Mapping climate adaptation corridors for biodiversity—A regional-scale case study in Central America

the species on which our conductance surface is based already live across lowland and high- land habitats in our region (i.e., wide temperature gradients). We next applied least cost path modeling based on our conductance surface using the ‘shortestPath’ function in the ‘gdistance’ R package [56]. Whereas many landscape connectiv- ity studies map least cost paths between each pair of start and end nodes, this was computa- tionally impractical given our large spatial extent and hundreds of nodes. Therefore, from each of the 475 start nodes, we identified the 5 nearest end nodes (based on Euclidean distance) using the ‘st_nn’ R function [46] and mapped least cost paths for each of these 5 node pairs, resulting in a total of 2375 unique least cost paths. Ecological characteristics, conservation status, and prioritization of potential climate adaptation corridors Whereas least cost paths inherently represent the path of least resistance from one location to another, individual pathways vary in their relative capacity to function as climate adaptation corridors. Therefore, we quantified several ecological characteristics and the conservation sta- tus of potential corridors to facilitate comparison among and prioritization of different poten- tial corridors (S4 Table in S1 Appendix). Corridor variables included length, elevational range, Fig 2. Comparison of area, elevational range, protection, and mean forest biomass between start and end nodes. Figure encompasses all start nodes

(n = 475) and only the end nodes in which potential climate adaptation corridors terminated (n = 192). https://doi.org/10.1371/journal.pone.0304756.g002 PLOS ONE Mapping climate adaptation corridors for biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0304756 May 31, 2024 6 / 19