Distribution models for koalas in South Australia using citizen science-collected data

Area of Science:

• Ecology
• Conservation Biology
• Citizen Science

Background:

• Koalas (Phascolarctos cinereus) are a threatened species in Australia, facing challenges from habitat loss, climate change, and disease.
• Accurate distribution and population data are crucial for koala conservation but are often difficult to obtain due to sampling challenges.

Purpose of the Study:

• To develop and apply statistical models to predict koala habitat suitability in South Australia using citizen science data.
• To estimate koala population size in key regions, specifically the Adelaide-Mount Lofty Ranges.

Main Methods:

• Utilized data from the Great Koala Count, a citizen science project, to build generalized linear mixed-effects models.
• Incorporated environmental predictors (vegetation, water, elevation, temperature) and human disturbance factors (roads).
• Employed random pseudo-absences and distance offsets to address biases inherent in citizen-collected data.

Main Results:

• Models identified temperature range, elevation, and rainfall as key predictors of koala occurrence.
• High habitat suitability was predicted for Kangaroo Island, Mount Lofty Ranges, and parts of the Eyre, Yorke, and Fleurieu Peninsulas.
• Estimated koala population in the Adelaide-Mount Lofty Ranges at 113,704 individuals (95% CI: 27,685–199,723).

Conclusions:

• Citizen science data, when rigorously analyzed, is a powerful tool for species distribution modeling and conservation planning.
• The study provides valuable, previously unavailable data on koala habitat suitability and population dynamics in South Australia.
• Future citizen science initiatives could be enhanced by smartphone app integration to better record search effort.