Investigating watershed-scale hydrologic connections and sources with dynamic-flux particle tracking and isotope measurements

Managing water resources in a changing climate requires a detailed understanding of subsurface flow paths from upland recharge to valley streamflow or groundwater storage. This may be nowhere truer than California, where Central Valley agriculture and a large coastal population depend on mountain precipitation that will shift from snow to rain. To better understand water's flow paths in California's Cosumnes watershed, we combine dynamic-flux particle tracking using EcoSLIM and radioactive and stable isotopic tracers. In EcoSLIM, particles trace out flow paths based on velocities from the integrated hydrologic model ParFlow—in this case there are >700 million particles in the 7000 km² watershed. From EcoSLIM we derive the source elevation, age, and precipitation phase of streamflow and evapotranspiration across the watershed. We compare EcoSLIM results with synoptic surveys of water isotope measurements collected from 16 nested sub-watersheds via a citizen-science collaboration.

The EcoSLIM results demonstrate that snow and rain contribute evenly to watershed streamflow, but that evapotranspiration comes primarily from rain (Figure 1). In the model, most snowmelt pulses quickly through the subsurface, exiting days after melting. As a result, snow from certain northern subwatersheds provides an outsized fraction of discharge. We illustrate how the simulation selects water from subsurface storage for discharge and evapotranspiration at multiple spatial and temporal scales, elucidating watershed processes not described by traditional watershed models.