Agricultural legacies, contemporary restoration, and seed limitation co-structure longleaf pine understory plant communities

Background/Question/Methods

Large portions of the terrestrial globe have a history of agricultural land use.  Compared to plant communities in habitats never used for agriculture, post-agricultural ecosystems often exhibit altered community composition for decades to centuries following the cessation of agriculture.  Despite their ubiquity and persistence, the mechanisms underlying these agricultural legacies remains poorly understood because few studies have manipulated factors thought to structure post-agricultural communities over relevant spatial scales.  Our objectives were to evaluate the importance of habitat quality and dispersal limitation in maintaining altered plant community composition due to past agricultural land use. We initiated a large-scale experiment in 2012 that manipulates habitat quality via overstory tree thinning (i.e., restoration) within 126, 1-ha longleaf pine woodland plots that do and do not support histories of agriculture (post-agricultural and remnant, respectively).  Within these experimental landscapes, dispersal limitation was eliminated by adding seeds of 12 understory plant species that are characteristically absent from post-agricultural woodlands.  Remnant and post-agricultural plots are immediately adjacent in space, allowing us to evaluate how overstory thinning affects spread of the diverse remnant understory plant community into post-agricultural plots (we previously showed little spread prior to restoration activities).

Results/Conclusions

Agricultural history, overstory thinning, and seed sowing each had impacts on longleaf pine woodland understory plant communities.  Thinning increased understory richness at 1, 100, and 1000m² sampling scales, by 8.4 – 30.9%, whereas the effects of agricultural history were only apparent at the 100m² scale (9.4% reduction, relative to remnants).  Thinning and agricultural legacies interacted to produce four distinct understory plant communities that differed in their composition.  However, although thinning increased richness and altering community composition in post-agricultural woodlands, this did not result in communities that more closely resembled communities in remnants.  Our seed addition experiment suggests that this is due to limited dispersal of remnant understory species into post-agricultural woodlands.  Sown plots supported 34.5% greater species richness than unsown plots; this effect of seed addition was greater following overstory thinning.  Together, these results show how restoration thinning can promote the establishment of arriving seeds, but that dispersal limitation – even over tens of meters – constrains the spread of remnant understory populations into post-agricultural woodlands, contributing to persistent legacies of past land-use on contemporary plant communities.