Jose Ramon Martinez Batlle , Michela Izzo Gioiosa

Many weather station networks lack sufficient representativeness, and their station density is often inadequate to capture spatial and climatic variability effectively. Optimal site selection is therefore essential to enhance spatial coverage and improve data quality. This study proposes a methodology for identifying optimal sites for a meteorological station network in the Dominican Republic, utilizing a multi-criteria decision-making framework based on the Analytic Hierarchy Process (AHP) and neighborhood analysis. Using the H3 library as a spatial indexing tool, zonal statistics were derived from geospatial variables, including seasonality, habitat heterogeneity, proximity to water bodies, slope, solar radiation, and elevation. Expert-defined weights were assigned to each variable based on their relative importance. Areas with high topographic and climatic variability were prioritized to maximize spatial representativeness. Results highlight thermal and precipitation seasonality, elevation, and solar radiation as the most influential variables, emphasizing the need to collect data in elevated areas with marked seasonality. Sites were evenly distributed across three density scenarios, ensuring robust climatic and topographic coverage while avoiding redundancy through proximity constraints to existing stations. The proposed network would provide essential data for meteorological and climatic research in the region. Future studies should assess the accessibility and feasibility of the selected sites and incorporate additional environmental variables into the framework.