Parameterization of a two-layer model for estimating vineyard evapotranspiration using meteorological measurements

The Shuttleworth and Wallace (SW) model with variable canopy resistance was evaluated to estimate evapotranspiration (ETv) from a drip-irrigated Merlot(Vitis vinifera L) vineyard trained on a vertical shoot-positioned (VSP) system. This vineyard is located in the Talca Valley, Region del Maule, Chile (35 degrees 25' LS; 71 degrees 32' LW; 125 m a.s.l). The performance of the SW model was evaluated using the eddy-covariance method on a 30 min time interval. Also, sub-models to estimate net radiation (Rn) and soil heat flux (G) were used in the SW model. A good agreement between observed and estimated values of Rn was found with a root mean square error (RMSE) of 33 W m(-2) and a mean absolute error (MAE) of 24 W m(-2). Also, the SW model was able to estimate latent heat flux with RMSE and MAE of 34 and 21 W m(-2), respectively. On a daily basis, results indicate that the SW model was able to predict the ETv with RMSE and MAE values of 0.51 and 0.41 mm d(-1), respectively. These results suggest that it is possible to directly estimate ETv over unstressed grapevines using meteorological data and soil moisture measurements. (C) 2009 Elsevier B.V. All rights reserved.