Applying meta-community theory for understanding long-term changes in arable plant communities of intensively managed farmland

Background/Question/Methods

Meta-community theory provides a robust framework for disentangling the contributions of local interactions and regional processes in the structuring and dynamics of local communities. However, empirical approaches are sparse in comparison to theoretical advances. Thus, gaps between theory and empirical approaches require incorporating different organisms and habitats and linking well understood local interactions (competition) with dispersal and habitat heterogeneity at wider spatial scales. Here, we apply concepts rooted in meta-community theory to understand long-term changes in the arable flora of the Rolling Pampa, the corn-belt of Argentina. We focus on maize, the warm-season crop with the longest cropping history in the region. Weeds in maize fields were periodically surveyed during the summer seasons between 2004 and 2011. Fencerow vegetation was also surveyed between 2003 and 2005. Previous research between the early 20^th century and 1995 on plant communities in remnant grasslands and maize fields for the study area was also included. Historical information about maize crop management was also compiled. Regional species richness (γ-diversity) was accumulated for particular periods associated with key technological changes. γ-diversity was additively partitioned into local (α-diversity) and turnover (β-diversity) components, where β-diversity = γ-diversity − α-diversity. Species were classified according to origin and functional traits.

Results/Conclusions

Remnant patches of Pampa grassland, though significantly reduced, still sustain many native species and have been enriched with one exotic species per year. Interestingly, fencerows nowadays retain numerous native species (ca. 40%), albeit their reduced area. Weed flora of maize crops was enriched in ca. 7 species per decade between 1930 and 1995, the year before the inception of glyphosate tolerant GM-soybean. Subsequently, maize weed flora lose nearly three species per year (γ = 99 to 49). β-diversity decline (85 to 41) mainly reflected regional richness loss; whereas α-diversity decrease was relatively lower (14 to 8). Land-use changes explain β-diversity decrease. Maize sown area considerably decreased since 1996. Improvements of maize crop management restricted the space available for weed species packing at local scale (fertilization, higher density, and crop protection). No-tillage hindered weed species requiring regular plowing to perpetuate their presence within fields (Datura ferox), whereas promoted small-seed weeds (Digitaria sanguinalis), wind-dispersed species (Conyza bonariensis), and woody species (Sida rhombifolia). Pasture plowing and fencerow removal to increase cropland changed the proportions of source and sink habitats consequently altering dispersal relationships. Our results suggest that mass effects and species-sorting were involved in determining meta-community dynamics of weed communities in the Pampas.