Zooplankton and the total phosphorus – chlorophyll <i>a</i> relationship: hierarchical Bayesian analysis of measurement error

Zooplankton grazing is important in resolving residual variation around the total phosphorus – chlorophyll a relationship. In empirical studies, zooplankton body size is often a better predictor of residual variation than zooplankton biomass. We investigate whether higher measurement error associated with zooplankton biomass may explain its lower predictive ability. We collected five replicate zooplankton biomass samples in 19 lakes, allowing us to quantify measurement error in volumetric zooplankton biomass with greater precision than in previous studies. A hierarchical Bayesian model was used to assess the predictive ability of volumetric zooplankton biomass and mean individual zooplankton length, corrected for measurement error. We found consistent effects of total zooplankton biomass, but not zooplankton length, on chlorophyll a. This finding does not appear to be related to the higher precision with which total zooplankton biomass was measured in our study, but rather to ecological factors. Interlake variation outweighed the effects of measurement error in estimating the strength of relationships between zooplankton variables and chlorophyll a. Our findings therefore suggest that studies to estimate zooplankton effects on phytoplankton should allocate resources to study a larger range of lakes over different time periods than to process replicate samples to reduce measurement error.