Semester

Spring

Date of Graduation

2008

Document Type

Thesis

Degree Type

MS

College

Eberly College of Arts and Sciences

Department

Biology

Committee Chair

James B. McGraw.

Abstract

Ailanthus altissima (tree of heaven) is a non-native invasive tree spreading within central Appalachia. This dioecious, deciduous, and allelopathic species copiously produces samaras, capable of traveling at least 200 m through primary wind-dispersal. Removal of A. altissima individuals prior to timbering and other forest disturbances may help prevent spread into forest interiors. To aid in species management, this study investigated the use of remote sensing to identify the location and abundance of samaras in mixed mesophytic forests through supervised classifications. From empirical measurements, the estimated number of seeds per classified unit area was determined and the relationship between quantified propagule sources and individual seed primary dispersal was spatially modeled using a cellular automata model. The predicted seed dispersion pattern was compared to empirical seed trap measurements. Secondary seed dispersal, germination and seedling survival parameters were determined through field experiments. Specifically, I found that remote sensing can be successfully used to distinguish A. altissima samara clusters from surrounding closed canopy vegetation. The identified total area of classified A. altissima samara clusters was determined to be linearly related to total canopy seed yield, however model predictions of seed dispersal patterns generally predicted greater numbers of seeds per seed trap compared to measured outcomes. Seed dispersal predictions using all classified A. altissima samara clusters provided no positive relationships with observed values (Site II, p=0.2997; negative relationship at Site III, p=0.0053). Manually delineating seed sources through photo interpretation resulted in a positive relationship between model estimates and observed seed rain (Site II, p=0.0225), but may still contain commission errors (Site III, p=0.7002). Microsite survey observations of seed germination indicate the frequency of safe sites was high in disturbed environments (canopy gaps), but seedling survival was negligible in the year this study was performed. Measurements of secondary dispersal on land and water (hydrochory) indicate the species can be dispersed with multiple dispersal agents. Hydrochory was shown to be able to contribute greatly to the species' long distance dispersal, as seeds can reach water bodies, stay viable for long periods of immersion, and can be dispersed downstream distances more than two orders of magnitude greater than primary dispersal. This was the first demonstration of hydrochory in this species. From these studies, it appears that polychory may contribute greatly to this species dispersal and I can infer the same polychory occurs in other terrestrial species, and may contribute to invasive success.

Share

COinS