Using a 3D hydrodynamic-biogeochemical model to compare estuarine nitrogen assimilation efficiency under anoxic and oxic conditions

Abstract

Estuarine ecosystems, an important interface between the land and the sea are valued for their high productivity as well as their potential to act as a sink for catchment-sourced nitrogen. The efficiency of nitrogen removal and storage in estuaries is dependent on a range of hydrodynamic and biogeochemical processes. In particular, the position of the freshwater/saltwater interface (commonly referred to as the 'salt wedge') and potential concurrent zone of oxygen deficiency in the benthic layer. The salt wedge in the Yarra River Estuary in Melbourne, Australia is both tidally and seasonally driven leading to transient patterns of hypoxia and anoxia causing dynamic shifts in nutrient recycling pathways. Understanding these dynamic nutrient cycles is considered of paramount importance as estuaries can act as major sinks for catchment nutrients that would otherwise end up in the ocean leading to nuisance algal blooms. Despite major implications for estuarine health, these extremely variant nitrogen cycling pathways in salt-wedge estuaries have rarely been quantified during anoxia and hypoxia. In this study we apply a coupled 3D curvilinear hydrodynamic-biogeochemical model to the Yarra River Estuary. Measured discharge, tidal and meteorological data were used to simulate the shifting zone of oxygen deficiency within the estuary over a period of 12 months from 1 July 2009 to 1 July 2010. Simulated concentrations of organic nitrogen, ammonium and nitrate and process rates of mineralisation, nitrification, denitrification and sediment fluxes were analysed in relation to upstream nutrient loads to select two distinct dates where estuarine nitrogen assimilation efficiency under anoxic and oxic conditions could be compared. Simulated data from the anoxic period (10 January 2010) showed strong anoxia in the bottom waters. Warm air temperatures and low riverine discharge characterised forcing data. By contrast the forcing data for the oxic period (21 September 2011) was during an extended period of high flow so that simulated oxygen concentrations were close to saturation. Model results indicated a shift in denitrification from the sediments to the water column under anoxic conditions and a concomitant net flux of ammonium from the sediments to the water column. During the oxic period sediment denitrification dominated and ammonium flux out of the sediments was minimal. As a result net nitrogen assimilation capacity of the estuary stayed positive throughout the model domain. In contrast the model predicted negative net assimilation of nitrogen under anoxic conditions. Although the total denitrification efficiency for the estuary of both periods was positive, (8.9% and 7.3% for low and high flow periods respectively) the spatial distribution patterns were distinctly different. The model has been used to highlight potential changes to the nitrogen dynamics in the estuary under altered catchment management practices. Rates of nitrification and denitrification in the sediments and shifts into the water column as well as fluxes of nitrogen across the sediment/water interface under anoxic/oxic conditions were analysed with respect to the nitrogen assimilation efficiency of the estuary. The results of these simulations demonstrate major implications for estuarine management in response to rapidly changing anthropogenic pressures.