LAUSR

Abstract Forest conversion can affect the diversity and community structure of soil bacteria and fungi. However, little is known about the impact of forest conversion on aboveground-belowground linkages and associated nutrient cycling. To address this research gap, we investigated the aboveground and belowground recovery, the specific responses of bacterial and fungal communities and the associated nutrient cycling in the conversion from artificial Chinese fir forests to broad-leaved conifer-mixed forests at 5-yr-, 10-yr- and 25-yr-old stands after selective logging. Bacterial and fungal community structures were analyzed using high-throughput sequencing of the 16S rRNA and ITS genes. Six microbial enzyme activities involved in the cycling of C, N and P were studied. Our results showed that the aboveground forest structure was recovering towards native evergreen broad-leaved forest after selective logging. The alpha-diversity of fungal community was significantly increased across the chronosequence of logging, but the alpha-diversity of bacterial community was not. Fungal community composition and structure changed during forest conversion, whereas the bacterial community showed slight changes in unique taxa. Neither bacterial nor fungal community structures changed towards the similar pattern as native evergreen broad-leaved forest across 25 years. These results indicated that fungi were more susceptible to forest conversion than bacteria. Furthermore, compared with bacteria, fungi were more strongly affected by soil properties, such as NH4+-N, pH, AK and TOC. Structural equation modeling indicated that fungi had strong impacts on C-, N- and P-cycling but that bacteria only affected C-cycling. Overall, the fungal community may play a more important role in the aboveground-belowground linkages and C-, N-, and P-cycling compared with the bacterial community during forest conversion.