Key messages

• Biodiversity is facing increasing pressure around the world due to climate change.
• An ecological modelling technique, Generalised Dissimilarity Modelling (GDM) provides insights into the change in species composition across geographic space and time.
• GDM has been used to compare changes in native vascular plant species composition between current and NARCliM future climate scenarios.
• Environmental predictors, such as climate, terrain and soil, that drive species distributions are transformed in the modelling process into measures of environmental dissimilarity and ecological distance.    
• By using the transformed predictors, researchers can make more accurate projections about how biological communities will respond to climate change.
• Transformed predictors can be used independently of the GDM model for spatial biodiversity analysis, such as determining the representativeness of the Protected Areas network, the expected species persistence, gaps in survey analysis, or uniqueness of species composition to the landscape of interest or to a focal point. 

Context

The effects of climate change are placing increased pressure on biodiversity across the globe. Climate-driven effects on biodiversity are expected to be variable, with some local gains, but the broad outlook for NSW is significant loss. Australian studies indicate climate change poses a level of risk to biodiversity comparable to losses resulting from land-use changes since European settlement¹. 

Ecological modelling provides insights into where and how biodiversity may be impacted by climate change and helps identify adaptation opportunities to minimise biodiversity loss. By developing multiple models across a range of plausible future climate scenarios, researchers can explore and develop a range of optimum adaptation options. 

This dataset is for technical users who have an understanding of Generalised Dissimilarity Modelling (GDM). 

The dataset is a set of spatial environmental predictors (climate, terrain, soil) for current (1990-2009) and future climates as projected by NARCliM 1.0 and 2.0 scenarios. The predictors are transformed during model fitting to account for how each environmental factor drives native vascular plant species compositional turnover across NSW. These GDM transformed variables have been used for a range of other data products available in the hub and are provided to modellers and researchers to use for new climate adaptation applications.
 

Key findings

In this study, Generalised Dissimilarity Modelling (GDM) was applied to measure species composition of native vascular plant species across NSW. The species data is compiled from comprehensive floristic surveys, allowing the model to capture community-level turnover in composition across the landscape. 

Biological differences between site-pairs are modelled using a dissimilarity metric, while environmental predictors are transformed to represent ecological distance.  

The transformed environmental layers produced by GDM can be used independently of the model itself.

This makes them a valuable resource for spatial biodiversity analysis, such as mapping ecological patterns or identifying areas of high conservation value.

A common use for the GDM transformed layers in climate adaptation work is to produce dissimilarity grids between the current climate and future climate scenarios. 

Uses and initial findings

Maps generated through GDM can identify areas of past, current or future change. Figure 1. is an example. It shows the projected dissimilarity between the current climate (1990-2009) and 2070. 

Darker shaded areas are more likely to remain climatically suitable for existing species with less change in species composition expected (i.e. lower turnover) whereas lighter shaded areas are expected to become less climatically suitable for existing species and experience greater change in composition (i.e. higher turnover).

The maps show that Central Western NSW in general, is projected to experience larger changes in species composition, making this an area of vulnerability and risk for existing biodiversity. This area comprises vegetation formations such as Dry Sclerophyll Forests (shrubby), Semi-arid Woodlands (shrubby) and Grassy Woodlands. 

Dissimilarity grids are provided as part of the BIAP package: Data packages for the Biodiversity Impacts and Adaptation Project | Future Compositional Dissimilarity Grids | SEED 

Adaptation strategies and applications of data

The use of GDM-transformed predictors can be used to create maps or predictions about how a species' habitat is likely to change in the future providing insights into conservation or restoration efforts and planning (see Figure 1).

These GDM-transformed predictors underpin a range of data products available in the Hub. These products include dissimilarity grids, biodiversity metrics, conservation and restoration benefit layers, and adaptation land management scenarios. 

Biodiversity metrics, calculated at regular intervals, provide insights to changes in biodiversity over time and can be used in a policy setting for tracking adaptation outcomes. 

Land management scenarios offer strategic guidance for conservation or restoration efforts. They can be used by State government departments, Local Land Services, Councils, NGOs, and restoration groups to identify and implement actions that optimise biodiversity outcomes under current and future climate conditions. 

Questions that can be investigated using these transformed data and applied to adaptation include:

• Representation of biodiversity in protected areas
  Is the current Protected Area network in NSW representative of species diversity under both present and projected future climates? If not, where should additional conservation efforts be focused to improve representation?
• Species persistence under habitat loss
  Where are species likely to persist over the long term given historical habitat loss? Persistence is defined as the proportion of species originally occurring in a location that are expected to remain. These areas could be regarded as “low-hanging fruit” for conservation prioritisation.
• Ecological uniqueness of locations
  What is the uniqueness of each location within a region? Preservation of unique communities contributes to overall biodiversity. Where are these unique areas located, and what strategies can be employed to preserve them?  

Further information

For application of GDM transform grids derived using NARCliM 1.0 data refer to the Biodiversity and Impacts and Adaptation Project (BIAP). Data and documentation provided here:  Data packages for the Biodiversity Impacts and Adaptation Project | Dataset | SEED. 

Biodiversity Impacts and Adaptation Project / 2016 State of NSW and Office of Environment and Heritage

References

1. Drielsma MJ, Love J, Williams KJ, Manion G, Saremi H, Harwood T and Robb J, ‘Bridging the gap between climate science and regional-scale biodiversity conservation in south-eastern Australia’, Ecological Modelling, 2017, 360:343–362. https://doi.org/10.1016/j.ecolmodel.2017.06.022