Evapotranspiration dynamics and partitioning from concurrent above and below canopy flux measurements in a Montane Sierra Nevada Forest

Evapotranspiration (ET) from the land surface to the atmosphere consists of transpiration (T) from plants and evaporation (E) from soil and vegetated surfaces. These biological and physical component fluxes respond differently to changes in temperature, water availability and atmospheric composition. ET can be measured directly at the ecosystem scale with the eddy covariance (EC) method but similar measurements are not currently available for the component fluxes E and T. Concurrent EC measurements above and below forest canopies provide a promising approach to partition ET into T and E. However, our understanding of the performance of such measurements is still very limited. To address these challenges, we measured and partitioned ET with three concurrent EC towers (1 above & 2 below canopy) in a montane forest at Sagehen Creek in the Sierra Nevada, California from late June 2017 to September 2020. We observed a total forest ET of 606 mm yr^-1 with 275 mm yr^-1 measured in the understory and a tree transpiration of 331 mm yr^-1. Below-canopy measurements replicated at two locations within the above-canopy footprint indicated only small spatial variability for understory ET near the creek at Sagehen. Interannual variability in ET above and below canopy was small during the water years 2018 to 2020, despite large variability in precipitation totals. Accordingly, vegetation water use was relatively stable across years and the P–ET water balance was mainly driven by variations in water supply. Partitioning the components of total forest ET at Sagehen with concurrent EC measurements showed that on average 67–74% originated from T (47% from trees and 20–27% from understory grasses), while 26–33% were from E (mostly from the understory). Our results demonstrate the strength of concurrent above- and below-canopy EC measurements for the partitioning of ET.