Global high-resolution ultrafine particle number concentrations through data fusion with machine learning

This dataset provides the first global, high-resolution (1 km²) estimates of particle number concentrations (PNC) and ultrafine particles (UFPs; <100 nm) for 2010–2019, derived through a machine learning framework integrating meteorological features, urban indices, and ground monitoring. It captures fine-scale spatial heterogeneity (e.g., urban hotspots) spanning 2010-2019. The NetCDF files include annual mean PNC, 95% coverage intervals, and UFPs under a CC-BY 4.0 license.