Estimating evapotranspiration from processing tomato using the surface renewal technique

Abstract

The use of the surface renewal (SR) technique for evapotranspiration (ET) estimates of processing tomato crop was examined. The objective was to study the effects of sampling rate, in the range 1–10 Hz, and measurement height, on the SR method performance. Measurements were carried out in a processing tomato crop at the Hula Valley of Northern Israel. Eddy covariance and surface renewal systems were deployed simultaneously for a period of 11 days on August 2010. Data from the first 6 measurement days was used for calibration and data of the rest 5 days for verification of ET data. Three parameters were chosen for optimising the performance of the SR technique: the sampling rate, the SR analysis time lag and sensor height. Results showed that during calibration the best combination of coefficient of determination ( R 2 = 0.86) and root mean square error (RMSE = 29.57 W m −2 ) between EC and SR half-hourly sensible heat fluxes was obtained for temperature fluctuations measured at the level of the canopy top, using a 0.25 s analysis time lag and 10 Hz sampling rate. For these operational conditions, and using the corresponding calibration coefficient α = 2.47, mean ratio between daily estimated (by SR) and deduced (from EC) ET during the verification period was 1.03 ± 0.02. Data analyses at lower sampling rates of 1, 2 and 5 Hz and for the same measurement height also provided good agreement between estimated and measured daily ET.