Predicting kill sites of an apex predator from GPS data in different multiprey systems.

Kill rates are a central parameter to assess the impact of predation on prey species. An accurate estimation of kill rates requires a correct identification of kill sites, often achieved by field-checking GPS location clusters (GLCs). However, there are potential sources of error included in kill-site identification, such as failing to detect GLCs that are kill sites, and misclassifying the generated GLCs (e.g., kill for nonkill) that were not field checked. Here, we address these two sources of error using a large GPS dataset of collared Eurasian lynx (Lynx lynx), an apex predator of conservation concern in Europe, in three multiprey systems, with different combinations of wild, semidomestic, and domestic prey. We first used a subsampling approach to investigate how different GPS-fix schedules affected the detection of GLC-indicated kill sites. Then, we evaluated the potential of the random forest algorithm to classify GLCs as nonkills, small prey kills, and ungulate kills. We show that the number of fixes can be reduced from seven to three fixes per night without missing more than 5% of the ungulate kills, in a system composed of wild prey. Reducing the number of fixes per 24 h decreased the probability of detecting GLCs connected with kill sites, particularly those of semidomestic or domestic prey, and small prey. Random forest successfully predicted between 73%-90% of ungulate kills, but failed to classify most small prey in all systems, with sensitivity (true positive rate) lower than 65%. Additionally, removing domestic prey improved the algorithm's overall accuracy. We provide a set of recommendations for studies focusing on kill-site detection that can be considered for other large carnivore species in addition to the Eurasian lynx. We recommend caution when working in systems including domestic prey, as the odds of underestimating kill rates are higher.

Evaluating expert-based habitat suitability information of terrestrial mammals with GPS-tracking data.

Aim: Macroecological studies that require habitat suitability data for many species often derive this information from expert opinion. However, expert-based information is inherently subjective and thus prone to errors. The increasing availability of GPS tracking data offers opportunities to evaluate and supplement expert-based information with detailed empirical evidence. Here, we compared expert-based habitat suitability information from the International Union for Conservation of Nature (IUCN) with habitat suitability information derived from GPS-tracking data of 1,498 individuals from 49 mammal species. Location: Worldwide. Time period: 1998-2021. Major taxa studied: Forty-nine terrestrial mammal species. Methods: Using GPS data, we estimated two measures of habitat suitability for each individual animal: proportional habitat use (proportion of GPS locations within a habitat type), and selection ratio (habitat use relative to its availability). For each individual we then evaluated whether the GPS-based habitat suitability measures were in agreement with the IUCN data. To that end, we calculated the probability that the ranking of empirical habitat suitability measures was in agreement with IUCN's classification into suitable, marginal and unsuitable habitat types. Results: IUCN habitat suitability data were in accordance with the GPS data (> 95% probability of agreement) for 33 out of 49 species based on proportional habitat use estimates and for 25 out of 49 species based on selection ratios. In addition, 37 and 34 species had a > 50% probability of agreement based on proportional habitat use and selection ratios, respectively. Main conclusions: We show how GPS-tracking data can be used to evaluate IUCN habitat suitability data. Our findings indicate that for the majority of species included in this study, it is appropriate to use IUCN habitat suitability data in macroecological studies. Furthermore, we show that GPS-tracking data can be used to identify and prioritize species and habitat types for re-evaluation of IUCN habitat suitability data.

GPS collars have an apparent positive effect on the survival of a large carnivore.

Are instrumented animals representative of the population, given the potential bias caused by selective sampling and the influence of capture, handling and wearing bio-loggers? The answer is elusive owing to the challenges of obtaining comparable data from individuals with and without bio-loggers. Using non-invasive genetic data of a large carnivore, the wolverine (Gulo gulo) in Scandinavia, and an open-population spatial capture-recapture model, we found a 16 (credible interval: 4-30) percentage points lower mortality probability for GPS-collared individuals compared with individuals without GPS collars. While the risk of dying from legal culling was comparable for collared and non-collared wolverines, the former experienced lower probability of mortality due to causes other than legal culling. The aforementioned effect was pronounced despite a potentially lower age-and therefore likely higher natural mortality-of collared individuals. Reports of positive effects of bio-loggers on the survival of individuals are uncommon and we argue that GPS collars could shield animals from poaching. Our results highlight the challenges of drawing population-level inferences for populations subjected to poaching when using data from instrumented individuals.

Scavenging on a pulsed resource: quality matters for corvids but density for mammals.

BACKGROUND: Human food subsidies can provide predictable food sources in large quantities for wildlife species worldwide. In the boreal forest of Fennoscandia, gut piles from moose (Alces alces) harvest provide a potentially important food source for a range of opportunistically scavenging predators. Increased populations of predators can negatively affect threatened or important game species. As a response to this, restrictions on field dressing of moose are under consideration in parts of Norway. However, there is a lack of research to how this resource is utilized. In this study, we used camera-trap data from 50 gut piles during 1043 monitoring days. We estimated depletion of gut piles separately for parts with high and low energy content, and used these results to scale up gut pile density in the study area. We identified scavenger species and analyzed the influences of gut pile quality and density on scavenging behavior of mammals and corvids (family Corvidae). RESULTS: Main scavengers were corvids and red fox (Vulpes vulpes). Parts with high energy content were rapidly consumed, mainly by corvids that were present at all gut piles shortly after the remains were left at the kill site. Corvid presence declined with days since harvest, reflecting reduction in gut pile quality over time independent of gut pile density. Mammals arrived 7-8 days later at the gut piles than corvids, and their presence depended only on gut pile density with a peak at intermediate densities. The decline at high gut pile densities suggest a saturation effect, which could explain accumulation of gut pile parts with low energy content. CONCLUSIONS: This study shows that remains from moose harvest can potentially be an important food resource for scavengers, as it was utilized to a high degree by many species. This study gives novel insight into how energy content and density of resources affect scavenging patterns among functional groups of scavengers.

Decomposing risk: landscape structure and wolf behavior generate different predation patterns in two sympatric ungulates.

Recolonizing carnivores can have a large impact on the status of wild ungulates, which have often modified their behavior in the absence of predation. Therefore, understanding the dynamics of reestablished predator-prey systems is crucial to predict their potential ecosystem effects. We decomposed the spatial structure of predation by recolonizing wolves (Canis lupus) on two sympatric ungulates, moose (Alces alces) and roe deer (Capreolus capreolus), in Scandinavia during a 10-year study. We monitored 18 wolves with GPS collars, distributed over 12 territories, and collected records from predation events. By using conditional logistic regression, we assessed the contributions of three main factors, the utilization patterns of each wolf territory, the spatial distribution of both prey species, and fine-scale landscape structure, in determining the spatial structure of moose and roe deer predation risk. The reestablished predator-prey system showed a remarkable spatial variation in kill occurrence at the intra-territorial level, with kill probabilities varying by several orders of magnitude inside the same territory. Variation in predation risk was evident also when a spatially homogeneous probability for a wolf to encounter a prey was simulated. Even inside the same territory, with the same landscape structure, and when exposed to predation by the same wolves, the two prey species experienced an opposite spatial distribution of predation risk. In particular, increased predation risk for moose was associated with open areas, especially clearcuts and young forest stands, whereas risk was lowered for roe deer in the same habitat types. Thus, fine-scale landscape structure can generate contrasting predation risk patterns in sympatric ungulates, so that they can experience large differences in the spatial distribution of risk and refuge areas when exposed to predation by a recolonizing predator. Territories with an earlier recolonization were not associated with a lower hunting success for wolves. Such constant efficiency in wolf predation during the recolonization process is in line with previous findings about the naive nature of Scandinavian moose to wolf predation. This, together with the human-dominated nature of the Scandinavian ecosystem, seems to limit the possibility for wolves to have large ecosystem effects and to establish a behaviorally mediated trophic cascade in Scandinavia.

When species' ranges meet: assessing differences in habitat selection between sympatric large carnivores.

Differentiation in habitat selection among sympatric species may depend on niche partitioning, species interactions, selection mechanisms and scales considered. In a mountainous area in Sweden, we explored hierarchical habitat selection in Global Positioning System-collared individuals of two sympatric large carnivore species; an obligate predator, the Eurasian lynx (Lynx lynx), and a generalist predator and scavenger, the wolverine (Gulo gulo). Although the species' fundamental niches differ widely, their ranges overlap in this area where they share a prey base and main cause of mortality. Both lynx and wolverines selected for steep and rugged terrain in mountainous birch forest and in heaths independent of scale and available habitats. However, the selection of lynx for their preferred habitats was stronger when they were forming home ranges and they selected the same habitats within their home ranges independent of home range composition. Wolverines displayed a greater variability when selecting home ranges and habitat selection also varied with home range composition. Both species selected for habitats that promote survival through limited encounters with humans, but which also are rich in prey, and selection for these habitats was accordingly stronger in winter when human activity was high and prey density was low. We suggest that the observed differences between the species result primarily from different foraging strategies, but may also depend on differences in ranging and resting behaviour, home range size, and relative density of each species. Our results support the prediction that sympatric carnivores with otherwise diverging niches can select for the same resources when sharing main sources of food and mortality.

Home range size variation in a recovering wolf population: evaluating the effect of environmental, demographic, and social factors.

Home range size in mammals is a key ecological trait and an important parameter in conservation planning, and has been shown to be influenced by ecological, demographic and social factors in animal populations. Information on space requirements is especially important for carnivore species which range over very large areas and often come into direct conflict with human interest. We used long-term telemetry-location data from a recovering wolf population in Scandinavia to investigate variation in home range size in relation to environmental and social characteristics of the different packs. Wolves showed considerable variation in home range size, which ranged from 259 to 1,676 km(2). Although wolf density increased fourfold during the study period, we found no evidence that intraspecific competition influenced range size. Local variation in moose density, which was the main prey for most packs, did not influence wolf home range size. Home ranges increased with latitude and elevation and decreased with increased roe deer density. Although prey biomass alone did not influence range size, our data suggest that there is a correlation between habitat characteristics, choice of prey species and possible hunting success, which currently combine to shape home range size in Scandinavian wolves.

Proximity-sensors on GPS collars reveal fine-scale predator-prey behavior during a predation event: A case study from Scandinavia.

Although the advent of high-resolution GPS tracking technology has helped increase our understanding of individual and multispecies behavior in wildlife systems, detecting and recording direct interactions between free-ranging animals remains difficult. In 2023, we deployed GPS collars equipped with proximity sensors (GPS proximity collars) on brown bears (Ursus arctos) and moose (Alces alces) as part of a multispecies interaction study in central Sweden. On 6 June, 2023, a collar on an adult female moose and a collar on an adult male bear triggered each other's UHF signal and started collecting fine-scale GPS positioning data. The moose collar collected positions every 2 min for 89 min, and the bear collar collected positions every 1 min for 41 min. On 8 June, field personnel visited the site and found a female neonate moose carcass with clear indications of bear bite marks on the head and neck. During the predation event, the bear remained at the carcass while the moose moved back and forth, moving toward the carcass site about five times. The moose was observed via drone with two calves on 24 May and with only one remaining calf on 9 June. This case study describes, to the best of our knowledge, the first instance of a predation event between two free ranging, wild species recorded by GPS proximity collars. Both collars successfully triggered and switched to finer-scaled GPS fix rates when the individuals were in close proximity, producing detailed movement data for both predator and prey during and after a predation event. We suggest that, combined with standard field methodology, GPS proximity collars placed on free-ranging animals offer the ability for researchers to observe direct interactions between multiple individuals and species in the wild without the need for direct visual observation.

Scavenging patterns of an inbred wolf population in a landscape with a pulse of human-provided carrion.

Scavenging is an important part of food acquisition for many carnivore species that switch between scavenging and predation. In landscapes with anthropogenic impact, humans provide food that scavenging species can utilize. We quantified the magnitude of killing versus scavenging by gray wolves (Canis lupus) in Scandinavia where humans impact the ecosystem through hunter harvest, land use practices, and infrastructure. We investigated the cause of death of different animals utilized by wolves, and examined how the proportion of their consumption time spent scavenging was influenced by season, wolf social affiliation, level of inbreeding, density of moose (Alces alces) as their main prey, density of brown bear (Ursus arctos) as an intraguild competitor, and human density. We used data from 39 GPS-collared wolves covering 3198 study days (2001-2019), including 14,205 feeding locations within space-time clusters, and 1362 carcasses utilized by wolves. Most carcasses were wolf-killed (80.5%) while a small part had died from other natural causes (1.9%). The remaining had either anthropogenic mortality causes (4.7%), or the cause of death was unknown (12.9%). Time spent scavenging was higher during winter than during summer and autumn. Solitary wolves spent more time scavenging than pack-living individuals, likely because individual hunting success is lower than pack success. Scavenging time increased with the mean inbreeding coefficient of the adult wolves, possibly indicating that more inbred individuals resort to scavenging, which requires less body strength. There was weak evidence for competition between wolves and brown bears as well as a positive relationship between human density and time spent scavenging. This study shows how both intrinsic and extrinsic factors drive wolf scavenging behavior, and that despite a high level of inbreeding and access to carrion of anthropogenic origin, wolves mainly utilized their own kills.

Predation of endangered Arctic foxes by Golden eagles: What do we know?

Dedicated conservation efforts spanning the past two decades have saved the Fennoscandian Arctic fox (Vulpes lagopus) population from local extinction, and extensive resources continue to be invested in the species' conservation and management. Although increasing, populations remain isolated, small and are not yet viable in the longer term. An understanding of causes of mortality are consequently important to optimize ongoing conservation actions. Golden eagles (Aquila chrysaetos) are a predator of Arctic foxes, yet little information on this interaction is available in the literature. We document and detail six confirmed cases of Golden eagle depredation of Arctic foxes at the Norwegian captive breeding facility (2019-2022), where foxes are housed in large open-air enclosures in the species' natural habitat. Here, timely detection of missing/dead foxes was challenging, and new insights have been gained following recently improved enclosure monitoring. Golden eagle predation peaked during the winter months, with no cases reported from June to November. This finding contrasts with that which is reported from the field, both for Arctic and other fox species, where eagle depredation peaked at dens with young (summer). While the seasonality of depredation may be ecosystem specific, documented cases from the field may be biased by higher survey efforts associated with the monitoring of reproductive success during the summer. Both white and blue color morphs were housed at the breeding station, yet only white foxes were preyed upon, and mortality was male biased. Mitigation measures and their effectiveness implemented at the facility are presented. Findings are discussed in the broader Arctic fox population ecology and conservation context.

Smartphone app reveals that lynx avoid human recreationists on local scale, but not home range scale.

Outdoor recreation is increasing and affects habitat use and selection by wildlife. These effects are challenging to study, especially for elusive species with large spatial requirements, as it is hard to obtain reliable proxies of recreational intensity over extensive areas. Commonly used proxies, such as the density of, or distance to, hiking paths, ignore outdoor recreation occurring on other linear feature types. Here we utilized crowdsourced data from the Strava training app to obtain a large-scale proxy for pedestrian outdoor recreation intensity in southeast Norway. We used the proxy and GPS-tracking data from collared Eurasian lynx (Lynx lynx) to investigate how recreation affects habitat selection at the home range scale and local scale by lynx during summer. We fitted resource selection functions at the two scales using conditional logistic regression. Our analysis revealed that lynx avoided areas of recreational activity at the local scale, but not at home range scale. Nonetheless, lynx frequently used areas associated with recreation, and to a greater degree at night than during the day. Our results suggest that local-scale avoidance of recreation and temporal adjustments of habitat use by lynx mitigate the need for a home range-scale response towards recreation. Scale-dependent responses and temporal adjustments in habitat use may facilitate coexistence between humans and large carnivores.

Timing and synchrony of birth in Eurasian lynx across Europe.

The ecology and evolution of reproductive timing and synchrony have been a topic of great interest in evolutionary ecology for decades. Originally motivated by questions related to behavioral and reproductive adaptation to environmental conditions, the topic has acquired new relevance in the face of climate change. However, there has been relatively little research on reproductive phenology in mammalian carnivores. The Eurasian lynx (Lynx lynx) occurs across the Eurasian continent, covering three of the four main climate regions of the world. Thus, their distribution includes a large variation in climatic conditions, making it an ideal species to explore reproductive phenology. Here, we used data on multiple reproductive events from 169 lynx females across Europe. Mean birth date was May 28 (April 23 to July 1), but was ~10 days later in northern Europe than in central and southern Europe. Birth dates were relatively synchronized across Europe, but more so in the north than in the south. Timing of birth was delayed by colder May temperatures. Severe and cold weather may affect neonatal survival via hypothermia and avoiding inclement weather early in the season may select against early births, especially at northern latitudes. Overall, only about half of the kittens born survived until onset of winter but whether kittens were born relatively late or early did not affect kitten survival. Lynx are strict seasonal breeders but still show a degree of flexibility to adapt the timing of birth to surrounding environmental conditions. We argue that lynx give birth later when exposed to colder spring temperatures and have more synchronized births when the window of favorable conditions for raising kittens is shorter. This suggests that lynx are well adapted to different environmental conditions, from dry and warm climates to alpine, boreal, and arctic climates. This variation in reproductive timing will be favorable in times of climate change, as organisms with high plasticity are more likely to adjust to new environmental conditions.

Extreme home range sizes among Eurasian lynx at the northern edge of their biogeographic range.

Eurasian lynx (Lynx lynx) have a wide distribution across Eurasia. The northern edge of this distribution is in Norway, where they reach up to 72 degrees north. We conducted a study of lynx space use in this region from 2007 to 2013 using GPS telemetry. The home range sizes averaged 2,606 (± 438 SE) km2 for males (n = 9 ranges) and 1,456 (± 179 SE) km2 for females (n = 24 ranges). These are the largest home ranges reported for any large felid, and indeed are only matched by polar bears, arctic living wolves, and grizzly bears among all the Carnivora. The habitat occupied was almost entirely treeless alpine tundra, with home ranges only containing from 20% to 25% of forest. These data have clear implications for the spatial planning of lynx management in the far north as the current management zones are located in unsuitable habitats and are not large enough to encompass individual lynx movements.

Scavenger community structure along an environmental gradient from boreal forest to alpine tundra in Scandinavia.

Scavengers can have strong impacts on food webs, and awareness of their role in ecosystems has increased during the last decades. In our study, we used baited camera traps to quantify the structure of the winter scavenger community in central Scandinavia across a forest-alpine continuum and assess how climatic conditions affected spatial patterns of species occurrences at baits. Canonical correspondence analysis revealed that the main habitat type (forest or alpine tundra) and snow depth was main determinants of the community structure. According to a joint species distribution model within the HMSC framework, species richness tended to be higher in forest than in alpine tundra habitat, but was only weakly associated with temperature and snow depth. However, we observed stronger and more diverse impacts of these covariates on individual species. Occurrence at baits by habitat generalists (red fox, golden eagle, and common raven) typically increased at low temperatures and high snow depth, probably due to increased energetic demands and lower abundance of natural prey in harsh winter conditions. On the contrary, occurrence at baits by forest specialists (e.g., Eurasian jay) tended to decrease in deep snow, which is possibly a consequence of reduced bait detectability and accessibility. In general, the influence of environmental covariates on species richness and occurrence at baits was lower in alpine tundra than in forests, and habitat generalists dominated the scavenger communities in both forest and alpine tundra. Following forecasted climate change, altered environmental conditions are likely to cause range expansion of boreal species and range contraction of typical alpine species such as the arctic fox. Our results suggest that altered snow conditions will possibly be a main driver of changes in species community structure.

Tracking neighbours promotes the coexistence of large carnivores.

The study of competition and coexistence among similar interacting species has long been considered a cornerstone in evolutionary and community ecology. However, understanding coexistence remains a challenge. Using two similar and sympatric competing large carnivores, Eurasian lynx and wolverines, we tested the hypotheses that tracking among heterospecifics and reactive responses to potential risk decreases the probability of an agonistic encounter when predators access shared food resources, thus facilitating coexistence. Lynx and wolverines actively avoided each other, with the degree of avoidance being greater for simultaneous than time-delayed predator locations. Wolverines reacted to the presence of lynx at relatively short distances (mean: 383 m). In general, lynx stayed longer, and were more stationary, around reindeer carcasses than wolverines. However, when both predators were present at the same time around a carcass, lynx shortened their visits, while wolverine behavior did not change. Our results support the idea that risk avoidance is a reactive, rather than a predictive, process. Since wolverines have adapted to coexist with lynx, exploiting lynx-killed reindeer carcasses while avoiding potential encounters, the combined presence of both predators may reduce wolverine kill rate and thus the total impact of these predators on semi-domestic reindeer in Scandinavia. Consequently, population management directed at lynx may affect wolverine populations and human-wolverine conflicts.

Activity patterns of Eurasian lynx are modulated by light regime and individual traits over a wide latitudinal range.

The activity patterns of most terrestrial animals are regarded as being primarily influenced by light, although other factors, such as sexual cycle and climatic conditions, can modify the underlying patterns. However, most activity studies have been limited to a single study area, which in turn limit the variability of light conditions and other factors. Here we considered a range of variables that might potentially influence the activity of a large carnivore, the Eurasian lynx, in a network of studies conducted with identical methodology in different areas spanning latitudes from 49°7'N in central Europe to 70°00'N in northern Scandinavia. The variables considered both light conditions, ranging from a day with a complete day-night cycle to polar night and polar day, as well as individual traits of the animals. We analysed activity data of 38 individual free-ranging lynx equipped with GPS-collars with acceleration sensors, covering more than 11,000 lynx days. Mixed linear additive models revealed that the lynx activity level was not influenced by the daily daylight duration and the activity pattern was bimodal, even during polar night and polar day. The duration of the active phase of the activity cycle varied with the widening and narrowing of the photoperiod. Activity varied significantly with moonlight. Among adults, males were more active than females, and subadult lynx were more active than adults. In polar regions, the amplitude of the lynx daily activity pattern was low, likely as a result of the polycyclic activity pattern of their main prey, reindeer. At lower latitudes, the basic lynx activity pattern peaked during twilight, corresponding to the crepuscular activity pattern of the main prey, roe deer. Our results indicated that the basic activity of lynx is independent of light conditions, but is modified by both individual traits and the activity pattern of the locally most important prey.