Dynamic versus static occupancy: How stable are habitat associations through a breeding season?

Abstract

Most studies of habitat use by animals assume that there is little movement by individuals once they have settled. This assumption of static occupancy is especially true in studies of birds, even though many bird species are known to abandon failed nest sites in search of better habitat or to move up a habitat gradient as the season proceeds. If birds move into different habitats as the season progresses, studies assuming static use of habitat may provide misleading or incomplete inference into habitat use. We tested the hypothesis that birds use different habitats as the breeding season progresses by analyzing point counts conducted early and late in the breeding season within Tuskegee National Forest, AL. For 15 species of conservation concern, we compared models that assumed static occupancy with models that estimated apparent movement between early and late‐season surveys. Models that estimated movement outcompeted static models for every species. Furthermore, patterns of movement provided inference that would not have been gained through a static modeling approach, with species moving into habitats traditionally known to be of high quality, and moving out of those assumed to be of poor quality. Our results suggest that studies of habitat use would benefit from estimating movement within a season.

Distinguishing distribution dynamics from temporary emigration using dynamic occupancy models

“…Nevertheless, numerous studies have interpreted these dynamic rates as evidence of within‐season dispersal (e.g. Betts et al., ; McClure & Hill, ; Otto & Roloff, ; Rota et al., ). Indeed, Rota et al.…”

“…Because multiple studies (e.g. McClure & Hill, ; Rota et al., ) compare static and dynamic occupancy models to test for evidence of dynamics, we compared model pairs with a common approach (Dail & Madsen, ; Rota et al., ; Self & Liang, ). Static occupancy models are special cases of their dynamic counterparts where dynamic parameters are constrained to be zero.…”

“…Multiple researchers have compared static and dynamic occupancy models fit to data collected analogously, and concluded there was evidence for within‐breeding season territorial changes in bird communities (e.g. Betts et al., ; McClure & Hill, ; Rota et al., ). Analysing this dataset allowed us to replicate such results.…”

Microclimate predicts within‐season distribution dynamics of montane forest birds

“…We found that most species of forest birds examined exhibited apparent within‐season distribution dynamics. Our results add to growing evidence that within‐season site occupancy may be less static than traditionally assumed (Betts et al ., ; McClure & Hill, ). For example, McClure & Hill () also found dynamic occupancy models outperformed static occupancy models in a south‐eastern U.S. forest bird community suggesting birds were shifting sites within a breeding season.…”

“…Our results add to growing evidence that within‐season site occupancy may be less static than traditionally assumed (Betts et al ., ; McClure & Hill, ). For example, McClure & Hill () also found dynamic occupancy models outperformed static occupancy models in a south‐eastern U.S. forest bird community suggesting birds were shifting sites within a breeding season. Radio tracking (Klemp, ; Gow & Stutchbury, ), territory mapping (Brambilla & Rubolini, ) and mark–recapture studies (Gilroy et al ., ) have also demonstrated within‐season movements and site shifts in birds.…”