A multiscale analysis of breeding bird species in north-central Tennessee

The major research question of how breeding bird/land cover associations vary with spatial scale, was addressed herein, through a combination of techniques and theories from many different disciplines. The scope of which covered such areas as Geography, Ecology, Digital Image Processing, and Geographic Information System technology.; For four consecutive breeding seasons, I recorded breeding-bird species presence data from randomly selected, geographically referenced roadside point-count locations that were separated by at least 0.6 km. A total of 2053 sites were censused for three minutes.; These data were transcribed into a relational database and Geographic Information System (GIS). The breeding birds species were separated into four categories: all species, resident species (RS), short-distance migrant species (SD), and Neotropical migrant species (NT). The land-cover was processed from a Landsat TM scene imaged August 30, 1993 and was classified into: forest, field/pasture, hardsurfaces (urban), water, edge, grass/agriculture, and shrubby. Both distribution mapping and multivariate statistics were used to analyze patterns of species distributions and occurrences as compared to the variation in distributions and proportions of the different land-cover types at four spatial scales.; Urbanized areas were comparatively species poor as compared to more rural areas containing native vegetation. RS breeding birds were found not to associate strongly with the land-cover categories. Field and pasture areas contained the highest relative proportion of SD breeding birds when these species were compared to the other land-cover categories. Forested land cover was found to be positively correlated with NT breeding birds and urbanized landscapes were negatively correlated. Furthermore, the correlations between NT breeding birds and land cover were found to occur at different spatial scales.; Forested land cover appears to be an important factor to maintaining breeding NT diversity within the study area. Urbanized landscapes appear to be relatively unfavorable to NT species as well as breeding bird richness. Furthermore, these conditions were found to be independent of spatial scale.