Accumulating evidence suggests that ecological communities undergoing change in response to either anthropogenic or natural disturbance regimes exhibit macroecological patterns that differ from those observed in similar types of communities in relatively undisturbed sites. In contrast to such cross-site comparisons, however, there are few empirical studies of shifts over time in the shapes of macroecological patterns. Here we provide a dramatic example of a plant community in which the species-area relationship and the species-abundance distribution change markedly over a period of six years. These patterns increasingly deviate from the predictions of the Maximum Entropy Theory of Ecology (METE), which successfully predicts macroecological patterns in relatively static systems. Information on the dynamic state of an ecosystem inferred from snapshot measurements of macroecological community structure can assist in extending the domain of current theories and models to disturbed ecosystems.

Accumulating evidence suggests that ecological communities undergoing change in response to either anthropogenic or natural disturbances exhibit macroecological patterns that differ from those observed in similar types of communities in relatively undisturbed sites. In contrast to such cross-site comparisons, however, there are few empirical studies of shifts over time in the shapes of macroecological patterns. Here, we provide a dramatic example of a plant community in which the species-area relationship and the species-abundance distribution change markedly over a period of six years. These patterns increasingly deviate from the predictions of the maximum entropy theory of ecology (METE), which successfully predicts macroecological patterns in relatively static systems. The error in the species-area relationship prediction additionally correlates over time with increased stress measured as mortality minus recruitment, providing a link between demography and the failure of macroecological theory. Information on the dynamic state of an ecosystem inferred from snapshot measurements of macroecological community structure can potentially assist in identifying causes and consequences of disturbance and extending the domain of current theories and models to disturbed ecosystems.

No abstract