Context Vehicle-strike has been identified as a key threatening process for koala (Phascolarctos cinereus) survival and persistence in Australia. Roads and traffic act as barriers to koala movement and can impact dispersal and metapopulation dynamics. Given the high cost of wildlife mitigation structures such as purpose-built fauna-specific underpasses or overpasses (eco-passages), road construction and management agencies are constantly seeking cost-effective strategies that facilitate safe passage for fauna across roads. Here we report on an array of detection methods trialled to verify use of retrofitted road infrastructure (existing water culverts or bridge underpasses) by individual koalas in fragmented urban landscapes in south-east Queensland. Aims The study examined whether the retrofitting of existing road structures at six sites facilitated safe passage for koalas across roads. Our primary objective was to record utilisation of retrofitted infrastructure at the level of the individual. Methods We used a combination of existing monitoring methods such as GPS/VHF collars, camera traps, sand plots, and RFID tags, along with a newly developed animal-borne wireless identification (WID) tag and datalogging system, specifically designed for this project, to realise the study aims. Key results We were able to verify 130 crossings by koalas involving a retrofitted structure or a road surface over a 30-month period by using correlated data from complementary methods. We noted that crossings were generally uncommon and mostly undertaken by only a subset of our tagged individuals at each site (21% overall). Conclusions An important element of this study was that crossing events could be accurately determined at the level of the individual. This allowed for detailed assessment of eco-passage usage, rather than the more usual approach of simply recording species’ presence. Implications This study underscores the value of identifying the constraints of each individual monitoring method in relation to site conditions. It also highlights the benefits of contingency planning to limit data loss (i.e. using more than one method to collect data). We suggest an approach that uses complementary monitoring methods has significant advantages for researchers, particularly with reference to improving understanding of whether eco-passages are meeting their prescribed conservation goals.

The ability to predict the frequency with which koalas (Phascolarctos cinereus) cross roads is an important step in developing mitigation strategies aimed at maintaining metapopulation viability for this species. In south-east Queensland, Australia, rapid urban development, including fragmentation and vehicle-related mortality, has contributed to a dramatic decline in local koala populations. Assessment of wildlife crossing structures that mitigates the impact of roads demonstrates these to be beneficial for many species. However, building enough structures to enable animals to move freely across impacted landscapes may be prohibitive in complex urban landscapes. The focus for this study was to consider the role of differing characteristics for species at risk. We examined the frequency of road crossings by individual koalas among six subpopulations in south-east Queensland. We explored the influence of potential predictor variables including: age, sex, and distance from roads. We established that road crossings were limited to a subset of individuals, with only 18 of 51 (~35%) koalas studied ever crossing a road. We found that koalas were disproportionately more likely to cross a road if they had initially been found within a distance of 100 m of that road. Males were more likely to cross roads than females, as were koalas aged less than five years. We suggest that successful wildlife road mitigation can be improved by better understanding the road-crossing behaviour of species such as the koala.

The use of animal‐borne devices in wildlife studies is increasing due to the miniaturization of sensors and reductions in component costs. Retrieval of animal‐borne devices is often advantageous for several reasons, including data retrieval and animal welfare considerations; however, commercially available drop‐off units are often prohibitively expensive and/or unreliable. Here, we present ‘OpenDropOff’, an open‐source, low‐cost electronic drop‐off unit for remotely detaching animal‐borne sensors from individuals. OpenDropOff devices function by using heat, produced by running an electrical current through resistance wire (nichrome), to melt nylon (e.g. fishing) line, detaching sensors from animals at or after a pre‐determined time. OpenDropOff can be constructed using commercially available components (costing < USD $15 in total; approximately 10% or less of many commercially available drop‐off units) and can be customized to fit project‐ and species‐specific needs. Fourteen OpenDropOff devices have been field deployed on koalas (Phascolarctos cinereus) (n = 11), swamp wallaby (Wallabia bicolor) (n = 1), Australian brush turkey (Alectura lathami) (n = 1) and Australian white ibis (Threskiornis molucca) (n = 1), with the only detachment failure occurring after a device was damaged during the koala’s unrelated death. OpenDropOff devices provide a wider range of researchers with a reliable, cost‐effective tool for detaching animal‐borne sensors. Their use could allow researchers to more efficiently allocate project resources, and their flexibility provides an opportunity to mitigate animal welfare concerns surrounding the use of animal‐borne devices.

Methods in Ecology and EvolutionThis article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as