Statistical analysis
The difference in soil properties and ammonia oxidizers abundances was assessed by one-way variance analysis (ANOVA) using SPSS Statistics 20 (IBM, USA). Pearson correlation analyses evaluated the relationships among soil N pools, N transformation rates, and microbial parameters. Differences at p < 0.05 based on the least significant differences (LSD) test were statistically significant. All statistical tests were performed using R v.3.5.3 (R Core Team, 2019). Principal component analysis (PCA) in the R package ‘vegan’ (Oksanen et al. 2016) was used to assess ammonia-oxidizer community composition differences across the different treatments. The microbial co-occurrence network was inferred from the Sparse Correlations for Compositional data (SParCC) correlation matrix constructed with the WGCNA package (Langfelder and Horvath, 2012). The nodes in the network represent OTUs, while the edges connecting these nodes represent correlations between OTUs. Network properties were calculated with the “igraph” R package, and we generated network images with Gephi (http://gephi.github.io/) and Cytoscape (http://www.cytoscape.org/).