Comparing pollen- and plant-derived climatic niches of pollen taxa in mountainous East Asia: Implications for pollen-based paleoclimate reconstruction

Fossil pollen is a key proxy for reconstructing Quaternary vegetation and climate change. However, the quantitative characteristics of climatic niches at the pollen-taxon level and their consistency with plant-based estimates remain insufficiently evaluated. Direct comparisons between pollen- and plant-derived niches are particularly scarce. In this study, we quantified and compared the realized climatic niches of 18 major arboreal pollen taxa within a shared climate space using surface pollen assemblages and GBIF-derived modern plant occurrence records from the southern part of the Korean Peninsula. Climatic niches were estimated using a probability density–based approach, and niche breadth was standardized as relative niche width (RNW). RNWs derived from the two data sources exhibited contrasting patterns across temperature- and precipitation-related variables. For temperature variables (BIO1, BIO10, BIO11), pollen-based RNWs were generally comparable to or broader than plant-based estimates, whereas for precipitation variables (BIO12, BIO16, BIO17), plant-based RNWs were predominantly broader. These patterns reflect the interaction between elevational gradients in the pollen data, pollen transport processes, and differences in the spatial structure of climatic variables. Despite differences in niche breadth, niche optima showed high concordance between pollen- and plant-based estimates. Temperature-related optima were clearly differentiated among taxa along the regional thermal gradient, whereas precipitation-related optima exhibited substantial overlap and clustering within the high-precipitation portion of the background climate space. Our results quantitatively demonstrate that pollen data robustly capture thermal niche differentiation in mountainous humid temperate regions, while precipitation signals are more susceptible to niche overlap and spatial sampling effects. The standardized niche framework presented here provides a basis for evaluating uncertainty in pollen-based paleoclimate reconstructions and enhancing their robustness.