An integrated model to predict and prevent hypoxia in floodplain-river systems

Available online 5 March 2021

Hypoxia can occur following rewetting of floodplains and cause severe impacts on aquatic biota and biogeochemical processes. The likelihood of such events is influenced by a number of factors including temperature, the mass of plant litter on the floodplain (which is influenced by the duration between inundation events), the volume of water available for dilution of oxygen-demanding dissolved organic matter, and the exchange of water to dilute and disperse that material. Using constructed infrastructure to generate managed inundations on floodplains increases the likelihood of hypoxic “blackwater” events relative to unregulated floods, as larger areas of floodplain are inundated at lower flow rates. A model (the “DODOC plugin”) was developed for the Source hydrological modelling software to inform risk mitigation strategies for these managed inundation events. This development enables the interaction between complex hydrology and floodplain inundation on the resulting release of dissolved organic carbon (DOC), and subsequent consumption of dissolved oxygen (DO), to be represented. Key functionality of the plugin includes the ability to represent (i) spatial variability in organic litter build up and degradation, (ii) DOC leaching from litter when inundated, (iii) DO consumption arising from microbial decomposition of the DOC, and (iv) reaeration processes from autotrophic productivity and turbulence as water passes over water level regulating structures. The model is configurable on both river channels (links) and floodplains (storages) to represent changes in DO from both natural and managed inundation events at the scale of an individual floodplain up to multiple floodplains and river reaches. The plugin was parameterised to successfully simulate DOC (R² = 0.84–0.93) and DO (R² = 0.74–0.92) along an approx. 100 km study reach of the River Murray in South Australia, once the different behaviour of the labile and refractory components of the DOC was represented in the model. A number of hypothetical operational scenarios were tested using the model to demonstrate parameter sensitivity and to inform planning of managed inundations. The development of the DODOC plugin demonstrates that complex water quality processes can be integrated into the Source (or other) hydrological software, to represent cumulative implications of floodplain inundation events and to minimise the risk of hypoxia.

Luke M. Mosley, Todd Wallace, Joel Rahman, Tom Roberts, Matt Gibbs