Role of N2-fixation in constructed old-field communities under different regimes of [CO2], temperature, and water availability

Nitrogen fixation was measured in constructed old-field ecosystems that were exposed for 3 years to different combinations of elevated atmospheric [CO2] and temperature (300 ppm and 3 degrees C above ambient, respectively), and ambient or reduced soil moisture (corresponding to 25 or 2 mm rainfall per week). The old-fields included seven planted herbaceous annual and perennial species, including two legumes (Trifolium pratense and Lespedeza cuneata). Potential asymbiotic N-2-fixation by soils, measured in laboratory incubations, was significantly less under the dry treatment but was estimated to contribute little overall to annual ecosystem N budgets. Foliar N concentrations declined significantly under elevated [CO2]. Effects of the three environmental factors on the mean (+/- SE) fraction of legume N derived from atmospheric N-2 (FNdfa) varied from year-to-year, and FNdfa ranged from 0.64 +/- 0.05 to 0.94 +/- 0.03 depending on species and growing season. High rates of symbiotic N-2-fixation (4.6-12 g N m(-2) y(-1)) that annually contributed from 44% to 51% to the aboveground N stock in the old-field community was an important process driving changes in species composition over the 3-year experiment. Lespedeza biomass increased over time at the expense of several other species, including the other N-fixer, Trifolium. The dominance of Lespedeza in our ecosystem was due to high symbiotic N-2-fixation rates, as well as shading effects on other species. The high symbiotic N-2-fixation rates were largely independent of manipulations of [CO2], temperature, and water. The relatively high percentage of the aboveground N stock in this ecosystem contributed by symbiotic N-2-fixation suggests that non-legume species may have benefited indirectly via reduced community demands on soil N supplies. Species-specific traits were important in the constructed ecosystems, indicating that multi-species studies are required for understanding complex interactions among environmental factors and dynamic changes in community composition.