Date of Award

12-2007

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Entomology

Committee Chair/Advisor

Zungoli, Patricia A.

Committee Member

Adler , Peter H.

Committee Member

Benson , Benson P.

Committee Member

Bridges , William C.

Committee Member

Turnbull , Matthew W.

Abstract

Fragmented habitats are a common result of human activities that involve the clearing of native vegetation for various land uses. Habitat fragmentation has two primary and inseparable components: habitat loss and insularization. Landscape-level connectivity in fragmented habitats is diminished for many species, and edge characteristics become more important as the edge-to-interior ratio increases. The study of the effects of edges on biodiversity has compelled ecologists to study them for over a century and has influenced the design and management of natural preserves. Invasion of habitats by non-native species is a phenomenon that coincides with the increased human activity typical of fragmented habitats. Biological invasions are second only to habitat loss in terms of effects on native biota and synergistically compound the impacts of habitat fragmentation. The primary goal of this research was to describe the effects of habitat boundaries between undisturbed forests and artificially maintained grasslands on ant communities. A secondary goal was to ecologically assess the invasive potential of Pachycondyla chinensis (Emery), a little-studied adventive species in North America.
Ant communities were sampled across the forest-grassland habitat boundary in the Great Smoky Mountains National Park (GSMNP), North Carolina and Tennessee, USA. Different ant communities were found on either side of this boundary. The native forest-ant community is intact within 15 m of edge zones. Twenty-seven significant indicator species were identified within four

distinct ant assemblages. Ecotonal effects were observed only in forests when individual species abundance was compared at different distances into undisturbed forest habitat. Analysis of variance revealed differential abundance, with respect to habitat boundaries, of significant indicator species in both grassland and forest habitats. Differences in rarely collected species as visualized with rank abundance plots, occurred in the forest only. Total species richness was greatest near edge zones in both forest and grassland habitats. However, within-habitat ecotonal effects were not found in analysis of averaged, per sample species richness or diversity across distances from edges. Analysis of averaged per sample ant abundance revealed an ecotonal effect in the grassland where values increased with distance from the forest edge. Several biotic and abiotic habitat characteristics were identified that correspond with patterns in habitats and ant distributions. These characteristics have not been causally linked with trends in ant communities and should be interpreted as starting points for future hypothesis testing.
Pachycondyla chinensis (Emery), native to southeastern Asia, was detected in the United States more than 70 years ago. Basic knowledge regarding the biology and ecology of this species is lacking, especially in its expanded range in North America. This work documents the first collection of P. chinensis within the forests of GSMNP. In South Carolina, USA, habitats were surveyed, colonies were excavated, and ant communities were sampled along transects to determine the nest characteristics and formicid-community ecology of P. chinensis. Colonies of P. chinensis were found just beneath the soil under stones, logs, and stumps, and ranged in size from thirty-nine to several thousand individuals, most with multiple dealated females per nest. In forest habitats sampled in South Carolina, P. chinensis is a dominant ant in the community in terms of abundance and frequency. Pachycondyla chinensis and Paratrechina faisonensis (Forel), the two ants with the highest relative importance values and similar nesting habits, had a significant negative association. Pachycondyla chinensis was not found in open-field habitats. Although P. chinensis is not well established in GSMNP, the forests are similar, in terms of ant assemblage composition, to heavily infested forests in South Carolina. The future establishment of dense populations of P. chinensis within the GSMNP is likely. Following a substantial lag phase since the original introduction to North America, P. chinensis has become well established in urban habitats, where it reaches high population densities and is a true invasive species able to dominate naturally forested habitats. The known range of the species is summarized from published and unpublished sources, with significant range expansion noted in the United States and unpublished records from the Australasian and Oriental zoogeographic regions.
To validate sampling techniques for ground-foraging ants in temperate North America, capture data were compared for Winkler litter extraction and pitfall trapping at three periods in the calendar year. Sampling was conducted in the forests and grasslands of GSMNP. Individual species were biased towards capture with both collection techniques, and as expected their combined use generated more comprehensive species lists than did the use of either technique alone. These findings agree with literature, citing Winkler litter extraction and pitfall trapping as complementary techniques. However, ecological studies often use a single collection technique due to resource limitations. In forest habitats, Winkler litter extraction is recommended if a single technique will be used. The majority of species showing bias in forests were captured more frequently with Winkler litter extraction across all subsample sizes. However, Camponotus species, an ecologically important genus, were biased towards pitfall trapping. In grassland habitats, an equal number of species showing bias was divided between sampling techniques. Pitfall trapping captured more total species in the grassland. However, at small sample sizes (< 40), Winkler litter extraction performed better. Additionally, species of Ponerinae were biased towards Winkler litter extraction in both forest and grassland habitats. Seasonal sampling revealed that the majority of species were captured in July-August versus December-January or March-April and that the reduced species assemblages of the latter two pairs were perfectly nested subsets of summer sampling with one exception. Stenamma meridionale Smith was not present in July-August sampling and had peak frequency in December-January.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.