The impact of nutrient state and lake depth on top-down control in the pelagic zone of lakes: A study of 466 lakes from the temperate zone to the arctic

摘要

Using empirical data from 466 temperate to arctic lakes covering a total phosphorus (TP) gradient of 2-1036 mug L-1, we describe how the relative contributions of resource supply, and predator control change along a nutrient gradient. We argue that (a) predator control on large-bodied zooplankton is unimodally related to TP and is highest in the most nutrient-rich and nutrient-poor lakes and generally higher in shallow than deep lakes, (b) the cascading effect of changes in predator control on phytoplankton decreases with increasing TP, and (c) these general patterns occur with significant variations-that is, the predation pressure can be low or high at all nutrient levels. A quantile regression revealed that the median share of the predator-sensitive Daphnia to the total cladoceran biomass was significantly related unimodally to TP, while the 10% and 90% percentiles approached 0 and 100%, respectively, at all TP levels. Moreover, deep lakes (more than 6 m) had a higher percentage of Daphnia than shallow (less than 6 m) lakes. The median percentage of Daphnia peaked at 0.15 mg L-1 in shallow lakes and 0.09 mg L-1 in deep lakes. The assumption that fish are responsible for the unimodality was supported by data on the abundance of potential planktivorous fish (catch net(-1) night(-1) gill nets with the different mesh sizes [CPUE]). To elucidate the potential cascading effect on phytoplankton, we examined the zooplankton phytoplankton biomass ratio. Even though this ratio was inversely related to CPUE at all TP levels, we found an overall higher ratio in oligotrophic lakes that declined toward low values (typically below 0.2) in hypertrophic lakes. These results suggest that planktivorous fish have a more limited effect on the grazing control of phytoplankton in oligotrophic lakes than in eutrophic lakes, despite similar predator control of large-bodied zooplankton. Accordingly, the phytoplankton yield, expressed as the chlorophyll a-TP ratio, did not relate to CPUE at low TP, but it increased significantly with CPUE at high TP. We conclude that the chances of implementing a successful restoration program using biomanipulation as a tool to reduce phytoplankton biomass increase progressively with increasing TP, but that success in the long term is most likely achieved at intermediate TP concentrations.