Quantifying dissolved organic carbon concentrations in upland catchments using phenolic proxy measurements

Concentrations of dissolved organic carbon (DOC) in soil and stream waters in upland catchments are widely monitored, in part due to the potential of DOC to form harmful by-products when chlorinated during treatment of water for public supply. DOC can be measured directly, though this is expensive and time-consuming. Light absorbance in the UV–vis spectrum is often used as a surrogate measurement from which a colour-carbon relationship between absorbance and DOC can be derived, but this relationship can be confounded by numerous variables. Through the analysis of data from eight sites in England and Wales we investigate the possibility of using the concentration of phenolic compounds in water samples as a proxy for DOC concentration. A general model using data from all the sites allowed DOC to be calculated from phenolics at an accuracy of 81–86%. A detailed analysis at one site revealed that a site-specific calibration was more accurate than the general model, and that this compared favourably with a colour-carbon calibration. We therefore recommend this method for use where estimates of DOC concentration are needed, but where time and money are limiting factors, or as an additional method to calculate DOC alongside colour-carbon calibrations. Tests demonstrated only small amounts of phenolic degradation over time; a loss of 0.92 mg L−1 after 8 months in storage, and so this method can be used on older samples with limited loss of accuracy.