An assessment of the suspended sediment rating curve approach for load estimation on the Rivers Bandon and Owenabue, Ireland
Civil and Environmental Engineering | Engineering
This paper presents an assessment of the suspended sediment rating curve approach for load estimation on the Rivers Bandon and Owenabue in Ireland. The rivers, located in the South of Ireland, are underlain by sandstone, limestones and mudstones, and the catchments are primarily agricultural. A comprehensive database of suspended sediment data is not available for rivers in Ireland. For such situations, it is common to estimate suspended sediment concentrations from the flow rate using the suspended sediment rating curve approach. These rating curves are most commonly constructed by applying linear regression to the logarithms of flow and suspended sediment concentration or by applying a power curve to normal data. Both methods are assessed in this paper for the Rivers Bandon and Owenabue.
Turbidity-based suspended sediment loads are presented for each river based on continuous (15 min) flow data and the use of turbidity as a surrogate for suspended sediment concentration is investigated. A database of paired flow rate and suspended sediment concentration values, collected between the years 2004 and 2011, is used to generate rating curves for each river. From these, suspended sediment load estimates using the rating curve approach are estimated and compared to the turbidity based loads for each river. Loads are also estimated using stage and seasonally separated rating curves and daily flow data, for comparison purposes.
The most accurate load estimate on the River Bandon is found using a stage separated power curve, while the most accurate load estimate on the River Owenabue is found using a general power curve. Maximum full monthly errors of − 76% to + 63% are found on the River Bandon with errors of − 65% to + 359% found on the River Owenabue. The average monthly error on the River Bandon is − 12% with an average error of + 87% on the River Owenabue. The use of daily flow data in the load estimation process does not result in a significant loss of accuracy on either river. Historic load estimates (with a 95% confidence interval) were hindcast from the flow record and average annual loads of 7253 ± 673 tonnes on the River Bandon and 1935 ± 325 tonnes on the River Owenabue were estimated to be passing the gauging stations.
► Rating curve based loads are compared to continuous turbidity based loads. ► Rating curves estimate the annual load accurately (mean error of < 12%). ► Monthly uncertainties are greater (> 30%) especially for months with small loads. ► Rating curve separation did not increase the accuracy of the SS load significantly. ► Mean historic SS yields are 17.1–18.8 t km− 2 year− 1.
Seán T. Harrington, Joseph R. Harrington, An assessment of the suspended sediment rating curve approach for load estimation on the Rivers Bandon and Owenabue, Ireland, Geomorphology, Volume 185, 2013, Pages 27-38, doi.org/10.1016/j.geomorph.2012.12.002.