Comparison of the energy fluxes inferred from eddy covariance and optical scintillometer at the agricultural field during different parts of the season

This study summarizes seasonal monitoring of turbulent energy fluxes from eddy covariance (EC) and large aperture optical scintillometer, measuring in parallel. The site (agricultural field of app. 16.5 ha) is located in north-eastern Austria, Danube river lowland (48.21N, 16.622E); in 2019 covered subsequently by a winter wheat field, straw, and bare soil. The EC together with ancillary measurement was located at the 2.7 m height at the center of the field. The radiation balance components measurements consist of a 4-channel net radiometer for net radiation (Rn) installed at 3.5 m and three soil heat flux plates for soil heat flux (G) monitoring (0.05 m below surface), including thermocouples for quantification of the heat storage above the soil heat flux plates. The scintillometer transmitter and receiver units were fixed at 4 m height masts, facing towards each other from the NW and SE corners of the field, with a measurement path length of 570 m diagonally across the field. The EC method enables the determination of fluxes within a footprint centered around the point of measurement in the middle of the field, whereas the scintillometer provides an estimation of sensible heat flux (H_SC), derived from air refractive index fluctuation integrated over the measurement path length. The scintillometer-based latent heat (LE_SC) is calculated as a residuum from available energy (Rn-G) and H_SC, provided by the scintillometer. As the EC method provides direct measurements of sensible heat (H_EC) and latent heat (LE_EC) fluxes we use it as a reference method. During the period March to June (green canopy) the comparison of the EC-based turbulent fluxes (H_EC+LE_EC) and the available energy (Rn-G) showed a very good agreement, resulting in the energy balance closure of 0.97 (R² = 0.94). This suggests the ability of the EC method to capture all scales of eddies responsible for energy transport at this site as well as the good accuracy and robustness of the measurement setup. During the period March to June (green canopy), the H_EC, LE_EC, H_SC, and G fluxes accounted for 23 % (R² = 0.55), 60 % (R² = 0.72), 32 % (R² = 0.56), and 12 % (R² = 0.62) of the Rn flux, respectively. The comparison of methods indicates that H_SC overestimated H_EC by 20 % (R² = 0.78). However, during the latter part of the season (straw and bare soil) we found that under highly unstable atmospheric stratification, H_SC is even more overestimating H_EC and sometimes is even exceeding the available energy. The main reason for such behavior can be the choice of the universal stability function in the computation of H_SC. Recommendations for universal functions that are correcting this artifact in H_SC will be discussed.

Acknowledgment: This study was supported by SustES - Adaptation strategies for sustainable ecosystem services and food security under adverse environmental conditions (CZ.02.1.01/0.0/0.0/16_019/0000797).