A Diel Method of Estimating Gross Primary Production: 1. Validation With a Realistic Numerical Model of Chesapeake Bay

Abstract

A method for estimating gross primary production (GPP) is presented and validated against a numerical model of Chesapeake Bay that includes realistic physical and biological forcing. The method statistically fits a photosynthesis-irradiance response curve using the observed near-surface time rate of change of dissolved oxygen and the incoming solar radiation, yielding estimates of the light-saturated photosynthetic rate and the initial slope of the photosynthesis-irradiance response curve. This allows estimation of GPP with 15-day temporal resolution. The method is applied to the output from a numerical model that has high skill at reproducing both surface and near-bottom dissolved oxygen variations observed in Chesapeake Bay in 2013. The rate of GPP predicted by the numerical model is known, as are the contributions from physical processes, allowing the proposed diel method to be rigorously assessed. At locations throughout the main stem of the Bay, the method accurately extracts the underlying rate of GPP, including pronounced seasonal variability and spatial variability. Errors associated with the method are primarily the result of contributions by the divergence in turbulent oxygen flux, which changes sign over the surface mixed layer. As a result, there is an optimal vertical location with minimal bias where application of the method is most accurate.