COS 81-7 - Agricultural land-use and individual life history control spatial patchiness in post-agricultural forest communities

Wednesday, August 9, 2017: 10:10 AM
C120-121, Oregon Convention Center
Marion A. Holmes and Glenn R. Matlack, Environmental and Plant Biology, Ohio University, Athens, OH
Background/Question/Methods

Development of spatial pattern in plant distributions reflects multiple processes of community assembly in second-growth forests. During succession, pattern develops reflecting processes of colonization, vegetative and sexual propagation, and environmental filtering by underlying physical gradients. Agricultural land use may influence distribution of forest herbs by homogenizing the physical environment and imposing life history filters that select against slow-dispersing species; post-agricultural forests serve as our laboratory for understanding community assembly and pattern formation during secondary succession. We tested two hypotheses: 1) spatial pattern reflects plant life history traits or conformity to environmental gradients, and 2) pattern becomes stronger through successional time due to increasing influence of deterministic processes such as environmental sorting.

Forest stands were grouped into four age classes, and minimally-disturbed mature forest patches were included as a control. Stand age was determined using aerial photographs. Herbaceous vegetation and environmental variables recorded in nested 2 meter, 1 meter, and .5 meter plots to analyze spatial pattern at multiple scales. In a subset of common forest herb species spatial pattern was described in terms of coefficient of variation of stem number at each scale.

Results/Conclusions

Development of spatial pattern was linked to plant life history traits and environmental gradients. Distributions were clustered at fine scales early in the chronosequence, consistent with stochastic colonization.Clustering became more pronounced at intermediate scales by in the 40-60 year age class, reflecting clonal propagation in species such as Podophyllum peltatum. Distributions of species such as Arisaema triphyllum and Pilea pumila were clustered at the site scale in mid-late age classes. The long-established control stands were characterized by low densities and very highly clustered distributions of shade-tolerant herbs such as Asarum canadense. Sites have transitioned from stochastic to biotic pattern in the course of succession. We infer a process of community assembly by stochastic colonization, nucleation, and environmental filtering spanning more than a century, and mediated by individual life histories.