Comparing wildlife habitat suitability models based on expert opinion with camera trap detections
Cindy Hurtado1,2, Victoria Hemming3, Cole Burton1
1Department of Forest Resources Management, Faculty of Forestry, University of British Columbia, Vancouver, British Columbia, Canada.
Summary
Wildlife habitat suitability models (HSMs) for specialist species showed better accuracy than generalist species. Expert experience and feedback improved model accuracy, especially for understudied species, highlighting the need for local knowledge in conservation efforts.
Area of Science:
- Ecology
- Conservation Biology
- Wildlife Management
Background:
- Expert knowledge is crucial for developing wildlife habitat suitability models (HSMs) used in conservation.
- The consistency and accuracy of expert-based HSMs have been questioned, necessitating validation against empirical data.
- The analytic hierarchy process is one method used for eliciting expert knowledge in HSM development.
Purpose of the Study:
- To assess the correspondence between expert-based HSMs and empirical data (camera-trap detections) for four felid species.
- To evaluate the influence of study species (specialist vs. generalist) and expert attributes on model accuracy.
- To determine if aggregating expert responses and using iterative feedback improves HSM performance.
Main Methods:
- Generated 160 expert-based HSMs using the analytic hierarchy process for two forest specialists (ocelot, margay) and two generalists (Pampas cat, puma).
- Utilized generalized linear models to compare HSM predictions with camera-trap survey data.
- Assessed the impact of participant experience, iterative feedback, and group size on model correspondence (AUC).
Main Results:
- HSMs for specialist species demonstrated higher correspondence with detections (AUC >0.7) than for generalists (AUC <0.7).
- Model correspondence improved with increased participant experience in the study area, particularly for the Pampas cat.
- Iterative feedback and aggregating expert judgments enhanced model accuracy, with optimal group size leveling off at five experts.
Conclusions:
- Habitat specialization positively correlates with the accuracy of expert-based HSMs when compared to empirical data.
- Incorporating local expert knowledge and validating models with empirical data are essential, especially for understudied and generalist species.
- Aggregating expert opinions and utilizing feedback loops can improve the reliability of conservation models.
Related Concept Videos
02:31Habitat Fragmentation
17.7K
Habitat fragmentation describes the division of a more extensive, continuous habitat into smaller, discontinuous areas. Human activities such as land conversion, as well as slower geological processes leading to changes in the physical environment, are the two leading causes of habitat fragmentation. The fragmentation process typically follows the same steps: perforation, dissection, fragmentation, shrinkage, and attrition.
17.7K
00:30What are Populations and Communities?
34.2K
Overview
34.2K