Determination of effective sapwood areas of Common oaks (Quercus robur) and analysis of uncertainties for estimation of the water balance component evapotranspiration in lowland floodplain forests

Water balance observations in forest stands are challenging. Sap flow measurements in trees are promising for a direct measurement of transpiration in individual trees. The estimation of stand level transpiration from individual sap flow measurements requires scaling to the individual scale, mainly by estimation of the associated sap wood area, and then to the stand level. All involved steps are associated with uncertainties. Thus, this study was set up to answer four main questions:

• 1. Which uncertainties are involved in determining the stem-circumference-sapwood-area-relationship?
• 2. Do single-point measurements provide reliable sap flux density data or is the radial variability of the sapwood and thus the water transport too great?
• 3. Is it feasible to determine water balance components from sapflow measurements in quercus robur under the conditions of uncertainties and sources of error?
• 4. How do sap flow estimates of evapotranspiration compare to alternative methods, such as lysimeters, soil water profile observations or passive capillary wick samplers?

The study was carried out at a long term soil monitoring site in the floodplain area of the Parthe river in the lowlands of Leipzig, Germany, with Quercus robur as the site dominant tree species.

A site-specific relationship between circumference and sap flow area for Quercus robur was established based on the colour change method (methyl-orange) and drill cores from 20 trees of varying circumference. The results show that the main uncertainties of estimating sapwood area come from deviation of the sapwood area from an optimal circular ring (± 31,2%) and the variability of the sapwood depths (± 9,2%). Furthermore, analysis of sap flow velocities at various depths in the trunk, shows that there is a radial heterogeneity of the axial water transport and thus a single-point measurement can lead to both a possible over- and underestimation of the sap flow under certain circumstances. When scaling the transpiration from tree to stand level, a comparative water balance equation was set up with the aid of infiltration meters in order to investigate the significance of the estimated transpiration.

The investigated low land stand of Quercus robur shows a distinct, but uncertain, relationship between circumference and sapwood area, which is unique compared to relationships of other oak species. In conclusion the results of the study show that the transpiration using sap flow und sap wood measurements cannot be obtained with high sufficient precision due to uncertainties: (1) in the relationship between circumference and sapwood area, (2) in sap flux densities within individual trees and (3) sap flow measurements themselves. As a direct result it is shown, that scaling transpiration from individual trees to the stand level needs to consider the associated uncertainties and leads to comparable results with other estimates.