Persistent effects of land-use history on myrmecochorous plant and epigeic ant assemblages across an ecoregional gradient in New York State

Mariano G. Arias and Martin Dovciak, Department of Environmental Biology, State University of New York College of Environmental Science and Forestry, Syracuse, NY

Background/Question/Methods
The presence of ant-dispersed plants (myrmecochores) can indicate residual forests unaffected by past stand-replacing disturbances such as clear-cut logging, fire, or past agricultural land-use. In temperate deciduous forests of northeastern US, past forest land use and habitat fragmentation are major drivers of forest understory plant species composition, and important threats particularly to slow migrants such as myrmecochores. To inform myrmecochore conservation and to better understand the importance of mutualist ants for their distribution, we studied how past land-use affected community patterns of both ant-dispersed plants and their ant mutualists across different forest land-use histories (residual vs. secondary forests) along a broad climatic gradient in New York State (spanning 250 km and elevations from 276 to 616 m above sea level). We surveyed myrmecochore cover and the density of epigeic ant workers by species, and characterized vegetation structure, climatic drivers and edaphic characteristics on 18 sites with paired residual and secondary forest stands. We used Non-metric Multidimensional Scaling to identify main drivers of myrmecochore-ant species composition and to interpret compositional changes across the ecoregional gradient. We used Analysis of Similitude to evaluate compositional similarities across and within ecoregions and land-uses, and Generalized Linear Models to identify main drivers of species richness.
Results/Conclusions
Regional species turnover and composition of myrmecochore and ant assemblages responded to the ecoregional gradient based on forest land-use history. In residual stands, changes in myrmecochore and ant species composition across the gradient were correlated to each other, and species turnover resulted in distinct myrmecochore-ant assemblages at each end (cool vs. warm) of the gradient. No such regional patterns were observed in myrmecochore-ant assemblages in secondary stands, where compositional changes in myrmecochore and ant species were independent of each other and neither species composition varied across the ecoregional gradient. Within ecoregions, keystone seed disperser ants (e.g., Aphaenogaster) were collected at similar densities across forest land-use histories suggesting that land-use history may not be a significant factor affecting ant seed dispersal. Further, no differences in species richness of either myrmecochores or epigeic ants were found across ecoregions or land-use histories. In conclusion, myrmecochore and ant assemblages in residual stands clearly varied in unison across the ecoregional gradient, suggesting their mutual relationship and similar coupling to the environmental gradient. In contrast, myrmecochore and ant assemblages in secondary stands seemed to be decoupled, both from each other and from the ecoregional gradient, as past land-use homogenized both of these assemblages across the gradient.