In grassland ecosystems, most of the carbon (C) occurs below-ground. Understanding changes in soil fluxes induced by elevated atmospheric CO2 is critical for balancing the global C budget and for managing grassland ecosystems sustainably. In this review, we use the results of short-term (1–2 years) studies of below-ground processes in grassland communities under elevated CO2 to assess future prospects for longer-term increases in soil C storage.

Results are broadly consistent with those from other plant communities and include: increases in below-ground net primary productivity and an increase in soil C cycling rate, changes in soil faunal community, and generally no increase in soil C storage. Based on other experimental data, future C storage could be favoured in soils of moderate nutrient status, moderate-to-high clay content, and low (or moderateIy high) soil moisture status. Some support for these suggestions is provided by preliminary results from direct measurements of soil C concentrations near a New Zealand natural CO2-venting spring, and by simulations of future changes in grassland soils under the combined effects of CO2 fertilization and regional climate change.

Early detection of any increase in soil C storage appears unlikely in complex grassland communities because of (a) the difficulty of separating an elevated CO2 effect from the effects of soil factors including moisture status, (b) the high spatial variability of soil C and (c) the effects of global warming. Several research imperatives are identified for reducing the uncertainties in the effects of elevated atmospheric CO2 on soil C.