Temporal and spatial variations in the oxygen-18 content of leaf water in different plant species

Temporal variations in the delta 18 oxygen (delta 18O) content of water transpired by leaves during a simulated diurnal cycle fluctuated around the delta 18O content of the source water. Reconstructed variations in the delta 18O values of leaf water differed markedly from those predicted by conventional models. Even when transpiring leaves were maintained under constant conditions for at least 3 h, strict isotopic steady-state conditions of leaf water (equality of the 18O/16O ratios in the input and transpired water) were rarely attained in a variety of plant species (Citrus reticulata, Citrus paradisi, Gossypium hirsutum, Helianthus annuus, Musa musaceae and Nicotiana tabacum). Isotopic analysis of water transpired by leaves indicated that leaves approach the isotopic steady state in two stages. The first stage takes 10 to 35 min (with a rate of change of about 3.3 ppt h-1), while in the second stage further approach to the isotopic steady state is asymptotic (with a rate of change of about 0.4 ppt h-1), and under conditions of low transpiration leaves can last for many hours. Substantial spatial isotopic heterogeneity was maintained even when leaves were at or near isotopic steady state. An underlying pattern in this isotopic heterogeneity is often discerned with increasing 18O/16O ratios from base to tip, and from the centre to the edges of the leaves. It is also shown that tissue water along these spatial isotopic gradients, as well as the average leaf water, can have 18O/16O ratios both lower and higher than those predicted by the conventional Craig and Gordon model. We concluded, first, that at any given time during the diurnal cycle of relative humidity the attainment of an isotopic steady state in leaf water cannot be assumed a priori and, secondly, that the isotopic enrichment pattern of leaf water reflects gradual enrichment along the water-flow pathway (e.g. as in a string of pools), rather than a single-step enrichment from source water, as is normally assumed.