http://maps.usm.edu/migratorybirds/download.html

Model Output Data Text

This project was conducted using the most current publicly available data that provide approximately equivalent coverage across the entire study area, which is the land surrounding Lakes Michigan, Huron, Erie and Ontario and connecting waters. This includes parts of seven US states (Wisconsin, Illinois, Indiana, Michigan, Ohio, Pennsylvania and New York) and the Canadian province of Ontario, extending over 1000 km east to west and over 600 km north to south. Developing conforming data across the national boundary was a significant challenge. While we were able to develop GIS layers which were analogous for both sides of the international border, their ages, scales, formats, and resolutions were different.

As was done in similar studies previously, migratory birds were divided into three groups: landbirds (including raptors), shorebirds, and waterfowl. Landscape characteristics or attributes considered to be of importance to each of these bird groups were developed, with the understanding that these attributes needed to be framed in way that allowed them to be depicted and analyzed with currently available GIS and data. While this may have reduced the resolution or specificity of some of the attributes, the analysis involved remained complex

Once the attributes of importance to the bird groups were developed (e.g. distance from shore) they were assigned a value based on their meeting certain criteria. For each bird group up to 5 attributes were developed, each of which was assigned a score from 0 to 1. The scores for each attribute for each bird group were summed and a final attribute score ranging from 0 (low) to 5 (high) was produced. Thus it is possible to identify those areas best matching the profile of high value to migratory birds by determining which areas scored highest.

To better match the scale of the analyses to areas of habitat likely to be meaningful in the context of bird conservation, the entire project was carried out with 100m pixels using the Great Lakes Albers (NAD 1983) projection.