Flow Partitioning Modelling Using High-Resolution Isotopic and Electrical Conductivity Data

Water-stable isotopic (WSI) data are widely used in hydrological modelling investigations. However, the long-term monitoring of these tracers at high-temporal resolution (sub-hourly) remains challenging due to technical and financial limitations. Thus, alternative tracers that allow continuous high-frequency monitoring for identifying fast-occurring hydrological processes via numerical simulations are needed. We used a flexible numerical flow-partitioning model (TraSPAN) that simulates tracer mass balance and water flux response to investigate the relative contributions of event (new) and pre-event (old) water fractions to total runoff. We tested four TraSPAN structures that represent different hydrological functioning to simulate storm flow partitioning for an event in a headwater forested temperate catchment in Western, Oregon, USA using four-hour WSI and 0.25-h electrical conductivity (EC) data. Our results showed strong fits of the water flux and tracer signals and a remarkable level of agreement of flow partitioning proportions and overall process-based hydrological understanding when the model was calibrated using either tracer. In both cases, the best model of the rainstorm event indicated that the proportion of effective precipitation routed as event water varies over time and that water is stored and routed through two reservoir pairs for event and pre-event. Our results provide great promise for the use of sub-hourly monitored EC as an alternative tracer to WSI in hydrological modelling applications that require long-term high-resolution data to investigate non-stationarities in hydrological systems.

Keywords: high-resolution tracer data; rainfall-runoff modelling; hydrograph separation; electrical conductivity/specific conductance; tracer hydrology; water stable isotopes