Impact of Upper-Level Atmospheric Circulation and El Niño-Southern Oscillation Phases on Particulate Matter Concentrations in Tropical South America

Biomass burning has been identified as a major cause of poor regional air quality and the dominant source of particulate matter (PM) in the Amazon basin. In this study, we analyse the impact of the upper-level jet on PM2.5 (PM with an aerodynamic diameter ≤ 2.5 μm) concentrations in tropical South America (SA) from December to February during the period 2003-2022, using the Copernicus Atmosphere Monitoring Service (CAMS) and ERA5 reanalyses. Furthermore, we investigate the response of air pollutants to the joint modulation of the upper-level jet and El Niño-Southern Oscillation (ENSO). First, a climatological analysis shows that PM in the region is largely composed of organic matter and black carbon, with the highest concentrations and temporal variability in Colombia and northeastern Brazil. Regarding the link with the upper-level circulation, we find that PM2.5 concentrations in northeastern Brazil are reduced on days when the subtropical jet (STJ) is absent, due to increased convection and precipitation over the region. This improvement in air quality is independent of the ENSO phase. Conversely, a prominent STJ inhibits convection and contributes to dry conditions that favour increased biomass burning and elevated pollutant concentrations in the lower troposphere. At a 3-day persistence of these STJ conditions, there is a 90% probability of exceeding the World Health Organisation threshold of 15 μg m-3. In addition, the co-occurrence of a prominent STJ with an El Niño phase acts synergistically to increase pollutant concentrations, as both reduce precipitation in northeastern Brazil. In combination with La Niña, this upper-level pattern does not modulate PM2.5 concentrations because the wet conditions favoured by this ENSO phase prevail, reducing biomass burning. This study provides new insights into the modulation of air quality by the upper-level atmospheric circulation in tropical SA. The results have the potential to improve short-term predictability and can serve as a first step towards the development of a warning system.