Semester of Degree Completion
Richard D. Andrews
Two discriminant models were derived from 40 variables measured in 12 white-tailed deer (Odocoileus virginianus) winter concentration areas and 12 non-concentration areas in east-central Illinois. The first model correctly classified 100% of these areas based on area of refuge, area of upland hardwoods with <50% crown closure, area of bottomland forest with <50% crown closure, distance of unimproved roads, and total topographic relief. This model was tested on 6 winter concentration areas in west-central Illinois and 6 winter concentration areas in northern Illinois. The first discriminant model correctly classified 91.7% of these areas.
The second model originated from the same set of variables, however the refuge area variable was removed in an attempt to classify winter concentration areas without knowledge of refuge areas. This model correctly classified 91.7% of sites in east-central Illinois, and 75% of the areas in west-central and northern Illinois.
Refuge accounted for nearly 59% of the explained variation between winter concentration areas and non-concentration areas. This component of winter habitat was found in all winter concentration areas examined.
These models offer land managers a statistical method of evaluating winter white-tailed deer habitat based on a low number of measurable variables. Winter habitat is presently adequate in Illinois. Changes in land use and/or harvest regulations may create a greater need to locate, preserve, or establish winter deer habitat.
Brewer, Paul A., "Winter Concentration Areas of White-tailed Deer (Odocoileus virginianus) in Illinois: A Discriminant Analysis" (1985). Masters Theses. 2780.