Exploring source water mixing and transient residence time distributions of outflow and evapotranspiration with an integrated hydrologic model and Lagrangian particle tracking approach

Abstract

Understanding the time water takes as it moves from rain or snowmelt through the terrestrial system to arrive as stream discharge, or evapotranspiration (ET) is an important hydrologic quantity. We develop a Lagrangian particle tracking method to capture transient residence times from source to either ET or outflow in an integrated hydrologic model. This method is parallel and efficiently captures time evolution of parcels of water in the model and tracks the source of water for hydrograph or ET separation. We demonstrate this model using hypothetical hillslope simulations driven by snow or rain dominated forcing and two different land cover types. We show that land cover and forcing both impact the outflow residence time distribution, which spans many years. We also introduce the idea of ET residence time distributions and show that while mean ET residence times are typically less than 1 year, land cover affects this quantity and simulated ET processes draw from much older water (many years old) depending on location on the hillslope or seasonal cycle. Finally, we study source water contribution to outflow and ET and explore assumptions about a residence time-based definition of older, pre-event, or groundwater end member. We show that simulated plant processes may switch to more opportunistic and younger sources of water, changing the composition of outflow.