HABITAT CHARACTERIZATION OF SECONDARY CAVITY-NESTING BIRDS IN OKLAHOMA

We analyzed vegetation structure at potential and actual nest sites of secondary cavity-nesting birds in south-central Oklahoma. Habitats co nsisted of old fields with remnants of tallgrass prairie and patches o f post oak-blackjack oak Quercus stellata, Q. marilandica) woodland. D uring the 1989 and 1990 breeding seasons, 194 sites with nest boxes we re analyzed, including those occupied by Bewick's Wrens (Thryomanes be wickii), Carolina Chickadees (Parus carolinensis), Tufted Titmice (P. bicolor), Eastern Bluebirds (Sialia sialis), House Sparrows (Passer do mesticus), and some not used by birds. Principal-components analysis w as employed to describe general vegetational gradients and stepwise di scriminant analysis to delineate differences in nest-site habitats amo ng species. Through use of appropriate indices and Monte Carlo simulat ions, niche breadth and overlap were assessed relative to a habitat gr adient (principal component I) extending from open areas to woodlands. Chance expectations were established assuming that the nest boxes rep resented a limited resource, albeit one that was not fully utilized du ring the course of the study. Eastern Bluebirds and House Sparrows cho se nest boxes in open areas with few trees, Bewick's Wrens selected bo xes in wooded areas with junipers and few deciduous trees other than o aks, Carolina Chickadees most often were found in areas with junipers and oaks, and nest boxes used by Tufted Titmice were broadly distribut ed, not showing association with any particular habitat type. Niche ov erlap for Eastern Bluebirds and House Sparrows was more pronounced tha n expected by chance. These two species showed less overlap with Bewic k's Wrens, Carolina Chickadees, and Tufted Titmice than expected given simulation results. For the House Sparrow and Eastern Bluebird, which were restricted to open habitats, niche breadth was significantly les s than expected by chance. Likewise, niche breadth for the Bewick's Wr en, with the majority of its nests being in semiopen areas, was less t han predicted. For Carolina Chickadees and Tufted Titmice, nest-box us e relative to the habitat gradient represented by principal component I was not different from random expectations. Our findings indicate th at the introduced House Sparrow potentially can negatively influence n esting success of Eastern Bluebirds given that preferences for nest si tes of the two species correspond so closely. Direct observations of H ouse Sparrow and Eastern Bluebird interactions indicate that in some c ases bluebirds are detrimentally affected.