BRANTA — Richard A. Stanton, Jr.
Habitat selection of Brown-headed Nuthatches at multiple spatial scales
Institution: University of Missouri-Columbia, US
Supervisors: Dylan C. Kesler and Frank R. Thompson III
Details: MSc 2013 (Completed)
Subject Keywords: Range extension, habitat restoration, prescribed fire, optimal foraging, resource selection, Sitta pusilla, savannah, woodland
Species Keywords: Brown-headed Nuthatch, Sitta pusilla
Resources shape the movements and space use of birds. In turn, birds differ in their relative fitness, in part as a consequence of movement and space use decisions. The saga of each individual plays out across time and space, generating the dynamic pattern known as a species' geographic range. Then, changes in geographic range dimensions alter the selective environment encountered by individuals, potentially driving evolutionary change in movement modes. Thus, understanding resource selection requires knowledge of both individual behavior and landscape patterns of patch occupancy.
This thesis describes resource selection from both perspectives. We conducted two concurrent studies in a cooperatively breeding bird (Brown-headed Nuthatches Sitta pusilla). The first study documented space use within individual home ranges while the second documented patch occupancy across a heterogeneous range extension front. The range extension front followed prescribed fire and stand thinning treatments intended to restore pine forests to savannah-woodland condition. The individuals we tracked were in sites where restoration efforts had already been deemed successful.
Together these studies enabled us estimate the relative value of several resources to nuthatches and test competing hypotheses regarding the range extension process in resident birds. These studies were also developed in order to characterize the extent of suitable habitat in Missouri where the species was extirpated, and to predict changes in the extent of habitat that may occur in response to climate change and habitat restoration efforts.
We radio-tracked 22 Brown-headed Nuthatches, and related their space use and home range size to available resources within their respective home ranges using linear mixed models and lognormal regression. Nuthatch home ranges typically had two centers of activity, and areas of high use were associated with recently-killed (fresh) snags, recent prescribed fire, pine dominance, and grassy herbaceous cover in descending order of importance based upon their respective model-averaged standardized coefficients. For example, fresh snags appeared to be 1.7 times more important to Brown-headed Nuthatches than grassy herbaceous cover. Total stocking had a weak negative association with nuthatch use: 47% as important as fresh snags. These associations controlled for the influence of the nest site as a central place; thus the intensity of use associated with these resources reflected their perceived value to nesting nuthatches. Home range size was negatively related to the amount of pine, used for foraging, and snags, suitable for nesting. Apparent cooperative breeding was ubiquitous (ÔëÑ82%). Suitable nest sites may have been limiting, explaining both the incidence of cooperative breeding and part of the variation in nuthatch home range size. Grassy herbaceous cover was associated with larger home ranges, which was surprising since it has been associated with high quality foraging habitat in ecologically similar Red-cockaded Woodpeckers. We found no evidence that growing season prescribed burns had an impact on resource selection.
We also modeled patch occupancy across a range extension front created by habitat restoration. We ran repeated call-response surveys at 284 sites, and counted all nuthatches detected. Model results indicated that patch-level measures of recently-killed snags suitable for foraging and soft, 'punky- snags suitable for nesting were strongly associated with probability of nuthatch occupancy (188% and 115% increase in predicted ╬¿ from minimum to maximum observed values of fresh and punky snag density with other covariates fixed at their respective mean values, respectively). Tree stocking percent and years since fire had a negative relationship with nuthatch occupancy (99% and 27% predicted decrease in╬¿ from minimum to maximum observed values of tree stocking percent and percent grassy herbaceous cover with other covariates fixed at their respective mean values). Percent pine, grassy herbaceous cover, and percent of burned pine in the landscape within a 2 km radius had no apparent relationship with nuthatch occupancy. These results indicate that managing for more freshly-killed snags in pine savannah-woodland sites may be warranted for increasing nuthatch occupancy, while grassy herbaceous cover may have little value for nuthatches. In each case, further confirmatory study is needed to establish whether these associations have verifiable, biologically significant links to nuthatch survival and reproduction.
Importantly, model results also indicated that high quality habitat near the range limit was less likely to be occupied while nuthatch detections showed that marginal habitat was occupied. This pattern is contrary to the standard model of range extension and habitat selection; however, it is consistent with a range extension front that has been shaped by constrained dispersal. Dispersal can be highly variable within species and some range extensions might be driven by individuals at the extreme of the dispersal distance distribution. We documented a range extension front apparently driven by the behavior of average birds. This pattern may be quite common among birds given that many species are sedentary and that Allee effects are common in peripheral animal populations.
We emphasize the importance of further developing and testing a theoretical model that explicitly relates individual behavior to species distributions. Improvement and resolution of this model will be a critical component of achieving 'predictive ecology,- a desperately-needed means of managing global change.