Effects Of Forest Edges On Population Dynamics In A Successional System

Abstract

Abandoned agricultural land is often bound by remnant woodlands which may locally influence successional dynamics. The primary objective of this study was to examine the effect of an old-growth forest edge on plant population dynamics within a successional system. Data from the Buell-Small Succession Study (BSS) in New Jersey were used to analyze population dynamics in relation to the forest edge during the first 50 years of succession following agriculture. It was hypothesized that edge responses would be driven by either differential colonization (dispersal and establishment) or differential performance (increase within plots).

By looking at individual species' edge responses, I worked toward understanding the role the forest edge plays in regulating successional patterns. Individual species edge patterns were determined for 75 species by looking at the year of peak cover (highest percent cover over the first 50 years) and determining the edge association by performing an analysis of covariance to quantify the field and distance from the forest edge effects for each species. The role of mechanistic processes, differential colonization and differential performance, in generating species' edge responses were analyzed by performing a logistic regression or analysis of covariance, respectively.

Large amounts of variation were found among species in both the direction and magnitude of edge responses. Of the species analyzed, approximately 27% were more abundant closer to the forest edge, 39% were more abundant farther from the edge, and 34% had no discemable edge response. Edge responses were not directly associated with plant traits. Mechanistic analyses of species edge responses found that some were driven by differential dispersal, some by differential performance, and some by both mechanisms. Though both were important, differential colonization (53% of species) influenced spatial patterns nearly twice as often as differential performance (29% of species). Despite its prevalence, differential colonization was not associated with any of • the plant traits. In contrast, differential perfonnance was associated with clonality, • dispersal mode, and peak year in succession.

There was no overriding pattern that connected edge response with species traits that would allow prediction of an association with distance from the edge for groups of species. However, species varied individualistically in response to edges and no one mechanism ( differential colonization or performance). Overall, the presence of habitat boundaries had a large but unpredictable effect on population dynamics and spatial patterns within the successional system. Understanding succession of plant communities is important for management of this ecosystem altered by cultivation and fragmentation.

Date of Award	2010
Original language	American English
Awarding Institution	Eastern Illinois University
Supervisor	Scott J. Meiners (Supervisor)