PARENT SESSION
Oral Session # 92: Biogeography IV: Dispersal.
Presiding: J Choo
Friday, August 8. 8:30 AM to 12:00 PM, SITCC Meeting Room 104.
Data-driven models of bluebird movement in and near corridors.
Bolker, Benjamin*,1, Levey, Douglas1, Tewksbury, Joshua1, Sargent, Sarah2, Okuyama, Toshinori1, Seavy, Nathaniel1, 1 University of Florida, Gainesville, FL2 Allegheny College, Meadville, PA
ABSTRACT- The function and design of habitat corridors in ecological landscapes to mitigate the effects of habitat fragmentation continues to excite interest. Previous results from a landscape-scale, replicated corridor experiment in South Carolina showed that open corridors of second growth in a matrix of forest facilitate colonization of connected habitat patches by bird-dispersed fruits (yaupon holly, Ilex vomitoria) relative to unconnected patches; they fail to show significant evidence for a "drift-fence" effect where corridors increase colonization of patches by intercepting animals moving through matrix habitat. We used follow data from individual Eastern Bluebirds (Sialia sialis) to parameterize a statistical model of bird movement in order to probe the mechanisms underlying the observed effects of corridors on seed dispersal. Our best-fit model included random movement within patches and corridors, a tendency for birds in the matrix to follow the nearest edge, and correlated directions of subsequent moves in the matrix. We simulated the movement of individuals, starting from the center of a patch and following for the typical gut passage time (45 minutes), to obtain average proportions of simulated seeds deposited in each patch type. The model suggests that our previous supposed mechanism for corridor functions -- that birds prefer to move within corridors -- is wrong. Instead, birds move within the matrix but use the corridor edge as a movement cue. This edge-based mechanism also suggests that drift-fence effects should occur, in contrast to previous results. In fact, our results provide evidence for a small drift-fence effect, and suggest that the failure to find it in the experimental results results from a lack of power.
Key words: Sialia sialis, movement, corridor, statistical model