Seasonal, Interannual, and Longitudinal Patterns in Estuarine Metabolism Derived from Diel Oxygen Data Using Multiple Computational Approaches

Abstract

Twenty-three station-years of diel oxygen data for the James River Estuary were analyzed to characterize longitudinal, seasonal, and interannual patterns of gross primary production (GPP) and ecosystem respiration (ER). We compared two commonly used methods for deriving metabolism (bookkeeping and Bayesian) to determine whether the observed patterns were robust with respect to computational methodology. The two methods revealed similar longitudinal patterns of increasing GPP and ER, and decreasing net ecosystem metabolism (NEM), with increasing salinity. Seasonal patterns in GPP and ER tracked water temperature and solar radiation, except during high discharge events when metabolism declined by 40%. The bookkeeping method yielded higher estimates of GPP and ER in the higher end of the range, and smaller estimates in the low end of the range, thereby accentuating seasonal and longitudinal differences. Inferences regarding net autotrophy and heterotrophy were robust, as both methods yielded positive estimates of NEM at the chlorophyll maximum (tidal fresh segment) and negative values for the saline portion of the estuary. Inferences regarding the relative importance of allochthonous inputs (based on inferred ER at GPP = 0) differed between the two methods. Values derived by the bookkeeping method indicated that respiration was largely supported by autochthonous production, whereas the Bayesian results indicated that autochthonous and allochthonous inputs were equally important. Overall, our findings show that methodological differences were small in the context of longitudinal, seasonal, and interannual variation but that the bookkeeping method yielded a wider range of values for GPP and ER relative to the Bayesian estimates.