A first-order analysis of the potential role of CO2 fertilization to affect the global carbon budget: a comparison of four terrestrial biosphere models

We compared the simulated responses of net primary production, heterotrophic respiration, net ecosystem production and carbon storage in natural terrestrial ecosystems to historical (1765 to 1990) and projected (1990–2300) changes of atmospheric CO 2 concentration of four terrestrial biosphere models: the Bern model, the Frankfurt Biosphere Model (FBM), the High-Resolution Biosphere Model (HRBM) and the Terrestrial EcosystemModel (TEM). The results of the model intercomparison suggest that CO 2 fertilization of natural terrestrial vegetation has the potential to account for a large fraction of the so-called ‘‘missing carbon sink’′ of 2.0 Pg C in 1990. Estimates of this potential are reduced when the models incorporate the concept that CO 2 fertilization can be limited by nutrient availability. Although the model estimates differ on the potential size (126 to 461 Pg C) of the future terrestrial sink caused by CO 2 fertilization, the results of the four models suggest that natural terrestrial ecosystems will have a limited capacity to act as a sink of atmospheric CO 2 in the future as a result of physiological constraints and nutrient constraints on NPP. All the spatially explicit models estimate a carbon sink in both tropical and northern temperate regions, but the strength of these sinks varies over time. Differences in the simulated response of terrestrial ecosystems to CO 2 fertilization among the models in this intercomparison study reflect the fact that the models have highlighted different aspects of the effect of CO 2 fertilization on carbon dynamics of natural terrestrial ecosystems including feedback mechanisms. As interactions with nitrogen fertilization, climate change and forest regrowth may play an important role in simulating the response of terrestrial ecosystems to CO 2 fertilization, these factors should be included in future analyses. Improvements in spatially explicit data sets, whole-ecosystem experiments and the availability of net carbon exchange measurements across the globe will also help to improve future evaluations of the role of CO 2 fertilization on terrestrial carbon storage. DOI: 10.1034/j.1600-0889.1999.00017.x