Quantifying the influences of atmospheric stability on air pollution in Lanzhou, China, using a radon-based stability monitor

Commercially-available "stability monitors" based on in situ atmospheric radon progeny measurements remain underutilised as a tool for urban pollution studies, due in part to difficulties experienced in relating their standard output directly to the atmospheric mixing state in a consistent manner. The main confounding factor has been a lack of attention to the fact that the observed near-surface atmospheric radon concentration includes large synoptic and fetch-related components in addition to the local stability influence. Here, a technique recently developed for stability classification using a research-quality dual-flow-loop two-filter radon detector is adapted for use with a commercially-available radon-based stability monitor. Performance of the classification scheme is then tested in Lanzhou, China, a topographically-complex region renowned for low mean annual wind speeds (0.8 m s1) and winter stagnation episodes. Based on an 11-month composite, a factor of seven difference is estimated between peak NOx concentrations in the city's industrial region and a rural background location under stable conditions. The radon-based scheme is evaluated against the Pasquil-Gifford "radiation" (PGR) scheme, and assigns pollutant concentrations more consistently between defined atmospheric stability states than the PGR scheme. Furthermore, the PGR scheme consistently underestimates all peak pollutant concentrations under stable conditions compared with the radon-based scheme, in some cases (e.g. CO in the industrial region) by 25%.