Using field deployable sensors to identify the inter-event predictability of Dissolved Organic Matter mobilisation in an urban river

Emerging sensor technology offers new opportunities to monitor different fractions of Dissolved Organic Matter (DOM) in high resolution. Concentration-discharge (C-Q) relationships (e.g. hysteresis or c-q slopes), derived from high frequency observation can offer insight into source mobilization and reactive transport processes of DOM. However, few studies have explored patterns in urban catchments, where understanding of storm event DOM responses under different hydrometeorological conditions remains elusive. To bridge this gap, we collected 2-years (15 min resolution) fluorescence data (humic-like fluorescence [HLF: Ex. 325 nm/ Em 470 nm] and tryptophan-like fluorescence [TLF: Ex 275 nm/ Em 350 nm]) in an urban headwater stream (Birmingham, UK). We used c-q slopes and two indices, the hysteresis index (HI) and flushing index (FI), to explore the inter-event variability in DOM dynamics. In addition, we assessed the hydrometeorological factors (e.g., antecedent conditions, temperature, discharge and rainfall characteristics) that govern DOM mobilisation and transport using statistical multiple linear regression. Our findings reveal pronounced seasonal variation in the behaviour of TLF and HLF. In warmer periods, the chemodynamic characteristics of both fluorescence peaks become evident. We observed a consistent counter-clockwise hysteresis pattern accompanied by flushing behaviour. The magnitude of discharge, antecedent temperature, and rainfall intensity were identified as key drivers of HLF and TLF flushing and hysteresis dynamics. Conversely, during colder months, a shift in DOM mobilisation was observed. For TLF, source limitation was apparent, characterized by clockwise hysteresis and a notable dilution. In contrast, HLF exhibited a more variability during this period, with complex hysteresis patterns and a combination of solute flushing and dilution. The magnitude of discharge and antecedent wetness were identified as the key factors influencing the solute behaviour in this cooler period. Our research indicates that the responses of DOM in urban rivers exhibit distinct responses to hydrometeorological conditions which were relatively predictable (i.e. low stochasticity within particular event types). However, variability in DOM composition and magnitude was pronounced between event types which has implications for managing urban rivers, specifically ensuring ecological health and resilience are maintained in the face of increasing climatic extremes.