Abstract Detail


Holmes, Marion [1].

Pasture trees shape post-agricultural forest composition by facilitating vertebrate-mediated seed dispersal.

Deciduous forests in eastern North America are largely second-growth that has recruited on abandoned agricultural land during the 20th century. The conditions of land at the time of abandonment, including presence of existing vegetation and structural features such as hedgerows or retained pasture trees, influence successional trajectories in regrowth forest. Open-grown pasture trees contribute to diversity within post-agricultural forests by creating habitat variation and influencing seed deposition by providing animal habitat. Forest specialists and vertebrate-dispersed species are typically clustered around single trees due to successional nucleation. In eastern deciduous forests of North America, previous work has demonstrated that pasture trees contribute to within-stand diversity in post-agricultural forests for decades after canopy closure. Less is known about how their presence might shape forest composition at the landscape scale. To answer these questions, I compared forest canopy composition between post-agricultural second-growth forests at sites with and without pasture trees. I tested the hypothesis that forest composition will differ between sites with and without pasture trees because they facilitate colonization by vertebrate-dispersed species; therefore, sites with pasture trees will have higher richness and abundance of nut-bearing and fleshy-fruited tree species. Sites without pasture trees will instead be dominated by wind-dispersed species.
Field sampling took place in Athens County, Ohio, USA. The study area is dominated by mixed mesophytic forest patches with extensive histories of agricultural clearance. I selected 24 post-agricultural forests with (n = 13) and without pasture trees (n = 11) at the time of abandonment. Sites were located on upland ridgetops and formerly cultivated sites were avoided to limit variation associated with different landscape positions and types of agriculture. I sampled the forest community using the point-centered quarter method along two 50-meter transects per site, and recorded all tree species present within each site. I compared species richness, multivariate composition, and richness and abundance of trees with different dispersal syndromes between treatments.
Tree community composition differed significantly between sites with and without pasture trees. Sites with them had significantly higher richness and abundance of nut-bearing species, which was reflected in overall composition. Sites without pasture trees had a significantly higher proportion of wind-dispersed species and fewer nut-bearing trees, consistent with the hypothesis that pasture trees facilitate animal-dispersed species through perch effects. Fleshy-fruited species were uncommon and did not differ significantly between treatments, which suggests that the influence of pasture trees may be stronger for certain types of animal dispersal than others. The different communities found in second-growth forests with and without pasture trees persist for decades, and likely represent alternate successional trajectories that likely influence forest composition and response to additional layers of disturbance into the future.
Forest plants amid overlapping landscape mosaics: land-use history and soils together shape native shrub and vine communities in second-growth forests

1 - 232 Copeland Ave, Braddock, PA, 15104, United States

land-use history
Community composition
community assembly
seed dispersal
vertebrate dispersal.

Presentation Type: Oral Paper
Number: EC12006
Abstract ID:559
Candidate for Awards:None

Copyright © 2000-2022, Botanical Society of America. All rights reserved