Generalized nonlinearity in animal ecology: Research, review, and recommendations.

Generalized linear models (GLMs) are an integral tool in ecology. Like general linear models, GLMs assume linearity, which entails a linear relationship between independent and dependent variables. However, because this assumption acts on the link rather than the natural scale in GLMs, it is more easily overlooked. We reviewed recent ecological literature to quantify the use of linearity. We then used two case studies to confront the linearity assumption via two GLMs fit to empirical data. In the first case study we compared GLMs to generalized additive models (GAMs) fit to mammal relative abundance data. In the second case study we tested for linearity in occupancy models using passerine point-count data. We reviewed 162 studies published in the last 5 years in five leading ecology journals and found less than 15% reported testing for linearity. These studies used transformations and GAMs more often than they reported a linearity test. In the first case study, GAMs strongly out-performed GLMs as measured by AIC in modeling relative abundance, and GAMs helped uncover nonlinear responses of carnivore species to landscape development. In the second case study, 14% of species-specific models failed a formal statistical test for linearity. We also found that differences between linear and nonlinear (i.e., those with a transformed independent variable) model predictions were similar for some species but not for others, with implications for inference and conservation decision-making. Our review suggests that reporting tests for linearity are rare in recent studies employing GLMs. Our case studies show how formally comparing models that allow for nonlinear relationships between the dependent and independent variables has the potential to impact inference, generate new hypotheses, and alter conservation implications. We conclude by suggesting that ecological studies report tests for linearity and use formal methods to address linearity assumption violations in GLMs.

Characteristics of wild moose (Alces alces) vocalizations.

Moose are a popular species with recreationists but understudied acoustically. We used publicly available videos to characterize and quantify the vocalizations of moose in New Hampshire separated by age/sex class. We found significant differences in peak frequency, center frequency, bandwidth, and duration across the groups. Our results provide quantification of wild moose vocalizations across age/sex classes, which is a key step for passive acoustic detection of this species and highlights public videos as a potential resource for bioacoustics research of hard-to-capture and understudied species.

Animal tracking moves community ecology: Opportunities and challenges.

Individual decisions regarding how, why and when organisms interact with one another and with their environment scale up to shape patterns and processes in communities. Recent evidence has firmly established the prevalence of intraspecific variation in nature and its relevance in community ecology, yet challenges associated with collecting data on large numbers of individual conspecifics and heterospecifics have hampered integration of individual variation into community ecology. Nevertheless, recent technological and statistical advances in GPS-tracking, remote sensing and behavioural ecology offer a toolbox for integrating intraspecific variation into community processes. More than simply describing where organisms go, movement data provide unique information about interactions and environmental associations from which a true individual-to-community framework can be built. By linking the movement paths of both conspecifics and heterospecifics with environmental data, ecologists can now simultaneously quantify intraspecific and interspecific variation regarding the Eltonian (biotic interactions) and Grinnellian (environmental conditions) factors underpinning community assemblage and dynamics, yet substantial logistical and analytical challenges must be addressed for these approaches to realize their full potential. Across communities, empirical integration of Eltonian and Grinnellian factors can support conservation applications and reveal metacommunity dynamics via tracking-based dispersal data. As the logistical and analytical challenges associated with multi-species tracking are surmounted, we envision a future where individual movements and their ecological and environmental signatures will bring resolution to many enduring issues in community ecology.

An apex carnivore's life history mediates a predator cascade.

Apex predators can shape communities via cascading top-down effects, but the degree to which such effects depend on predator life history traits is largely unknown. Within carnivore guilds, complex hierarchies of dominance facilitate coexistence, whereby subordinate species avoid dominant counterparts by partitioning space, time, or both. We investigated whether a major life history trait (hibernation) in an apex carnivore (black bears Ursus americanus) mediated its top-down effects on the spatio-temporal dynamics of three sympatric mesocarnivore species (coyotes Canis latrans, bobcats Lynx rufus, and gray foxes Urocyon cinereoargenteus) across a 15,000 km2 landscape in the western USA. We compared top-down, bottom-up, and environmental effects on these mesocarnivores using an integrated modeling approach. Black bears exerted top-down effects that varied as a function of hibernation and were stronger than bottom-up or environmental impacts. High black bear activity in summer and fall appeared to buffer the most subordinate mesocarnivore (gray foxes) from competition with dominant mesocarnivores (coyotes and bobcats), which were in turn released by black bear hibernation in winter and early spring. The mesocarnivore responses occurred in space (i.e., altered occupancy and site visitation intensity) rather than time (i.e., diel activity patterns unaffected). These results suggest that the spatio-temporal dynamics of mesocarnivores in this system were principally shaped by a spatial predator cascade of interference competition mediated by black bear hibernation. Thus, certain life history traits of apex predators might facilitate coexistence among competing species over broad time scales, with complex implications for lower trophic levels.

A rare 300 kilometer dispersal by an adult male white-tailed deer.

Despite the key roles that dispersal plays in individual animal fitness and meta-population gene flow, it remains one of the least understood behaviors in many species. In large mammalian herbivores, dispersals might span long distances and thereby influence landscape-level ecological processes, such as infectious disease spread. Here, we describe and analyze an exceptional long-distance dispersal by an adult white-tailed deer (Odocoileus virginianus) in the central United States. We also conducted a literature survey to compare the dispersal to previous studies. This dispersal was remarkable for its length, duration, and the life history stage of the dispersing individual. Dispersal is typical of juvenile deer seeking to establish postnatal home ranges, but this dispersal was undertaken by an adult male (age = 3.5). This individual dispersed ~300 km over a 22-day period by moving, on average, 13.6 km/day and achieving a straight-line distance of ~215 km, which was ~174 km longer than any other distance recorded for an adult male deer in our literature survey. During the dispersal, which occurred during the hunting season, the individual crossed a major river seven times, an interstate highway, a railroad, and eight state highways. Movements during the dispersal were faster (mean = 568.1 m/h) and more directional than those during stationary home range periods before and after the dispersal (mean = 56.9 m/h). Likewise, movements during the dispersal were faster (mean = 847.8 m/h) and more directional at night than during the day (mean = 166.4 m/h), when the individual frequently sheltered in forest cover. This natural history event highlights the unpredictable nature of dispersal and has important implications for landscape-level processes such as chronic wasting disease transmission in cervids. More broadly, our study underscores how integrating natural history observations with modern technology holds promise for understanding potentially high impact but rarely recorded ecological events.

Dynamic rodent behavioral response to predation risk: implications for disease ecology.

Prey modify their behavior in response to variation in predation risk, and such modifications can affect trophic processes such as disease transmission. However, variation in predation risk is complex, arising from direct risk from the predator itself and indirect risk due to the environment. Moreover, direct risk typically stems from multiple predators and varies over timescales (e.g., a predator nearby vs. its seasonal activities). We implemented a field-based experiment to disentangle these sources of risk and relate them to antipredator behavior in rodents. We modeled rodent occurrence and activity as a function of short- and long-term risk from a primary predator, red foxes (Vulpes vulpes), long-term risk from a second predator, coyotes (Canis latrans), and environmental variables. We found that long-term red fox activity strongly reduced rodent occurrence and that cues of nearby red fox presence decreased rodent activity by > 50%. In addition, this activity reduction was dynamic in that varied according to the background level of long-term red fox activity. Importantly, rodents did not respond to environmental variables (moonlight, temperature, and habitat) or long-term coyote activity. These results bear upon recent work that suggests predators can alter tick-borne disease dynamics via induced antipredator behavior of rodents, which are hosts for pathogens and ticks. Specifically, our study corroborates the hypothesis that red foxes act as important proximal agents in regulating tick-borne diseases by reducing rodent activity. More generally, this study highlights the need to consider the dynamic nature of prey antipredator response across landscapes with variable long-term predation risk.

Spatial variation in leopard (Panthera pardus) site use across a gradient of anthropogenic pressure in Tanzania's Ruaha landscape.

Understanding large carnivore occurrence patterns in anthropogenic landscapes adjacent to protected areas is central to developing actions for species conservation in an increasingly human-dominated world. Among large carnivores, leopards (Panthera pardus) are the most widely distributed felid. Leopards occupying anthropogenic landscapes frequently come into conflict with humans, which often results in leopard mortality. Leopards' use of anthropogenic landscapes, and their frequent involvement with conflict, make them an insightful species for understanding the determinants of carnivore occurrence across human-dominated habitats. We evaluated the spatial variation in leopard site use across a multiple-use landscape in Tanzania's Ruaha landscape. Our study region encompassed i) Ruaha National Park, where human activities were restricted and sport hunting was prohibited; ii) the Pawaga-Idodi Wildlife Management Area, where wildlife sport hunting, wildlife poaching, and illegal pastoralism all occurred at relatively low levels; and iii) surrounding village lands where carnivores and other wildlife were frequently exposed to human-carnivore conflict related-killings and agricultural habitat conversion and development. We investigated leopard occurrence across the study region via an extensive camera trapping network. We estimated site use as a function of environmental (i.e. habitat and anthropogenic) variables using occupancy models within a Bayesian framework. We observed a steady decline in leopard site use with downgrading protected area status from the national park to the Wildlife Management Area and village lands. Our findings suggest that human-related activities such as increased livestock presence and proximity to human households exerted stronger influence than prey availability on leopard site use, and were the major limiting factors of leopard distribution across the gradient of human pressure, especially in the village lands outside Ruaha National Park. Overall, our study provides valuable information about the determinants of spatial distribution of leopards in human-dominated landscapes that can help inform conservation strategies in the borderlands adjacent to protected areas.

The many faces of fear: a synthesis of the methodological variation in characterizing predation risk.

Predators affect prey by killing them directly (lethal effects) and by inducing costly antipredator behaviours in living prey (risk effects). Risk effects can strongly influence prey populations and cascade through trophic systems. A prerequisite for assessing risk effects is characterizing the spatiotemporal variation in predation risk. Risk effects research has experienced rapid growth in the last several decades. However, preliminary assessments of the resultant literature suggest that researchers characterize predation risk using a variety of techniques. The implications of this methodological variation for inference and comparability among studies have not been well recognized or formally synthesized. We couple a literature survey with a hierarchical framework, developed from established theory, to quantify the methodological variation in characterizing risk using carnivore-ungulate systems as a case study. Via this process, we documented 244 metrics of risk from 141 studies falling into at least 13 distinct subcategories within three broader categories. Both empirical and theoretical work suggest risk and its effects on prey constitute a complex, multi-dimensional process with expressions varying by spatiotemporal scale. Our survey suggests this multi-scale complexity is reflected in the literature as a whole but often underappreciated in any given study, which complicates comparability among studies and leads to an overemphasis on documenting the presence of risk effects rather than their mechanisms or scale of influence. We suggest risk metrics be placed in a more concrete conceptual framework to clarify inference surrounding risk effects and their cascading effects throughout ecosystems. We recommend studies (i) take a multi-scale approach to characterizing risk; (ii) explicitly consider 'true' predation risk (probability of predation per unit time); and (iii) use risk metrics that facilitate comparison among studies and the evaluation of multiple competing hypotheses. Addressing the pressing questions in risk effects research, including how, to what extent and on what scale they occur, requires leveraging the advantages of the many methods available to characterize risk while minimizing the confusion caused by variability in their application.

AIC and the challenge of complexity: A case study from ecology.

Philosophers and scientists alike have suggested Akaike's Information Criterion (AIC), and other similar model selection methods, show predictive accuracy justifies a preference for simplicity in model selection. This epistemic justification of simplicity is limited by an assumption of AIC which requires that the same probability distribution must generate the data used to fit the model and the data about which predictions are made. This limitation has been previously noted but appears to often go unnoticed by philosophers and scientists and has not been analyzed in relation to complexity. If predictions are about future observations, we argue that this assumption is unlikely to hold for models of complex phenomena. That in turn creates a practical limitation for simplicity's AIC-based justification because scientists modeling such phenomena are often interested in predicting the future. We support our argument with an ecological case study concerning the reintroduction of wolves into Yellowstone National Park, U.S.A. We suggest that AIC might still lend epistemic support for simplicity by leading to better explanations of complex phenomena.

Spatial patterns of African ungulate aggregation reveal complex but limited risk effects from reintroduced carnivores.

The "landscape of fear" model, recently advanced in research on the non-lethal effects of carnivores on ungulates, predicts that prey will exhibit detectable antipredator behavior not only during risky times (i.e., predators in close proximity) but also in risky places (i.e., habitat where predators kill prey or tend to occur). Aggregation is an important antipredator response in numerous ungulate species, making it a useful metric to evaluate the strength and scope of the landscape of fear in a multi-carnivore, multi-ungulate system. We conducted ungulate surveys over a 2-year period in South Africa to test the influence of three broad-scale sources of variation in the landscape on spatial patterns in aggregation: (1) habitat structure, (2) where carnivores tended to occur (i.e., population-level utilization distributions), and (3) where carnivores tended to kill ungulate prey (i.e., probabilistic kill site maps). We analyzed spatial variation in aggregation for six ungulate species exposed to predation from recently reintroduced lion (Panthera leo) and spotted hyena (Crocuta crocuta). Although we did detect larger aggregations of ungulates in "risky places," these effects existed primarily for smaller-bodied (<150 kg) ungulates and were relatively moderate (change of 4 individuals across all habitats). In comparison, ungulate aggregations tended to increase at a slightly lower rate in habitat that was more open. The lion, an ambush (stalking) carnivore, had stronger influence on ungulate aggregation than the hyena, an active (coursing) carnivore. In addition, places where lions tended to kill prey had a greater effect on ungulate aggregation than places where lions tended to occur, but an opposing pattern existed for hyena. Our study reveals heterogeneity in the landscape of fear and suggests broad-scale risk effects following carnivore reintroduction only moderately influence ungulate aggregation size and vary considerably by predator hunting mode, type of predation risk, and prey species.

A new 'view' of ecology and conservation through animal-borne video systems.

Over the past three decades, technological advances for monitoring wild animals have expanded the ability of ecologists to study animal behavior and space use. Currently, researchers are deploying animal-borne video and environmental data collection systems (AVEDs), which enable researchers to see what the animal sees in the field. AVEDs record fine-scale movements as well as features of the surrounding environment and thus provide essential context for understanding animal decisions and interactions with other individuals. These fine-scale data are often crucial for understanding potential conservation threats to species of concern. Here, we discuss the development and research potential offered by AVEDs. The benefits of AVEDs are greatest in hypothesis-driven studies that require a fine-scale perspective that other technologies cannot offer.