Diagnosing drivers of PM 2.5 simulation biases in China from meteorology, chemical composition, and emission sources using an efficient machine learning method

作者：Shuai Wang, Mengyuan Zhang, Yueqi Gao, Peng Wang, Qingyan Fu, Hongliang Zhang · 发表于：Geoscientific model development · 年份：2024 · DOI：10.5194/gmd-17-3617-2024 · 被引用次数：18 · 研究领域：Atmospheric chemistry and aerosols、Air Quality and Health Impacts、Air Quality Monitoring and Forecasting

Chemical transport models (CTMs) are widely used for air pollution modeling, which suffer from significant biases due to uncertainties in simplified parameterization, meteorological fields, and emission inventories. Accurate diagnosis of simulation biases is critical for the improvement of models, interpretation of results, and management of air quality, especially for the simulation of fine particulate matter (PM 2.5 ). In this study, an efficient method with high speed and a low computational resource requirement based on the tree-based machine learning (ML) method, the light gradient boosting machine (LightGBM), was designed to diagnose CTM simulation biases. The drivers of the Community Multiscale Air Quality (CMAQ) model biases are compared to observations obtained by simulating PM 2.5 concentrations from the perspectives of meteorology, chemical composition, and emission sources. The source-oriented CMAQ was used to diagnose the influences of different emission sources on PM 2.5 biases. The model can capture the complex relationship between input variables and simulation bias well; meteorology, PM 2.5 components, and source sectors can partially explain the simulation bias. The CMAQ model underestimates PM 2.5 by −19.25 to −2.66 µg m −3 in 2019, especially in winter and spring and during high-PM 2.5 events. Secondary organic components showed the largest contribution to the PM 2.5 simulation bias for different regions and seasons (13.8 %–22.6 %) of all components. Relat...