An Empirical Study of the Wound Effects on Sap Flow Measured with Thermal Dissipation Probes

Abstract:

Sap flow sensors are common to assess the contribution of tree transpiration to ecosystem evapotranspiration (ET). Thermal dissipation (TD) is one of the most popular methods for sap measurements but the insertion of TD probes in the tree stems imply wounding of the wood tissue and a consequent alteration of the sap flow. But the determination of sap flux density (SFD) is based on an empirical function developed for freshly drilled holes and it does hence not account for the wound effect.

 Here we investigate the effect of wound healing on sap flow measurements with TD probes. Our objectives were (1) the establishment of correction factors to account for the wound effect and (2) the determination of the point in time after installation when the correction factors become applicable. For that we performed an experiment in which TD probes were installed successively in diffuse- and ring-porous trees (Fagus sylvatica and Quercus petraea, resp.) during the growing season. The trees were logged in fall and additional sensors were installed afterwards in the logged stems. SFDs measured by the different TD sensors were compared with gravimetric estimates in the laboratory.

 Gravimetric flow compared well with SFD estimates from freshly installed sensors without wound formation, with only a slight underestimation by the TDs. In contrast, older sensors, submitted to wound reactions, underestimated SFD by up to 40%. However, sensors with 5, 11 and 22 week old wounds showed no significant differences, which implies that wound healing occurs in the first weeks after scission. Similar sap flow underestimations due to wound effects were observed in both species, oak and beech.

 This study highlights the relevance of accounting for tree wound reactions for accurate estimation of tree transpiration based on thermal dissipation sensors. We provide a correction factor for the classical Granier TD sensors that can be used from the first weeks after installation in similar species. This enables more accurate estimation of tree transpiration at the stand level and eliminates confounding effects with tree responses due to changing environmental conditions over the season.