Least cost path analysis for urban greenways planning: a test with moths and birds across two habitats and two cities.

International audience

English. 1.One of the major planning tools to respond to urban landscape fragmentation is the development of ecological corridors, that is, interconnected networks of urban green and blue spaces. Least‐cost paths (LCP) appear to be an easy and appropriate resistance‐based modelling method to respond to urban planners’ needs. However, the ecological validation of urban corridors using LCP is rarely performed and needs to be generalized to different species, habitats and cities.2.We developed an experimental design to test the efficiency of LCP predictions to detect highly connecting landscape contexts that facilitate individual movements compared to movements in less connecting landscape contexts. We deliberately assigned LCP analysis parameters based on the scientific literature and expert knowledge to test a method potentially easy to use for urban stakeholders. To extend the validation, we applied our LCP model to two biological taxa with different habitat requirements: grassland‐dwelling moths and forest‐dwelling passerines, and to two medium‐sized cities.3.We used mark–release–recapture (MRR) methods for moths and playback recall protocols for passerines to compare the patterns of individual movement between two contrasted connectivity contexts determined by the presence and absence of modelled LCPs. MRR protocol estimated movement rates between herbaceous patches and the two contrasted connectivity contexts. Playback recall protocol consisted in attracting individuals from wooded patches to the two contrasted connectivity contexts. A movement was considered facilitated, when displacement was rapidly engaged and individuals moved a long distance from their wooded patch.4.Moth and passerine movement patterns differed between the two connectivity contexts: moth recapture rates were higher in highly connecting contexts than in less connecting contexts. For passerine birds, responses to playback recalls were faster and movement distance longer in highly connecting contexts. All results support the hypothesis that both taxa were more prone to move in corridors modelled by LCP.5.Synthesis and applications. The convergence of the results for different biological models and across cities strengthens the relevance of LCP analysis for planning urban greenways and provides guidelines for landscape planners in the development of these corridors to favour the movement and survival of multiple urban species.