Abstract: Biological nitrogen (N) fixation is an important source of N in terrestrial ecosystems, but the response of soil microbial N fixation rate to N deposition in different forest ecosystems still remains uncertain. We conducted a field N addition experiment to simulate atmosphere N deposition in subtropical Pinus taiwanensis and Castanopsis faberi forests. We set up three levels of nitrogen addition using urea as the N source: 0 (control), 40 (low N), and 80 g N·hm^-2·a^-1(high N) to examine the chemical properties, microbial biomass C, enzyme activities, and nifH gene copies of top soils (0-10 cm). We also measured the microbial N fixation rate using the ¹⁵N labeling method. Results showed that N addition significantly reduced the soil microbial N fixation rate in the P. taiwanensis and C. faberi forests by 29%-33% and 10%-18%, respectively. Nitrogen addition significantly reduced N-acquiring enzyme (i.e., β-1, 4-N-acetylglucosaminidase) activity and nifH gene copies in both forest soils. There was a significant positive correlation between the microbial N fixation rate and soil dissolved organic C content in the P. taiwanensis forest, but a significant negative relationship between the rate of soil microbial nitrogen fixation and NH₄⁺-N content in the C. faberi forest. Overall, soil microbial N fixation function in the P. taiwanensis forest was more sensitive to N addition than that in the C. faberi forest, and the factors affecting microbial N fixation varied between the two forest soils. The study could provide insights into the effects of N addition on biological N fixation in forest ecosystems, and a theoretical basis for forest management.

Key words: N addition, forest type, free-living N₂ fixation, ¹⁵N labeling method, nifH, arbuscular mycorrhizal fungi