ECOHYDROLOGIC SEPARATION ALTERS INTERPRETED HYDROLOGIC STORES AND FLUXES IN A HEADWATER MOUNTAIN CATCHMENT

Recent studies have demonstrated that in some catchments, compartmentalized pools of water supply either plant transpiration (poorly mobile water) or streamflow and groundwater (highly mobile water), a phenomenon referred to as ecohydrologic separation. The omission of ecohydrologic separation in hydrological models is expected to influence estimates of water residence times and plant water availability. However, no study has tested this expectation or investigated how ecohydrologic separation alters interpretations of stores and fluxes of water within a catchment. In this study, we compare two hydrochemical lumped rainfall-runoff models, one which incorporates ecohydrologic separation and one which does not, for a second-order watershed at the H.J. Andrews Experimental Forest (Oregon, USA), the site where ecohydrologic separation was first observed. The models are calibrated against stream discharge and chloride concentration. Results demonstrate that model structural variations result in small differences between calibrated parameters. However, differences in mobile water storage volumes and fluxes influence interpreted residence times and plant water availability, suggesting the importance of ecohydrologic separation when considering catchment-scale water and solute transport.