Predicting biomass responses to global warming is crucial to clarify the debate on the positive feedback between climate warming and terrestrial carbon cycle. Unfortunately, current state-of-the-art models suffer great uncertainty in simulating and predicting the biomass responses to global warming. To advance our mechanistic understandings on the responses of carbon processes to climate change and evaluate the model performance of ecosystem and land surface models, it is necessary to incorporate the information provided by the global change experiments at different scales from leaf to individual to ecosystem levels. However, as the spatial scales are usually mismatched between ecosystem models and manipulative experiments, a sophisticated investigation on the scaling rules of ecophysiological responses to global warming among different levels could be helpful to evaluate and improve model performance. Therefore, in this study, applying a series of statistic techniques on a global dataset of plant biomass responses to warming based on 481 published papers, we identified critical drivers of biomass responses and investigated the potential scaling relationships at different levels from leaf to ecosystem.
Results/Conclusions
Our results showed that from leaf to individual levels, metabolic theory - derived allometric effects on leaf morphology and biomass allocation were dampened by warming, leading to an emergent scaling relationship between leaf area and plant biomass in responses to warming. Meanwhile, plant evolutionary history (i.e., interspecific relatedness and intraspecific variation) explained more variance in warming effects on biomass than ecological and experimental factors did, suggesting the important role of phylogenetic niche conservatism (PNC). The PNC was mediated by leaf economic spectrum traits at the family but not the species level. When the species-level responses were scaled up to ecosystem-level, the responses of ecosystem biomass to warming depended on biodiversity. Specifically, experimental warming increased ecosystem biomass by 28.7% in low-diversity ecosystems, but had no significant effect in high-diversity systems. This buffering effect of biodiversity might be due to the insurance effects because species in high-diversity systems tended to respond to warming less synchronously. Overall, our study questioned the applicability of metabolic theory under global warming, revealed the importance of evolutionary history in explaining biomass responses that had been overlooked, and highlighted the regulation of biodiversity on ecosystem responses to climate change. All these aspects should be incorporated into next generation of ecosystem and land surface models for predicting changes of ecosystem functioning under global warming.