Modelling canopy conductance and stand transpiration of an oak forest from sap flow measurements

In this study, transpiration was estimated from half-hourly sap flow measurements in a 35-year-old sessile oak stand (Quercus petraea) from 1990 until 1993 under various soil water conditions. The canopy conductance, calculated from the Penman-Monteith equation, was first analysed in relation to climatic variables: global radiation (R(g)) and vapour pressure deficit (VPD). The maximum canopy conductance (g(cmax)) was modelled with a nonlinear multiple regression over a period of non-limiting soil water content, and of maximal leaf area index (LAI) with a r(2) similar to 0.80. Limitations of gc due to soil water deficit (relative extractable water [REW]) and canopy development (LAI) were then taken into account in the model by using multiplicative limiting functions of REW and LAI. A general canopy conductance model was then proposed. Finally, this relationship was re-introduced in the Penman-Monteith equation to predict dry canopy transpiration. Simulated transpiration was in good agreement with sap flow measurements during the year following the calibration (r(2) = 0.92 in the control plot, 0.86 in the dry plot). The omega decoupling coefficient was close to 0.1 on a seasonal basis, indicating that transpiration was highly dependent on VPD.