High‐frequency storm event isotope sampling reveals time‐variant transit time distributions and influence of diurnal cycles

High‐frequency sampling of stable water isotopes in precipitation and stream water during winter and summer storm events was carried out in a 2·3 km 2 lowland agricultural catchment. During peak flows of monitored events, the responses of δ 2 H and δ 18 O were comparable and inferred the dominance ( ca 70%) of ‘old’ pre‐event water. Transit Time Distribution (TTD) inferred by a gamma function were fitted (Nash–Sutcliffe = 0·8) and were also similar for δ 2 H and δ 18 O. However, the shape (α) and scaling (β) parameters were markedly different for summer and winter events. Consequently, when antecedent wetness was high, mean transit times were in the order of days; when drier, they increased to months. Moreover, while the responses of δ 2 H and δ 18 O exhibited similar gradual recovery to pre‐event conditions during winter hydrograph recessions, they differed dramatically on summer recessions. Time series analysis showed that δ 2 H isotope content was correlated with the diurnal cycle of air temperature, suggesting an evaporative fractionation pattern which could be reproduced by a temperature‐based first‐order autoregressive model. The heavier δ 18 O isotope showed no evidence for such diurnal variability. The study highlights the utility of high‐frequency stable isotope sampling to explore the time‐variant nature of TTDs. Furthermore, it shows that the time of sampling in a diurnal cycle may have crucial significance for interpreting stream isotope signatures, particularly δ 2 H.