A trait-based ecosystem model suggests that long-term responsiveness to rising atmospheric CO₂ concentration is greater in slow-growing than fast-growing plants

Atmospheric carbon dioxide concentration (C_a) has a direct and measurable effect on plant growth. However, it does not affect all plant species equally, which could lead to shifts in competitive dominance of species in ecosystems. We used a dynamic plant carbon-nitrogen model to systematically examine how species traits affect the long-term C_a responsiveness of C₃ plants when growing as established monocultures in the field. The model was tested against responses of 7 C₃ herbaceous species growing in a free-air C_a enrichment (FACE) experiment (BioCON) in Minnesota, USA. Model simulations showed that several species traits affected the C_a response strongly, giving rise to a number of testable hypotheses about interspecific differences in responsiveness to C_a. The largest responses to rising C_a were obtained for species with low carbon-use efficiency (net primary production: gross primary production ratio), low foliar carbon allocation, low stomatal conductance, low instantaneous photosynthetic nitrogen use efficiency and low specific leaf area. In general, our model predicted that, for established plants growing in resource-limited field conditions, species with slow growth rates would be most responsive to elevated C_a. This prediction was supported by data from the BioCON experiment. Our model also predicts that, for young plants growing in non-resource-limited conditions, species with high growth rates will be most responsive to elevated C_a. This difference in species ranking under different resource availabilities is largely explained by the indirect effects of C_a on leaf area. Leaf-area feedbacks favour fast-growing species the most during leaf-area expansion, but following stand maturation they favour slow-growing species the most. These results imply that species that respond strongly to elevated C_a in short-term (non-resource-limited) glasshouse experiments are unlikely to also be the most responsive in resource-limited field conditions, and therefore that we cannot directly extrapolate from glasshouse experiments to predict which species will be most responsive to elevated C_a in the long term.