Strong legacy of agricultural land use on soils and understory plant communities in longleaf pine woodlands

Abstract

Post-agricultural ecosystems are increasingly prevalent in contemporary landscapes. Understanding the ecological impacts of agricultural legacies, which may last for decades to millennia following agricultural abandonment, is critical for interpreting contemporary patterns of biodiversity and for informing strategies for the restoration of post-agricultural ecosystems. However, the study of agricultural legacies is complicated by non-random land-use decisions: areas selected for agriculture commonly differ from non-agricultural areas in ways that may confound interpretation of agricultural legacies (e.g., agricultural areas may have more productive soils). We evaluated legacies of tillage agriculture in longleaf pine (Pinus palustris) woodlands of the Southeastern United States, using a paired-plot study design (n=29 paired remnant and post-agricultural woodlands) to minimize the confounding influence of non-random past land-use decisions. Agricultural legacies were prevalent, but not due to differences in the types of sites selected for agriculture. Nearly 60years after agricultural abandonment and afforestation, numerous soil attributes (i.e., N, P, organic matter, soil water holding capacity) remain altered in post-agricultural woodlands, relative to adjacent remnants. Overstory tree abundance and understory plant species richness were similar in remnant and post-agricultural woodlands, but we found a pronounced agricultural legacy on over- and understory plant community composition. Understory compositional differences were striking, particularly given the proximity of post-agricultural woodlands to immediately adjacent remnant source populations and the time (>50years) for source populations to spread. This difference in understory community composition could be caused by the persistent environmental differences we observed or by dispersal limitation, as evidenced by greater species richness in post-agricultural sites with greater historical landscape connectivity. We suggest that future experimental work will further assist in a mechanistic understanding of how agricultural legacies impede understory plant recovery and that this understanding will benefit the formulation of strategies for the restoration of post-agricultural longleaf pine woodlands and other forested systems.