Shallow fluvial lakes are heterogeneous ecosystems in which marked spatio-temporal variation renders difficult the analysis of key ecological processes, such as growth. In this study, we used generalized additive modelling of the RNA/DNA ratio, an index of short-term growth, to investigate the influence of environmental variables and spatio-temporal variation on growth of yellow perch (Perca flavescens) in Lake St. Pierre, Quebec, Canada. Temperature and water level had seemingly stronger effects on short-term growth than seasonal change or spatial variation between and along the lakeshores. Consistent with previous studies, the maximum RNA/DNA ratio was found at 20.5 °C, suggesting that our approach provides a useful tool for estimating thermal optima for growth in the field. The RNA/DNA ratio showed a positive relationship with water level, as predicted by the flood pulse concept, a finding with implications for ecosystem productivity in fluvial lakes. The RNA/DNA ratio was more variable along the north than the south shore, possibly reflecting exposure to more differentiated water masses. The negative influence of both high temperatures and low water levels on growth points to potential impacts of climatic change on fish production in shallow fluvial lakes.
Short-term forecast of Yangtze River water level based on Long Short-Term Memory neural network
