Linking high‐frequency DOC dynamics to the age of connected water sources

We combined high‐frequency dissolved organic matter fluorescence (FDOM) data with stable isotope observations to identify the sources and ages of runoff that cause temporal variability in dissolved organic carbon (DOC) within a peat‐dominated Scottish catchment. FDOM was strongly correlated (r2 ∼ 0.8) with DOC, allowing inference of a 15 min time series. We captured 34 events over a range of hydrological conditions. Along with marked seasonality, different event responses were observed during summer depending on dry or wet antecedent conditions. The majority of events exhibited anticlockwise hysteresis as a result of the expansion of the riparian saturation zone, mobilizing previously unconnected DOC sources. Water ages from the main runoff sources were extracted from a tracer‐aided hydrological model. Particularly useful were ages of overland flow, which were negatively correlated with DOC concentration. Overland flow age, which ranged between 0.2 and 360 days, reflected antecedent conditions, with younger water generally mobilizing the highest DOC concentrations in summer events. During small events with dry antecedent conditions, DOC response was proportionally higher due to the displacement and mixing of small volumes of previously unconnected highly concentrated riparian soil waters by new precipitation. During large events with wet antecedent conditions, the riparian saturation zone expands to organic layers on the hillslopes causing peaks in DOC. However, these peaks were limited by dilution and supply. This study highlights the utility of linking high‐frequency DOC measurements with other tracers, allowing the effects of hydrologic connectivity and antecedent conditions on delivery of DOC to streams to be assessed. Key Points Stream DOC dynamics are strongly linked to variations in the age of source waters Nonstationary flow‐DOC relationships reflect changing hydrologic connectivity FDOM is a reliable proxy for long‐term, high‐resolution DOC time series