[Description of methods used for collection/generation of data] The high-resolution (5 m) microclimatic models of above- and below-ground winter minimum temperatures (Tn) and summer maximum temperatures (Tx) were developed using temperature records collected by a network of miniaturized field dataloggers (iButton® and TMS-TOMST), environmental predictors derived from LiDAR-PNOA and Sentinel-2 data, and generalized additive models (GAMs).
(1) Temperature monitoring and processing: Dataloggers were deployed across environmentally heterogeneous locations within the PNOMP. Above-ground temperatures recorded by iButtons were measured at approximately 5 cm above the ground in open habitats and 1.20 m in forested environments. TMS sensors measured at approximately 2 cm above the ground in open habitats and 15 cm in forested environments. Below-ground temperatures from both sensor types were recorded at approximately 8 cm soil depth. Raw temperature series consisted of at least six measurements per day. Daily Tmin and Tmax values were extracted and subsequently averaged by month when at least 22 valid daily records were available. Monthly values were then averaged across years for sensors with a minimum of three complete annual cycles, generating a representative “year-type” for each monitoring location. Seasonal winter Tn (December-February) and summer Tx (June-August) were subsequently derived from these reference climatologies. The final dataset included 100 above-ground and 85 below-ground monitoring locations.
(2) Environmental predictors: Predictor variables representing the main drivers of local climatic variability were generated from two complementary data sources. Topographic and forest structural variables were derived from the second ALS-LiDAR survey of Spain (PNOA, 2021) at 5 m spatial resolution. Multispectral land-cover variables were derived from Sentinel-2 imagery at 10 m spatial resolution and subsequently resampled to 5 m.
(3) Model calibration and spatial prediction: Seasonal winter Tn and summer Tx values derived from the reference year-type were used as response variables. These data were combined with the corresponding environmental predictors to calibrate generalized additive models (GAMs). The resulting models were used to generate continuous raster layers of microclimatic temperatures at 5 m spatial resolution across the entire PNOMP. Model performance (R²) was 0.44 for above-ground winter Tmin, 0.84 for below-ground winter Tmin, 0.65 for above-ground summer Tmax, and 0.73 for below-ground summer Tmax., Identifying microclimates is essential for understanding the exposure of low-mobility organisms to local thermal conditions and the impacts of climate change. However, widely available regional and global climatic products may not adequately represent the thermal variability that emerges at fine spatial scales in environmentally and topographically complex landscapes. This dataset provides four very high-resolution (5 m) microclimatic models representing winter minimum temperatures (Tn) and summer maximum temperatures (Tx) in both above- and below-ground environments for the Ordesa and Monte Perdido National Park (PNOMP), located in the Central Pyrenees of northeastern Spain. The models were developed using temperature records collected over at least three complete years from a network of miniaturized field dataloggers (iButton® and TMS-TOMST) distributed across 100 above-ground and 85 below-ground monitoring locations. Predictor variables were derived from LiDAR-PNOA data and Sentinel-2 imagery and were designed to capture the influence of topography, vegetation structure, snow cover, and other biophysical factors controlling local climatic variability. The dataset consists of continuous raster layers at 5 m spatial resolution that characterize microclimatic patterns across the PNOMP., Proyecto VUBI (Proyectos de I+D+i Gobierno de Aragón, E35_24); Proyecto REFUGIA (Agencia Estatal de Investigación, Gobierno de España, PID2021-129056OB-I00); PTI Salud Global CSIC (Comisión Europea - NextGenerationEU, SGL2104002); Proyecto LIFE PYRENEES4CLIMA, WP3 (Comisión Europea)., Peer reviewed