Legume-based cropping systems that re-couple carbon (C) and nitrogen (N) cycles have been shown to reduce N losses and increase soil N storage compared to conventional agricultural practices. Soil N availability is a key factor constraining the distribution of legumes in natural ecosystems, yet the relationship between endogenous soil N pools and biological nitrogen fixation (BNF) has rarely been studied in agroecosystems. Two field studies were conducted to understand: 1) the effects of a management-induced soil fertility gradient on BNF; 2) how BNF of different legume species or species mixtures respond to the fertility gradient; and 3) if there are integrative soil measurements that may improve estimates of BNF. Research plots were established on grain farm fields in New York in 2004 and 2006. Fields represented a gradient of N availability due to management differences, ranging from exclusive use of Haber-Bosch N to almost exclusive use of legume N inputs. Plant treatments included soybean (Glycine max), field pea (Pisum sativum), and perennial red clover (Trifolium pratense) grown either as monocultures or in mixture with grasses. BNF was measured using the ¹⁵N natural abundance method and several soil N pools were quantified ranging in microbial accessibility from extremely labile to primarily recalcitrant pools.

Results/Conclusions

BNF decreased with increasing soil N availability. Inorganic soil N and occluded particulate organic matter N, a labile pool of soil organic matter known to respond to management practices, show the strongest negative correlations with BNF (e.g., for soybean: r = -0.68 and r = -0.65, respectively). The strength of this relationship varied with legume species, and followed the order: soybean > red clover > field pea, with soybean BNF being the most sensitive to inorganic N availability. Red clover and field peas grown in mixtures with grasses maintained higher rates of N fixation than did monocultures, even in high fertility soils. The mean percent of plant N derived from fixation was 78% for mixed plots and 66% for monocultures. These results suggest that through legume species selections and the use of species mixtures, BNF can be intentionally managed to both retain soil N and maintain agroecosystem net primary productivity at different stages of soil fertility development.