Habitat Use by Cottontail Rabbits across Multiple Spatial Scales in Mississippi

Abstract

Relative importance of different land-cover types is often inferred from studies of habitat selection in the context of availability. From 1 February 1997 to 31 January 1999, we examined seasonal (breeding and nonbreeding) habitat selection of cottontail rabbits (Sylvilagus floridanus) in an agricultural landscape in east-central Mississippi at multiple spatial scales (second-, third-, and fourth-order habitat selection; Johnson 1980). We characterized available land-cover types as woody, rowcrop, or grass. The composition of seasonal rabbit home ranges (95% Adaptive Kernel [HR]) and core areas (50% Adaptive Kernel [CA]; second-order selection) exhibited selection for grass patches, followed by woody and rowcrop patches. However, genders differed across diel periods for the composition of HR. Males selected woody patches during diurnal hours, and females selected grassy patches, whereas they both selected grassy patches during the nocturnal hours. Within HR, allocation of time among habitats (third-order selection) differed between seasons and genders. Males selected grass patches followed by woody and rowcrop during the breeding season, and females used them in proportion to availability, whereas they both used them in proportion during the nonbreeding season. Genders used HR land-cover types in proportion to availability for both diel periods. Two predictive models of fourth-order selection poorly discriminated (56.9% and 60.7%) between used and random locations. Cottontails exhibited habitat selection at the HR (second-order) and males at the patch scale (third-order); however, cottontails randomly used within a patch (fourth-order). Our results support the characterization of cottontails as vegetative community generalists at the patch scale (fourth-order); however, cottontails select among land-cover types at larger spatial scales (second- and third-order).