Does leaf water efficiency vary among eucalypts in water-limited environments?

There is a need to generalize water use behavior of eucalypts to facilitate bioengineering and landscape remediation programs in a wide range of Australian environments. A critical question can be stated as a null hypothesis: tree water use per unit leaf area (leaf efficiency) is independent of eucalypt species. This is implicitly equivalent to the hydrological equilibrium hypothesis that states that leaf area is a function of climate, at least in cases where transpiration and growth are limited by soil water. Failure to reject this null hypothesis simplifies (a) the selection of tree species for water balance management, (b) the generation of regional-scale expectations of leaf area index, and (c) the estimation (monitoring) of the effectiveness of plantations in controlling site water balance. The hypothesis was tested with tree water use data collected in natural multi-species stands across Australia, including sites in the wet-dry season tropical woodlands of the Northern Territory, the Mediterranean climate forests of Western Australia, and a woodland system in southern New South Wales receiving an even distribution of rainfall throughout the year. We also tested the hypothesis in a multi-species tree plantation growing on a saline gradient. In each case, we could not reject the hypothesis of constant leaf efficiency among eucalypts. In every case there was a common, strong, linear relationship among tree leaf area and mean daily water use by all tree species in a sample. Single factor (species) analysis of variance did not detect significant differences between leaf water efficiencies of species. For the jarrah forest (Eucalyptus marginata J. Donn ex Sm., E. calophylla R. Br. ex Lindl.), the null hypothesis held in both spring (wet) and autumn (dry) conditions. The null hypothesis held in the mixed species woodland of New South Wales (E. macrorhynca F.J. Muell. ex Benth., E. blakelyi Maiden., E. polyanthemos Schauer.) under summer and autumn conditions, and across five species in the wet-dry tropical woodland (E. miniata A. Cunn. ex Schauer, F. tetrodonta F.J. Muell., E. porecta S.T. Blake, Erythrophleum chlorostachys F.J. Muell., and Treminalia ferdinandiana Exell.). The null hypothesis also held for a plantation of E. occidentalis Endl. and provenances of E. camaldulensis Dehnh. growing on a shallow saline gradient; i.e., leaf water efficiency remained constant across species and varieties despite obvious effects of salinity on the size of individual canopies. We conclude that there is little evidence for rejecting the hypothesis that leaf efficiency does not vary significantly among sympatric eucalypt species in rainfall-limited (soil-water-limited) systems. These findings open the way for useful bioengineering generalities about the hydrological role of trees in the Australian landscape.