Species distribution and habitat attributes guide translocation planning of a threatened short-range endemic plant

The success of plant translocations depends on defining habitat attributes critical for establishment and survival, and on locating this habitat in the landscape. We used species distribution modelling, coupled with fine-scale characterisation of local habitat attributes to characterise and identify potential sites for translocation of Tetratheca erubescens J.P.Bull (Elaeocarpaceae), a threatened shrub restricted to cliff faces on one banded ironstone range in semi-arid south-west Western Australia. Here, we 1) constructed a maximum entropy species distribution model (SDM) from known occurrence locations and environmental data, and projected this onto a broader surrounding landscape to seek suitable unoccupied habitat; 2) characterised the local habitat attributes (LHA) of both occupied and predicted but unoccupied habitat (locality types), determining potential characteristics of habitat suitability at the scale of individual plants (microsites) not captured by the landscape-scale models and; 3) assessed unoccupied areas of modelled high habitat suitability to identify ‘potential translocation sites’ that may have the characteristics to support T. erubescens. The SDM resolved two management considerations by identifying suitable habitat outside the known extent of threatened species to locate areas to search for additional natural populations (none found) or to verify as ‘potential translocation sites’ (24 sites identified). Using field surveys to assess LHA at a finer-scale, we detected differences between plant microsites and other random points in occupied habitat. As such, T. erubescens was more likely to grow in wider rock cracks, located in relatively water gaining points in the cliff profile, in soils with organic (rather than mineral) content. Our approach systematically addressed the complex challenges involved in identifying and selecting sites for translocation of a threatened plant species, a first for a semi-arid environment. We used the outcomes of this study to present a conceptual model for practitioners and regulators that outlines our approach for identifying ‘potential translocation sites’ and ranking their suitability within the context of three assessment filters: species requirements; management-operational constraints and regulatory considerations. This application is relevant to translocation programs that seek to return plant species into natural habitat areas.