Climate-Informed Renewable Resource–Demand Screening with Constraint-Aware Geospatial Graphs

Abundant wind and solar resources are often distant from demand centers, and apparent resource value actually changes after land, water, and climate exposure are considered. Geospatial suitability studies commonly rank individual locations, while power-system models solve downstream capacity, transmission, or dispatch decisions. We study an intermediate output: an auditable list of directed source–demand candidates. Public climate, Earth-observation, population, terrain, and protected-area layers are harmonized on a 0.25° grid. Explicit thresholds select 2,780 supply cells and 3,976 demand-proxy cells, and a fixed six-component score ranks 31,808 candidate edges. The 500 highest-ranked edges have a mean internal score of 0.714. Low- and medium-observed-risk groups have nearly equal mean edge scores (0.510 and 0.509) but different adverse-month proxies (0.263 and 0.400).In practical reverse-relation matching, the Qinghai–Henan province-proxy pair ranks as the 18th candidate edge nationwide, strongly validating utility of the graph. The deterministic graph framework provides auxiliary for climate-informed candidate screening.