Estimating half-saturation constants of microzooplankton grazing in the sea

To obtain the grazing half-saturation constant (K) of natural microzooplankton assemblages, we used three

nonlinear grazing models (rectilinear, Holling type II, and Holling type III) to fit the detailed data (phytoplankton

net specific growth rate vs. dilution factor) of individual dilution experiments that show significant concave

curves. In the dataset consisting of 528 experiments, 96 experiments show significant concave curves, and the

associated chlorophyll a (Chl a) concentrations are significantly higher than those of experiments showing linear

or convex curves. Experiments showing concave curves likely reflect that these microzooplankton assemblages

were under top-down control. The three models perform equally well in fitting the data. The K values estimated

from these 96 experiments vary over three orders of magnitude and are log–log linearly related with ambient Chl a

concentrations, but not correlated with temperatures. Estimates of K of natural microzooplankton assemblages

tend to be smaller than estimates from laboratory cultures. For the experiments not showing concave curves, it is

hard to obtain a robust estimate of K.