Assessing the accuracy of suspended sediment rating curves for a small basin

Abstract

Sediment rating curves are often used to estimate suspended sediment loads where the sampling program is insufficient to define the continuous record of sediment concentration. Use of this technique will involve errors in the values of sediment load produced, and comparison with measured daily values has been employed by several workers to assess the magnitude of these errors. Comparisons are more difficult for small‐ and medium‐sized catchments because of the general lack of direct measurements of loads. Recording turbidity meters could be employed to provide a continuous record of sediment concentration which can be used to calculate sediment loads. Results are presented for the river Creedy in which the values of sediment load calculated from the continuous concentration record have been compared with estimates derived from rating curves. The rating‐curve data have been grouped according to season and stage tendency to provide various rating relationships. Values of annual sediment load estimated by using a rating curve could involve errors of up to +280%, whereas the errors for monthly loads could range between +900% and −90%. Careful consideration should be given to possible error terms before rating‐curve estimates of sediment load are used in statistical and other analyses.