Field verification of the surface renewal method for evapotranspiration measurement in arid agricultural area: The uncertainty of calibration parameter

Accurate measurement of evapotranspiration (ET) in farmland is crucial for irrigation management. The surface renewal (SR) method offers a low-cost and efficient alternative to the established ET measurement techniques such as eddy covariance (EC) method and Bowen-ratio energy balance (BREB) method. However, the SR method still needs extensive tests in arid agricultural areas, while the uncertainty of the calibration parameter (α) in the theory remains a major challenge. In this perspective, two years of field observations were conducted at maize fields with EC flux towers in the Hetao Irrigation District, a typical arid agricultural area in Northwest China. Two EC flux towers were set up to monitor the half-hourly sensible heat flux and latent heat flux for comparison. Each year, we installed high-frequency temperature sensors (thermistors made of Pt) at 4.5 m on the towers to capture the raw data needed for the SR method. The half-hourly sensible heat flux was calculated firstly with the SR method, and the latent heat flux was then calculated as residual of energy closure equation. According to the results, the comparisons were satisfactory, with R² > 0.70, RMSE < 35 W m⁻², MAE ≈ 20 W m⁻² for half-hourly sensible heat flux, and R² > 0.9 for half-hourly latent heat flux. The comparisons of ET results at different cases indicated the reliability of the SR method, with R² > 0.9, and RMSE and MAE are approximately 0.5 mm d^-1 and 0.6 mm d^-1. The default value (α = 1) was found to be feasible. We found the atmospheric stability could affect the accuracy of estimated sensible heat flux, and latent heat flux showed higher accuracy under higher air temperature conditions. Furthermore, it was observed that constant α is positively correlated with the time interval (r) of high-frequency temperature traces, while variable α under Castellví method’s framework would eventually converge to a certain range at unstable atmospheric conditions. The findings prove the applicability of the SR method for ET measurement in arid agricultural area and provide a solid experimental basis for its further advancement.