WaterSubject

A comparative limnological study was carried out to present a snapshot of crustacean zooplankton communities and their relations to environmental factors to test whether there is a consistent relationship between crustacean biomass and trophic indicators among lake groups with similar trophic conditions. The study lakes showed a wide range of trophic status, with total phosphorus (TP) ranging from 0.008 to 1.448 mg L^?1, and chlorophyll a from 0.7 to 146.1 ?g L^?1, respectively. About 38 species of Crustacea were found, of which Cladocera were represented by 25 taxa (20 genera), and Copepoda by 13 taxa (11 genera). The most common and dominant species were Bosmina coregoni, Moina micrura, Diaphanosoma brachyurum, Cyclops vicinus, Thermocyclops taihokuensis, Mesocyclops notius and Sinocalanus dorrii. Daphnia was rare in abundance. Canonical correspondence analysis showed that except for four species (D. hyalina, S. dorrii, C. vicinus and M. micrura), almost all the dominant species had the same preference for environmental factors. Temperature, predatory cyclopoids and planktivorous fishes seem to be the key factors determining species distribution. TP was a relatively better trophic indicator than chlorophyll a to predict crustacean biomass. Within the three groups of lakes, however, there was no consistent relationship between crustacean biomass and trophic indicators. The possible reason might be that top-down and bottom-up control on crustaceans vary with lake trophic state. The lack of significant negative correlation between crustacean biomass and chlorophyll a suggests that there was little control of phytoplankton biomass by macrozooplankton in these shallow subtropical lakes.

reported that pelagic cladocerans of large lakes in the eastern hemisphere were mainly composed of Diaphanosoma, Daphnia, Ceriodaphnia, Bosmina and Moina. These genera have a great significance in terms of occurrence and biomass in our study lakes. They are also common genera in temperate and tropical waterbodies. With respect to community structure, found in St. Lawrence Great Lakes that cyclopoids (Cyclops, Mesocyclops) and cladocerans (Daphnia and Bosmina) dominated eutrophic lakes. Although cyclopoids were the dominant copepods in our lakes, the dominant cladocerans differed from the Great Lakes. The cladocerans in most of our lakes were dominated by Bosmina, Moina and Diaphanosoma. Daphnia was only dominant in five lakes in spring. The three dominant cladocerans had relatively high occurrence and dominance in our lakes, although being less competitive in exploiting resources than Daphnia demonstrated in the PEG-model that temperature, food resources and fish predation had great importance in structuring the zooplankton community. There are studies showing that Bosmina, Moina and Diaphanosoma are all resistant to planktivorous fish predation. However, it was not the case for Daphnia . On the other hand, cyclopoid copepods, despite having much lower biomass, can also account for a greater portion of the predation mortality on herbivorous zooplankton than planktivorous fish . The biplot of the CCA analysis showed that Daphnia deviated from the predatory cyclopoids (C. vicinus, T. taihokuensis and M. notius) but Bosmina, Moina and Diaphanosoma peaked with them. Thus, the disappearance of Daphnia and the increasing dominance of invulnerable cladocerans might be an outcome of combined predation arising from predatory copepods and planktivorous fishes.

In addition, D. hyalina, S. dorrii and C. vicinus peaked at low temperature, while other species at relatively high temperature. In our lakes, the above three species and M. micrura were found at an average temperature of 20.0, 21.9, 17.9 and 28.7 °C, respectively. observed that D. hyalina occurred below a temperature of 20 °C and the population could survive in winter but disappeared in May or June. also found temperature played a role in determining the dominance of daphnids in subtropical Florida lakes. C. vicinus was a predominant early spring species in eutrophic lakes and was replaced by two typical thermophilic species, T. taihokuensis and M. notius. documented that C. vicinus preyed on M. notius and even replaced it to be dominant in winter and spring. Moreover, C. vicinus has a higher rate of development at a temperature below 25 °C. Therefore, the seasonal temperature changes may be another reason responsible for the shift of crustacean community.

In the present study, the measured environmental factors only explained 38.4% of the total variance of crustacean species distribution, suggesting that many other undetermined factors within lakes are also of great importance in structuring the crustacean community. These undetermined factors might include bacteria, invertebrate predator such as Chaoborus, and planktivorous fishes.

Response of crustacean biomass to trophic indicators

TP was a relatively better predictor of the biomass of crustacean groups than chlorophyll a in our study, as suggested by . However, in the study of, the ability of TP to predict crustacean biomass in subtropical region was only based on the significant relationship between cladocerans and TP in five Florida lakes. The number of lakes included in the study was low. Therefore, the conclusion needs to be further justified. In our 29 subtropical lakes, the pooled biomass of crustaceans, cladocerans and copepods were all significantly correlated with TP but not with chlorophyll a. Therefore, our study supplied a more comprehensive evidence to support the power of TP in predicting zooplankton biomass. TP is also considered to be better for reflecting the total food materials edible than chlorophyll a.

In the three groups of lakes, however, only a few strong correlations were found. Some studies also fail to find significant relationships, and the possible reason is attributed to differences in predation pressure exerted by fishes or invertebrates, or the lack of nannoplanktonic food resources. The different response of lake groups to trophic indicators suggests that top-down and bottom-up control on crustaceans also varied with trophic state in our lakes. The second group of lakes was associated with better prediction of crustacean biomass, suggesting that crustacean biomass may be bottom-up controlled. The first group of lakes was highly eutrophic and seven of the eight lakes were stocked with abundant filter-feeding fishes. Exceptionally, copepods were negatively correlated with phytoplankton biomass. Although studies exist showing that copepods have a strong impact on the size structure of phytoplankton, they show no impact on the decline of phytoplankton biomass . Thus, the most probable explanation for the negative interaction might be fish predation, which depresses copepod biomass to a low level but is associated with high phytoplankton biomass. Anyhow, no consistent relationship between crustacean biomass and trophic indicators was found among the three lake groups.

In the present study, the lack of significant negative correlation between crustacean biomass and chlorophyll a suggests that there is little control of phytoplankton biomass by macrozooplankton, as found in subtropical Florida lakes demonstrated that phytoplankton biomass in south Florida lakes was controlled by bottom-up, rather than top-down forces. Another study in our lakes also provided evidence for the above conclusion that nutrients are more important than crustacean zooplankton in controlling the phytoplankton biomass in shallow subtropical lakes (unpublished data).

This research was supported by a National Natural Science Foundation of China (Grant No. 30225011). Great thanks are given to Baozhu Pan, Haijun Wang and Xiaomin Liang for their generous helps in the field sampling. The authors are also grateful to two anonymous reviewers for their valuable comments on this manuscript.