Assessing the validity and sensitivity of microbial processes within a hydrodynamic model.

Eutrophication due to excess anthropogenic nutrients in waterways is a significant issue worldwide. The pressure-stressor-response of a waterway to excessive nutrient loading is reliant on numerous physical and biological factors, including hydrodynamics and microbial processing. While substantial progress has been made towards simulating these mechanisms there are limited multi-disciplinary studies that relate the physical hydrodynamics of a site with the ecological response from linked laboratory and field studies. This paper presents the development of a coupled hydrodynamic and aquatic ecosystem response model, expanded to include an integrated microbial loop, that allows the explicit representation of heterotrophic bacteria growth and dissolved organic nutrient mineralisation. A unique long-term water quality dataset at an estuary in south-eastern Australia was used to validate and assess the model's sensitivity to complex biophysical processes driving the observed water quality variability. Results indicate that explicit time-varying bacterial mineralisation rates provide a substantially improved understanding of the broader aquatic ecosystem response than assigned fixed bulk rate parameter values, which are typically derived from non-local literature. Implementation of a microbial loop at the study site indicated that the model is sensitive to the boundary conditions, in particular catchment loads, with both net transport rates and the net growth rates of heterotrophic bacteria demonstrating different responses. Under average flow conditions, a smaller net transport and reduced nutrient availability has a pronounced effect of lowering net growth rates through the applied limitation factors. During high flow conditions, freshwater inflows increased net transport and nutrient loads, which resulted in higher net growth rates. Further, temporal variability in water temperature had a compounding effect on the model's response sensitivity. This approach has broader application in other riverine systems subject to eutrophication, and in interrogating linkages in hydrodynamic and microbial mediated processes (e.g., productivity). Future studies are recommended to better understand the sensitivity of aquatic ecosystem response models to microbial net growth rate kinetics at different temperatures and from top-down predation (e.g., zooplankton grazing).