Grazing exclusion on capacity of symbiotic nitrogen fixation estimated with natural 15N abundance and the 15 N dilution techniques

Abstract Grazing exclusion is the commonest degradation-recovering practice for alpine grasslands on Qing-Tibetan Plateau (QTP). A general response of legumes is the reduction in abundance, cover or biomass (esp. in relative scale), which is supposed to decrease in biological nitrogen fixation (BNF). Here, we want to estimate whether the capacity of BNF in legumes (measured by unit biomass) would be changed by grazing exclusion. Two 15N isotope techniques were adopted to compare the BNF capacity of Astragalus arnoldii (a legume) between inside and outside animal-exclusion fence. For natural abundance (NA) method, plants were sampled at three timepoints, while for isotope dilution (ID) method, plants were collected after one day (24h), one month and one year of labeling. The two consistent in estimating percentage of N derived from atmosphere (%Ndfa) (77.52 ± 1.96% in NA vs. 72.39 ± 2.59% in ID), except an underestimation in ID for the first-time sampling (24h after labeling), seemingly owing to insufficient recovery time for reference plants. No significant difference in %Nfda inside (74.36 ± 2.36%) and outside (75.57 ± 2.26%) of the fence was found. Given that the biomass of legumes inside the fence (33.96 ± 2.71 g m-2) was significantly lower than that outside (19.62 ± 1.25 g m-2), we conclude that if total BNF were reduced by grazing exclusion, it would be due to the population size of legumes rather than the capacity.