SEARCH

SEARCH BY CITATION

Keywords:

  • Spatial distribution;
  • spatial modelling;
  • terrestrial vertebrates;
  • environmental hierarchy spatial scale;
  • potential distributions;
  • Leadbeater’s Possum;
  • south-eastern Australia

Aim

A hierarchical framework is presented for modelling the spatial distribution of terrestrial vertebrate animals.

Location

The location of the study is the montane ash forests of the Central Highlands of Victoria, south-eastern Australia.

Methods

The framework is illustrated using as a case study the distribution of Leadbeater’s Possum [Gymnobelideus leadbeateri McCoy, 1867, (Marsupialia: Petauridae)], a small arboreal marsupial. The framework is based upon quantifying the environmental response of a species in terms of a five-level environmental hierarchy defined by scales (global-, meso-, topo-, micro- and nano-scales) that represent natural breaks in the distribution and availability of the primary environmental resources. Animal response is examined in terms of a species’ distribution as observed in four biological units (the species in toto, meta-population/population, group/colony, and individual organism). We define the spatial occurrence and abundance of the target species in each of these units as its ‘distributional behaviour’.

Results

Predictions of the potential spatial distribution of Leadbeater’s Possum are presented at meso-, topo-, micro- and nano-scales. These spatial predictions utilize Geographical Information System (GIS)-based spatial models of long term mean monthly climate and terrain-modified surface radiation, together with vegetation cover and individual tree attributes from air-photo interpretation and field survey.

Main conclusions

Ideally, species’ responses at each level in the environmental hierarchy should be empirically derived using statistical models based on field observation of a species’ distribution and abundance. Spatial modelling of species’ responses becomes problematic at finer scales because of the lack of suitable environmental data. The key characteristics of the modelling framework are generic, but the influence of additional scales and processes will be important in other ecosystems and species.