A new method for fingerprinting sediment source contributions using distances from discriminant function analysis

Abstract

Mixing models are frequently used as part of sediment fingerprinting to quantify sediment source contributions. Much research effort has been devoted to improving these mixing models. The objective of this study was to develop and evaluate a new method using discriminant function analysis (DFA) to fingerprint sediment source contributions. It was hypothesized that the outcome of DFA, commonly used as a component of standard fingerprinting procedures, can potentially be used directly to quantify source contributions, avoiding mixing models altogether. This hypothesis was tested in the Bull Creek Watershed in Oklahoma State, USA. DFA results were compared with outcomes from the Collins mixing model and previous research. When conservative geochemical tracers were used, DFA results did not differ significantly from the mixing model results, indicating that DFA alone has the potential to accurately quantify sediment source contributions, while being simple to use. When using non-conservative tracers, however, the results from the two methods were significantly different. On the basis of a comparison with previous research, we suggest that DFA offers an intuitive method for characterizing sediment source contributions.