Graduation Year


Document Type




Degree Name

Doctor of Philosophy (Ph.D.)



Degree Granting Department


Major Professor

Glen E. Woolfenden, Ph.D.

Committee Member

John W. Fitzpatrick, Ph.D.

Committee Member

John M. Lawrence, Ph.D.

Committee Member

Earl D. McCoy, Ph.D.

Committee Member

Henry R. Mushinsky, Ph.D.


Vertebrate populations rarely are homogeneous. Instead they exist as arrays of demographically variable subpopulations that are linked through dispersal. Variation in habitat quality is a major cause of demographic variability among subpopulations, but few studies have examined habitat-specific demography empirically. Here I present a detailed field study o f habitat-specific demography in Blue Jays (Cyanocitta cristata) that occupied a habitat mosaic landscape in south-central Florida. By measuring habitat-specific demographic parameters, including density, reproductive output, survival, and dispersal, I tested for the existence of population sources and sinks associated with different habitats, and the applicability ofa series ofhabitat selection models using the observed habitat occupancy distribution.

Because estimates of density were integral to this study, I evaluated the efficacy of three common density-estimating techniques by comparing derived estimates with actual density of Blue Jays across a range ofhabitats. The 30-m fixed-radius technique performed better than the two other techniques and was used to estimate density for the remaining analyses.

Variation in Blue Jay demography across habitats gave rise to population sources and sinks (Pulliam 1988), providing the first empirical support for such population dynamics in a generalist bird species using several habitats at the scale ofa local landscape. However, the resulting habitat occupancy distribution was inconsistent with both the equal-competitors interference ideal free distribution (Sutherland 1983) and the phenotype-limited interference ideal free distribution (Parker and Sutherland 1986) habitat selection models. The observed habitat occupancy distribution ofBlue Jays may be consistent with the ideal despotic distribution (Fretwell and Lucas 1970), depending on the validity ofcertain predictions ofthat model. Ifso, this study may be the first to correctly identify an ideal despotic distribution in birds based on fitness (i.e., λ). I suggest that other empirical tests ofthese models have drawn incorrect conclusions about habitat occupancy distributions, either because of misinterpretation of the models, or because insufficient demographic data were evaluated.

The catholic use of habitats by Blue Jays, combined with patterns of breeding and dispersal in relation to age, suggest that habitat-specific demography may be sufficient to account for the relatively atypical social organization in this New World jay.

Included in

Biology Commons