Modeling phytoplankton production: problems with the Eppley curve and an empirical alternative

Papers reporting the results of dynamic simulation models of aquatic ecosystems tend to show predicted concentrations of the state variables. The phytoplankton compartment is typically represented as predicted biomass, expressed as the concentration of chlorophyll a, particulate carbon, or particulate nitrogen. While computed values of phytoplankton biomass generally agree with observations, many of these same models significantly underestimate primary production. Existing simulation models often base the calculation of primary production on the Eppley curve, which sets the maximum daily phytoplankton growth rate as a function of temperature. Despite the apparent wide applicability of the Eppley curve, an increasing number of culture and field studies have measured growth rates in excess of those predicted by the curve, which may explain why existing models often underestimate primary production. An alternate empirical formulation which predicts daily phytoplankton production from biomass, photic depth, and incident irradiance has been shown to apply in a variety of nutrient-rich estuarine systems. Despite the large number of systems in which these empirical models have been developed, they predict remarkably similar rates of daily and annual production. Furthermore, these empirical models predict rates of production in excess of those predicted by the Eppley curve. The empirical formulation therefore presents an alternative to the Eppley curve in dynamic ecosystem models, and may result in more accurate predictions of primary production by these models.