A project undertaken at the University of Technology, Sydney, and supervised by John Gollan
In the Pilbara region of northwest Western Australia, large-scale mining threatens climate-sensitive plants and animals (Fig.1). The ‘Short-range endemic invertebrates’, which are generally confined to the isolated cool moist gullies and gorges, are just one example (Fig. 2). However, traditional survey work for these species is problematic, largely due to low detection probabilities. Species with low detection probabilities can lead to false negatives i.e. the conclusion that a species is absent, when they are really present.
Site selection to find climatically sensitive species is also done using a ‘best-guess’ survey approach, often seeking guidance from coarse grained climate models or remote sensing information (e.g. satellite imagery). Consequently, the final determination of sites to survey is qualitative, and thus prone to bias and error. All of these issues can lead to misdirected conservation actions and undesirable environmental outcomes. Less subjective approaches for mapping and identifying critical habitat are urgently needed, otherwise there is the very real chance that species will be lost before they are even discovered.
The solution requires appropriately scaled and biologically relevant climate maps. Work on the temperate east coast of Australia has demonstrated that topoclimate (climate models that account for local climate forcing factors) can improve our understanding of where species and their habitat occur. But whether this approach is transferable to a semi-arid region remains untested. Furthermore, whether topoclimate can increase detection probabilities against current ‘best-practice’ survey approaches is also unproven.
This project will deploy miniature temperature and humidity loggers (Fig.3) across the entire Pilbara so that:
- Detection probabilities of climatically sensitive species and their habitat are significantly increased, and
- Subjectivity in the site selection process is removed and allows patch attributes (e.g. total size, shape, connectivity) to be calculated.
Achieving these objectives will provide the important first steps towards becoming part of ‘best-practice’ methodology in biodiversity survey and assessment.