Key results. Abundance estimates for the three species varied across the sampling periods, with the estimated number of Australian snubfin dolphins ranging from 136 (s.e. 62) to 222 (s.e. 48); from 48 (s.e. 17) to 207 (s.e. 65) for the Australian humpback dolphin; and 34 (s.e. 16) to 75 (s.e. 26) for the bottlenose dolphin.

Table 4. Estimates of the total marked population size (_m) and total population size (_total) of coastal dolphins in Port Essington harbour, with log-normal 95% lower and upper confidence intervals (by session) and coefficient of variance (CV) Standard errors are shown in parentheses

Context. The pampas deer (Ozotoceros bezoarticus) is an endangered species in Argentina. Scarce information exists about one of the four last populations that survive in Corrientes province, where direct counts estimated a population of <500 individuals.

Aims. To evaluate the status of the pampas deer population in Corrientes by applying a standardised methodology and to develop methodological recommendations for future deer monitoring.

Methods. We conducted six population censuses between 2007 and 2010, using line transects placed on roads throughout 1200 km² of grasslands in the Aguapey region, Corrientes, Argentina. From a moving vehicle, we counted every pampas deer group observed along transects. We used Distance 6.0 and its multiple-covariate distance sampling engine to estimate deer density, while exploring the potential effect of roads, habitat type, hour, season, observer experience and survey effort on deer occurrence and density estimation.

Key results. The occurrence of pampas deer was irrespective of transect location (minor or major road) but a greater number of animals was detected over transects in minor roads and in areas covered by grasslands with young pine plantations. We estimated a density of 1.17 individuals km^–2 (s.e. = 0.52), and habitat type was the most important covariate for density estimation. We estimated a total population of 1495 deer (95% CI = 951–2351, CV = 23.27%) for the Aguapey region in Argentina.

Conclusions. Corrientes hosts one of the largest populations of pampas deer in Argentina, with ∼1000 individuals. The fact that we estimated a larger population than did previous studies could be explained either by actual population growth during the past 10 years, or by the use of more exhaustive and sophisticated sampling design and data analysis.

Implications. Population surveys using covariate distance sampling on ground line transects can provide more realistic population estimates than do other simpler methods. Our population estimates and methods can be used as a baseline for future monitoring of this population, as long as factors such as sampling effort, type of roads for locating transects, and habitat type are considered in future analysis.

Context. Changed fire regimes are an important threatening process to savanna biodiversity. Fire-sensitive vegetation such as pindan and its fauna may be particularly susceptible to fire impacts. Invasion by alien species is an additional threatening process. The toxic anuran Rhinella marina is a well publicised invader of savannas. Little is known of impacts in many habitats.

Aims. To test the hypotheses (1) that fire responses among pindan fauna are greater than general savanna responses, and (2) that cane toad-invasion impacts will be reduced relative to riparian habitats.

Methods. Reptiles, frogs, invertebrates and mammals were surveyed seven times from 2008 to 2012, four times before and three times following R. marina invasion. Time since last fire was recorded during each survey. Vegetation change was measured.

Key results. Pindan vegetation structural recovery took 4–5 years, whereas fauna recovery took only 1 year. Ground active agamids, combined Scincidae, fossorial skinks and ground-layer invertebrates responded positively to recent fire. Skinks of Ctenotus spp. declined in size after fire. Short-term fauna responses reflect rapid re-establishment of herbaceous cover. Fauna responses were detected following R. marina invasion, including increases in frogs of Uperoleia spp. and skinks of Carlia spp., and decreases in Lerista griffini and ground-layer invertebrates. Insufficient data were available to test for responses among large predators; however, >50% lower Varanus spp. trap success occurred post-invasion. No invasion response was detected among small mammals.

Conclusions. Pindan fauna fire responses were similar to those of savannas. Fauna responses to Rhinella marina invasion were relatively minor compared with those previously reported in riparian habitats and this may be related to the lower abundance of the invader here than in previous studies in riparian or more fertile habitats.

Implications. The dominant obligate seeding tree in pindan woodland, A. tumida, requires >4 years with no high-intensity fires for re-establishment of the dominant tree. Fire management should aim to minimise extensive fires to reduce impacts on fire-sensitive fauna. Persistence of large predators after cane-toad invasion suggests possible refuge value of low-productivity pindan savannas.

Context . Global positioning system (GPS) collars are increasingly used to study animal ecology. However, understanding their real-world performance is important so that biases in data collection can be addressed. These biases may be species specific. Few studies have examined the influence of sex or temporal variation on locational fix success rates (FSR), and none has assessed GPS-collar performance on bison (Bison bison).

Aims . To test for sex- and temporal-based biases on the performance of GPS collars on free-ranging bison.

Methods . Data from 59 deployments of 37 GPS collars on 27 female and seven male bison was used to test for temporal variation and differences between sexes on collar-deployment success rates and four fix acquisition metrics.

Key results . Most (72.9%) of the 55 GPS collars retrieved malfunctioned before reaching their expected deployment schedule (2 years). We observed a significant difference in the performance of GPS collars deployed on male and females, with those on females functioning almost three times longer. All four fix metrics differed between sexes, with males having significantly better metrics. For females, nightly fix metrics were better than the daytime ones, whereas for males, there was no difference. We observed a significant difference in fix metrics between seasons for females, but not males.

Conclusions . Although the data from GPS collars may be useful in understanding animal ecology, understanding associated biases can aid in the design of field studies using, and the interpretation of results stemming from, GPS collars. Temporal variation in our study was apparent for females, but not males, and likely related to differences in habitat use and movements. Overall, sex had a significant influence on the success of GPS-collar deployments on bison, and should be examined for other species as well.

Implications . Bison damage to the collars was the main reason for collar failures. Researchers should plan for catastrophic failures and malfunctions when deploying GPS collars on bison, particularly adult males. Significant differences observed between daily and seasonal FSR for females were small (≤4.9%), and do not warrant post hoc treatment before analyses.

Context. Loss and degradation of habitat from urban development is a key threat to the squirrel glider (Petaurus norfolcensis), because its distribution coincides where most people live in Australia. Squirrel gliders are known to occur in or around urban fringes where native vegetation is retained; however, little is known about specific anthropogenic factors that may affect their persistence.

Aims. We investigated the relative influence of biophysical and urban factors on the use of large hollow-bearing Eucalyptus trees, which are a key resource for squirrel gliders. The study was located in a typical urban growth area located in southern New South Wales.

Methods. A stratified random sampling approach was used to survey squirrel gliders in urban and rural areas. Infrared, motion-sensor cameras were placed in 34 selected trees for 17 nights to record squirrel glider occupancy and activity. Data on urban (e.g. light and noise pollution levels, road and housing density) and biophysical (e.g. tree height, hollow-bearing tree density, vegetation cover) variables were recorded at each survey tree. Data were statistically analysed using general linear modelling approaches.

Key results. Squirrel gliders were detected more frequently in the rural matrix (23.4% of camera trap-nights) than in urban areas (9.5%). Model results showed that tree height, and the distance to neighbouring trees, had a significant influence on the occurrence and activity of squirrel gliders. Road density and light pollution were included in ‘best’ models to explain glider activity (a negative influence), and noise pollution negatively influenced glider occurrence. Although gliders used large trees in both urban and rural areas, activity generally decreased as levels of urbanisation increased.

Conclusions and implications. Access to and availability of key resources such as tall, hollow-bearing trees is critical for gliders to persist in urban environments. Squirrel gliders will tolerate human stressors such as roads, noise and light pollution to a certain extent, but impacts on population viability remain largely unknown. Novel solutions need to be developed to lessen the effects of anthropogenic factors (such as light and noise) on patches of native vegetation retained in urban areas for conservation purposes.

Context. Extrapolation of knowledge for threatened taxa between parts of their range that are disconnected and/or ecologically diverse can result in significant sources of error that undermine the effectiveness of conservation efforts.

Aims. We investigated the risks associated with extrapolation of ecological information across environmental gradients, using the quokka (Setonix brachyurus) as a case study. Information documented in the northern part of its range is currently used to manage this species across its range in south-western Australia. We examined the suitability of this approach by developing a habitat suitability model (HSM) for the quokka in the southernmost areas of its range and comparing this with existing knowledge for the species.

Methods. We surveyed 327 sites, representative of a range of ecotypes, for presence/absence of quokkas. Occupancy models were applied to select a subset of habitat variables that best predicted occupancy patterns.

Key results . Occupancy patterns were influenced by complex vegetation structure, low levels of woody debris and habitat patchiness. HSMs developed for quokkas in the north could not predict occupancy patterns in the south. Significant fragmentation of subpopulations was observed due to patchiness in the availability of suitable habitat.

Conclusions. The choice of predictor variables in HSMs that are not transferrable between regions could contribute to inappropriate management of habitat for quokkas and an increased risk of local extinctions. In addition, failure to consider processes that affect preferred habitat variables could contribute to the segregation of habitat patches and intervening distances that are too great for successful dispersal, immigration and recolonisation processes.

Implications. The extrapolation of HSMs between geographical areas can increase the risk of outcomes that are detrimental to the conservation of threatened species. Where such extrapolation is necessary, actions guided by the HSMs should be implemented in a management framework that can detect adverse effects, allow for inclusion of new ecological information and explicitly consider the limitations and assumptions of this approach. In addition, perceptions of habitat fragmentation need to include processes such as fire regimes and feral animals that affect the availability and connectivity of habitat and have the potential to adversely affect population viability.

Context . There has been concern that feral cats have negative impacts on the endangered endemic mammals of Amami-Ohshima Island, Japan, including the Amami rabbit, Pentalagus furnessi, Ryukyu long-tailed giant rat, Diplothrix legata, and Amami spiny rat, Tokudaia osimensis. However, no diet study of feral cat has been conducted to support the necessity of an urgent feasible feral-cat management for the island.

Aims. The aims of the present study were to analyse feral-cat diet on Amami-Ohshima Island by using scat analysis and estimate the potential predation impact of feral cats on endangered mammals on the island.

Methods. The diet of feral cats was studied using scat analysis. We estimated the number of prey, percentage of prey, frequency of occurrence (the percentage of scats in a sample containing a particular prey item), percentage of biomass (biomass of the same prey item divided by the total consumed biomass ×100) and daily consumed biomass (DCB).

Key results. Three endangered endemic mammals were the main prey species of the feral cat diet (65% of total DCB). The percentage contributions of these species on DCB were long-tailed giant rat (34.7%), Amami spiny rat (21.9%) and Amami rabbit (12%).

Conclusions. Mammals, especially endangered endemic mammals, were main prey species of feral cat on Amami Island. In Amami Island, where native and invasive rodents coexisted, feral cats consumed more native (56.6%) than invasive (22.2% for Rattus rattus) species.

Implications. Feral cats are likely to be having a significant impact on endangered endemic mammals on the island. To ensure the long-term survival of these endemic species, active management of the feral-cat population should be considered.

Context. Supplemental feeding of large mammalian herbivores is a common management tool mainly aimed at promoting healthy populations and at increasing productivity and trophy sizes. Such management measure may indirectly affect herbivore effects on plant communities through altered foraging patterns. The quantification of the ecological effects of large herbivore management is important for designing holistic management and conservation programs.

Aims. Here we aimed at quantifying the ecological effects of supplemental feeding of Iberian red deer, Cervus elaphus hispanicus, on the composition of and on the browsing effects on Mediterranean woody plant community.

Methods. An experiment was set up in a hunting rangeland located in central Spain, where female deer were kept in enclosures with either exclusive access to natural forages or with additional ad libitum access to a nutritionally rich concentrate. The experiment also included a control area where deer were absent.

Key results. We observed significant differences in browsing impacts among the supplemented, non-supplemented and control areas, and such effect varied for the different plant species. Plant species which nutritional content complemented that of fodder were more highly consumed, for instance, Erica spp., which digestible fibre content is higher and N content lower than that of provided fodder. The presence of deer and the concentrate supplied, instead, did not influence the relative abundances of shrub species.

Conclusions. Artificial supplemental feeding provided to ungulates led to increased browsing on plant species which nutritional composition complemented that of the supplement provided.

Implications. So as to alleviate herbivory impact on all shrubs, we suggest that composition of supplemental feeding should adjust both to the natural forage availability and quality and to ungulate requirements across seasons.

Context . Pindone is used to control rabbits in Australia and New Zealand, but the secondary non-target risks presented by pindone-poisoned rabbits are poorly known.

Aims . We aimed to generate new data on residual concentrations of pindone in poisoned rabbits for use in a review of the secondary poisoning risks posed to non-target animals in New Zealand.

Methods . Laboratory rabbits were offered pellet bait containing 0.25 g kg^–1 pindone in three trials to simulate a range of bait uptake scenarios: single exposure and low or high chronic exposure. Residual pindone concentrations measured in body tissues of laboratory rabbits that had ingested known exposures of pindone were compared with those in wild rabbit carcasses collected after three pindone-baiting operations. Residues in the faeces of some laboratory rabbits were also measured.

Key results . Highest concentrations of pindone residues were in the liver and fat tissue of poisoned rabbits, with consistently lower concentrations in muscle tissue. A dose–response relationship between pindone exposure and liver residue concentrations was found only at the highest chronic exposures. Rabbit carcasses collected after field-baiting operations had generally higher pindone residue concentrations than did laboratory rabbits that had ingested known lethal amounts of bait. Unmetabolised pindone was excreted in rabbit faeces.

Conclusions . The occurrence of the highest residual pindone concentrations in rabbit liver was consistent with the known tissue distribution of anticoagulants in mammals; however, the co-occurrence of similar-range pindone concentrations in rabbit fat has not been previously described. Re-ingestion of soft faecal pellets (caecotrophy) in rabbits that have eaten pindone bait may function as a secondary exposure to increase or prolong their oral exposure to pindone. Some rabbits poisoned following field pindone-baiting operations are likely to have consumed well in excess of a lethal amount of bait.

Implications . Concentrations of residual pindone in fat and liver of poisoned rabbits suggest that secondary poisoning hazard to some non-target predators and scavengers is high. The lack of field-based assessments of the non-target impacts of pindone is a marked information gap that needs to be addressed.