Quantifying turbulent energy fluxes and evapotranspiration in agricultural field conditions: A comparison of micrometeorological methods

Accurate estimation of energy fluxes and evapotranspiration (ET) in agricultural systems is critically needed, especially for water resource sustainability, soil moisture monitoring and irrigation. Numerous micrometeorological methods have become commercially available. Considering the eventual trade-off between cost and accuracy, it is important to evaluate these methods to provide recommendations for practical purposes. Therefore, we tested five different techniques at one field in the region of Central Europe dominated by rainfed farming but suffers from drought spells. In an intensive campaign, we used eddy covariance (EC), large aperture and surface layer scintillometers, the Bowen ratio energy balance (BREB), and the surface renewal (SR) methods to estimate the sensible (H) and latent (λET) heat fluxes of winter wheat and bare soil with harvest residues during two months in summer 2015. At the half-hourly level, the methods showed varying agreement under different field conditions. While H estimated by EC and scintillometry agreed well, there was an underestimation of λET by EC compared to the other methods, most likely due to energy balance non-closure. The λET estimated by the BREB method showed good agreement with the λET obtained by scintillometry when the Bowen ratio (β) was high and with the λET obtained by EC when β→ 0. Our study confirmed good reliability of scintillometers across wide range of meteorological conditions. Although the SR method provided the most inferior agreement with other methods at half-hourly basis, it was deemed to be valuable when longer averaging periods were used. Over the entire experiment, mean daily ET estimated by scintillometry (2.6 mm d−1), BREB (2.3 mm d−1), and SR (2.9 mm d−1) showed reasonable agreement while EC (1.6 mm d−1) significantly underestimated. This indicates that low cost methods (BREB and SR) are sufficient for water management purposes when a daily and longer time scales are important. Further, parallel deploying of BREB and SR can provide additional diagnostics and increase the confidence in ET estimates.