Modelling the surface conductance of a broad-leaf canopy: effects of partial decoupling from the atmosphere

The transpiration of a mature beech (Fagus sylvatica L,) forest was measured over a whole season by the heat pulse velocity technique and the results analysed in terms of a new analytical canopy conductance model, which takes into account the effects of partial decoupling from the atmosphere on the local humidity environment experienced by the canopy, Stand daily transpiration ranged from 0.62 to 2.97 mm d(-1), with a seasonal mean value of 1.97 mm d(-1). Maximum canopy conductance was 18.5 mm s(-1), with a mean estimated value of 5.0 mm s(-1); computed values were little affected by the assumption of neutral atmospheric conditions. The decoupling coefficient Omega varied greatly on a daily and seasonal basis, with an average value of 0.28. As a result of partial decoupling, the estimated vapour pressure deficit at the notional canopy surface exceeded the values measured above the canopy by 380 Pa on average. When correctly expressed in terms of humidity at the canopy surface, the model explained 80% of the variance in half-hourly transpiration measurements. Upon cross-validation it still explained 72% of the variance, as compared to only 40% when correction for partial decoupling was not introduced. A baseline canopy conductance of 0.7 mm s(-1), not modulated by the environment, was estimated. The implications of the model are discussed for the representation of canopy conductance and transpiration of broad-leaf forests.