The present study demonstrates one solution to a problem faced by managers of species of conservation concern – how to develop broad-scale maps of populations, within known general distribution limits, for the purpose of targeted management action. We aimed to map the current populations of the koala, Phascolarctos cinereus, in New South Wales, Australia. This cryptic animal is widespread, although patchily distributed. It principally occurs on private property, and it can be hard to detect. We combined a map-based mail survey of rural and outer-urban New South Wales with recent developments in estimating site occupancy and species-detection parameters to determine the current (2006) distribution of the koala throughout New South Wales. We were able to define the distribution of koalas in New South Wales at a level commensurate with previous community and field surveys. Comparison with a 1986 survey provided an indication of changes in relative koala density across the state. The 2006 distribution map allows for local and state plans, including the 2008 New South Wales Koala Recovery Plan, to be more effectively implemented. The application of this combined technique can now be extended to a suite of other iconic species or species that are easily recognised by the public.

Dog walking is among the world's most popular recreational activities, attracting millions of people to natural areas each year with diverse benefits to human and canine health. But conservation managers often ban dog walking from natural areas fearing that wildlife will see dogs as potential predators and abandon their natural habitats, resulting in outcry at the restricted access to public land. Arguments are passionate on both sides and debate has remained subjective and unresolved because experimental evidence of the ecological impacts of dog walking has been lacking. Here we show that dog walking in woodland leads to a 35% reduction in bird diversity and 41% reduction in abundance, both in areas where dog walking is common and where dogs are prohibited. These results argue against access by dog walkers to sensitive conservation areas.

Extracting species calls from passive acoustic recordings is a common preliminary step to ecological analysis. For many species, particularly those occupying noisy, acoustically variable habitats, the call extraction process continues to be largely manual, a time-consuming and increasingly unsustainable process. Deep neural networks have been shown to offer excellent performance across a range of acoustic classification applications, but are relatively underused in ecology. We describe the steps involved in developing an automated classifier for a passive acoustic monitoring project, using the identification of calls of the Hainan gibbon Nomascus hainanus, one of the world's rarest mammal species, as a case study. This includes preprocessing-selecting a temporal resolution, windowing and annotation; data augmentation; processing-choosing and fitting appropriate neural network models; and post-processing-linking model predictions to replace, or more likely facilitate, manual labelling. Our best model converted acoustic recordings into spectrogram images on the mel frequency scale, using these to train a convolutional neural network. Model predictions were highly accurate, with per-second false positive and false negative rates of 1.5% and 22.3%. Nearly all false negatives were at the fringes of calls, adjacent to segments where the call was correctly identified, so that very few calls were missed altogether. A post-processing step identifying intervals of repeated calling reduced an 8-h recording to, on average, 22 min for manual processing, and did not miss any calling bouts over 72 h of test recordings. Gibbon calling bouts were detected regularly in multi-month recordings from all selected survey points within Bawangling National Nature Reserve, Hainan. We demonstrate that passive acoustic monitoring incorporating an automated classifier represents an effective tool for remote detection of one of the world's rarest and most threatened species. Our study highlights the viability of using neural networks to automate or greatly assist the manual labelling of data collected by passive acoustic monitoring projects. We emphasize that model development and implementation be informed and guided by ecological objectives, and increase accessibility of these tools with a series of notebooks that allow users to build and deploy their own acoustic classifiers.

For Critically Endangered "species of extreme rarity," there is an urgent need to clarify the potential survival of remnant populations. Such populations can be difficult to detect using standard field methods. Local ecological knowledge (LEK) represents an important alternative source of information, but anecdotal reports of rare or possibly extinct species can contain uncertainty and error. The Hainan gibbon (Nomascus hainanus), the world's rarest primate species, is confirmed to only survive as a tiny remnant population in Bawangling National Nature Reserve, China, but unverified gibbon sightings have been reported from other forest areas on Hainan. We conducted a large-scale community interview survey to gather new data on patterns of primate LEK from 709 respondents around seven reserves across Hainan, to investigate the possibility of gibbon survival outside Bawangling and assess whether LEK can provide useful information for conservation management of cryptic remnant populations. Comparative LEK data for gibbons and macaques are consistent with independent data on the relative status of these species across Hainan. Local awareness and experience of gibbons was low across Hainan, including at Bawangling, but we recorded recent anecdotal gibbon reports from most reserves. A follow-up field survey at Limushan Provincial Nature Reserve did not detect gibbons, however, and documented intensive wildlife exploitation within this reserve. All other surveyed landscapes showed some statistically lower levels of respondent awareness, experience, or sighting histories of gibbons compared to Bawangling, and are therefore considered biologically unlikely to support gibbons. Unverified LEK data can provide important insights into the possible status of cryptic remnant populations when assessed carefully and critically in relation to data from known populations.

Targeted management actions informed by robust data are needed to conserve species of extreme rarity, and assessing the effectiveness of different field methods for detection and monitoring of such species is a conservation priority. Gibbons are typically detected by their daily song through passive listening surveys, but lone gibbon individuals and lowdensity populations are less likely to sing, making detection difficult or impossible using standard survey techniques. Call playback represents an alternative potential method for detecting gibbon presence, but there has been no empirical evaluation of the usefulness of this method in the field. We investigated the efficacy of call 1,*