Separating Daily 1 km PM2.5 Inorganic Chemical Composition in China since 2000 via Deep Learning Integrating Ground, Satellite, and Model Data

Jing Wei, Zhanqing Li, Xi Chen, Chi Li, Yele Sun, Jun Wang, Alexei Lyapustin, Guy Pierre Brasseur, Mengjiao Jiang, Lin Sun, Tao Wang, Chang Hoon Jung, Bing Qiu, Cuilan Fang, Xuhui Liu, Jinrui Hao, Yan Wang, Ming Zhan, Xiaohong Song, Yuewei Liu

Research output: Journal article publicationJournal articleAcademic researchpeer-review

33 Citations (Scopus)

Abstract

Fine particulate matter (PM2.5) chemical composition has strong and diverse impacts on the planetary environment, climate, and health. These effects are still not well understood due to limited surface observations and uncertainties in chemical model simulations. We developed a four-dimensional spatiotemporal deep forest (4D-STDF) model to estimate daily PM2.5 chemical composition at a spatial resolution of 1 km in China since 2000 by integrating measurements of PM2.5 species from a high-density observation network, satellite PM2.5 retrievals, atmospheric reanalyses, and model simulations. Cross-validation results illustrate the reliability of sulfate (SO42-), nitrate (NO3-), ammonium (NH4+), and chloride (Cl-) estimates, with high coefficients of determination (CV-R2) with ground-based observations of 0.74, 0.75, 0.71, and 0.66, and average root-mean-square errors (RMSE) of 6.0, 6.6, 4.3, and 2.3 μg/m3, respectively. The three components of secondary inorganic aerosols (SIAs) account for 21% (SO42-), 20% (NO3-), and 14% (NH4+) of the total PM2.5 mass in eastern China; we observed significant reductions in the mass of inorganic components by 40-43% between 2013 and 2020, slowing down since 2018. Comparatively, the ratio of SIA to PM2.5 increased by 7% across eastern China except in Beijing and nearby areas, accelerating in recent years. SO42- has been the dominant SIA component in eastern China, although it was surpassed by NO3- in some areas, e.g., Beijing-Tianjin-Hebei region since 2016. SIA, accounting for nearly half (∼46%) of the PM2.5 mass, drove the explosive formation of winter haze episodes in the North China Plain. A sharp decline in SIA concentrations and an increase in SIA-to-PM2.5 ratios during the COVID-19 lockdown were also revealed, reflecting the enhanced atmospheric oxidation capacity and formation of secondary particles.

Original languageEnglish
JournalEnvironmental Science and Technology
DOIs
Publication statusAccepted/In press - 2023

Keywords

  • China
  • deep learning
  • PM composition
  • remote sensing
  • secondary inorganic aerosols

ASJC Scopus subject areas

  • General Chemistry
  • Environmental Chemistry

Fingerprint

Dive into the research topics of 'Separating Daily 1 km PM2.5 Inorganic Chemical Composition in China since 2000 via Deep Learning Integrating Ground, Satellite, and Model Data'. Together they form a unique fingerprint.

Cite this