SEARCH

SEARCH BY CITATION

Keywords:

  • Australia;
  • beta diversity;
  • biogeographic breaks;
  • bioregionalization;
  • phytogeography;
  • spatial turnover;
  • zoogeography

Abstract

Aim  We introduce a method to quantify shared breaks in aggregate biotic distributions and their relationships to geographic variables. The method is based on quantification of distributional taxic and abiotic data that can be applied over multiple spatial scales. We aim to show biogeographic breaks and varying transition zones at a fine level of detail (5-km resolution) and develop an approach to assess existing bioregionalization schemes.

Location  Global applicability, using an example from New South Wales in south-eastern Australia.

Methods  Moving window analyses, rotated in 15° increments through 360°, are used to assess the degree of anisotropic spatial turnover between sets of gridded cells containing georeferenced species observations. Patterns of biotic turnover are compared with equivalent analyses for elevation and lithology. Identified breaks are assessed against an existing bioregionalization scheme (Interim Biogeographic Regionalisation of Australia, IBRA).

Results  There was fine-scale concordance between turnover patterns and several IBRA bioregions. Breaks in turnover of flora and fauna corresponded with the boundaries of the Hunter Valley and Sydney Basin regions, particularly the boundary between the Brigalow Belt South and Sydney Basin. Low-turnover zones were quantified; prominent examples are the Sydney Cataract and Wyong bioregions. Turnover along many boundaries was gradational, confirming that mapped breaks are not abrupt. A previously unidentified break was identified in the South East Corner bioregion. Spatial turnover patterns were similar between biota and were reflected in mean correlation coefficients between turnover in each group: mammals–reptiles (= 0.70, P << 0.01); mammals–flora (= 0.56, P << 0.01); and reptiles–flora (= 0.51, P << 0.01). Generally, patterns of abiotic turnover reflected biotic turnover, although mean turnover correlations were weaker than between biota.

Main conclusions  Using this method we were able to characterize taxic breaks and overlaps in detail and at a spatially fine resolution. For our study region, we confirm the overall integrity of the IBRA framework, but suggest that it may benefit from revision in some respects.