Abstract Precipitation processes are critical for removing pollutants from the atmosphere, yet in many mid‐latitude continental regions, the effects of rainout are difficult to distinguish from the broader influence of atmospheric transport. In this study, we analyze 582 cloud and rain samples collected during summers 1996–2014 from Mount Washington, New Hampshire. We use the sulfate ion concentration (SO 4 2− ) of each sample as a proxy for anthropogenic pollution loading and the water isotopic composition (δD) as a tracer of the water‐cycle history associated with each sample's air mass. Since the δD signal records both exchange with near‐surface air (a source of moisture and pollutants) and rainout, we use it to evaluate how these same source and sink processes influence pollution concentrations. To isolate the effects of rainout from source, we compare the ln(SO 4 2− ) variability explained by sample δD with that explained by a more traditional back‐trajectory analysis. Using trajectory cluster, sample type (cloud or rain), and time as predictors in a multivariate regression, we explain 40% of the observed ln(SO 4 2− ) variability. In comparison, substituting δD for cluster, or using δD and back‐trajectory information in combination, increases the explained variance to 51% and 56%, respectively. After accounting for sample type and time, roughly 14% of the remaining variability in ln(SO 4 2− ) is due to precipitation effects. These results quantitatively demonstrate the importance of cloud and precipitation processes in determining pollution concentrations during air mass transport.
Richards et al. (Fri,) studied this question.