Abstract
Aim Recent research has shown that much of the variability in leaf gas exchange and leaf longevity can be related to variations in the surface : volume ratio of leaves. The aim of this paper was to develop a theoretical framework and a practical method to extend that result to the vegetation at the continental scale. Location The study was conducted in Australia. Methods We propose that vegetation is composed of a mixture of three basic leaf types, ‘turgor’ (T), ‘mesic’ (M) and ‘sclerophyll’ (S) leaves. Changes in the relative proportions of T, M and S leaves within a vegetation type are visualized using a ternary diagram and differences in vegetation structure are shown to be easily mapped onto the ternary diagram. We estimate the proportions of T, M and S leaves using readily available data. The total amount of PAR absorbed by the vegetation (fPAR) is estimated using continental‐scale satellite observations. The total fPAR is then decomposed into that absorbed by T, M and S leaves. The relative absorption of PAR by T leaves is estimated from the temporal dynamics in the satellite signal, while the relative proportions of M and S leaves are estimated using climatic (solar radiation, rainfall) data. Results When the availability of light, nutrients and water were near‐optimal, the vegetation was composed of predominantly M leaves. In low nutrient environments S leaves predominated. T leaves were dominant in disturbed environments. Conclusions The theoretical framework is used to predict that elevated atmospheric CO 2 would tend to increase the proportion of M and S leaves in an ecosystem and the resulting change means that the proportion of T leaves would decrease. In terms of the TMS scheme, this implies that elevated CO 2 has the same net effect on the vegetation as a decrease in disturbance.