Data from: State-space modeling reveals habitat perception of a small terrestrial mammal in a fragmented landscape
Gardiner, Riana et al. (2019), Data from: State-space modeling reveals habitat perception of a small terrestrial mammal in a fragmented landscape, Dryad, Dataset, https://doi.org/10.5061/dryad.kk2sd41
1.Habitat loss is a major cause of species loss, and is expected to increase. Loss of habitat is often associated with fragmentation of remaining habitat. Whether species can persist in fragmented landcsapes may depend on their movement behaviour, which determines their capability to respond flexibility to changes in habitat structure and spatial distribution of patches. 2.Movement is frequently generalised to describe a total area used, or segmented to highlight resource use, often overlooking finer-scale individual behaviours. We applied hidden Markov models (HMM) to movement data from 26 eastern bettongs (Bettongia gaimardi) in fragmented landscapes. HMMs are able to identify distinct behaviour states associated with different movement patterns, and discover how these behaviours are associated with habitat features. 3.Three distinct behaviour states were identified and interepretated as denning, foraging and fast-travelling. The probability of occurrence of each state, and of transitions between them, were predicted by variation in tree-canopy cover and understorey vegetation density. Denning was associated with woodland with low canopy cover but high vegetation density, foraging with high canopy cover but low vegetation density, and fast-travelling with low canopy cover and low vegetation density. 4.Bettongs did move outside woodland patches, often fast-travelling through pasture and using smaller stands of trees as stepping-stones between neighbouring patches. Males were more likely to fast-travel and venture outside woodland patches, while females concentrated their movement within woodland patches.5. Our work demonstrates the value of using animal movement to understand how animals respond to variation in habitat structure, including fragmentation. Analysis using HMMs was able to characterize distinct habitat types needed for foraging and denning, and identify landscape features that facilitate movement between patches. Future work should extend the use of individual movement analyses to guide management of fragmented habitat in ways that support persistence of species potentially threatened by habitat loss.
Tasmanian Midlands bioregion