Meta-analysis of lotic nutrient amendment experiments: detecting and quantifying subtle responses

Abstract

Ecological theory predicts that multiple nutrients, added independently, could each stimulate algal community biomass. In contrast to this theoretical prediction, individual nutrient amendment experiments frequently detect significant increases in algal community biomass in response to addition of a single nutrient, whereas the independent addition of other nutrients often does not result in a statistically significant increase in biomass. However, because of the low replication and high variability typical of ecological field experiments, only large responses can be detected as statistically significant in any single experiment. It is therefore possible that smaller, yet still real and biologically important responses occur, but remain undetected because of the low statistical power of individual experiments. Meta-analysis of lotic nutrient (N and P) amendment experiments indicated that simultaneous stimulation of benthic algal community biomass by >1 nutrient was the rule, not the exception. Addition of a limiting nutrient typically doubled algal biomass, whereas addition of another nutrient generally increased algal biomass ∼1.25-fold. N was approximately equally likely as P to be limiting. The power of the typical experiment for detecting biomass stimulation resulting from limiting and “nonlimiting” nutrients was >85% and <50%, respectively. This study presents yet another line of evidence that multispecies communities are unlikely to be limited by a single nutrient, and illustrates the utility of meta-analysis for conducting statistically powerful syntheses of ecological experiments.