Evaporation from irrigated wheat estimated using radiative surface temperature: An operational approach

Abstract

Hourly values of latent energy flux ( LE) of wheat, calculated from the energy balance of the stand using infrared measurements of crop temperature to estimate sensible heat flux ( H), were compared with measurements made by a weighing lysimeter. Measurements spanned a period of 50 days in which the phenology of the crop changed from anthesis to physiological maturity, plant height changed from standing to lodged crop, soil water deficit increased from 0 to 150 mm over the last 20 days, and weather varied seasonally and diurnally. Under this wide range of conditions, the energy balance estimate accounted for 76% of the variance in LE, but its accuracy was insensitive to the correction of aerodynamic resistance ( r a) for atmospheric stability and wind speed effects. In comparison, the Penman and Priestley-Taylor models of evaporation accounted for only 58 and 41%, respectively, of the variance in LE. The superiority of the energy balance estimate was due to the significance of H in the energy balance under this wide range of conditions. The accuracy of the energy balance model was limited as infrared temperature appeared to differ from aerodynamic surface temperature at the virtual source/sink height. The difference was linearly related to H, such that under stable conditions the infrared temperature was above the aerodynamic temperature, while under unstable conditions it was below it. As a consequence, LE was overestimated under unstable and underestimated under stable conditions. Over a day, however, unstable conditions predominated before solar noon and stable conditions after solar noon. Therefore, when making daily estimates of LE, this source of error tended to be self-cancelling. Hence, the energy balance estimate accounted for 86% of the variance for LE averaged over the day, compared with 76% when individual hourly averages were used. The implication of these results for improving the operational use of evaporation estimates for the management of irrigated crops is discussed.