Effect of height and time lag on the estimation of sensible heat flux over a drip-irrigated vineyard using the surface renewal (SR) method across distinct phenological stages

• The SR method based on classical structure function analysis was tested using a three-dimensional sonic anemometer across distinct phenological stages.
• Twelve combinations of four r and three z were evaluated to estimate H using the SR method in a sparse vineyard indicated that α mainly depends on z.
• The comparison between HEC and HSR indicated that α changed during the different phenological periods. These variations could be associated to changes in canopy structure, growth parameters and water management.
• The best estimation of H, was obtained with the calibrated SR method when z1 (0.5 m over the canopy) and r3 (0.7 s) were used in the calculation.

For drip-irrigated vineyards, sensible heat flux (H) is a key parameter to estimate water requirements, when actual evapotranspiration (ETa) is computed as a residual from the surface energy balance. In this regard, a field experiment was carried out to study the effect of measurement height (z) and time lag (r) on the estimation of H over a drip-irrigated vineyard using classical formulation of surface renewal (SR) method. For vineyards, previous studies have indicated that the calibration factor (α) and the accuracy of the SR method depend on z, however the combined effect of z and r on the estimation of H has not been studied in detail for key phenological stages. In this study 12 combinations of 4 time lags (r1 = 0.2 s, r2 = 0.5 s, r3 = 0.7 s and r4 = 1.0 s) and 3 measurement heights (z1 = 0.5, z2 = 1.0 and z3 = 1.5 m above canopy) of high-frequency air temperature were evaluated to estimate H using the SR method (HSR) across distinct phenological stages. A three-dimensional sonic anemometer (CSAT3) was used to measure sensible heat (HEC) over the vineyard. Results indicated that the regression analysis between HSR and HEC was highly significant with determination coefficients (r2) between 0.70 and 0.93. Also, α values registered in this study varied from 0.67 to 1.01 for the different combinations and phenological periods. Calibrated HSR computed using z1 and r3 gave the best estimates of HEC in the validation period, with a root mean square error (RMSE) of 52.2 W m−2 and mean absolute error (MAE) of 35.2 W m−2 for all dataset analysis.