FUNCTIONAL RESPONSES IN HABITAT USE: AVAILABILITY INFLUENCES RELATIVE USE IN TRADE‐OFF SITUATIONS

Current methods for evaluating habitat selection from animal space‐use observations ignore possible interactions between time allocation patterns relative to different resources, their relative abundance, and their spatial arrangements. Habitat selection may occur in situations in which animals experience a trade‐off, e.g., between time used foraging in areas with abundant forage but poor protective cover, and time used for resting in areas with good protective cover but low forage abundance. We show how functional responses in habitat use (i.e., change in preference with availability of one of two main habitat types) may be tested. Given radio‐telemetry data for a sample of individuals, binomial logit models can be used to regress proportionate use of a habitat type p ( u ) against the proportion of that habitat available, p ( a ) . Given an appropriate fit to the data by a linear predictor on a logit scale, functional response will be indicated by a estimated slope parameter ≠ 1, while a slope = 0 will indicate a consistent use as availability changes. Habitat preference is inferred from the logit regression parameters when the fitted value of the proportion of use at a specified proportion of availability, is significantly greater than the proportional availability.