Context The koala is a threatened species in New South Wales and long-term datasets of koala rehabilitation provide a valuable source of insight into local threats. Aims To examine the long-term trends of the cause and fate of koalas admitted for rehabilitation to assist monitoring the recovery of koala populations and provide a new outlook on the limitations and strengths of rehabilitation records for koala conservation. Methods We used data from long-term records (1989–2020) of the Friends of the Koala wildlife rehabilitation group in north-east New South Wales to identify spatial and temporal trends in 5051 koala admissions. Key results Chlamydiosis was the most common cause of admission, affecting female koalas more often than males. Admissions of female koalas with chlamydiosis showed a two-fold increase over summer, and admissions for motor vehicle collisions and unsuitable environment increased in spring. The rescue locations of admissions show an increasing geographic spread over time, as well as an increasing density of records. Admission body scores were higher in koalas that were eventually released than in non-released koalas. Other main causes of admission were unsuitable environments, dog attacks, abandonment/orphaning and attacks from other species. Conclusions We have produced an analysis and interpretation of a long-term dataset of the relative importance of various threats facing the koala population in north-east New South Wales. However, admissions to rehabilitation do not capture all the landscape-scale problems confronting koalas locally, particularly climate change and habitat loss. Implications We have provided the baseline necessary to detect future changes in the causes for admission, rates of rehabilitation and post-release survival.

A challenge to understanding the impacts of roads on koalas is that their effects operate across multiple scales. To determine what conditions underlie koala roadkill, we looked at koala roadkill at two spatial scales – the entire state of NSW, and a local government area (Coffs Harbour) – for three road types (primary, arterial and local). We also subdivided the state data into three regions – coast, tableland and inland – to look at regional differences. Our analyses show that koala roadkill is ubiquitous across the distribution of koalas. Similar numbers of roadkilled koalas were recorded on each road type, but because of the vastly different total length of roads in the three types (local roads 122 755 km; arterial 79 706 km; primary 12 972 km) it is clear that an individual koala is more likely to be killed on a primary road than the other two road types. The pattern shown at the LGA scale is similar to that seen across the state. Habitat type adjacent to a road appears to have little influence on the likelihood of a koala being killed on the road. Mitigation measures can reduce koala roadkill, but there is a need to correctly assess the relative importance of koala roadkill compared with other factors that impact on koala populations when allocating resources to koala conservation. We conclude that roadkill is an ever-present threat which will remove an ever-increasing proportion of decreasing koala populations in NSW.

This 6-year study aimed to monitor a local koala population and koala protection/mitigation measures before, during and after a major upgrade to the Pacific Highway at Lindsay’s cutting near Coffs Harbour in northern coastal New South Wales. This represents the scale of the individual koala interacting with traffic on a primary road within a well-studied Local Government Area for koalas. We also collected records of koala road injury and death from local wildlife carers. The principal findings were that, before the upgrade or any mitigation provisions, koalas were regularly killed on this 1 km stretch of highway flanked by primary koala habitat (26 records 1991–97). Koala exclusion fences largely eliminated the roadkill here, but the benefits were highly localised. Koalas made occasional use of a wildlife underpass below the highway. Two problems were identified: a farm gate left open in the fence allowed koalas onto the highway, and koalas could walk around the fence ends onto the highway. On the positive side, we have shown the value of local-scale research and mitigation, such as koala exclusion fences, and the value of examining the koala population of the surrounding area to assess the effectiveness of this mitigation at a larger scale.

In the 1990s, the Pilliga forests were carrying the largest population of koalas west of the Great Dividing Range in New South Wales (NSW). Whereas the NSW koala population in its entirety was thought to be in decline, the Pilliga population stood out as potentially increasing. By 2007, anecdotal evidence suggested that the population was in decline. We undertook surveys of koalas in the Pilliga forests that repeated surveys undertaken between 1991 and 2011. We found that koalas had declined and were found in only 21% of sites in which they were observed in the initial surveys – by any measure, a 5-fold drop in occupancy in less than two decades is severe. Declines occurred evenly across the Pilliga, with persistence at a site seemingly related to a high initial density of koalas rather than to a slower rate of decline. Sites where koalas persisted were characterised as having higher temperatures and lower rainfall relative to other sites, being close to drainage lines with deeper soils and having a lower occurrence of fire. This pattern fits with the observation in the recent surveys that koalas were next to drainage lines in the western half of the Pilliga and fits with the suggestion that koalas show refugial persistence. Recovery from this point is not assured and will depend on how we manage the landscape, particularly with the threat of climate change. This will likely require active management within an adaptive management framework, such as restoration of refuges, and not simply habitat reservation.

Koala rescue from the bushfires over the extended 2019–2020 fire season in NSW was a high-profile media topic over many months. We sought to determine how many fire-affected koalas were rescued and admitted to rehabilitation in NSW, what was the fate of these koalas, what parts of the state were involved, and how many other koalas entered rehabilitation at the same time? These matters are not only of broad public interest, it is important information for the conservation and management of koalas. The fire season ended in March 2020. In June 2020 we wrote to and received replies from, the licensed rehabilitation groups in NSW that rescued koalas, seeking to quantify the impact of the 2019–2020 bushfires. The answers were: 209 koalas came into care due to the bushfires, and of these, 106 were either euthanised or died, 74 were released, and the remainder were still in care, but due for release soon. The number of koalas that came into care for reasons not related to the fires over the same time period was 307. From the point of view of each surviving individual koala, the rehabilitation was a success. What the rescue and rehabilitation has done is highlight the difficulties that koalas face in dealing with fire. This information will contribute to koala management at a local, State and Commonwealth level, particularly in relation to koala rehabilitation policies, and it will be increasingly important given the likelihood of more frequent fires.