Comparing ∆T(max) Determination Approaches for Granier-Based Sapflow Estimations

Granier-type thermal dissipation probes are common instruments for quantifying tree water use in forest hydrological studies. Estimating sapflow using Granier-type sapflow sensors requires determining the maximum temperature gradient (∆T(max)) between the heated probe and the reference probe below. ∆T(max) represents a state of zero sap flux, which was originally assumed to occur each night leading to a ∆T(max) determination on a daily basis. However, researchers have proven that, under certain conditions, sapflow may continue throughout the night. Therefore alternative approaches to determining ∆T(max) have been developed. Multiple ∆T(max) approaches are now in use; however, sapflow estimates remain imprecise because the empirical equation that transfers the raw temperature signal (∆T) to sap flux density (F(d)) is strongly sensitive to ∆T(max). In this study, we analyze the effects of different ∆T(max) determination approaches on sub-daily, daily and (intra-)seasonal F(d) estimations. On this basis, we quantify the uncertainty of sapflow calculations, which is related to the raw signal processing. We show that the ∆T(max) determination procedure has a major influence on absolute ∆T(max) values and the respective sap flux density computations. Consequently, the choice of the ∆T(max) determination approach may be a significant source of uncertainty in sapflow estimations.