Effects of the urban matrix on avian assemblages in forest fragments

Ryan R Smith, Purdue University

Abstract

Habitat loss due to anthropogenic development is one of the greatest threats to birds, particularly in urban and suburban areas. While it is clear that avian assemblages are adversely affected by unsuitable habitat in such situations, there is less known about which aspects of this type of development are the important ones affecting birds. In order to explore anthropogenic impacts on birds influenced by cities, I monitored bird diversity at 25 woodlot study sites in and around Fort Wayne, Indiana, over the course of a breeding season. Point counts were conducted in each site and visited 3 times to obtain a species richness list. Study sites were characterized by area, the area to perimeter ratio, percent canopy cover, and mid- and overstory vegetation. The identity, number, and size of the vegetation was measured. The landscape matrix around the sites was characterized by distance to woodlots larger than the site itself, and distance to rivers, lakes/ponds, and wetlands. To obtain measurements of the urban matrix, the densities of habitable structures, distances of roads, and volume of vehicle traffic were measured in 5 concentric buffers ranging from 100 m to 3 km distance from the site. After removing collinear factors with variance inflation factors > 5, within each buffer size multiple linear regressions were chosen on the basis of best subsets analyses with all significant predictors. The models were ranked with Akaike Information Criterion with correction. Log transformed site area appeared in every model and had the largest positive effect on woodland bird richness. Greater mean area of an individual stem and distance to rivers were also positive predictors of richness in the 300 m buffer model. The densities roads, buildings, and vehicles adversely affected richness. Roads adversely affected richness in the 100 m buffer model. The effect of buildings extended to 300 m and had the largest negative impact, while vehicles appeared in the 300 m and 3000 m buffer models. The effect of vehicles, likely an indication of the degree of site isolation, was present out to 3 km demonstrating that urban and suburban development influences deciduous woodlot birds in a large radius around the site. The results from my study reinforce the findings from other researchers that site area is a very important factor for predicting bird diversity. Consequently, maintaining suitable habitat in patches of substantial size is the best way to aid birds. While of less importance but still of significance, forests with larger trees, associated with a forest’s age, should be maintained. The ramification of this is that forests must be left alone without removal of the larger trees in order to best provide for birds. Building development should be kept to at least 300 m from these patches as a greater density of habitable buildings nearer than this distance significantly caused a decrease in the bird richness. With these things in mind, site preservation for birds should focus on large mature sites surrounded by the least amount of urban development.

Degree

M.S.

Advisors

Kingsbury, Purdue University.

Subject Area

Biology|Ecology|Urban planning

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS