Relationship between soil microbial communities and nutrient parameters
The results of the Mantel test based on Spearman’s rank correlation revealed that the diversity of bacterial communities in invaded plots along the altitudinal gradient were influenced by all the environmental variables (p <0.05) except SAL, Zn and Temp (p >0.05), whereas in uninvaded plots only three parameters viz. SAL, Zn and Temp determine the bacterial community diversity (p <0.05) (Table 2). The most influential variable explaining community composition is TOC (rho =0.97,p < 0.001) in invaded plots and SAL in uninvaded plots (rho = - 0.81, p < 0.001) (Table 2). Similarly, the fungal community composition in invaded plots was influenced by all the environmental variables except SAL and Zn (p > 0.05), whereas in uninvaded plots EC, K, Zn, Temp and Alt did not show any significant correlation (p> 0.05). WC in invaded plots (rho = 0.92, p< 0.001) and Mn in uninvaded plots (rho =0.93, p< 0.001) were most significantly correlated with fungal community composition (Table 2).
Further, Canonical Correspondence analysis (CCA) was used to analyze the relationship between microbial community structure and environmental variables along the elevation gradient (Fig. 6). Based on Variance Inflation Factors (VIF) values obtained, it was observed that among the 15 environmental variables only five (pH, EC, K, Fe and temperature) were more significant (p <0.001) in determining the bacterial and fungal diversity among the plots along the altitudinal gradient. All the five environmental variables were found to be statistically significant based on Monte Carlo permutation F test (Appendix S3: Table S4). Constrained inertia or variance explained by the environmental variables in bacteria and fungi was found to be 1.039 and 1.605 respectively.
In bacteria, first two axes explained 70.9% of the variation (Axis 1~ 41.5% and Axis 2~ 29.4%). Axis 1 was positively related with pH and negatively associated with Fe. Axis 2 was positively correlated with EC and Temp and negatively correlated with K. The relative abundance of bacteria at the KZ_IN and KD_IN were related with pH and negatively associated with Fe. However, the bacterial communities of uninvaded plots at KZ and TM showed a positive relation with EC and Temp, whereas KD_UN were correlated with K. TM and KU sites were positively related with Fe (Fig. 6a). In fungi, first two axes explained 93.33%of variation (Axis 1~ 49.98% and Axis 2~ 43.35%). Axis 1 was positively correlated with EC and Temp and negatively associated with K. Axis 2 showed positive relation with pH and negatively related to Fe. The relative abundance of fungi in both plots at KZ site was correlated with EC and Temp, whereas both plots of KD site along with KU_IN were associated with K. Both plots of TM and KU_UN plot were positively related with Fe and negatively associated with pH (Fig. 6b). Our results revealed that there is a significant effect of soil physicochemical properties on the bacterial and fungal community abundance along the altitudinal gradient.