Long-Term Evolution of the Vertical Distribution of Key Atmospheric Pollutants Revealed by MAX-DOAS Observations in Hefei, China

_{Corresponding authors: Pinhua Xie (phxie@aiofm.ac.cn), Jin Xu (jxu@aiofm.ac.cn); Xin Tian (xtian@ahu.edu.cn);}

Long-term observations of the vertical distribution of atmospheric pollutants provide critical insights into the temporal evolution and vertical structure of atmospheric pollutants, yet such datasets remain scarce. In this study, continuous MAX-DOAS measurements conducted from 2014 to 2025 (excluding 2019) at the Science Island site in Hefei, China provide information on the vertical distribution of aerosols and concentrations of NO₂, HCHO, HONO, and SO₂.

Long-term observations reveal pronounced seasonal and interannual variability across all species. Aerosol optical depth (AOD) from MAX-DOAS shows overall consistency with collocated CE318 sun photometer, while near-surface NO₂ and SO₂ concentrations are consistent with measurements from nearby national air quality monitoring stations, supporting the reliability of the MAX-DOAS retrievals. NO₂ and SO₂ consistently exhibit wintertime maxima, reflecting enhanced emissions combined with suppressed atmospheric dispersion under stable boundary-layer conditions, whereas HCHO shows pronounced summertime maxima driven by intensified photochemical production. Over the study period, SO₂ displays a persistent long-term decline, consistent with sustained reductions in coal combustion and industrial emissions, while NO₂ decreases until 2022 and rebounds in 2023, likely associated with the recovery of traffic and industrial activities following COVID-19 lockdowns. Distinct seasonal characteristics are evident in the vertical distributions of different pollutants. NO₂ and SO₂ exhibit strong near-surface gradients, particularly in winter, indicating the dominant influence of local emissions and limited vertical mixing. In contrast, HCHO shows weaker vertical gradients and enhanced concentrations aloft during summer, highlighting the importance of secondary formation and vertical transport processes. Across all species, strong exponential decreases in concentrations within the lowest 1 km emphasize the combined control of surface emissions and boundary-layer mixing on pollutant distributions in the lower troposphere.

To identify the dominant processes influencing pollutants at different altitudes, these datasets of the vertical distribution of multi-species were jointly analyzed using Positive Matrix Factorization (PMF). The results indicate that pollutant variability near the surface is mainly controlled by local primary emissions and surface-related chemical processes, whereas secondary formation and regional influences play an increasingly important role at elevated levels. Interannual PMF analysis further reveals a systematic shift around 2017-2019, with declining contributions from near-surface emission-related processes and a strengthened influence of secondary and regional processes, reflecting long-term changes in dominant pollution drivers.

Overall, these results demonstrate that long-term MAX-DOAS observations provide valuable insights on both the temporal evolution and vertical structure of key atmospheric pollutants, revealing distinct controlling mechanisms and long-term trends associated with changes in emissions reflected in the vertical distribution of atmospheric pollutants.