Context. Predation is one of the key contributors to mortality in freshwater turtles. Confirming the identity of predators is an important step towards conservation management action. Throughout Australia, the red fox (Vulpes vulpes) is suspected to apply significant and unsustainable predation pressure to turtle populations, killing adults and depredating nests; however methods for confirming this are limited.

Aims. The present study used a range of methods to confirm predation of oblong turtle (Chelodina colliei) nests and adults by the introduced red fox.

Methods. First, depredated adult carapaces, and turtle egg-shell fragments from excavated nests were swabbed and analysed for trace DNA. Second, we used artificial turtle nests, monitored by camera traps, to analyse seasonal changes in the behaviour of foxes around sites where turtle nests are present, including over the nesting season. Last, we used scat analysis to identify the prevalence of turtle remains in fox diet.

Key results. Predominantly fox DNA was recovered from both adult carapaces and depredated eggs. In addition, camera traps recorded only foxes depredating artificial nests. Despite this evidence that foxes kill adults and excavated nests, we found that turtle remains were only a small part of the diet of foxes at this study site (hatchling or turtle egg shell were present in only 4% of 230 scats sampled). The diet of these foxes was largely anthropogenic-sourced foods, such as fruit (e.g. figs, grapes, melons; 81% of scats), sheep carrion (41%) and rodents (36%).

Conclusions. We conclude that DNA analysis, camera trapping and scat analysis are effective methods of identifying foxes as predators of adult turtle, and their nests. Furthermore, we found that anthropogenic foods (orchard crops, livestock or synanthropic species) may subsidise greater fox population size than might occur in their absence, thereby increasing potential pressure on these freshwater turtles.

Implications. Our findings give credence to the argument that foxes are effective predators of turtle adults and nests. In addition, the high proportion of anthropogenic food sources in the diet of foxes, and potential subsidisation, is an important consideration for land managers.

Context. Feral cats are implicated in the decline of terrestrial native mammals across northern Australia. Research in the Kimberley region of north-western Australia found feral cats strongly selected for fire scars when hunting, suggesting that intensifying fire regimes will have severe consequences for declining prey species.

Aims. We tested the generality of cat–fire interaction beyond the Kimberley, by measuring habitat selection of feral cats in relation to fire scars and habitat types in north-eastern Australia.

Methods. Our study was conducted at Piccaninny Plains Wildlife Sanctuary, Cape York Peninsula. We live-captured feral cats during the dry season of 2015, released them with GPS collars set to record fixes at 15-min intervals, and recaptured cats 4 months later. We created dynamic habitat maps of vegetation types, fire and wetlands, and compared cat habitat selection using discrete choice modelling. We also measured cat density from arrays of camera traps and examined cat diet by analysis of stomach contents.

Key results. We obtained GPS movement data from 15 feral cats. Feral cats selected strongly for recent fire scars (1 or 2 months old), but avoided fire scars 3 months old or older. Three long-distance movements were recorded, all directed towards recent fire scars. Cats also selected for open wetlands, and avoided rainforests. Density of cats at Piccaninny Plains was higher than recorded elsewhere in northern Australia. All major vertebrate groups were represented in cat diet.

Conclusions. We showed that feral cats in north-eastern Australia strongly select for recent fire scars and open wetlands. These results are consistent with those from the Kimberley. Together, these studies have shown that amplified predation facilitated by loss of cover is likely to be a fundamental factor driving mammal decline across northern Australia.

Implications. Reducing the frequency of intense fires may indirectly reduce the impact of feral cats at a landscape scale in northern Australia. We also suggest that managers target direct cat control towards open wetlands and recently burnt areas, which cats are known to favour.

Context. Many ectothermic animals survive winter by hibernating, either buried or in burrows. During their hibernation these animals are vulnerable to changes in soil structure and temperature caused by the activity of grazing mammals. This may be a particular risk to ectotherms living in native grassland, as this habitat is often used for livestock grazing. The endangered pygmy bluetongue lizard lives in burrows, in fragments of native grassland in South Australia, and these burrows are likely to be affected by sheep grazing during the lizards’ hibernation.

Aims. The current study aimed to determine the following effects on the persistence of burrows suitable to the pygmy bluetongue lizard: different levels of grazing, winter vegetation cover, burrow entrance size and whether a lizard was overwintering in the burrow.

Methods. During two winters, we applied different grazing treatments to six experimental paddocks and determined whether suitable lizard burrows located in autumn were still suitable to lizards in the following spring. For each burrow, we recorded whether a lizard was overwintering, the burrow entrance diameter and the vegetation cover around the burrow during the winter.

Key results. Increased grazing pressure led to decreased persistence of lizard burrows. We also found that burrows with an overwintering lizard had a greater chance of persisting, but found no direct effect of winter vegetation cover or entrance diameter.

Conclusions. The results show that although pygmy bluetongue lizards may be able to stabilise their own burrows, the more intense the grazing the lower the chance of lizard burrow persistence.

Implications. Management of sheep grazing is an important component in future conservation of the endangered pygmy bluetongue lizard. Grazing on grassland with a lizard population should be kept at a moderate level and hard grazing should be avoided.

Context. Fire is an important driver of species distributions globally. At the same time, biota also influence fire regimes. Animal activities that modify fuel characteristics may influence fire regimes and hence ecosystem function. However, apart from herbivory, animal effects on fuels and fire behaviour have rarely been studied.

Aims. We examined the effect of nest building by malleefowl (Leipoa ocellata) on litter fuel loads and fire behaviour in the fire-prone, semiarid mallee ecosystem of south-eastern Australia. Malleefowl nests consist of mounds constructed by raking large amounts of leaf litter from the surrounding area. Mound-building activity is likely to affect fuel loads and potentially affect fire behaviour in this environment.

Methods. Litter cover and mass were compared between paired mound and non-mound sites. Fire behaviour modelling was used to determine whether differences in fuel load were likely to translate into differences in fire behaviour. Additionally, in an area recently burnt by wildfire, the minimum diameter of burnt stems was compared between mound and non-mound sites to determine the effects of malleefowl activity on fire intensity.

Key results. Malleefowl nesting activity reduced litter fuel loads around mounds. Fire behaviour modelling and post-fire minimum stem diameter measurements showed this led to reduced fire intensity around mounds, even under extreme fire weather conditions.

Conclusions. The likelihood and intensity of fire is reduced around active malleefowl mounds. Malleefowl nesting contributes to more heterogeneous burn patterns in mallee vegetation. This in turn may contribute to the formation of fire refuges.

Implications. Few studies have examined the effects of animal activities on fire. This study demonstrates that non-trophic interactions of fauna with fuels may influence fire regimes. Species that reduce or disrupt the continuity or connectivity of fuels could have similar effects to malleefowl in fire-prone regions. Further examination of the interactions of animals and fire regimes will contribute to a better understanding and conservation management of fire-prone ecosystems.

Context. Driven hunts exemplify the most representative form of big-game hunting in southern Europe.

Aims. We analysed hunter preferences for driven hunts and the marginal willingness to pay for their characteristics.

Methods. We conducted a discrete-choice experiment for driven hunts, taking into account the number of deer that could be hunted, the possibility of free-range wild-boar hunting, the presence of trophies, and other characteristics of driven hunts, such as congestion and travel time.

Key results. The highest influential driven-hunt characteristic on the utility of big-game hunters is the presence of trophy specimens, whereas for the small-game hunter it would be free-range wild-boar hunting.

Conclusions. Small-game hunters are reluctant to participate in the big-game market because of cultural factors and not because of budgetary restrictions.

Implications. Wildlife management and marketing of driven hunts can be improved taking into account the hunter preferences.

Context. Black-tailed prairie dogs (Cynomus ludovicianus) often come into conflict with humans in urban natural areas adjacent to private property by destroying plants and denuding the landscape. There is a diversity of attitudes related to urban prairie dog management in Colorado and interest in non-lethal methods is increasing. Therefore, the development of non-lethal methods that allow the existence of prairie dogs and mitigate damage related to population density is needed.

Aims. The efficacy of DiazaCon (20,25-diazacholesterol dihydrochloride, a cholesterol analogue) as an oral contraceptive bait for reducing reproduction in black-tailed prairie dogs was tested. Persistence of DiazaCon in black-tailed prairie dogs 11 months after initial application was also determined.

Methods. Prairie dogs were baited with DiazaCon-coated rolled oats or control bait for 10 days over a 20-day period in November 2007 on one control and one treatment site. Sites were randomly assigned to a treatment group. Visual counts were used 8 months post-treatment (June/July 2008) to determine pup : adult ratios on both sites. Prairie dogs on both sites were trapped 11 months post-treatment (September/October 2008) to gather mass and age data to assess juvenile : adult ratios and to collect blood samples for determination of serum cholesterol and desmosterol concentrations.

Key results. DiazaCon treatment reduced the number of pups per adult by 95.5% when compared with the control site in June/July 2008 (0.09 ± 0.02, n = 10, number of observation days) versus 2.0 ± 0.2 (n = 10). Free serum cholesterol levels in treated animals (370.7 µg mL^–1) were lower than control animal levels (411.6 µg mL^–1, P < 0.015), while treated desmosterol levels (14.5 µg mL^–1) were higher than controls (0.35 µg mL^–1, P < 0.02) 11 months after ingestion of DiazaCon bait. However, these differences would likely not have been sufficient to prevent successful breeding in the upcoming breeding season.

Conclusions. DiazaCon is an effective contraceptive in black-tailed prairie dogs. It is probably reversible and will likely affect breeding success for only one breeding season. DiazaCon may provide an additional tool to help manage urban populations of prairie dogs in locations where lethal control is unacceptable.

Implications. The need for further research into secondary hazards to non-target animals, such as predators, is discussed. Recommendations for field application are also discussed.

Context. Throughout the world, declines in large mammalian carnivores have led to the release of smaller meso-mammalian predators. Coyotes (Canis latrans) have increased in abundance, distribution and ecological influence following the extirpation of apex predators in North America. Coyotes have had substantial influence on many ecosystems in recently colonised portions of their range, but those influences can vary across land cover types. Thus, understanding the relationship between coyote abundance and land cover may enhance our ability to predict spatial variation in the ecological effects of coyotes.

Aims. Our objective was to examine the influence of landscape attributes on eastern coyote abundance to ultimately facilitate predictions of spatial variation in the effects of coyotes on prey populations, ecological communities and human interests.

Methods. We collected count data from repeated visits to 24 sites by eliciting howl responses from coyotes. We fit abundance models to howl-response data to examine the effects of landscape composition and configuration on coyote abundance in a mixed forest/agricultural ecosystem in south-western Georgia, USA.

Key results. Our investigation revealed that coyote abundance was positively associated with grasslands that were predominantly used for livestock production, and negatively associated with patch diversity.

Conclusions. Our results supported the prediction that coyotes would be positively associated with open habitats and that they are well adapted for areas structurally similar to the plains of central North America, where the species originated. In addition, these results suggest that aspects of fragmentation, such as patch diversity, can negatively affect coyote abundance. Our results highlight the importance of patch type and landscape juxtaposition on the abundance of coyotes in complex heterogeneous landscapes.

Implications. Our results further our understanding of the spatial variation in coyote abundances across a recently colonised portion of the species range. Combining howl-response surveys with abundance modelling is a promising approach for studying the associations between population dynamics of vocal canids and landscape structure over large spatial scales.

Context . Understanding habitat use and selection by threatened ungulates is a crucial prerequisite to prioritise management areas and for developing effective conservation strategies.

Aims . The aim of our research was to determine the habitat use and selection of takins (Budorcas taxicolor) in the middle range of the Qinling Mountains, China.

Methods . The study was conducted from August 2013 to August 2015. Global positioning system (GPS) radio-tracking was used to monitor 10 collared takins to gain their location information. The Manly–Chesson selectivity index and Bonferroni-adjusted 95% confidence intervals were applied to determine which habitats were selected.

Key results . Habitat use and selection by takins showed obvious individual differences. At the landscape scale, all of the four most common habitat types were preferred by takins. However, all takins avoided artificially planted larch forest, and farmland and village. Available habitats within the home ranges also mostly included the four common habitat types. At the home-range scale, all individuals had significant habitat selectivity during the entire tracking period and each season. The habitat use and selection within the home range varied obviously with season and showed sexual differences to a certain extent.

Conclusions . The habitat selection by takins is scale-dependent. At the landscape scale, takins are most likely to occur at sites covered by forest. At both landscape and home-range scales, our results indicated that takins need more diverse forest habitats, but none of the four most common forest habitats is essential for survival of this species.

Implications . The present work has provided more insight into the habitat use and habitat selection of takins in mountainous forest landscapes. Many measures such as maintaining a diversity of forest habitats, avoiding habitat alteration by invasion of exotic plants, and increasing the area of available habitats by relocating the villages from within to outside of the reserve are recommended to conserve this large species.

Context . Small mammalian herbivores, such as European rabbits (Oryctolagus cuniculus), can have significant unwanted impacts on agriculture and horticulture and can attract birds of prey and avian scavengers to airports, increasing the risk of bird strike. Sustainable wildlife management tools that have high efficacy and animal welfare are needed to mitigate these impacts.

Aim . We assessed perennial ryegrass (Lolium perenne) and tall fescue (Festuca arundinacea) cultivars associated with selected Epichloë endophytes, originally developed for bird management at airports, to determine their feeding-deterrent properties towards invasive rabbits.

Methods . A pellet feed trial using caged domestic rabbits and a pen trial using wild rabbits were conducted to assess rabbit responses to food pellets with novel endophyte-infected or uninfected grass seed and to plots planted with endophyte-infected or uninfected grasses, respectively.

Key results . Caged rabbits ate significantly less food pellets containing endophyte-infected grass straw than pellets containing endophyte-free grass straw. Wild rabbits consumed significantly less herbage from plots planted with endophyte-infected grasses than those of the equivalent endophyte-free grass cultivars. Temporal patterns of rabbit feeding behaviour support the hypothesis that deterrence in rabbits resulted from post-ingestion feedback associated with grass–endophyte secondary metabolites.

Conclusions . Although more research is required on the exact mechanism of action in rabbits, our results suggest novel endophyte-infected grass associations may substantially reduce rabbit feeding behaviour and possibly rabbit numbers in areas where these grasses are sown.

Implications . Novel grass–endophyte associations have great potential for deterring problem wildlife at airports and other amenity areas.

Context. Grazing by domestic stock can potentially influence ecosystems positively or negatively, depending on the grazing regime and intensity. The pygmy bluetongue lizard (Tiliqua adelaidensis) is an endangered species, restricted to fragmented grasslands in the Mid North region of South Australia, predominantly grazed by sheep. These lizards refuge in vertical burrows. They do not dig their own burrows, instead relying on co-existing spiders for this essential resource.

Aims. We investigated how sheep grazing influenced spider burrows, and predicted the following: that grazing sheep might trample and destroy shallow burrows; that burrows in grazed habitat with less vegetative cover would have higher temperatures in summer; and that, by way of compensation, spiders would dig deeper burrows in grazed habitat.

Methods. We monitored 12 30 m × 30 m plots monthly, over two austral spring/summer periods (September–March 2012–14). We recorded the number, depth and occupant of all spider burrows in those plots in each survey. We also measured temperatures inside artificial burrows in grazed and ungrazed habitat over a 3-day period in September 2014.

Key results. Mean burrow depth increased with grazing. This was not because individual burrows became deeper, but because the shallowest burrows were lost where there were sheep. Burrows were significantly hotter in grazed habitat, but lizards consistently chose deeper, cooler burrows. In the subset of burrows that were occupied by lizards, there was no difference in depth between grazed and ungrazed treatments. We detected no difference in spider digging activity between grazed and ungrazed plots.

Conclusions. Sheep grazing can destroy shallow burrows, probably through trampling, and because sheep produce burrow filling debris. However, grazing did not affect deeper burrows occupied by pygmy bluetongue lizards.

Implications. This short study indicated that moderate levels of grazing have low impact on deeper burrows suitable for pygmy bluetongue lizards and for lizard persistence. It explains how sheep and lizards have co-occurred for so long, but leaves open the question of how the loss of shallow burrows affects populations of the burrow-digging spiders that are an essential component of the lizard environment.