Scale-dependant legacies of agricultural land-use on soils and understory plant communities in longleaf pine woodlands

Background/Question/Methods

Agricultural land-use represents a major ecosystem disturbance, which can result in long-lasting ecological legacies following cessation. Although past agricultural land use can have significant effects on the richness and diversity of plant communities, it remains unclear whether these differences coincide with, or are driven by, agriculture-mediated changes in soil attributes.  Here we evaluate differences in understory plant diversity and soil environmental heterogeneity within and among paired post-agricultural and remnant longleaf woodland stands.  We quantified plant community and soil environmental characteristics in 29 paired post-agricultural and remnant woodland sites at the Savannah River Site, near Aiken, SC.  Post-agricultural stands were released from agriculture and converted to pine plantation ~70 years ago, whereas remnants show no signs of past agricultural activity.

Results/Conclusions

The composition and beta-diversity, but not the richness of understory plant communities differed with agricultural history.  These differences coincide with separation of mean soil attribute values between land-use histories, with levels of soil moisture holding capacity (SM), organic matter (OM), potassium (K) and sulpher (S) being lower, and phosphorus (P) being higher in post-agricultural stands compared to remnants.  Furthermore, we show that the influence of agriculture on the spatial variation of P depends on the scale of measurement: compared to remnant stands, P was more homogeneous within post-agricultural stands, but more heterogeneous among stands. Conversely, among-stand variability of pH was higher in remnants, whereas within-stand variation was similar between land-use histories. In this system, biotic and edaphic legacies of past land-use continue to persist even after ~70 years since agricultural abandonment and reforestation. Our results suggest that post-agricultural woodlands may support lower beta diversity due to within-stand homogenization of soil resources. However, conclusions about whether agriculture homogenizes the spatial distribution of edaphic variables largely depend on the scale of analysis.  Moreover, our work also suggests that restoration of plant communities in post-agricultural habitats will require strategies that deal explicitly with altered soil characteristics.