Influence of springtime atmospheric circulation types on the distribution of air pollutants in the Arctic

Abstract

The transport and distribution of short-lived climate forcers in the Arctic are influenced by the prevailing atmospheric circulation patterns. Understanding the coupling between pollutant distribution and dominant atmospheric circulation types is therefore important, not least to understand the processes governing the local processing of pollutants in the Arctic, but also to test the fidelity of chemistry transport models to simulate the transport from the southerly latitudes. Here, we use a combination of satellite-based and reanalysis datasets spanning over 12 years (2007–2018) and investigate the concentrations of NO2, O3, CO and aerosols and their co-variability during eight different atmospheric circulation types in the spring season (March, April and May) over the Arctic. We carried out a self-organizing map analysis of mean sea level pressure to derive these circulation types. Although almost all pollutants investigated here show statistically significant sensitivity to the circulation types, NO2 exhibits the strongest sensitivity among them. The circulation types with low-pressure systems located over the northeast Atlantic show a clear enhancement of NO2 and aerosol optical depths (AODs) in the European Arctic. The O3 concentrations are, however, decreased. The free tropospheric CO is increased over the Arctic during such events. The circulation types with atmospheric blocking over Greenland and northern Scandinavia show the opposite signal in which the NO2 concentrations are decreased and AODs are smaller than the climatological values. The O3 concentrations are, however, increased, and the free tropospheric CO is decreased during such events. The study provides the most comprehensive assessment so far of the sensitivity of springtime pollutant distribution to the atmospheric circulation types in the Arctic and also provides an observational basis for the evaluation of chemistry transport models.