Initial evaluation of a state-wide multi-method survey to estimate koala occupancy and abundance

• The efficacy of three survey methods were evaluated for their effectiveness for surveying koalas. • Acoustic recorders were most effective for occupancy whereas drones required increased effort and cost. • However, drones were superior at estimating abundance. • The combined use of acoustic recorders and drones is therefore recommended to enhance the precision of abundance estimates. Knowledge of species distribution and abundance is crucial for effective threatened species management and conservation but is rarely obtained from purpose-built large-scale surveys. The koala ( Phascolarctos cinereus ) is a nationally threatened species; however, its current distribution is poorly known, and uncertainty has surrounded the most suitable methods to address this at large spatial scales. To implement a state-wide koala survey in New South Wales (NSW), Australia, we sought to address this issue by evaluating and minimising trade-offs between the number of survey sites and site-specific survey effort using a multi-method survey approach. Here, we evaluate the comparative efficacy of three koala survey methods- diurnal transects, drones, and passive acoustic recorders (acoustic recorders) implemented across a wide range of habitats in NSW. We evaluated how these methods, individually and combined, influenced koala detection, occupancy, and relative abundance-related performance criteria to optimise the state-wide koala survey. Under the deployed survey effort, acoustic recorders performed optimally for assessing occupancy, achieving high detection of koalas and avoiding false absences within sites with high certainty (P* ≥ 0.99). To achieve similar performance, drone surveys and diurnal transects required survey effort (and cost) to be increased 19 and 206-fold, respectively. Due to considerable overlap in within-site detection between acoustic recorders and drones and the poor performance of diurnal transects, there is little advantage to using these methods concurrently to estimate koala occupancy. However, drone surveys were highly effective at estimating relative abundance of koalas compared to diurnal transects. Overall, the combined use of acoustic recorders and drones is highly advantageous because, where drones fail to detect koalas, acoustic recorder detections help eliminate false site absences, enhancing the precision of abundance estimates. Their combined use enables interoperability with other survey programs, fostering comparison and standardisation of important koala population measures. Ongoing optimising of koala survey and monitoring methods will improve the capacity to provide important insights for koala management and conservation.