Sediment and nutrient budgets are inherently dynamic: evidence from a long-term study of two subtropical reservoirs

• Published in: Hydrology and earth system sciences Vol. 20; no. 12; pp. 4881 - 4894
• Main Authors: O'Brien, Katherine R, Weber, Tony R, Leigh, Catherine, Burford, Michele A
• Format: Journal Article
• Language: English
• Published: Katlenburg-Lindau Copernicus GmbH 13.12.2016; European Geosciences Union; Copernicus Publications
• Subjects: Budgets; Catchment area; Catchment models; Catchments; Climate change; Creeks & streams; Drought; Environmental impact; Environmental Sciences; Flood damage; Flood predictions; Floods; Fluxes; Hydrologic data; Loads (forces); Mineral nutrients; Nitrogen; Nutrient concentrations; Nutrient flow; Nutrient loading; Nutrients; Phosphorus; Reservoirs; Rivers; Sediment; Sediments; Solid suspensions; Suspended particulate matter; Total suspended solids; Water shortages; Queensland Australia; Australia; flooding; NUTRIMENT; sediment; INONDATION; nutrients
• ISSN: 1607-7938; 1027-5606; 1607-7938
• DOI: 10.5194/hess-20-4881-2016

Accurate reservoir budgets are important for understanding regional fluxes of sediment and nutrients. Here we present a comprehensive budget of sediment (based on total suspended solids, TSS), total nitrogen (TN) and total phosphorus (TP) for two subtropical reservoirs on rivers with highly intermittent flow regimes. The budget is completed from July 1997 to June 2011 on the Somerset and Wivenhoe reservoirs in southeast Queensland, Australia, using a combination of monitoring data and catchment model predictions. A major flood in January 2011 accounted for more than half of the water entering and leaving both reservoirs in that year, and approximately 30 % of water delivered to and released from Wivenhoe over the 14-year study period. The flood accounted for an even larger proportion of total TSS and nutrient loads: in Wivenhoe more than one-third of TSS inputs and two-thirds of TSS outputs between 1997 and 2011 occurred during January 2011. During non-flood years, mean historical concentrations provided reasonable estimates of TSS and nutrient loads leaving the reservoirs. Calculating loads from historical mean TSS and TP concentrations during January 2011, however, would have substantially underestimated outputs over the entire study period, by up to a factor of 10. The results have important implications for sediment and nutrient budgets in catchments with highly episodic flow. First, quantifying inputs and outputs during major floods is essential for producing reliable long-term budgets. Second, sediment and nutrient budgets are dynamic, not static. Characterizing uncertainty and variability is therefore just as important for meaningful reservoir budgets as accurate quantification of loads.