Morpho-chemical fingerprinting and distance-dependent attenuation of airborne particulate matter from abandoned mine tailings in Northwestern Mexico

Abandoned mine tailings are a persistent source of airborne particulate matter (PM), containing potentially toxic elements (PTE), particularly in arid regions where wind erosion facilitates particle mobilization. However, source attribution and spatial dispersion of tailings-derived PM remain challenging because of their heterogeneous composition and elemental overlap with other PM sources. This study combines morpho-chemical fingerprinting of mine tailings and efflorescent salts using automated SEM-EDX single particle characterization with a three-step classification approach integrating machine learning, source-informed compositional screening, and statistical fingerprint classification. Three dry-season field campaigns (2022-2024) were conducted using Sigma-2 passive samplers at nine locations spanning the tailings surroundings, agricultural land, and a populated area. A total of 3,017 particles from oxidized and non-oxidized tailings and efflorescent salts were analyzed to establish source-specific fingerprints and quantify mine-derived particles in airborne PM. Concentrations of mine-derived particles were highest near the abandoned tailings and decrease sharply with distance, although mine-derived particles were still detected at approximately 1.1 km from the source. Power-law models accounted for 85.8% and 82.8% of the variability in PM10-2.5 and PM80-1 concentrations, respectively, while sensitivity analyses showed that the distance decay relationship was not dependent on any single sampling location. These findings underscore the value of integrating source-specific morpho-chemical fingerprinting with spatial modeling to assess atmospheric transport and dry deposition pathways of abandoned mine tailings.