Improving Estimation of Evapotranspiration During Soil Freeze-Thaw Cycles by Incorporating a Freezing Stress Index and a Coupled Heat and Water Transfer Model into the FAO Penman-Monteith Model

Qiang Xu,Xiaofei Yan,David A. Grantz,Xuzhang Xue,Yurui Sun,Peter Schulze Lammers,Zhongyi Wang,Qiang Cheng
DOI: https://doi.org/10.1016/j.agrformet.2019.107847
IF: 6.2
2020-01-01
Agricultural and Forest Meteorology
Abstract:Evapotranspiration (ET) plays an important role in water and energy balance at the surface-atmosphere interface. It is widely reported that near-surface soil water content (SWC) or soil water potential (SWP) significantly affects ET and this parameter has been incorporated into the FAO Penman-Monteith (FAO-PM) model for prediction of ET during crop growth seasons. However, there is little information on the effect of SWC or SWP on prediction of ET during soil freeze-thaw cycles in winter. We present an experiment conducted at a demonstration farm with a crop of winter wheat, over two winters near Beijing, China. A lysimeter system equipped with a weather station was used to measure the ET flux and meteorological data. Unfrozen soil water content (USWC) and soil temperature (T-soil) were measured using dielectric tube sensors (DTS) and digital temperature sensors, respectively. SWP was determined by measured USWC and a soil moisture characteristic (SMC) curve derived from the soil freezing characteristic (SFC) curve and the Clapeyron equation in frozen soil. Detailed measurements in year 1 showed that the FAO-PM model exhibited a complex error pattern, underestimating ET from unfrozen soil but overestimating ET from stable frozen soil. To address these errors, we define a freezing stress index (K-sf) as a function of SWP. Incorporation of K-sf as a modifier of the standard crop coefficient in the FAO-PM model improved prediction of ET (RMSE declined from 0.424 to 0.187 mm day(-1), in year 1). These data revealed a correlation between SWP near the soil surface (<10 cm) and measured ET over freezing and thawing cycles. We incorporated a coupled heat and water transfer (CHWT) model into the improved FAO-PM model to predict USWC, SWP and ET throughout the winter of year 2 from current meteorological data as upper boundary conditions, initial measurements of ET, USWC and Tsoil as initial conditions, and Ksf and soil hydraulic properties optimized in year 1. The results showed that ET estimation was significantly improved, with RMSE reduced from 0.323 to 0.281 mm day(-1) using the combined (FAOPM-K-sf-CHWT) model. Model error primarily derived from an underestimation of simulated SWP during soil freezing process. The combined model extends the utility of the FAO-PM model to non-cropping seasons including winter in temperate climates and reduces the data requirements for accurate prediction of ET from intermittently frozen soil.
What problem does this paper attempt to address?