Biological nitrogen fixation in California annual grassland: Responses to elevated carbon dioxide and effects on community nitrogen status and productivity

The capacity of many terrestrial ecosystems to increase carbon uptake and storage under elevated atmospheric CO₂ is influenced by the availability of limiting resources, particularly nitrogen (N). Biological N-fixation is a major source of N to many terrestrial systems, and the response of N-fixing organisms to elevated CO₂ is likely to affect N-availability. I measured the growth and N-fixation responses of two annual grassland legumes to elevated CO₂, and measured the effects of legume presence on community productivity and nutrient pools under ambient and elevated CO ₂, and across range of N-addition levels. Elevated CO₂ increased the growth of the legume Lotus wrangelianus, but did not significantly stimulate growth of the legume Vicia sativa, which was more abundant than L. wrangelianus in the communities. Nitrogen addition caused a decline in the biomass of both legume species. I estimated N-inputs from biological N-fixation using the 15N pool dilution method, and found that legumes fixed as much as 4.4 g N m −2 at low N-addition, and fixed 1.7 g N m−2 at high N-addition. Elevated CO₂ did not significantly affect the amount of N fixed. Communities with legumes had 28–55% more N in aboveground plant tissue than communities without legumes, with the greatest difference at low N-addition. However, N pools in legume tissue more than accounted for the differences in total community N, suggesting that legume presence did not increase N availability to non-fixers. Total aboveground productivity was comparable in communities grown with and without legumes, and there was no difference between communities in how productivity responded to elevated CO₂ or N addition. Communities with legumes had less phosphorus (P) in aboveground plant tissue than communities without legumes, particularly when grown under elevated CO₂. Non-fixing species increased P acquisition in response to elevated CO₂ when grown without legumes, but not when they were grown with legumes. These results suggest that P acquisition by non-fixers was limited by competition with legumes, and that legumes may have allocated more P to below-ground roots and nodules, reducing aboveground P pools at the community level, relative to communities composed only of non-fixers.