Land use intensity differently influences soil communities across a range of arable fields and grasslands

• Impacts of land use intensity (LUI) on the diversity of soil communities depend on land use type. • Different agricultural practices drive soil communities composition in arable vs grassland soils. • Understudied taxa like protists are a key element of soil network responses to LUI. • Land use type, rather than LUI, influences the complexity of the soil metacommunity network. While there is an emerging body of research showing the consequences of land use intensity on soil biodiversity, most studies focus on biodiversity responses to a single or a limited number of agricultural practices in controlled settings or at a single field site, neglecting that multiple practices are simultaneously applied by farmers in real agroecosystems. The combined effects of various agricultural practices have, until now, been largely overlooked in agroecosystems. Here, we conducted a field soil sampling campaign on 87 farms with two land use types (39 arable fields and 48 grasslands) to investigate the relationship between land use intensity, determined by various agricultural practices, and multiple soil communities (bacteria, fungi, protists and invertebrates) at regional scales. We found that land use intensity influenced the diversity and community composition of various soil taxa differently, and these impacts strongly depended on land use type. Soil fungi were most susceptible to land use intensity in both arable fields and grasslands. Specifically, irrigation and pest control were the main practices shaping soil communities in arable fields, while phosphorus and nitrogen fertilization were the main practices structuring soil communities in grasslands. Furthermore, an increase in land use intensity led to greater soil network complexity in arable fields. Our findings reveal regional on-farm patterns of land use intensity effects on various soil communities and identified key agricultural practices that structure soil communities. A key strength of this study is that these patterns can be generalized because the samples were collected from 87 farmlands where multiple agricultural practices were implemented simultaneously. Overall, this work provides a comprehensive perspective on the different responses of multiple soil communities and their associations to land use intensity in agroecosystems.