Estimating actual evapotranspiration from stony-soils in montane ecosystems

Quantification of evapotranspiration (ET) is crucial for understanding the water balance and for efficient water resources planning. Agricultural settings have received most attention regarding ET measurements while less knowledge is available for actual ET (ETA) in natural ecosystems, many of which have soils containing significant amounts of stones. This study is focused on modelling ETA from stony soil, particularly in montane ecosystems where we estimate the contribution of stone content on water retention properties in soil. We employed a numerical model (HYDRUS-1D) to simulate ETA in natural settings in northern Utah and southern Idaho during the 2015 and 2016 growing seasons based on meteorological and soil moisture measurements at a range of depths. We simulated ETA under three different scenarios, considering soil with (i) no stones, (ii) highly porous stones, and (iii) negligibly porous stones. The simulation results showed significant overestimation of ETA when neglecting stones in comparison to ETA measured by eddy covariance. ETA estimates with negligibly porous stones were lower for all cases due to the decrease in soil water storage compared with estimates made considering highly porous stones. Assumptions of highly porous or negligibly porous stones led to reductions in simulated ETA of between 10% and 30%, respectively, compared with no stones. These results reveal the important role played by soil stones, which can impact the water balance by altering available soil moisture and thus ETA in montane ecosystems.