Soil microbial community shifts and ecosystem multifunctionality in response to integrated cover cropping and nitrogen management in Mississippi corn production systems.

内容摘要

Integrating cover crops (CCs) with nitrogen (N) management can enhance soil health, yet their combined effects on soil microbiomes and multifunctionality remain insufficiently characterized in short-term, multi-location corn systems in Mississippi. We evaluated how CC identity (single species and mixtures) and reduced N fertilization shape soil microbial communities and soil multifunctionality across a three-year field experiment (2021-2025) conducted at two locations in Mississippi, USA (Starkville and Newton). Treatments included six CCs with a no-cover control under two N rates (0 and 112 kg N ha-1). At corn V4 growth stage, soil samples (0-15 cm) were analyzed for physicochemical and biological indicators (pH, total C and N, POXC, glomalin, and enzyme activities), bacterial and fungal communities (16S rRNA V4 and ITS2 amplicon sequencing), and N-cycling functional genes (amoA and nifH via qPCR). Across years and locations, CCs, especially ryegrass, consistently increased soil biological indicators and elevated a composite multifunctionality index, while N fertilization reduced soil pH and exerted comparatively smaller effects on multifunctionality. Microbial community structure was primarily driven by interannual variation and location factors, with N fertilization consistently increasing bacterial α-diversity, and CC effects were most evident in specific years and were strongest under ryegrass. Network analyses indicated higher bacterial network complexity under 0 N, whereas fungal networks showed greater connectivity under N addition. Bacterial diversity and key bacterial taxa were positively associated with multifunctionality, and structural equation modeling indicated that microbial diversity and N-cycling functional groups mediate management effects on multifunctionality. Overall, ryegrass enhanced microbial-linked ecosystem services under reduced N inputs, though these shifts did not translate into consistent corn yield gains.