(ii) Factors affecting above- and belowground plant pathogens
Among different plant community-level indices, environmental factors and
soil properties, linear mixed-effects models result showed that onlyProneness (P < 0.001, MarginalR2 = 0.348) had a significant positive effect
on PL , while no variable had a significant effect onsfpOTUs and sfpRA (Fig. S1.7, S1.8, S1.9). Among various
univariable and multivariable linear mixed-effects models, only
univariable of Pronenss (P < 0.001) andEveness (P = 0.043) had significant effects on PL ,
and the combination of these factors (Proneness + Eveness )
provided the best fit (AICc = 82.510,w AICc = 0.269, R2 = 0.438)
compared to the null model (AICc = 93.719,w AICc = 0.001), accounting for c. 44% of
the deviance explained in PL (Fig. 1a; Table S1.4).sfpOTUs were directly associated with Elevation (P= 0.005), and strongly influenced by Soil PCA1 (P = 0.021)
(Fig. 1b). The combination of Elevation and Soil PCA1provided the best fit (AICc = 76.842,w AICc = 0.294, R2 = 0.534) forsfpOTUs among various multivariable models (Table S1.5).sfpRA had no statistically
significant association with any ecological variable tested, and was
therefore not included in the SEM (Table S1.6)