| Abstract
|
:
|
The Habitat Suitability Index (HSI) model for fox squirrels was evaluated in Oklahoma during 1994-95. Although the HSI model succeeded in characterizing the overall quality of habitat on the study site as good to optimal for fox squirrels, it failed to discriminate within habitat differences. Characteristics of the study site and squirrel population influenced validation efforts. Lack of poor quality habitat reduced the HSI model's ability to discriminate among sample plots. Factors affecting reliability of HSI model calibration and validation, in general, such as animal distribution and abundance, spatial and temporal application of models, and research objectives, are discussed. Specifically, this test of the fox squirrel HSI model was explored in the context of squirrel population dynamics and model resolution in 1 cover type. Squirrel removals in 1989-90 could have influenced the relationship between HSI values and presence of squirrels in 1994-95. The fox squirrel HSI model may be more capable of distinguishing differences between habitat types, rather than within a habitat type. Augmenting the HSI model with local variables, such as diameter at breast height of tree species, improved the model's ability to discriminate within a habitat type. Density, weight, age and sex ratios, and home range estimates of the fox squirrel population were determined from trapping and radio-telemetry data. The squirrels had relatively low weights, supporting previous findings concerning a size cline in western populations and low summer weights in populations inhabiting hot, humid climates. Sex ratios favored males in most age classes, possibly indicating male dispersal during trapping periods. Home range HSI values were significantly (P < 0.001) lower than woodland sample plot HSI values. This result supported the conclusion that the HSI model failed to discriminate differences within a habitat.
|