Snowpack disrupts relationship between young water fraction and isotope amplitude ratio; approximately one fifth of mountain streamflow less than one year old

Abstract

AbstractPrevious “fraction of young water” (Fyw) estimates based on relative annual isotopic amplitudes in precipitation (Ap) and streamflow (As) produced low Fyw values in mountain catchments, which is contrary to extensive research that reports rapid water transmission in mountains. This study investigated this discrepancy by testing the effect of snow accumulation on the model that underpins the Fyw method. A Monte‐Carlo analysis of simulations for 20,000 randomly‐generated catchment model configurations used 10 years of precipitation inputs for the Upper Elbow River catchment in the Rocky Mountains (Alberta, Canada) to model discharge with and without snowpack storage of winter precipitation. Neither direct nor modified precipitation input produced a 1:1 relationship between As/Ap and Fyw, undermining the applicability of the original Fyw method in mountain watersheds with large seasonal snow accumulation. With snowpack‐modified input a given As/Ap ratio corresponds to a range of Fyw values, which can still provide semi‐quantitative information. In the small (435 km2) Elbow River catchment a Fyw range of 7–23% supports previous findings of rapid transmission in mountain catchments. Further analysis showed that the improved discharge prediction (Nash–Sutcliffe efficiency > 0.9) correlates with higher Fyw values and demonstrated that the interannual shifts in δ18O can be used to estimate of new water (<1 year) fraction in winter streamflow, and the estimate of 20% for the Elbow River further supports rapid transmission in mountain catchments.