Measurement differences between turbidity instruments, and their implications for suspended sediment concentration and load calculations: A sensor inter-comparison study

The use of turbidity for indicating environmentally detrimental levels of suspended and colloidal matter in freshwater systems, and for defining acceptable water quality standards in national and European drinking water regulations, is well established. Turbidity is therefore frequently adopted as a surrogate for suspended sediment concentrations (SSC), or as a relative and objective measure of water clarity in monitoring programmes. Through systematic, controlled experimentation, we tested the response of 12 commercially available turbidity sensors, of various designs, to gauge their measurement consistency when benchmarked against pre-prepared sediment suspensions of known SSC. Results showed that despite calibration to a Formazin standard, sensor responses to identical SSC solutions (in the range of 20 -1000 mg L-1) varied considerably. For a given SSC, up to five-fold differences in recorded turbidity were recorded across the tested instruments. Furthermore, inconsistent measurements were identified across instruments, regardless of whether they operated using backscatter or side-scatter optical principles. While the findings may have implications for compliance with turbidity-based water quality standards, they are less likely to be an issue when turbidity is being used as a surrogate for SSC, provided that instrument use remains constant and that instrument drift is not an issue. In this study, a field comparison of a subset of four study sensors showed that despite very different absolute turbidity readings for a given SSC, well correlated and reliable turbidity - SSC ratings were established (as evidenced by r(2) coefficients from 0.92 to 0.98). This led to reasonably consistent suspended sediment load estimates of between 64.7 and 70.8 tonnes for a rainfall event analysed. This study highlights the potential for issues to arise when interpreting water turbidity datasets that are often assumed to be comparable, in that measurement inconsistency of the type reported here may remain unknown to water resource decision makers and practitioners. (C) 2017 Elsevier Ltd. All rights reserved.