|Leakey, Andrew d b|
Submitted to: Global Biogeochemical Cycles
Publication Type: Peer reviewed journal
Publication Acceptance Date: 5/13/2013
Publication Date: N/A
Citation: Interpretive Summary: Global change is going to continue to have an impact on the carbon and water cycles of ecosystems. A tremendous amount of research has been to understand how plants respond to global changes, but less known is the impact of these changes scaled from the leaf or plant to the whole ecosystem or globe. Ecosystem models provide the only opportunity to assess the impact of global change on whole ecosystem functions. The research in this experiment uses the data measured over a wide range of environmental conditions and from a range of in-field experiments to develop ecosystem models for corn and soybean with accurate representation of the ecosystem responses to rising atmospheric carbon dioxide. In addition to improving the understanding of future atmospheric conditions on ecosystem functions for two very important food crops, this research shows the importance of accounting for known changes in plant function to climate change on large-scale carbon and water cycles.
Technical Abstract: The physiological response of vegetation to increasing atmospheric carbon dioxide concentration ([CO2]) modifies productivity and surface energy and water fluxes. Quantifying this response is required for assessments of future climate change. Many global climate models account for this response; however, significant uncertainty remains in model simulations of this vegetation response and its impacts. Data from in situ field experiments provide evidence that previous modeling studies may have overestimated the increase in productivity at elevated [CO2], and the impact on large-scale water cycling is largely unknown. We parameterized the Agro-IBIS dynamic global vegetation model with observations from the SoyFACE experiment to simulate the response of soybean and maize to an increase in [CO2] from 375 ppm to 550 ppm. The two key model parameters that were found to vary with [CO2] were the maximum carboxylation rate of photosynthesis and specific leaf area. Tests of the parameterized model showed a good fit to site-level validation data for all variables except latent heat flux over soybean and sensible heat flux over both crops. Simulations driven with historic climate data over the central U.S. show that increased [CO2] results in decreased latent heat flux and increased sensible heat flux from both crops when averaged over 30 years. Thirty-year average soybean yield increased everywhere (~10%); however, there was no statistically significant increase in maize yield except during dry years. Without these parameterizations, soybean simulations at 550 ppm overestimated leaf area and yield. Our results highlight important model parameterizations that, if not performed in other models, could result in biases when projecting future crop-climate-water relationships.