (Howard & Koehn 1985). Seasonal changes affecting water temperature impacted the feeding activity of both species during the study period. In this study, the mean water temperature in autumn and spring was 19.2°C and 17.8°C, respectively. In Lake Bafa, the feeding activity of S. abaster decreased in autumn and increased in spring. The S. acus samples contained considerably high levels of food in spring. The low water temperature in the lake during spring was due to the annual precipitation, and the increase in the number of planktic organisms in the lake during this season led to increased feeding activity. Didenko et al. (2018) emphasized that the feeding activity of fish species frequently changed depending on water temperature. The gut fullness ratios of S. abaster and S. acus differed by sex. The gut fullness ratio in female samples was higher than that in male samples. In addition to differences in the food composition depending on species, the feeding ecology of the sexes also differs (Vizzini, Mazzola 2004). Stefee et al. (1989) reported that female samples from two different pipefish species indicated that females were more successful than males in terms of feeding. This is mostly attributable to the lower ability of male fish to capture prey due to providing brood care.
The number of studies investigating the food consumption and dominant prey groups in S. abaster and S. acus with respect to length groups has increased since the 2000s (Vizzini & Mazzola 2004; Taskavak et al. 2010; Gürkan & Uncumusaoğlu 2016; Didenko et al. 2018). In Lake Bafa, S. abaster and S. acus preferred different prey groups depending on their length groups. Moreover, considering the increasing length values, the S. acus individuals fed more intensively than the S. abaster individuals. While there were fish with empty guts among the S. abaster samples from the short length group, Length Group II contained amphipods and C. aquaedulcis among eight prey groups and Length Group III contained amphipods among four prey groups. It is clear that the sampled fish preferred larger prey as their length increased. S. acus from the short-length samples preyed on C. aquaedulcis and S. acus from Length Group II preyed on amphipods and C. aquaedulcis, while fish from the large-length samples preyed on amphipods. Pipefish have restricted mouth gape, which affects their choice of prey groups (Ryer & Orth 1987; Oliviera et al. 2007; De Lussanet & Muller 2007).
Considering the seasons, amphipods and chironomid larvae were the most dominant and important prey groups for sampled S. abaster both in spring and autumn. S. acus consumed amphipods and low levels of chironomid larvae. In addition to being important bioindicators of the nutritional efficiency of a lake (Tasdemir & Ustaoglu 2005), chironomid larvae are the ideal prey group for pipefish.
Conclusion
The results of the study showed that zooplankton was the main prey group in the alluvial lake environment, and the differences in feeding habits of the two species were attributable to the differences in their mouth apparatus. However, the effect of pipefish on the food ecology of the lake was not significant. On the other hand, maintaining the existing aquatic vegetation in the lake will help in the survival of the fish species and allow more detailed studies.
