Soil moisture regulates the biological response of elevated atmospheric CO2 concentrations in a coupled atmosphere biosphere model

Terrestrial biosphere models/land surface models are routinely used to study the effects of CO2 doubling and climate change. The objective of this study is to show that the biological response associated with CO2 doubling is important, and that the effects intrinsically depend on the soil moisture state. Therefore, using a coupled biosphere–atmosphere model, we tested the hypothesis that the biological effects of CO2 changes in biosphere models are significantly coupled to the hydrological feedback via soil moisture availability in a terrestrial biosphere/land surface model. The results from a 15-day simulation of a photosynthesis-based land surface model, dynamically coupled to an atmospheric boundary layer and surface energy balance scheme, were analyzed to test the hypothesis. The objective was to analyze the biological effects of CO2 doubling under high as well as limiting soil moisture conditions for prescribed changes to the vegetation/land use type. The approachwas to analyze the results from a coupled land surface-atmospheremodel obtained by changing the biome type for each run. Sensitivity for all of the nine global vegetation type changes, as defined through the Simple Biosphere Model ver. 2 (SiB2) land cover classification, were analyzed for evapotranspiration and net carbon assimilation. The results indicated that: (i) the soil moisture (and its interaction with CO2) has a direct (first-order) effect on the biological effects of CO2 changes and the terrestrial ecosystem response; (ii) the biological impacts associated with CO2 changes in a biospheric model should be interpreted in consideration of the soil moisture status; and droughts or high soil moisture availability can enhance or completely balance or even reverse the effects associated with CO2 changes; (iii) for each vegetation type, the model results indicated a different response to soil moisture and CO2 changes; and resolving the direct and indirect effects explicitly, both C3 and C4 vegetation, appeared to be significantly affected by the biological effects of CO2 changes, and (iv) the explicit coupling between soilmoisture/hydrological state and the CO2 changes need to be explicitly considered in projecting climate change impacts. The study results also indicated that feedback pathways can be efficiently determined by dissociating the direct and the interactive effects of CO2 impacts. © 2006 Elsevier B.V. All rights reserved.