Field scale recalibration of the sap flow thermal dissipation method in a Mediterranean vineyard

The measurement of transpiration at plant level is a challenging topic in crop water requirements research, e.g. for irrigation scheduling and for the partitioning of plant transpiration and soil evaporation in modelling crop evapotranspiration process. The thermal dissipation method (TD) is widely applied for transpiration estimates of tree species, however some studies reported underestimations of transpiration measurements performed with the TD method when using the equation proposed by Granier. In particular, large underestimation of transpiration flux was reported in vineyards. The underestimation is ascribed to the equation, and its parameters, used to calculate the sapwood flux density from the heat transfer equations in the original Granier's model and site-specific calibration is required. In this study, using measurements performed in a 2-year field experiment, the TD method was recalibrated at field scale in a vineyard cultivated in a Mediterranean region (Sicily, southern Italy), comparing transpiration flux determined by the TD method with transpiration calculated by the partitioning flux-variance similarity (FVS) method. FVS is a relatively new and innovative method for partitioning water fluxes in transpiration and evaporation through EC data. In this study, transpiration calculated through the equation proposed by Granier was, on average, 26% lower than transpiration determined by FVS. The results show that a new exponential function to calculate the sapwood flux density performed better than the power function in the original equation proposed by Granier. The results also showed that species and site-specific relationships are needed to accurately determine transpiration through the TD method. The ratio T/ET when T is calculated by the recalibrated TD function was equal to 0.77.