Variable strategies to solve risk-reward tradeoffs in carnivore communities

Prediction of in-hospital mortality risk for patients with acute ST-elevation myocardial infarction after primary PCI based on predictors selected by GRACE score and two feature selection methods

Introduction

Accurate in-hospital mortality prediction following percutaneous coronary intervention (PCI) is crucial for clinical decision-making. Machine Learning (ML) and Data Mining methods have shown promise in improving medical prognosis accuracy.

Methods

We analyzed a dataset of 4,677 patients from the Regional Vascular Center of Primorsky Regional Clinical Hospital No. 1 in Vladivostok, collected between 2015 and 2021. We utilized Extreme Gradient Boosting, Histogram Gradient Boosting, Light Gradient Boosting, and Stochastic Gradient Boosting for mortality risk prediction after primary PCI in patients with acute ST-elevation myocardial infarction. Model selection was performed using Monte Carlo Cross-validation. Feature selection was enhanced through Recursive Feature Elimination (RFE) and Shapley Additive Explanations (SHAP). We further developed hybrid models using Augmented Grey Wolf Optimizer (AGWO), Bald Eagle Search Optimization (BES), Golden Jackal Optimizer (GJO), and Puma Optimizer (PO), integrating features selected by these methods with the traditional GRACE score.

Results

The hybrid models demonstrated superior prediction accuracy. In scenario (1), utilizing GRACE scale features, the Light Gradient Boosting Machine (LGBM) and Extreme Gradient Boosting (XGB) models optimized with BES achieved Recall values of 0.944 and 0.954, respectively. In scenarios (2) and (3), employing SHAP and RFE-selected features, the LGB models attained Recall values of 0.963 and 0.977, while the XGB models achieved 0.978 and 0.99.

Discussion

The study indicates that ML models, particularly the XGB optimized with BES, can outperform the conventional GRACE score in predicting in-hospital mortality. The hybrid models' enhanced accuracy presents a significant step forward in risk assessment for patients post-PCI, offering a potential alternative to existing clinical tools. These findings underscore the potential of ML in optimizing patient care and outcomes in cardiovascular medicine.

Intraspecific encounters can lead to reduced range overlap

Direct encounters, in which two or more individuals are physically close to one another, are a topic of increasing interest as more and better movement data become available. Recent progress, including the development of statistical tools for estimating robust measures of changes in animals' space use over time, facilitates opportunities to link direct encounters between individuals with the long-term consequences of those encounters. Working with movement data for coyotes (Canis latrans) and grizzly bears (Ursus arctos horribilis), we investigate whether close intraspecific encounters were associated with spatial shifts in the animals' range distributions, as might be expected if one or both of the individuals involved in an encounter were seeking to reduce or avoid conflict over space. We analyze the movement data of a pair of coyotes in detail, identifying how a change in home range overlap resulting from altered movement behavior was apparently a consequence of a close intraspecific encounter. With grizzly bear movement data, we approach the problem as population-level hypothesis tests of the spatial consequences of encounters. We find support for the hypotheses that (1) close intraspecific encounters between bears are, under certain circumstances, associated with subsequent changes in overlap between range distributions and (2) encounters defined at finer spatial scales are followed by greater changes in space use. Our results suggest that animals can undertake long-term, large-scale spatial changes in response to close intraspecific encounters that have the potential for conflict. Overall, we find that analyses of movement data in a pairwise context can (1) identify distances at which individuals' proximity to one another may alter behavior and (2) facilitate testing of population-level hypotheses concerning the potential for direct encounters to alter individuals' space use.

Spatio-temporal interactions between the red fox and the wolf in two contrasting European landscapes

Relationships among carnivore species are complex, potentially switching from competition to facilitation on a context-dependent basis. Negative associations are predicted to increase with latitude, due to limited resources emphasising competition and/or intra-guild predation. Accordingly, a stronger negative correlation between large- and meso-carnivore abundances should be expected at higher latitudes, with a substantial spatio-temporal partitioning favouring interspecific coexistence. Human presence may influence spatio-temporal relationships between (meso)carnivore species, as it can be perceived as a risk factor, but anthropogenic food can also provide an important additional food resource. Using camera-trap data, we studied the spatio-temporal associations between two of the most widespread carnivores in Europe, i.e., the red fox and wolf. We compared their monthly/daily spatio-temporal partitioning between two different landscapes: Białowieża Primeval Forest (Poland) and the Mediterranean Maremma Regional Park (Italy). We predicted a stronger interspecific partitioning, as well as more attraction of red foxes to humans in the northern site (Poland). Temporal activity patterns of the two carnivores overlapped in both sites, and their detection rates were positively associated, even though in weaker way in Poland. We observed a positive spatial association of red foxes with human activity in Białowieża, but not in Maremma. This association occurred only at a monthly temporal scale and disappeared at a daily scale, suggesting some disturbance in the shorter term. Our results provided partial support to our predictions and suggest that, despite the ecological differences between our study areas, only weak differences in wolf-fox relations were observed, suggesting that red fox responses to wolves may be relatively comparable over large spatial scales.

Badgers remain fearless in the face of simulated wolf presence near their setts

Many mesocarnivores are fossorial and use burrow systems to avoid predators. But fossorial animals cannot stay safely underground forever; they must also risk emerging overground to forage and find mates. To make this trade-off effectively and maximise their own fitness, it is imperative they assess how risk varies in space and time and adapt their denning behaviour accordingly. We used the badger in Białowieża Forest, Poland, as a model for investigating how the denning behaviour of a fossorial mesocarnivore varies in response to short-term large carnivore risk. To this end, we experimentally simulated perceived wolf presence outside 10 badger setts using audio playbacks of wolves (their howls). We assayed two behavioural measures of fear: badger emergence time from setts on the day playbacks were broadcast and their presence in setts on the day after. We found that neither badger emergence time nor next-day sett use varied in response to wolf playbacks. The results of the present study contrast with a previous study of ours that found badgers used setts in areas with high landscape level perceived wolf risk less often than those in lower-risk areas. Together, these papers' findings suggest that different spatiotemporal scales of perceived risk can have differential effects on badger behaviour. We conclude that rather than take risk avoidance measures at all risky times and places, badgers likely display a diversity of reactions to large carnivore presence that depend on the context and spatiotemporal scale of the risk being perceived.

Nearby night lighting, rather than sky glow, is associated with habitat selection by a top predator in human-dominated landscapes

Artificial light at night (ALAN) is increasing in extent and intensity across the globe. It has been shown to interfere with animal sensory systems, orientation and distribution, with the potential to cause significant ecological impacts. We analysed the locations of 102 mountain lions (Puma concolor) in a light-polluted region in California. We modelled their distribution relative to environmental and human-disturbance variables, including upward radiance (nearby lights), zenith brightness (sky glow) and natural illumination from moonlight. We found that mountain lion probability of presence was highly related to upward radiance, that is, related to lights within approximately 500 m. Despite a general pattern of avoidance of locations with high upward radiance, there were large differences in degree of avoidance among individuals. The amount of light from artificial sky glow was not influential when included together with upward radiance in the models, and illumination from moonlight was not influential at all. Our results suggest that changes in visibility associated with lunar cycles and sky glow are less important for mountain lions in their selection of light landscapes than avoiding potential interactions with humans represented by the presence of nearby lights on the ground. This article is part of the theme issue 'Light pollution in complex ecological systems'.

Topcats and underdogs: intraguild interactions among three apex carnivores across Asia's forestscapes

Intraguild interactions among carnivores have long held the fascination of ecologists. Ranging from competition to facilitation and coexistence, these interactions and their complex interplay influence everything from species persistence to ecosystem functioning. Yet, the patterns and pathways of such interactions are far from understood in tropical forest systems, particularly across countries in the Global South. Here, we examined the determinants and consequences of competitive interactions between dholes Cuon alpinus and the two large felids (leopards Panthera pardus and tigers Panthera tigris) with which they most commonly co-occur across Asia. Using a combination of traditional and novel data sources (N = 118), we integrate information from spatial, temporal, and dietary niche dimensions. These three species have faced catastrophic declines in their extent of co-occurrence over the past century; most of their source populations are now confined to Protected Areas. Analysis of dyadic interactions between species pairs showed a clear social hierarchy. Tigers were dominant over dholes, although pack strength in dholes helped ameliorate some of these effects; leopards were subordinate to dholes. Population-level spatio-temporal interactions assessed at 25 locations across Asia did not show a clear pattern of overlap or avoidance between species pairs. Diet-profile assessments indicated that wild ungulate biomass consumption by tigers was highest, while leopards consumed more primate and livestock prey as compared to their co-predators. In terms of prey offtake (ratio of wild prey biomass consumed to biomass available), the three species together harvested 0.4-30.2% of available prey, with the highest offtake recorded from the location where the carnivores reach very high densities. When re-examined in the context of prey availability and offtake, locations with low wild prey availability showed spatial avoidance and temporal overlap among the carnivore pairs, and locations with high wild prey availability showed spatial overlap and temporal segregation. Based on these observations, we make predictions for 40 Protected Areas in India where temporally synchronous estimates of predator and prey densities are available. We expect that low prey availability will lead to higher competition, and in extreme cases, to the complete exclusion of one or more species. In Protected Areas with high prey availability, we expect intraguild coexistence and conspecific competition among carnivores, with spill-over to forest-edge habitats and subsequent prey-switching to livestock. We stress that dhole-leopard-tiger co-occurrence across their range is facilitated through an intricate yet fragile balance between prey availability, and intraguild and conspecific competition. Data gaps and limitations notwithstanding, our study shows how insights from fundamental ecology can be of immense utility for applied aspects like large predator conservation and management of human-carnivore interactions. Our findings also highlight potential avenues for future research on tropical carnivores that can broaden current understanding of intraguild competition in forest systems of Asia and beyond.

Diet selection in the Coyote Canis latrans

The Coyote (Canis latrans) is one of the most studied species in North America with at least 445 papers on its diet alone. While this research has yielded excellent reviews of what coyotes eat, it has been inadequate to draw deeper conclusions because no synthesis to date has considered prey availability. We accounted for prey availability by investigating the prey selection of coyotes across its distribution using the traditional Jacobs' index method, as well as the new iterative preference averaging (IPA) method on scats and biomass. We found that coyotes selected for Dall's Sheep (Ovis dalli), White-tailed Deer (Odocoileus virginianus), Eastern Cottontail Rabbit (Sylvilagus floridanus), and California Vole (Microtus californicus), which yielded a predator-to-preferred prey mass ratio of 1:2. We also found that coyotes avoided preying on other small mammals, including carnivorans and arboreal species. There was strong concordance between the traditional and IPA method on scats, but this pattern was weakened when biomass was considered. General linear models revealed that coyotes preferred to prey upon larger species that were riskier to hunt, reflecting their ability to hunt in groups, and were least likely to hunt solitary species. Coyotes increasingly selected Mule Deer (O. hemionus) and Snowshoe Hare (Lepus americanus) at higher latitudes, whereas Black-tailed Jackrabbit (L. californicus) were increasingly selected toward the tropics. Mule Deer were increasingly selected at higher coyote densities, while Black-tailed Jackrabbit were increasingly avoided at higher coyote densities. Coyote predation could constrain the realized niche of prey species at the distributional limits of the predator through their increased efficiency of predation reflected in increased prey selection values. These results are integral to improved understandings of Coyote ecology and can inform predictive analyses allowing for spatial variation, which ultimately will lead to better understandings about the ecological role of the coyote across different ecosystems.

Dynamic balancing of risks and rewards in a large herbivore: Further extending predator-prey concepts to road ecology

Animal behaviour is shaped by the ability to identify risks and profitably balance the levels of risks encountered with the payoffs experienced. Anthropogenic disturbances like roads generate novel risks and opportunities that wildlife must accurately perceive and respond to. Basic concepts in predator-prey ecology are often used to understand responses of animals to roads (e.g. increased vigilance, selection for cover in their vicinity). However, prey often display complex behaviours such as modulating space use given varying risks and rewards, and it is unclear if such dynamic balancing is used by animals in the context of road crossings. We tested whether animals dynamically balance risks and rewards relative to roads using extensive field-based and GPS collar data from elk in Yoho National Park (British Columbia, Canada), where a major highway completely bisects their range during most of the year. We analysed elk behaviour by combining hidden Markov movement models with a step-selection function framework. Rewards were indexed by a dynamic map of available forage biomass, and risks were indexed by road crossings and traffic volumes. We found that elk generally selected intermediate and high forage biomass, and avoided crossing the road. Most of the time, elk modulated their behaviour given varying risks and rewards. When crossing the highway compared with not crossing, elk selected for greater forage biomass and this selection was stronger as the number of highway crossings increased. However, with traffic volume, elk only balanced foraging rewards when they crossed a single time during a travel sequence. Using a road ecology system, we empirically tested an important component of predator-prey ecology-the ability to dynamically modulate behaviour in response to varying levels of risks and rewards. Such a test articulates how decision-making processes that consider the spatiotemporal variation in risks and rewards allow animals to successfully and profitably navigate busy roads. Applying well-developed concepts in predator-prey theory helps understand how animals respond to anthropogenic disturbances and anticipate the adaptive capacity for individuals and populations to adjust to rapidly changing environments.

Habitat modification destabilizes spatial associations and persistence of Neotropical carnivores

Spatial relationships between sympatric species underpin biotic interactions, structure ecological communities, and maintain ecosystem health. However, the resilience of interspecific spatial associations to human habitat modification remains largely unknown, particularly in tropical regions where anthropogenic impacts are often greatest. We applied multi-state multi-species occurrence models to camera trap data across nine tropical landscapes in Colombia to understand how prominent threats to forest ecosystems influence Neotropical carnivore occurrence and interspecific spatial associations, with implications for biotic interactions. We show that carnivore occurrence represents a delicate balance between local environmental conditions and interspecific interactions that can be compromised in areas of extensive habitat modification. The stability of carnivore spatial associations depends on forest cover to mediate antagonistic encounters with apex predators and structurally intact forests to facilitate coexistence between competing mesocarnivores. Notably, we demonstrate that jaguars play an irreplaceable role in spatially structuring mesocarnivore communities, providing novel evidence on their role as keystone species. With increasing global change, conserving both the extent and quality of tropical forests is imperative to support carnivores and preserve the spatial associations that underpin ecosystem stability and resilience.

Interactions between carnivore species: limited spatiotemporal partitioning between apex predator and smaller carnivores in a Mediterranean protected area

BACKGROUND

There is need of information on ecological interactions that keystone species such as apex predators establish in ecosystems recently recolonised. Interactions among carnivore species have the potential to influence community-level processes, with consequences for ecosystem dynamics. Although avoidance of apex predators by smaller carnivores has been reported, there is increasing evidence that the potential for competitive-to-facilitative interactions is context-dependent. In a protected area recently recolonised by the wolf Canis lupus and hosting abundant wild prey (3 ungulate species, 20-30 individuals/km2, together), we used 5-year food habit analyses and 3-year camera trapping to (i) investigate the role of mesocarnivores (4 species) in the wolf diet; (ii) test for temporal, spatial, and fine-scale spatiotemporal association between mesocarnivores and the wolf.

RESULTS

Wolf diet was dominated by large herbivores (86% occurrences, N = 2201 scats), with mesocarnivores occurring in 2% scats. We collected 12,808 carnivore detections over > 19,000 camera trapping days. We found substantial (i.e., generally ≥ 0.75, 0-1 scale) temporal overlap between mesocarnivores-in particular red fox-and the wolf, with no support for negative temporal or spatial associations between mesocarnivore and wolf detection rates. All the species were nocturnal/crepuscular and results suggested a minor role of human activity in modifying interspecific spatiotemporal partitioning.

CONCLUSIONS

Results suggest that the local great availability of large prey to wolves limited negative interactions towards smaller carnivores, thus reducing the potential for spatiotemporal avoidance. Our study emphasises that avoidance patterns leading to substantial spatiotemporal partitioning are not ubiquitous in carnivore guilds.

Understanding carnivore interactions in a cold arid trans-Himalayan landscape: What drives co-existence patterns within predator guild along varying resource gradients?

Predators compete for resources aggressively, forming trophic hierarchies that shape the structure of an ecosystem. Competitive interactions between species are modified in the human-altered environment and become particularly important where an introduced predator can have negative effects on native predator and prey species. The trans-Himalayan region of northern India has seen significant development in tourism and associated infrastructure over the last two decades, resulting in many changes to the natural setting of the landscape. While tourism, combined with unmanaged garbage can facilitate red fox (Vulpes vulpes), it also allows free-ranging dogs (Canis lupus familiaris), an introduced mesopredator to thrive, possibly more than the native red fox. We look at the little-known competitive dynamics of these two meso-carnivores, as well as their intra-guild interactions with the region's top carnivores, the snow leopard (Panthera uncia) and the Himalayan wolf (Canis lupus chanco). To study interactions between these four carnivores, we performed multispecies occupancy modeling and analyzed spatiotemporal interactions between these predators using camera trap data. We also collected scat samples to calculate dietary niche overlaps and determine the extent of competition for food resources between these carnivores. The study found that, after controlling for habitat and prey covariates, red fox site use was related positively to snow leopard site use, but negatively to dog and wolf site use. In addition, site use of the dog was associated negatively with top predators, that is, snow leopard and Himalayan wolf, while top predators themselves related negatively in their site use. As anthropogenic impacts increase, we find that these predators coexist in this resource-scarce landscape through dietary or spatiotemporal segregation, implying competition for limited resources. Our research adds to the scant ecological knowledge of the predators in the region and improves our understanding of community dynamics in human-altered ecosystems.

Risk effects cascade up to an obligate scavenger

The effects of predation risk on prey populations have been studied extensively; yet, how risk is manifested in a trophically linked guild-scavengers-has been overlooked. Risk could be particularly consequential for obligate scavengers that are vulnerable while foraging and rely on carrion provisioned by, and shared with, apex predators. We investigated whether Andean condors (Vultur gryphus) respond to predation risk in a landscape where the main source of carrion are camelids killed by pumas (Puma concolor). We hypothesized that condors would exhibit different behavioral responses to predation risk while they search, encounter, and exploit carrion. We explored condor habitat selection while flying by tracking nine birds with satellite transmitters and monitored via camera traps 41 natural carcasses and 25 experimental carrion stations. We found that condors searched for carrion in areas with a high probability of occurrence of puma kills. However, condors avoided exploiting carrion in areas featuring tall vegetation and steep slopes-selected by pumas to stalk prey-suggesting that condors manage risk primarily through the identification of safe foraging sites prior to landing. Our finding that condors avoided foraging near stalking cover for pumas highlights the importance of risk effects beyond predator-prey interactions, particularly for obligate scavengers.

Dominant carnivore loss benefits native avian and invasive mammalian scavengers

Scavenging by large carnivores is integral for ecosystem functioning by limiting the build-up of carrion and facilitating widespread energy flows. However, top carnivores have declined across the world, triggering trophic shifts within ecosystems. Here, we compare findings from previous work on predator decline against areas with recent native mammalian carnivore loss. Specifically, we investigate top-down control on utilization of experimentally placed carcasses by two mesoscavengers—the invasive feral cat and native forest raven. Ravens profited most from carnivore loss, scavenging for five times longer in the absence of native mammalian carnivores. Cats scavenged on half of all carcasses in the region without dominant native carnivores. This was eight times more than in areas where other carnivores were at high densities. All carcasses persisted longer than the three-week monitoring period in the absence of native mammalian carnivores, while in areas with high carnivore abundance, all carcasses were fully consumed. Our results reveal that top-carnivore loss amplifies impacts associated with carnivore decline—increased carcass persistence and carrion access for smaller scavengers. This suggests that even at low densities, native mammalian carnivores can fulfil their ecological functions, demonstrating the significance of global carnivore conservation and supporting management approaches, such as trophic rewilding.

A seasonal pulse of ungulate neonates influences space use by carnivores in a multi-predator, multi-prey system

The behavioral mechanisms by which predators encounter prey are poorly resolved. In particular, the extent to which predators engage in active search for prey versus incidentally encountering them has not been well studied in many systems and particularly not for neonate prey during the birth pulse. Parturition of many large herbivores occurs during a short and predictable temporal window in which young are highly vulnerable to predation. Our study aims to determine how a suite of carnivores responds to the seasonal pulse of newborn ungulates using contemporaneous global positioning system (GPS) locations of four species of predators and two species of prey. We used step-selection functions to assess whether coyotes, cougars, black bears, and bobcats encountered parturient adult female ungulates more often than expected by chance in a low-density population of mule deer and a high-density population of elk. We then assessed whether the carnivore species that encountered parturient prey more often than expected by chance did so by shifting their habitat use toward areas with a high probability of encountering neonates. None of the four carnivore species encountered GPS-collared parturient mule deer more often than expected by chance. By contrast, we determined that cougar and male bear movements positioned them in the proximity of GPS-collared parturient elk more often than expected by chance which may provide evidence of searching behavior. Although both male bears and cougars exhibited behavior consistent with active search for neonates, only male bears used elk parturition habitat in a way that dynamically tracked the phenology of the elk birth pulse suggesting that maximizing encounters with juvenile elk was a motivation when selecting resources. Our results suggest that there is high interspecific and intersexual variability in foraging strategies among large mammalian predators and their prey.

Wolf risk fails to inspire fear in two mesocarnivores suggesting facilitation prevails

Large carnivores not only supress mesocarnivores via killing and instilling fear, but also facilitate them through carrion provisioning. Hence, mesocarnivores frequently face a trade-off between risk avoidance and food acquisition. Here we used the raccoon dog and red fox in Białowieża Forest, Poland as models for investigating how large carnivores shape mesocarnivore foraging behaviour in an area with widespread large carnivore carrion provisioning. Using a giving up density experiment we quantified mesocarnivore foraging responses to wolf body odour across a landscape-scale gradient in wolf encounter rates. At locations with higher wolf encounter rates, raccoon dogs depleted feeding trays more than at plots with lower wolf encounter rates. Simulating wolf presence by adding wolf body odour caused raccoon dogs to deplete feeding trays more at locations with low wolf encounter rates, but less at locations with high wolf encounter rates. Fox foraging costs did not vary with the application of wolf body odour or wolf encounter rates. The frequency that the mesocarnivores visited experimental foraging patches was unaffected by wolf body odour or landscape level encounter rates. These results provide further evidence that large carnivore suppression can play a subordinate role to facilitation in determining mesocarnivore behaviour. The varying raccoon dog response to wolf odour across the landscape-scale gradient in wolf encounter rates shows how mesocarnivore-large carnivore interactions can be context-dependent. We suggest that rather than testing the effects of single risk cues on prey behaviour, future studies should focus on understanding how context modifies the ecological impacts of large carnivores.

Temporal refuges of a subordinate carnivore vary across rural-urban gradient

Animals exhibit variation in their space and time use across an urban-rural gradient. As the top-down influences of apex predators wane due to human-driven declines, landscape-level anthropogenic pressures are rising. Human impacts can be analogous to apex predators in that humans can drive increased mortality in both prey species and carnivores, and impact communities through indirect fear effects and food subsidies. Here, we evaluate the time use of a common mesocarnivore across an urban-rural gradient and test whether it is influenced by the intensity of the use of a larger carnivore. Using multiple camera-trap surveys, we compared the temporal response of a small carnivore, the raccoon (Procyon lotor), to the larger coyote (Canis latrans) in four study areas across Michigan that represented a gradient of pressure from humans. We found that raccoon time use varied by study area and was most unique at the rural extreme. Raccoons consistently did not shift their activity pattern in response to coyotes in the study area with the highest anthropogenic pressures despite the considerable interannual variation, and instead showed stronger responses to coyotes in more rural study areas. Temporal shifts were characterized by raccoons being more diurnal in areas of high coyote activity. We conclude that raccoons may shift time use in the presence of coyotes, dependent on the level of anthropogenic pressure. Our results highlight that the variation in raccoon time use across the entirety of the urban-rural gradient needed to be considered, as anthropogenic pressures may dominate and obscure the dynamics of this interaction.

Cats and dogs: A mesopredator navigating risk and reward provisioned by an apex predator

Successfully perceiving risk and reward is fundamental to the fitness of an animal, and can be achieved through a variety of perception tactics. For example, mesopredators may "directly" perceive risk by visually observing apex predators, or may "indirectly" perceive risk by observing habitats used by predators. Direct assessments should more accurately characterize the arrangement of risk and reward; however, indirect assessments are used more frequently in studies concerning the response of GPS-marked animals to spatiotemporally variable sources of risk and reward. We investigated the response of a mesopredator to the presence of risk and reward created by an apex predator, where risk and reward likely vary in relative perceptibility (i.e., degree of being perceptible). First, we tested whether coyotes (Canis latrans) use direct or indirect assessments to navigate the presence of mountain lions (Puma concolor; risk) and kills made by mountain lions (reward) in an area where coyotes were a common prey item for mountain lions. Second, we assessed the behavioral response of coyotes to direct encounters with mountain lions. Third, we evaluated spatiotemporal use of carrion by coyotes at kills made by mountain lions. Indirect assessments generally outperformed direct assessments when integrating analyses into a unified framework; nevertheless, our ability to detect direct perception in navigating to mountain lion kills was likely restricted by scale and sampling limitations (e.g., collar fix rates, unsampled kill sites). Rather than responding to the risk of direct encounters with mountain lions, coyotes facilitated encounters by increasing their movement rate, and engaged in risky behavior by scavenging at mountain lion kills. Coyotes appear to mitigate risk by using indirect perception to avoid mountain lions. Our predator-predator interactions and insights are nuanced and counter to the conventional predator-prey systems that have generated much of the predation risk literature.