Development and field validation of a regional, management-scale habitat model: A koala Phascolarctos cinereus case study
Law, B, Caccamo, G, Roe, P, Truskinger, A, Brassil, T, Gonaslves, L, McConville, A & Stanton, M 2017, Ecology and Evolution, vol. 7, no. 1, pp. 7475-7489.
Predictive habitat suitability modelling based on environmental predictor variables can accurately identify areas of koala occupancy. Using the koala as a case study, the value of regional-level distribution modelling as a tool for conservation and land management in relation to threatened, rare or cryptic species is demonstrated.
The predictive koala habitat suitability model was developed using the Maximum Entropy Approach (MaxEnt) which models species distribution based on presence records. The model was informed by presence records of koalas, corrected for detectability and sampling bias, and spatial data for 14 biotic and abiotic predictor variables. At the intermediate pixel size of 250m, each variable was represented by a grid overlaying the study region with every pixel assigned a habitat suitability score for that variable. The layers were overlaid to produce an overall score for each pixel. Actual koala occupancy was then assessed using acoustic monitoring of koala bellows and quantification of browse tree availability to assess the validity of the MaxEnt model. Overall the models agreed, with areas exhibiting strong koala presence assigned high habitat suitability in the predictive model and areas where koalas were absent assigned low habitat suitability scores. The four most important drivers of habitat suitability in the MaxEnt model were wildfire frequency, soil type, floristic composition and elevation. Areas of lowest habitat suitability for the koala were those with a history of frequent wildfire, even if all other habitat suitability variable scores were maximised. Tenosol or Rudosol soil types were also indicators of low habitat suitability. High habitat suitability was positively correlated with number of primary browse species and low to mid elevations. Importantly, the MaxEnt model was far superior for predicting koala occupancy than any individual variable, highlighting the importance of considering how environmental variables interact to shape habitat quality.
For species that are rare or cryptic, spatial models that reliably predict the species’ distribution over a region are important tools for conservation and land management. Although useful for particular purposes, distribution maps that cover large areas but are coarse in resolution are not suitable for informing targeted actions. Conversely, fine-resolution maps may be suitable for local-scale management but do not illustrate the broader context of those actions. An intermediate, regional model such as that presented here is highly suitable for land managers in the assessment of habitat quality in areas characterised by a mosaic of habitat qualities or land uses.
There are two main implications of the outcomes of this study. The first is that acoustic monitoring in conjunction with occupancy modelling is a reliable and cost-effective method for quantifying koala presence. In this study, the authors detected greater numbers of koalas in northern New South Wales than had been previously recorded and suggest that the status of the koala in this area should be re-evaluated. The second implication is that spatial planning and prioritisation strategies can support the most efficient allocation of management resources towards conservation initiatives, especially in regions characterised by a variety of habitats, land uses and stakeholder interests.
Summarised by Joanna Horsfall
Disclaimer: The summary of this report is provided for reference purposes only and does not represent the findings or opinions contained in the original report. Although every effort has been made to bring forward the main elements of the report, this review is no substitute for the full the report itself. Should you have any concerns or perceive any errors please contact us and we shall endeavour to rectify and improve the review.