The hydrologic community uses geochemical tracers to determine the age distribution of water exiting a catchment, with transit time distributions (TTDs) important for understanding groundwater storage and mixing. New water-tagging capabilities within models track precipitation events as they move through simulated storages. Here, we present a ‘sequential precipitation input tagging’ (SPIT) framework to tag all input precipitation events at regular intervals over an extended period (monthly tags over seven years). SPIT is applied at six National Ecological Observatory Network sites to calculate TTDs and derive from these mean transit times (MTT), fractions of young water (Fyw), and hydrologic tracer concentrations (δQ-δ18O and δ2H) within a water-tagging enabled version of the Weather Research and Forecast hydrologic model. Throughout seven simulation years, the fraction of simulated discharge derived from tagged events increased each year, with the final year’s tagged stream water fraction (TSWF) ranging 21% to 100%. When the TSWF was ≥75%, simulated MTTs range 190 days to 850 days and Fyw 1% to 24%, with a root mean squared error (RMSE) of 456 days and 14.5%. The RMSE for δ18O is 1.08‰ and δ2H 6.58‰. Low TSWF values early in the simulation period highlights the need to apply SPIT over many years to fully understand the TTD. At daily timescales, model MTT and Fyw exhibit a power-law relationship with precipitation, discharge, and groundwater. The successful implementation of SPIT within a tracer-enabled version of an operational hydrologic model allows for a reproducible approach to calculate water transit times and hydrologic tracers.

The timescales associated with precipitation moving through watersheds reveal processes that are critical to understanding many hydrologic systems. Measurements of environmental stable water isotope ratios (δ 2H and δ 18O) have been used as tracers to study hydrologic timescales by examining how long it takes for incoming precipitation tracers become stream discharge, yet limited measurements both spatially and temporally have bounded macroscale evaluations so far. In this observation driven study across North American biomes within the National Ecological Observation Network (NEON), we examined δ 18O and δ 2H stable water isotope in precipitation (δP) and surface water (δQ) at 26 co-located sites. With an average 54 precipitation samples and 139 surface water samples per site, assessment of local meteoric water lines (LMWL) and local surface water line (LSWL) showed geographic variation across North America. Taking the ratio of estimated seasonal amplitudes of δP and δQ to calculate young water fractions ( Fyw), showed a Fyw range from 1% to 93% with most sites having Fyw below 20%. Calculated mean transit times (MTT) based on a gamma convolution model showed a range from 0.10 to 13.2 years, with half of the sites having MTT estimates lower than 2 years. Significant correlations (r) were found only between the Fyw and watershed area, longest flow length, and the longest flow length/slope, whereas the only significant correlation observed for MTT was with site latitude. The estimated Fyw and MTT provide information describing hydrologic processes at NEON sites, however limited correlations of Fyw and MTT with the environmental characteristics we analyzed demonstrate that these quantities are primarily driven by site or area specific factors. The analysis of isotope data presented here provides important constraints on isotope variation in North American biomes and the timescales of water movement through NEON study sites.