Assimilating catchment processes with monitoring data to estimate sediment loads to the Great Barrier Reef

Quantifying riverine sediment loads, and the uncertainty around these estimates, is important for monitoring the impact of land use on ecologically sensitive receiving waters such as the Great Barrier Reef lagoon. We used a Bayesian Hierarchical Modelling approach that assimilates information from a process model for runoff, a mechanistically motivated statistical model for sediment generation and observed runoff and sediment load data. The approach was trialled on a 10-year dataset collected from a 14-km 2 sub-catchment in the Burdekin basin, Australia. Using our model, we were able to estimate daily sediment concentrations, discharges and loads (with credible intervals) over a 10-year period, including 3 years where there were long periods of missing observational data. We found that for the high-frequency monitoring undertaken at the study site, credible intervals around sediment loads were narrow. Credible intervals were substantially wider in years where observational data were not available and load estimates relied on the underlying processes and neighbouring observations. The method presented here is the first attempt at assimilating discharge and concentration measurements with process models for the purpose of sediment load estimation. The potential for quantifying loads entering the Great Barrier Reef lagoon is promising, particularly for ephemeral streams that are typical of arid and semi-arid Australia