Population and community consequences of perceived risk from humans in wildlife

Human activities catalyze risk avoidance behaviors in wildlife across taxa and systems. However, the broader ecological significance of human-induced risk perception remains unclear, with a limited understanding of how phenotypic responses scale up to affect population or community dynamics. We conducted a comprehensive literature review of non-consumptive effects (NCE; population effects) and trait-mediated indirect effects (TMIE; community effects) of anthropogenic disturbances. This dataset includes all papers identified from the comprehensive review of the different types of human-induced behavioral and physiological phenotypic change and their influence on vital rates and population parameters in wildlife. All papers in this database tested for a human-induced NCE or TMIE in wildlife but not all found evidence for an effect. Many of the papers did not explicitly measure the presumed phenotypic change linking human activity to vital rates or population parameters. The authors, paper title, journal, publication year, type of human disturbance, species, system, phenotypic response measured, demographic response measured, if a demographic effect was found, and whether an NCE or TMIE was tested are all included in the dataset. In addition, we include the source of the paper in our dataset (i.e. whether it came up in our Web of Science search, as a citing paper of Frid and Dill (2002), or in a review paper on human-induced fear in wildlife; column A). The papers in which multiple NCE or TMIE pathways were tested may have multiple values in a single cell. Papers are sorted alphabetically by author. Evidence for human-induced NCEs and TMIEs is mixed, with half of published studies finding a relationship between human activities, phenotypic change, and population outcomes. Strong research biases in taxa, systems, human disturbance type, and demographic measures prevent unified inference about the prevalence of population responses to human activities. Coexistence with and conservation of wildlife requires additional research linking human-induced phenotypic change to population and community outcomes.

Methods
To evaluate the evidence linking perceived risk from humans and associated phenotypic responses to downstream ecological consequences, we comprehensively reviewed the literature on human-induced NCEs and TMIEs. Papers evaluated in our comprehensive review were identified from three sources: 1) two Web of Science searches; 2) papers citing Frid and Dill (2002; https://doi.org/10.5751/ES-00404-060111*), 3) relevant papers found within review papers identified from (1) and (2). The specifics of the Web of Science searches are provided below. 

We initially scanned all papers for three criteria in a progressive manner; to advance, each paper had to be empirical, examine an effect of anthropogenic disturbance, and reference a topic related to risk. Papers meeting all three criteria were further filtered to those that evaluated a human-induced risk effect and tested for an effect beyond a phenotypic response (i.e. a change in fitness, fecundity, survival, density, abundance, or population growth). For these papers, we recorded the nature of the response, type of human-induced cue, species, system, and if a demographic effect was found. The scoring of these papers was led by one author and evaluated for accuracy by two other authors (including the lead author). Where disagreements arose, papers were further co-reviewed, with final decisions made by the lead author. Of 1769 papers reviewed, 92 tested for an NCE or TMIE, and only 57 linked this effect to an explicitly measured phenotypic response.

Web of Science Search 1

"nonconsumptive" OR "non-consumptive" OR "ecology of fear" OR "landscape of fear" OR "trait mediated" OR "trait-mediated" OR "behaviorally mediated" OR "behaviorally-mediated" OR "interaction modification" OR "interaction-modification" OR "non-trophic interaction" OR "nontrophic interaction"

AND:

human OR anthropogenic OR recreat* OR hunt* OR disturbance OR "human footprint" OR roads OR "energy development" or infrastructure

Web of Science Search 2

"sublethal" OR "sub-lethal"

AND:

human OR anthropogenic OR recreat* OR hunt* OR disturbance OR "human footprint" OR roads OR "energy development" or infrastructure

AND:

predat* OR risk OR fear

NOT:

pestic* OR herbic* OR toxi* OR chemic* OR salin* OR drug* OR radiat* OR nitr* OR lead OR caffeine OR pharma* OR plastic OR hypox* OR mercury

NOT:

"nonconsumptive" OR "non-consumptive" OR "ecology of fear" OR "landscape of fear" OR "trait-mediated" OR "trait mediated" OR "behaviorally mediated" OR "behaviorally-mediated" OR "interaction modification" OR "interaction-modification" OR "Non-trophic interaction" OR "nontrophic interaction"

*Citations of: Frid, A. & Dill, L.M. (2002). Human-caused disturbance stimuli as a form of predation risk. Conservation Ecology, 6(1):11.