The Ecology of Fear: Linking Theory to Management Practice

ROOM: HCCC, Room 25B
Recent research has highlighted the importance of fear in driving animal behavior and shaping ecosystem dynamics. Predation risk can generate a “landscape of fear” for prey species, which influences the spatial and temporal patterns of prey distribution and foraging behavior. Similarly, many wildlife species fear human activity or disturbance, even when it poses no actual threat. Fear-induced changes in animal behavior can have significant consequences for physiology, fitness, and demography, often known as risk or non-lethal effects. Yet the magnitude of fear-induced effects of predators remains poorly quantified outside of experimental systems, making it challenging for managers to apply fear to solve management problems. Considering the nature and magnitude of fear effects in wildlife may improve management for game, keystone, pest, and endangered species. This symposium will bridge the divide between theory and management application of how fear structures wildlife communities and human-wildlife interactions in the modern world. We examine how humans have reshaped predator-prey interactions and landscapes of fear through predator removal and habitat modification, and consider how fear can be restored to ecosystems through both bottom-up and top-down ecosystem management. We also discuss how fear can be manipulated as a tool for managing undesirable human-wildlife interactions and conflict. Finally, we address the consequences of anthropogenic landscapes of fear and explore management strategies for mitigating the negative effects of human activity on fearful wild animals. Our diverse speakers discuss the implications and current limitations of fear for wildlife management and suggest directions for future research to increase its applicability.

12:50PM When Does Fear Matter in Wildlife Management?
  Kaitlyn M. Gaynor; Justine A. Smith
Animals experience varying levels of risk as they navigate heterogeneous landscapes, and their behavioral responses to perceived risk can ultimately structure ecosystems. The concept of “fear” has recently become central to describing patterns of risk, risk perception, and response. In this talk, we introduce our symposium, “The Ecology of Fear: Linking Theory to Management Practice.” We first situate the concept of fear within the existing predator-prey literature and propose a theoretical framework to guide the symposium. In this framework, we explore how the physical landscape, species distributions, and human activities generate patterns of risk. We examine how perceived risk, or fear, influences wildlife distributions and anti-predator behavior. By understanding how animals incorporate fear into decision-making, we can better predict the outcomes of management decisions and understand the relative importance of fear as a management consideration across species and systems. Using our framework as a guide, we explore how management of habitat, predators, prey, and human activity can reshape fear dynamics and change wildlife behavior and population distributions in accordance with management objectives. For example, we discuss how producers can manipulate fear to reduce human-wildlife conflict using nonlethal measures; how behavioral responses to hunting can inform management of game species; and how enhancing fear may mitigate effects of invasive predators to naïve prey. Finally, we introduce the case studies that will be featured in this symposium, positioning these talks in our fear framework.
1:10PM Manipulating the Landscape of Fear and Nutrition to Reach Management Objectives for Ungulates
  Sophie Gilbert; Kayte Groth; John Guthrie; Jon Horne; Shane Roberts; Mark Hurley
As humans and wildlife come into ever-closer contact in the changing modern world, managers are looking for new tools to decrease the costs and increase the benefits of wildlife to stakeholders and the public in human-modified landscapes. Predator-prey theory offers promising pathways for designing real-world tools for managers. However, translating the mechanisms that generate and maintain “landscapes of fear” induced by predators into practical “management levers” that are useful to managers requires rigorous testing, which is seldom carried out. In an ongoing series of experiments in agricultural landscapes in Idaho, we manipulated both the landscape of fear and landscape of nutrition using multiple mechanisms, and measured elk and white-tailed deer response using 15-minute GPS relocations, camera traps, and drone surveys. We present study design, current results, and future directions for research. We anticipate providing several new tools for managers seeking to change ungulate behavior in human-modified landscapes.
1:30PM Fear, Fire, and Fawning: Foraging Tradeoffs in White-tailed Deer
  Michael J. Cherry; Daniel A. Crawford; L. Mike Conner
Predators can induce individual prey to express behavioral traits but identifying when those traits influence population‐level processes remains a challenge. Maternal investment is an intuitive link between predator‐mediated traits of individuals and population‐level processes because maternal investment often decreases with predation risk, and influences processes such as neonatal growth and recruitment. In frequently burned systems, ungulates typically move into recently burned areas to exploit increased forage quality, but the effect of the removal of cover on predation risk has received less attention. When the quality of safety and food are inversely related in space, prey face tradeoffs that could scale to population-level processes if selection of safer, but less productive areas represents a nutritional consequence, particularly during maternal investment. We examined the effects of fire on the spatial ecology of white-tailed deer and coyotes during fawning in a frequently burned pine savanna. Adult female deer selected forests with greater time since fire and avoided recently burned areas. Fawns selected bed sites that provided greater concealment cover. Adult female deer increased vigilance in more recently burned areas and habitat patches with lower concealment cover and a higher probability of use by coyotes. Our work describes fire-mediated foraging tradeoffs in deer characterized by risk adverse behaviors, such as the avoidance of high‐quality forage in recently burned forests and open areas. Future work should examine if these antipredator behaviors affect prey nutritional condition, demographic rates or density.
1:50PM The Landscape of Fear in Yellowstone: Is It Real, and If So, Does It Even Matter?
  Michel T. Kohl; Peter J. Mahoney; Lacy M. Smith; Sarah H. Hoy; P. J. White; Travis Wyman; Daniel R. Stahler; Douglas W. Smith; Abby Nelson; Daniel R. MacNulty
The Landscape of Fear (LOF) has been proposed as a unifying concept in ecology that explains animal behavior, population dynamics, and trophic interactions. Moreover, it has been suggested that the LOF rivals the importance of direct killing in structuring food webs and modifying ecosystem function. Yet, recent work in Yellowstone National Park demonstrates the elk LOF is a highly dynamic surface that fluctuates with the daily activity pattern of wolves. This provides opportunities for prey to adequately manage predation risk at a fine scale. Alternatively, prey may forgo fine-scale anti-predator behaviors in favor of large-scale spatial avoidance. We tested this hypothesis using long-term elk data from northern Yellowstone. Prior to wolf reintroduction in 1995-96, >60% of the elk population resided in the upper reaches of the northern Yellowstone winter range (hereafter upper sector), but since 2012-16, elk numbers in the upper sector declined to <15% of the overall population. We tested whether this large-scale shift in elk distribution was due to behavioral or demographic processes. We found that winter-site fidelity was constant across all study periods including periods of no wolves, peak wolf abundance, and low wolf abundance. This suggests that elk did not shift their home ranges in response to wolves, and thus, is inconsistent with a behavioral response to wolves. In contrast, elk within the upper sector were older, and experienced lower survival and lower recruitment than elk in the lower sector, consistent with a demographic effect. Together, these results suggest that a fine-scale spatio-temporal response to risk may not generate large-scale distributional changes. As such, we demonstrate that demographic processes, rather than fear effects, may be the primary driver of prey spatial structures at a large-scale.
2:10PM Fear in a Landscape Dominated By Introduced Predators
  Thomas Newsome
There is growing recognition that fear can shape animal behaviour and ecosystem dynamics. The fear of being eaten, in particular, is thought to influence the strength and direction of trophic cascades over multiple trophic levels. Examples of such effects have typically come from strongly interacting mammalian predator and prey dyads that have a long history of co-evolution. However, invasive mammalian predators have been introduced either deliberately or inadvertently into almost every continent on earth, and at least 738 species have suffered negative impacts as a result, including population declines and extinctions. This begs the question of whether prey species can recognise invasive mammalian predators, and how long it takes for fear to evolve into a process that enables prey to escape the threat of going extinct from hyperpredation. In this talk I will provide an overview of how fear is or isn’t shaping animal behaviour and ecosystem dynamics in systems with invasive mammalian predators. In doing so, I will draw on examples globally to assess whether there is any evidence that prey species can adapt to the arrival of a new predator, and thus whether fear itself can evolve rapidly into a dominant force that shapes animal behaviour and ecosystem dynamics in the Anthropocene. The results have important implications for understanding how to mitigate the impacts of invasive predators on vulnerable prey communities throughout the world.
2:30PM Refreshment Break
3:20PM Effects of Non-Motorized Recreation on Mammals in a Fragmented Urban Reserve System
  Courtney L. Larson; Sarah E. Reed; Adina M. Merenlender; Kevin R. Crooks
Outdoor recreation is often a principal reason for the establishment of and continued support for protected areas, and it is commonly assumed that recreation is compatible with wildlife conservation. Despite increasing evidence that recreation has negative impacts on animal communities, questions remain about the magnitude and shape of the effects of recreation. In urban natural areas, participation in outdoor recreation can be high, putting pressure on remaining natural habitat fragments that support populations of native species. We examined the habitat use of the mammal community in San Diego, California across a gradient of recreation intensity. We selected 14 reserves that are part of a Multiple Species Conservation Plan and installed motion-triggered cameras at 92 sampling points to detect human recreation activity (hiking, biking, dog-walking, and horseback riding) and mammalian habitat use. We sampled for more than 10,000 camera days and amassed greater than 2 million photos, including detections of ten native mammal species. We are using an occupancy framework to model the probability of habitat use and detection probability, with visitation rates as a covariate along with other variables such as vegetation composition and nearby housing density. We are identifying specific thresholds of recreation that are associated with shifts in habitat use. Understanding the critical levels of recreation activity will enable managers to identify and target areas where high visitation rates are negatively impacting wildlife, facilitating management interventions to divert or concentrate visitation into other areas.
3:40PM Fear of Humans Affects Terrestrial Food Webs
  Justin P. Suraci; Michael Clinchy; Liana Zanette; Christopher C. Wilmers
As human land use continues to encroach on wildlife habitat, understanding how human presence affects terrestrial communities will be critical to managing our impacts. Humans are a major source of mortality for many wildlife populations, particularly large carnivores, meaning that fear of humans may be widespread. Indeed, many large carnivores exhibit fear responses to humans, and resulting changes in large carnivore behavior could have cascading effects by altering their interactions with their prey. Yet it remains unknown whether lower trophic level species will benefit from human suppression of carnivores, or be similarly impacted by the fear of humans. We conducted a landscape-scale field experiment to test the effects of the fear of humans on terrestrial food webs, presenting human or control vocalizations over replicate 1-km2 patches of habitat in the Santa Cruz Mountains of central California. We tested for changes in large carnivore (Puma, Puma concolor) habitat use via GPS collars on seven individuals, and monitored for changes in local abundance and behavior among lower trophic level species using camera traps (mesocarnivores) and live trapping grids (small mammals). Pumas significantly altered their habitat use during human treatments relative to controls, avoiding areas of high perceived risk. We found similar suppressive effects of fear of humans on mesocarnivore behavior, with species variously reducing overall activity levels, reducing foraging intensity, and shifting towards more nocturnal activity during the human treatment, relative to controls. Small mammal trapping results suggest that changes in mesocarnivore behavior in response to perceived human risk may additionally affect their interactions with their small mammal prey. This work indicates that the fear of humans may be pervasive across terrestrial food webs, with implications for management of human activity in protected areas meant to preserve wildlife habitat.
4:00PM Experimental Evidence for Fear as a Tool to Manage Human-Wildlife Conflict
  Julie K. Young; Stewart Breck
Fear plays a prominent role in the life history of wild animals by influencing how animals react to novel stimulants and potentially dangerous encounters with dominant species. Our aim is to take advantage of inherent and learned fear-based behavior to optimize the development and use of non-lethal tools to reduce human-wildlife conflict. These non-lethal tools typically prevent conflict by using fear to change or prevent wildlife from performing specific behavior. Fear can be stimulated using novel objects deemed frightening by particular species (e.g., fladry for wolves) or by using guarding animals (e.g., dogs) or people in a threatening way. We use data from three experimental studies and two observational studies on human-carnivore conflict to evaluate the role of fear in shaping predator behavior and reducing conflict. Results suggest that while fear can be a useful tactic to reduce conflict, behavioral differences within and between species limit efficacy of non-lethal tools that rely solely on fear-induced behavioral responses.
4:20PM Designing Predation Risk Studies for Improved Inference
  Laura R. Prugh; Kelly J. Sivy; Peter Mahoney; Taylor Ganz; Mark Ditmer; Madelon van de Kerk; Sophie Gilbert; Robert Montgomery
A recent review of predation risk studies highlighted a lack of standardization in approaches used to quantify predation risk. We provide an overview of four major study design considerations involved in assessing predation risk, highlighting how different design choices can impact the strength and scope of inference. First, spatial and temporal scales need to be appropriate to the systems and questions of interest. Considerations include seasonal changes in the nutritional landscape and body condition of prey, movement patterns (e.g., migrations), diurnal activity cycles, and dynamics of the nutritional landscape. Second, multiple signals of risk need to be considered: animals are often evaluating risk from multiple predators simultaneously, and they use multiple senses (vision, hearing, and smell) to evaluate predation risk. Experiments that manipulate signals of predation risk (e.g., auditory playbacks or application of predator scent) are promising approaches, but the dosage of cues needs to be carefully considered. Third, the choice of response variable in predation risk studies can strongly affect inference. Common response metrics include measures of behavior, habitat selection, physiology, and demography. We examine pros and cons of common methods to quantify these responses, highlighting the potential for experimental approaches (e.g., giving-up density trials) to improve strength of inference for behavioral responses. Fourth, emerging technologies may substantially improve our ability to assess risk. We discuss several promising technologies, such as animal-borne video, unmanned aerial vehicles, and physiological sensors. We conclude with general recommendations for study design, with the goal of furthering efforts to standardize and increase rigor of predation risk studies.
4:40PM Panel Discussion

Organizers: Justine A. Smith, University of California, Berkeley, CA; Kaitlyn M. Gaynor, University of California, Berkeley, Cal Sophie L. Gilbert, University of Idaho, Moscow, ID
Supported by: Wildlife Damage Management Working Group

Location: Huntington Convention Center of Cleveland Date: October 9, 2018 Time: 12:50 pm - 5:00 pm