The long-term legacy of human land use activities on natural communities has been well established. While we know a fair amount about how past disturbance drives plant communities, less is known about how communities of belowground organisms, such as fungi and invertebrates, are structured by disturbance and how belowground responses to land use history can relate aboveground. However, land that is converted from agriculture to forest may lack the diversity of soil mutualists that many forest plant species depend on for nutrient acquisition or seed dispersal. In order to understand how altered interactions between soil biota due to prior land use can restructure plant communities, we surveyed forested sites within Stebbins Gulch, a 334 hectare mixed-mesophytic forest at The Holden Arboretum (Ohio, USA) consisting of a matrix of high quality mature hardwood forests and mid-successional forests that were used for agriculture until the 1930s. To understand the persistent effects of land use history on soil biota and associated plant communities, we surveyed plots of mature hardwood and post-agricultural forests, measuring the composition of soil fungi, earthworm, and ant communities as well as understory herbaceous plant and tree communities.
Results/Conclusions
Land use history had a large effect on fungal community structure, invasive earthworm abundance, and ant community composition. Even 80 years after disturbance, forests recovered from agriculture had a lower abundance of seed-dispersing ants (p=0.05) compared to mature forests. Disturbed forests also had a higher abundance of litter-feeding earthworms (p=0.01) and sites with a high abundance of epi-endogenic earthworms had increased leaf-litter turnover (p=0.01) and decreased rates of ant-mediated seed dispersal (p=0.01). The higher abundance of seed-dispersing ants in mature forests coincided with a greater abundance of myrmecochorus plants (p=0.005). Tree community composition differences between disturbed and undisturbed sites correlated with fungal community differences. A high proportion of ericoid mycorrhizal fungi (35% of DNA sequences) was found in forests on disturbed sites, while mature forests were dominated by either ectomycorrhizal (75% of DNA sequences) or saprotrophic fungi (42% of DNA sequences). The predominance of ectomycorrhizal fungi in some mature forests coincided with high basal area of tree species known to form relationships with these fungi, such as Fagus grandifolia and Quercus rubra. Our data suggest that past agriculture can have long-lasting impacts on forest communities, establishing differences in cryptic soil communities that can likely drive differences in plant communities.