查看更多>>摘要：? 2022 The AuthorsInclusion of local, expert, or Indigenous knowledge about wildlife populations and their habitats can inform wildlife research, while also increasing knowledge holder engagement and support for wildlife conservation decisions. However, experiential wildlife knowledge accumulated over time through the personal observations of knowledge holders differs from other data based on systematic observations collected through standardized methodology such as telemetry locations or field surveys. Differences in the form and the function of these two types of wildlife information makes combining them into a single comprehensive analysis more easily encouraged than accomplished. Here, we systematically review primary literature that interweaves the experiential wildlife knowledge of diverse knowledge holders into quantitative, mixed methods analysis of terrestrial vertebrate populations and their habitats. Forty-nine studies that met our selection criteria were distributed around the globe and across terrestrial vertebrate species, but most frequently were situated in Australia, Canada, and United States and focused on large, harvested mammals including ungulates, carnivores, primates, and elephants. The most common descriptor of knowledge holders was hunters/trappers, with academic experts and community members also common. The most common analyses interweaved experiential wildlife knowledge as point observations in habitat models or as habitat covariates in habitat selection analyses. Local knowledge was also included, less frequently, in species distribution models, population models, and occupancy models. Most articles accounted for bias and uncertainty either in the knowledge elicitation stage through study design or knowledge holder selection, or in the analysis stage through regression methods. Most articles that assessed model success did so through comparison to independently collected telemetry locations or field survey data. There was wide variation in self-reported success, with the majority of authors offering neutral or positive assessments and many discussing study-specific factors contributing to model performance. Our overall assessment of these 49 studies, including 6 examples described in more detail, highlight several key challenges and solutions related to the inclusion of local, expert, and Indigenous knowledge into quantitative wildlife habitat and population analyses related to i) the incorporation of uncertainty, bias, reliability, and variation in experiential wildlife knowledge, ii) matching the scale of experiential wildlife knowledge to scale of study objectives, and iii) the appropriate use, communication, and application of experiential wildlife knowledge, including issues of consent, member checking, and knowledge co-production. We conclude with several recommendations intended to better standardize and communicate uncertainty, increase the involvement of knowledge holders in multiple stages of the research, improve validity assessment through multiple model comparisons and triangulation, and encourage more careful consideration of intellectual property protection and research ethics.
NSTL
Elsevier