Our modelling approach provides spatial information on the potential effects of climate change on key groups of organisms, such as plants or mammals, at a collective or ‘community’ level. This approach enables a ‘whole of biodiversity’ perspective in planning.
Species-based models are particularly useful for species of special concern. But the sheer number of species means we need a different approach to assess the implications of climate change for biodiversity, overall.
Our community-level modelling calculates ‘ecological similarity’, which then allows us to estimate the amount of change in species composition between two locations or points in time.
What is community-level modelling?
Community-level modelling combines data from multiple species in a single model. It produces information on spatial patterns in the distribution of biodiversity at a collective or community level.
We use a form of community-level modelling known as ‘generalised dissimilarity modelling’ (GDM). It analyses the relationship between many environmental variables (including climate) and the locations where species are observed to understand and project differences in species composition. Technical Note 2 in the Guide describes this approach in greater detail.
What is ecological similarity?
The models used in this guide estimate ‘ecological similarity’ – the similarity in predicted species composition between two locations or between one location at two points in time. Ecological similarity varies between 1 (all species are identical) and 0 (all species are different).
The measure of ecological similarity between a 1990 baseline and 2050 for all locations in Australia is the basis of most of the measures and maps in this Module.
Areas that will be less similar in terms of species composition in the future are likely to experience greater biodiversity change.
What data is used in projections of ecological similarity?
We developed projections of ecological similarity for four terrestrial biological groups:
- vascular plants (ferns, gymnosperms and angiosperms)
Together these groups highlight the potential contrasts in response to climate change by different groups of organisms.
To facilitate planning for multiple futures, we used two contrasting climate scenarios. We modelled two global climate models:
- a mild MIROC5
- a hot CanESM2
These were modelled to 2050 using the high emissions Representative Concentration Pathway (RCP) of 8.5. These two climate scenarios reflect a milder and a hotter likely future in the next few decades if emissions remain on their current trajectory. Further details on the climate scenarios can be found in Technical Note 3 in the Guide.
We used data on where native species are found collated from herbaria and museums across Australia by the Atlas of Living Australia. We also used best-available, high resolution, national environmental layers capturing spatial variation in the environment including soils, climate and landform.
The models use fine-grained data (250m resolution) so they are applicable to local and regional planning.