Data and code for: Range-Wide Habitat Suitability, Climate-Change Exposure and Field Stand Structure in Two Andean Polylepis Species

Complete analysis pipeline, field data and derived result tables supporting the manuscript submitted to Forests (MDPI), manuscript ID forests-4412160. Habitat suitability for two co-occurring high-Andean tree species, Polylepis rugulosa and Polylepis tarapacana, is modelled with MaxEnt and Random Forest calibrated across the full South-Central Andean range and clipped to the Tacna region of southern Peru. Predictions are validated by internal spatial block cross-validation, by an external test against 228 field-mapped stand polygons, and by a functional test against measured stand structure on 35 inventory plots (883 stems). Future suitability is projected under CMIP6 (four to five GCMs x four SSPs x four horizons). The deposit contains the R scripts for every step (occurrence cleaning and thinning, predictor selection, model calibration, spatial block cross-validation, CMIP6 projection and ensembling, external polygon validation, and the stand-structure regressions), the configuration files, the renv lockfile pinning every package version, the cleaned plot and stem inventory, the digitised stand polygons, and the derived result tables. It also contains the revision analyses: the dispersion-corrected and spatially explicit refits, the nine-candidate model sensitivity analysis, and the per-GCM agreement check. Environmental rasters (~40 GB) are not duplicated: they are regenerated deterministically from GBIF, WorldClim 2.1, SoilGrids 2.0 and SRTM by the deposited scripts.