Masters Theses

Date of Award

5-1996

Degree Type

Thesis

Degree Name

Master of Science

Major

Wildlife and Fisheries Science

Major Professor

David A. Buehler

Committee Members

Ralph W. Dimmick, John W. Philpot

Abstract

Avian species evolved with and adapted to specific disturbance patterns and regimes. To compare the effects of natural and anthropogenic disturbances on bird communities, I sampled birds and habitat conditions in undisturbed, tornado-disturbed and clearcut forests. I measured vegetation in the two disturbance treatments and undisturbed forest to characterize habitat conditions in the University of Tennessee's Forestry Experiment Station in Oak Ridge, Tennessee. I censused birds in the breeding seasons (May 15- July 15) of 1994 and 1995 using 50 m radius point counts and spot maps.

Overstory vegetation, herbaceous cover and coarse woody debris differed among treatments. Overstory vegetation was greatest in undisturbed forest, intermediate in tornado-disturbed areas and least in clearcuts (P ≤ 0.017). Density of herbaceous cover was greatest in clearcuts, intermediate in tornado sites and least in undisturbed forest (P ≤ 0.017). Coarse woody debris did not differ between clearcuts and tornado sites (P = 0.417, Bonferonni-adjusted α = 0.017), but was less abundant in undisturbed forest than in both disturbance treatments (P ≤ 0.017). Shrub and grass cover were greatest in clearcut areas, intermediate in undisturbed forest and least in tornado sites (P ≤ 0.017). Snag basal area was greatest in undisturbed forest, intermediate in tornado sites and least in clearcuts (P ≤ 0.017). Conifer basal area and litter depth did not differ between clearcuts and tornado sites (P > 0.017), but were significantly greater in undisturbed forest than the two disturbance treatments (P ≤ 0.017). Sapling density did not differ between tornado sites and undisturbed forest (P = 0.070), but was significantly greater in clearcuts than in undisturbed forests and tornado sites (P ≤ 0.017). Seedling density did not differ among the three treatments (P = 0.5867).

I observed 42 and 48 bird species on 50 m point counts in 1994 and 1995, respectively. I also recorded 37 and 38 species that had established breeding territories in 1994 and 1995, respectively.

Avian communities were more similar to one another on clearcuts and tornado sites than to those on undisturbed forests, based on Horn's Index of Community Similarity. Avian abundance, species richness and diversity did not differ between clearcut and tornado-disturbed areas (P > 0.017), but were significantly lower in undisturbed forests (P < 0.017). Equitability estimates did not differ among treatments (P = 0.1213).

Cavity nesting birds were most abundant in tornado sites, intermediate in undisturbed forest and least abundant in clearcuts (P ≤ 0.017). Apparently, damaged and dying trees in the tornado sites provided suitable nesting and foraging sites for cavity nesters. Canopy nesters and foragers were most abundant in undisturbed forests, intermediate in tornado sites and least abundant in clearcuts (P ≤ 0.017). Residual overstory vegetation in the tornado sites may have provided habitat for these canopy nesting and foraging birds. Shrub nesters and foragers were most abundant in clearcuts, intermediate in tornado sites and least abundant in undisturbed forest (P ≤ 0.017). Dense understory vegetation may have provided nesting and foraging sites for these birds in the clearcuts and tornado sites. Ground nesting and foraging birds were most abundant in tornado sites, intermediate in clearcuts and least abundant in undisturbed forest (P ≤ 0.017). Coarse woody debris and dense understory vegetation in the clearcuts and tornado sites may have provided habitat for ground nesters.

Species compositions differed among the three treatments. Clearcuts primarily had early successional birds, undisturbed forest had late successional species and tornado sites had both types of species. The residual overstory vegetation and dense understory vegetation found in tornado sites may have allowed both early and late successional species to coexist. Species-habitat association models and species distributions suggest that birds were selecting specific habitat characteristics, such as overstory vegetation, rather than simply late or early successional habitats. Late successional species may have also occurred in tornado-disturbed sites because of other factors, such as site fidelity.

Forest managers may be able to mimic natural disturbance more closely by managing overstory vegetation and coarse woody debris during timber harvests. By managing the overstory vegetation, forest managers can control most other vegetation characteristics and the associated bird community. Future research is needed to determine if the different bird communities in the two disturbed sites will become similar as forest conditions redevelop post-disturbance.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS