Modeling the Response of Daily Evapotranspiration and its Components of a Larch Plantation to the Variation of Weather, Soil Moisture, and Canopy Leaf Area Index

Abstract An accurate prediction of forest evapotranspiration (ET) based on its components in response to a changing environment is essential for understanding interactions between the atmosphere, soil, and vegetation and for integrated forest‐water management. The ET components of a pure larch plantation and the environment were monitored over 2 years in northwest China. The response functions of each ET component to individual driving factors were determined using upper boundary lines, then coupled to form the ET component modules, fitted with measured data in 2016 (May–September), and validated with measured data in 2015 (June–September). Results showed that (1) the response of daily transpiration (T) to potential ET (ET ref ) followed a binomial equation, and the response of T to relative extractable water (REW) of the 0‐ to 60‐cm soil layer and canopy leaf area index (LAI) followed a saturated exponential growth function. The module was T = (−5.766 × 10 −4 ET ref 2 + 0.005ET ref –0.002) × (18.769 + 46.990 (1–exp(−8.555REW))) × (−14.662 + 17.428 (1–exp(−1.414LAI))). (2) The response of daily forest floor evapotranspiration (FE) to ET ref , volumetric soil moisture (VSM) of the 0‐ to 30‐cm soil layer, and LAI followed positive linear, saturated exponential growth and saturated exponential decay function. The module was FE = (6.697ET ref –2.770) × (6.927–11.243exp(−1.959VSM)) × (0.032 + 1.162exp(−2.407LAI)). (3) The canopy interception (I c ) of individual rainfall events was mainly affected by precipitation amount (P) and LAI. The module was I c = 0.446 LAI (1–exp(−0.112P)) + 0.097P. (4) The daily ET model was built up as ET = T + FE + I c , which had good performance in both the calibration and validation period. It can be concluded that the variation of daily forest ET can be accurately predicted using the easily measurable driving factors of weather, soil, and vegetation.