Reconciling expert judgement and habitat suitability models as tools for guiding sampling of threatened species

Abstract

Summary Up‐to‐date knowledge on species distribution is needed for efficient biodiversity conservation and management decision‐making. Implementing efficient sampling strategies to identify previously unknown locations of species of conservation‐concern is therefore a key challenge. Both structured expert judgement and habitat suitability models may help target sampling towards areas where chances to find the species are highest. However, practitioners often object to the use of models and believe they do not result in better decisions than the subjective opinion of experts, thus potentially constraining an optimal use of available methods and information. To illustrate the potential of habitat suitability models for guiding sampling strategies, we evaluated and compared the ability of experts and models to identify important areas for the conservation of a bird species (Lanius collurio) in Luxembourg. We conducted extensive fieldwork to find as many unknown bird territories as possible according to three independent sampling strategies: (i) a sampling strategy based on structured expert judgement, (ii) a sampling strategy based on the predictions of a habitat suitability model and (iii) a general‐purpose stratified random sampling strategy used as a baseline reference. Both the expert‐based and the model‐based sampling strategies substantially outperformed the general‐purpose sampling strategy in identifying new species records. In addition, the model‐based sampling strategy performed significantly better than the expert‐based sampling strategy. Synthesis and applications. This study explicitly shows that habitat suitability models can efficiently guide field data collection towards suitable areas for species of conservation‐concern. Results may facilitate the involvement of practitioners in the development of habitat suitability models with the objective of maximizing the robustness of modelling applications in conservation practice and management decision‐making.