Introduction

Mandarin fish, or Chinese perch, Siniperca chuatsi (Basilewsky), is an important commercial freshwater fish inhabiting in the waters from the south in the Zhujiang River system to the north in the Amur River system (Liang et al. 2001). Mandarin fish is a piscivore throughout its life and is adapted to the nocturnal stalking hunting strategy (Liang et al. 1998, 2001). In the past several decades, the resources of mandarin fish in the Yangtze lakes have declined dramatically due to overfishing and eutrophication. In recent years, artificial stocking of mandarin fish has become commercially significant to recover mandarin fish resources and utilize the abundant small-sized fish as forage species (Cui and Li 2005). This may resolve the conflict between fishery development and water quality conservation, based on the principle of trophic-cascading effects (Carpenter and Kitchell 1988; Liere and Gulati 1992). Previous studies have shown that stocking could be profitable at relatively low yields, reducing the pressure of fisheries on the aquatic environment (Liu et al. 1998). However, low recapture and survival rates of stocked mandarin fish compromise the success of stocking. The availability and palatability of food and the predation efficiency of fish have direct impacts on survival and growth in their early developmental stage of fish (Welker et al. 1994). Given the abundant small fishes in the stocked lakes (Xie et al. 2000; Ye et al. 2006; Li et al. 2010), our hypothesis is that low predation efficiency of juvenile mandarin fish under eutrophic conditions results in a low survival rate.

Increased turbidity can have considerable impacts on lake ecosystems (Bruton 1985; Kirk and Gilbert 1990; Lind 2003), and affects primary productivity and trophic status by reducing the depth of the photosynthetic zone (Bruton 1985). Furthermore, turbidity may influence interactions between predators and prey. Turbidity reduces visibility in the water column by scattering light from inorganic particles and absorption of light from organic suspended particles (Lythgoe 1979).

In China, eutrophication of lakes has been very severe in recent decades. Survey results show that the proportion of eutrophic lakes has risen from 41 % in the late 1970s to 77 % in the late l990s (Jin 2008). The algal blooms resulting from eutrophication ultimately lead to decreased transparency of water and increased algal turbidity caused by organic matter (Elmgren 1989; Bonsdorff et al. 1997). In addition, the Yangtze lakes are mostly shallow, and strong winds can mix the water with sediments and subsequently create clay turbidity (Scheffer 1998).

Increased turbidity can have negative effects on the foraging success of fish (Gregory 1993; Miner and Stein 1993; Gregory and Levings 1998; Radke and Gaupisch 2005; Pekcan-Hekim and Lappalainen 2006; Carter et al. 2010), but does not affect the foraging success of some species (Bergman 1988; Vandenbyllaardt et al. 1991; Reid et al. 1999; Granqvist and Mattila 2004), and can even enhance foraging efficiency (Gregory and Northcote 1993). Little is known on the combined effects of turbidity and light on predation of juvenile piscivorous fish, and no information is available on juvenile mandarin fish.

In this study, we hypothesize that increased turbidity inhibits foraging success in juvenile mandarin fish, thus affecting their survival and growth. We investigated the effects of turbidity and light intensity on foraging success rate of juvenile mandarin fish on crucian carp Carassius auratus in laboratory conditions. The aim of the present study was to determine whether increased turbidity and decreased light intensity negatively affected the foraging efficiency of the juvenile mandarin fish. Our findings will be used to improve management of wild and stocked mandarin fish populations.

Materials and methods

Predator and prey

We used crucian carp as prey, and mandarin fish as the predator. Both species were obtained from a state-owned hatchery and were transported to the experimental base National Research Centre for Freshwater Fisheries Engineering of China (NRCFFEC). The mandarin fish were acclimatized for 2 weeks in two 1300-L aquaria that were filled to a constant depth of 60 cm before the tests. Mandarin fish were fed daily on a diet of silver carp Hypophthalmichthys molitrix larvae. The crucian carp were acclimatized for 2 weeks in three aquaria (675 cm × 400 cm × 80 cm) that were filled to a constant depth of 60 cm and fed daily on commercial mash feed. During the acclimation period, natural temperature (28.2 ± 0.3 °C) and photoperiod were maintained.

Experimental design

Foraging success rates by juvenile mandarin fish were tested in two turbidity types (clay-induced turbidity and algal-induced turbidity) at five levels of turbidity, i.e., 0, 10, 20, 40 and 80 NTU (nephelometric turbidity units) combined with two light intensities imitating daylight (1200–1400 lx) and night (0.01–0.04 lx). Turbidity levels were selected based on our investigation in 30 Yangtze lakes, where turbidity rarely exceeded 80 NTU. There were 20 treatments and each treatment was replicated five times. Controls for each treatment evaluated the non-predatory mortality of crucian carp. All treatments including the controls were randomized among the experimental runs and within experimental arrays.

All experiments were done in a laboratory with transparent roof from July to August in 2009, and were conducted in 20 black plastic 180-L aquaria, with water depth 40 cm. The entire experimental area was enclosed to prevent outside disturbances. Algal-induced turbidity was generated by diluting concentrated algae; the dominant species of algae was Microcystis. Clay-induced turbidity was generated by mixing clay into clear water. The clay was taken from Taojia Lake near to NRCFFEC and filtered through a 50-μm mesh sieve. Slight air bubbling was used to prevent sedimentation of the suspended material during the experiments. The turbidity was measured with a HACH 2100P portable turbidimeter. The turbidity levels remained constant during the experiments and did not vary significantly among treatments and replicates. The five turbidity levels ranged from 0.12 to 0.53 NTU, 9.38–10.83 NTU, 19.35–20.67 NTU, 39.63–40.85 NTU, and 79.17–80.78 NTU, respectively. Fluorescent lights were fixed 120 cm above the aquariums and the light intensity was kept constant for each treatment. Light intensity was measured at the surface of each aquarium before and after each trial with a LI-COR Inc. LI-185B quantum sensor. The mean light intensity at daylight ranged from 1200 to 1400 lx and at night ranged from 0.01 to 0.04 lx.

The daylight treatments were conducted at daytime from 12:00 to 14:00 h and the night treatments at night from 21:00 to 23:00 h. The aquaria were covered with black plastic to prevent disturbance from outside when doing night experiments. After starvation for 24 h, three mandarin fish (mean total length 68 ± 0.7 mm, mean weight 3.97 ± 0.25 g) were placed in each aquarium to acclimatize 24 h before each experiment and were not fed during the acclimation phase. Two hours before the start of a trial, 30 crucian carp (mean total length 25 ± 1.4 mm, mean weight 0.15 ± 0.01 g) in a perforated plastic box were placed in the aquaria in each replicate to acclimate. Experiments were started when the crucian carp were released from the plastic boxes. After 2 h, mandarin fish were taken out and the remaining prey fish were counted. Before and after experiments, water temperature, pH and dissolved oxygen were measured with a WTW Multi 3500i Handheld tester. The water temperature in the experiments varied from 27.6 to 28.6 °C, pH from 8.1 to 8.7, and dissolved oxygen from 6.8 to 7.5 mg L-1. Predator and prey fish were used once only.

Data analysis

The relative foraging success rates were then determined for each treatment. Two-way analysis of variance (ANOVA), after confirmation of normality and homogeneity, was used to determine whether turbidity or light intensity influenced foraging success within each respective turbidity type. Tukey’s HSD post-hoc test was used to detect the significance of differences of means among the turbidity level treatments after ANOVA. Two-way ANOVA was also used to determine eventual differences in prey survival in control treatments. Student’s t-tests were used to detect differences between predation and control treatments. Factorial three-way ANOVA was used for detection of differences in foraging success among turbidity, light intensity and turbidity type for mandarin fish. Differences were considered to be significant at p < 0.05. The statistical analysis was performed using SPSS 13.0 (SPSS Inc., Chicago, IL, USA).

Results

Foraging success of mandarin fish in algal-induced turbidity was significantly lower than in clay-induced turbidity, but turbidity level and light intensity did not have significant influence (Table 1). There was no significant interaction among turbidity, light intensity and turbidity type, and water temperature (27.6 to 28.6 °C) had no significant impact on foraging success when tested as a covariate (Table 1).

Table 1 Three-way ANOVA of the effects of turbidity type, turbidity levels and light intensity on foraging success rates
Full size table

In algal-induced turbidity trials, there were no significant differences in foraging success rate at different turbidity levels within any light level, but foraging success rate in 80 NTU was lower than in other turbidity levels (Fig. 1a, Table 2a). In each turbidity level, there were no significant differences in foraging success rate between daylight and night light intensities, but foraging success rate at night level were higher than at daylight level in 0 NTU and 10 NTU turbidity levels and lower in 40 NTU and 80 NTU turbidity levels (Fig. 1a, Table 2a). No significant interaction was found between turbidity and light intensity (Table 2a).

Fig. 1
figure 1

Foraging success rates of juvenile mandarin fish on crucian carp at different turbidity levels and light intensity in (a) algal-induced turbidity and (b) clay-induced turbidity

Full size image
Table 2 Two-way ANOVA of the effects of turbidity levels and light intensity on foraging success rates
Full size table

In clay-induced trials, there was a slight but insignificant increase in foraging success rate with increasing turbidity at daylight level, and there was also no significant difference in foraging success with increasing turbidity at night level (Fig. 1b, Table 2b). In each turbidity level, there were no significant differences in foraging success rate between daylight and night light intensities, but foraging success rate at night level were higher than at daylight level in 0 NTU, 10 NTU and 20 NTU turbidity levels and lower in 40 NTU and 80 NTU turbidity levels (Fig. 1b, Table 2b). No significant interaction was found between turbidity and light intensity (Table 2b).

Survival rate of crucian carp in control treatments ranged from 97 % to 100 % and there were no significant differences in survival rate among turbidity levels within any level of light (Table 3, two-way ANOVA test, p = 0.53–0.91). For both turbidity types, there were significant differences in survival rate of crucian carp within levels of turbidity and light level between predation and control treatments (t-test, p = 0.00–0.017).

Table 3 Survival rates of crucian carp in control treatments
Full size table

Discussion

Turbidity can be a key factor which influence the interaction between predator and prey in aquatic ecosystems, and has a strong influence on the foraging success rate of visually-oriented fishes, especially for the piscivorous fishes (Abrahams and Kattenfeld 1997; Radke and Gaupisch 2005). The results in our study indicated that there was no negative influence on the foraging success rate of juvenile mandarin fish by increasing turbidity (0–80 NTU). This was inconsistent with previous studies that reported increasing turbidity would negatively influence the foraging success of predator (Gregory 1993; Gregory and Levings 1998; Radke and Gaupisch 2005; Pekcan-Hekim and Lappalainen 2006). Generally, increased turbidity can decrease reaction distance and reduce the visual range of a predator, thus reducing the foraging efficiency and prey selectivity of vision-oriented fish (Vinyard and O’Brien 1976; Utne-Palm 1997; Sweka 1999; Davies-Colley and Smith 2001; Shoup and Wahl 2009). However, some studies have also shown that increased turbidity had no effect on the foraging success (Vandenbyllaardt et al. 1991; Reid et al. 1999; Granqvist and Mattila 2004), and even enhanced the foraging efficiency (Gregory and Northcote 1993).

The effects of turbidity on foraging success of different fish species are not consistent. One possible reason for this inconsistency is the changed behaviours of predator and prey (Reid et al. 1999). Foraging success of adult rainbow trout Oncorhynchus mykiss on salmon fry Oncorhynchus nerka was affected by turbidity (Ginetz and Larkin 1976), but no effect was found when adult cutthroat trout was the predator (Gregory and Levings 1996). Predation rates by smallmouth bass Micropterus dolomieu decreased with increasing turbidity, and smallmouth bass consumed more golden shiners Nitemigonus crysoleucas (a pelagic fish) but fewer round goby Neogobius melanostomus (a benthic fish) with increasing turbidity (Carter et al. 2010). In addition, some other factors such as experimental period and the size of the experimental containers should be taken into account. For example, experiments with juvenile walleye, a species adapted to low light and turbid conditions, showed that feeding was inhibited at above 100 NTU in 1-h trials but not in 4-h trials (Vandenbyllaardt et al. 1991). Also, de Lafontaine and Leggett (1987) found that the foraging success rate of the juvenile predator in large containers (890 to 6346 L) was significantly lower than in the small container (262 L). The same result was found in the study by Pekcan-Hekim and Lappalainen (2006), who used Eurasian perch Perca fluviatilis as a predator in experimental trials.

Considering the distinctive feeding habits of mandarin fish and the experimental conditions, we consider that the lateral line system of mandarin fish and experimental container size were the main reasons for our result that increased turbidity had no effect on the foraging success. Some fish species can use non-visual senses such as chemical and mechanical senses for detecting prey (Wootton 1990; Vilhunen and Hirvonen 2003; Mogdans 2005). Predatory fish that locate their prey by non-visual senses may not be affected by turbidity. The lateral line, a specific mechanical sense organ of fish and aquatic amphibians, often plays an important role for predation when their vision was limited (Coombs et al. 1989; Montgomery and Milton 1993; Liang 1996; Carton and Montgomery 2004). Previous study has suggested that mandarin fish recognized their prey visually (Wu and Hardy 1983). However, another studies showed that mandarin fish were able to feed on live prey fish when either eyes or lateral lines were intact or functional, but could scarcely feed when deprived of both of these two senses (Liang 1996; Liang et al. 1998), which indicated mandarin fish could detect and locate their prey using lateral line system when their vision was limited in turbid water. In addition, any effects of turbidity on predation should be driven by affecting the encounter rates between predator and prey (Gregory and Levings 1998). Generally, it is unlikely that encounter rates between highly mobile predators and prey will decrease in all but very turbid conditions in an aquarium (Granqvist and Mattila 2004). A study of predation by Eurasian perch in 10 to 45-L plastic bags suggested that turbidity was an important environmental factor that had negative effects on the predation. Probability of predation did not depend significantly on volume, though comparisons of predation over a large range of container sizes could have been problematic (Pekcan-Hekim and Lappalainen 2006). The size of experimental containers in our study was 180 L, which was extremely limited compared to the lake environments into which mandarin fish are stocked. The limited size of tanks in our study might have impaired the ability of the crucian carp to escape from predation by mandarin fish, possibly compensating for the influence of turbidity on the foraging success rate.

Predation can also be affected by light intensity. Generally, the reactive distance and foraging success reduce with decreasing light intensity (Grecay and Targett 1996; Vogel and Beauchamp 1999). Our study showed that light intensity had no significant influence on foraging success of juvenile mandarin fish in two different turbidity types. This corresponded with another result showing that mandarin fish larvae (21 days from hatching) and yearlings were able to feed in complete darkness as well as in light (Liang et al. 2008). Several studies have reported that reactive distance for predator and prey increased rapidly as light level increased, reaching relatively constant reaction distances at higher light levels (Utne-Palm 1997; Mazur and Beauchamp 2003; Horppila et al. 2004). Blaxter (1986) thought that 0.1 lx was the average threshold for feeding in young larvae, indicating that the reactive distance of prey and predator would be constant when the light intensity higher than 0.1 lx. In this study, the light intensities were above or below this average threshold, but no significant differences were found on the foraging success rates of juvenile mandarin fish. This was possibly related to predator’s ability to find and locate prey by both vision and lateral line system.

The foraging success of fish can be affected by turbidity types, which may affect the visual contrast of prey because different particles and molecules (organic and inorganic matter) in the water scatter and absorb light differently (Lythgoe 1979). Our results show that the foraging success rate in algal-induced turbidity was significantly lower than in clay-induced turbidity. Similar results were found in the other study that the relative capture success of Eurasian perch on roach Rutilus rutilus (Radke and Gaupisch 2005). Some authors considered the effects of turbidity types induced by algae compared to clay were often confused because algae absorb a large amount of visible light compared to clay or bentonite that scatter light (Granqvist and Mattila 2004; Radke and Gaupisch 2005). However, the scattering properties of organic or inorganic matter in water but not absolute light intensity seem to be the main reason for the differences found between algae- and clay-induced turbidity (Benfield and Minello 1996; Vogel and Beauchamp 1999; Davies-Colley and Smith 2001; Granqvist and Mattila 2004).

Our study aimed to evaluate the effect of turbidity and light intensity on the foraging success of juvenile mandarin fish on crucian carp, and both had surprisingly little effect. This suggests that turbidity level and light intensity are probably not the main reason for the low survival rate of stocked juvenile mandarin fish in lakes. However, the lateral of mandarin fish, the limited experimental environments, and low habitat heterogeneity in our experiments may underestimated the negative effect of turbidity and low light levels compared to the environment in the stocked lakes. Therefore, eliminating the use of the lateral line, a larger experimental water body, and greater habitat heterogeneity (e.g., macrophyte species and density) should be considered in future studies.