Observed relationships between leaf H218O Péclet effective length and leaf hydraulic conductance reflect assumptions in Craig-Gordon model calculations

Stable oxygen isotope techniques may be a useful tool to investigate the pathways of water movement within leaves. However, implementation of such methods is limited due to uncertainty in the effective path length (L) for the Péclet effect in leaf water enrichment models. Previous studies have found relationships between L and physiological parameters such as transpiration rate (E) and leaf hydraulic conductance (k(leaf)) both within and between species. However, these studies relied on assumptions in their calculation of L, which were not directly tested. Eucalyptus paniculata Smith was used to evaluate the relationships between L, k(leaf) and E under differing water availability and a range of leaf temperatures. Coupled gas exchange and transpiration isotope measurements allowed previous assumptions to be directly tested. L was significantly and negatively related to both k(leaf) and E when the isotopic signature of water vapour was assumed to be in equilibrium with source water, was equivalent to the room vapour or equal to source water. However, the relationship between L and k(leaf) was non-significant when measured δ( 18)O of transpired vapour was used and disappeared entirely when non-steady-state leaves were excluded, and when evaporation site water was calculated from coupled gas exchange and transpiration isotope values. These results suggest that great care must be taken when calculating L, particularly regarding assumptions of isotopic steady state and δ( 18)O of vapour. Previous suggestions of changes in pathways for water movement as transpiration rate varied need to be reassessed in light of these observations.