MaxEnt models (the only method reported here at the moment) could be generated for 79 of the 101 species included in the study. The overwhelming majority of species (71 species, or 89.9% of species modelled) showed a decrease in overall climate suitability by 2050, with the other 8 (10.1%) increasing in overall climate suitability. To see the full table of results, click here.
Some species showed surprising increases in climate suitability. For example, Mormopterus eleryi, a newly described free-tail bat (Reardon et al., 2008), showed a substantial increase in high-level climate suitability. However, an increase in climate suitability does not mean that the species will always improve its conservation status under the effects of climate change. In the case of M. eleryi, for example, the region in which it now occurs and within which it has a potential increase in overall climate suitability, covers some of the one of most heavily modified landscapes in Australia. Improved climate suitability may not translate into a more secure future for this species.
A similar constraint on the use of climate suitability predictions applies to species that exhibited a sharp decline in overall climate suitability. As an example, consider the result for Scoteanax rueppellii. This large-bodied vespertilionid is found just on or east of the Great Dividing Range in south-east Australia. It is predicted to see a substantial decline in climate suitability by 2050, but some areas of moderate suitability remain. Reduced climate suitability will probably place increased stress on populations of the species, and increase risks from other major threatening processes such as habitat loss and changed fire regimes.
The impact of climate change has not been given much consideration when assessing conservation prospects for microbats in the region. For example, in three major reviews of bat status and conservation within Australia, two do not mention climate change (Milne and Pavey 2011; Armstrong 2011), and one mentions climate change very briefly in the final paragraph (Pennay et al. 2011). Even though the climate suitability models presented on this website have some significant caveats on their interpretation as noted above, it may be useful to consider the implications for conservation status of the reviewed taxa when my results are applied.
As a first approximation to assessing the influence climate change information might have on the conservation status of Australian microchiroptera, I tabulated the conservation status of each species of microchiroptera considered in The Action Plan for Australian Bats (Duncan et al. 1999) which applied the then current IUCN evaluation process at a continental scale. Separate conservation rankings are available in state and territory jurisdictions within the Australian federation, but they necessarily take a regional perspective on conservation status. A scatterplot of the data is shown in the following figure.
The codes for conservation status are:
- DD: Data Deficient
- LR(lc): Lower Risk (least concern)
- LR(nt): Lower Risk (near threatened)
- VU: Vulnerable
- EN: Endangered
- CR: Critically Endangered
To evaluate the relationship statistically, I gave the IUCN conservation codes a numerical value in the order listed above from 1 (DD) to 6 (CR) and computed a Spearman rank correlation with percent change in overall climate suitability values between current and 2050 climates was almost significantly different from zero at the α = 0.05 level (rS = -0.255, n = 58, p = 0.054). This almost significant result could be due to the inclusion of the DD category which, in reality, is not an effective statement of conservation status but a declaration of "no idea". Excluding the DD points gives a non-significant result (rS = -0.09, n = 52, p = 0.525) indicating the relationship is indistinguishable from random association.
These results suggest that incorporating predictions of change in climate suitability may have a strong influence on future evaluations of conservation status for the microchiroptera of Australia, and should be routinely considered.