Adjustments in the effective leaf mixing length of the water pathway in pinnate leaves with partially severed veins

Palmate leaf venation is characterized by a redundant vascular system compared to pinnate leaf venation. This vascular redundancy prevents hydraulic disruption of palmate leaves during vascular damage, keeping the leaf photosynthetically active. According to the current leaf water oxygen isotope model, transpiration rate and effective leaf mixing length of the water pathway have an effect in the leaf water oxygen isotope enrichment above source water. In pinnate leaves, which show little vascular redundancy, decrease in leaf hydraulic conductivity due to vascular breakage would lower transpiration rate, which according to the Peclét effect would result in O18 enrichment compared to palmate leaves. On the other hand, pinnate leaves with vascular damage could also have a longer leaf water pathway because only small veins would be available for water movement. If this is the case, in pinnate damaged leaves the isotopic effect of low transpiration would be counteracted by the effect of a long leaf water pathway and the expected differences in leaf water oxygen isotope enrichment between pinnate and palmate treated leaves would be minimized. Here we hypothesize that after a vascular breakage, leaf water oxygen isotope enrichment above source water will be similar between pinnate and palmate leaves despite differences in transpiration and hydraulic rupture. In pinnate leaves, the effective mixing length of the water pathway will increase, therefore the effect of low transpiration rates in the leaf water oxygen isotope enrichment would be minimized. To test this hypothesis we selected 6 trees that differed in their leaf venation type (3 palmate species and 3 pinnate species). For each species we used 5 different branches, each having 1 control and 1 treated leaf. The treatment encompassed severing of the middle vein 1 cm after the petiole-lamina junction; so that in palmate leaves only the central main vein was severed but not the lateral main veins. A nested anova was performed to interpret the results (species were nested within each venation type). We found no differences in the transpiration rates between palmate control and treated leaves. Pinnate treated leaves, however, had a significantly lower transpiration rate compared to the controls. As expected palmate and pinnate treated leaves did not differ in leaf water oxygen isotope enrichment above source water, nor did they differ relative to the control leaves. The effective leaf mixing length of the water pathway (L) from the xylem to the stomatal pore was calculated to be longer in pinnate severed leaves than palmate severed leaves and palmate and pinnate control leaves. Therefore, we conclude that in pinnate severed leaves, a longer L minimized the back diffusion of enriched water from the evaporative pool at the internal leaf surface, leading to no differences in bulk leaf water oxygen isotope enrichment between pinnate and palmate venation leaves after vascular damage. This conclusion is important considering that the leaf water oxygen isotope signature is incorporated in the cellulose oxygen isotope signature and the leaf signal should reflect the environment conditions independent of leaf damage in order to reconstruct past climate accurately.