Determining the key meteorological factors affecting atmospheric CO2 and CH4 using machine learning algorithms at a suburban site in China

• Published in: Urban climate Vol. 59; p. 102312
• Main Authors: Liu, Wanyu, Niu, Zhenchuan, Feng, Xue, Zhou, Weijian, Liang, Dan, Wang, Guowei, Liu, Lin
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.02.2025
• Subjects: Dominant emission sources; eXtreme gradient boosting; Machine learning algorithms; Random forest; Variable importance of meteorological factor; Machine learning algorithms; Random forest; Variable importance of meteorological factor; eXtreme gradient boosting; Dominant emission sources
• ISSN: 2212-0955; 2212-0955
• DOI: 10.1016/j.uclim.2025.102312

Atmospheric CO2 and CH4 were measured from March 2023 to February 2024 at a suburban site near the northern foot of the Qinling Mountains, China, aiming to determine the key meteorological factors that influencing the seasonal CO2 and CH4 dynamics using machine learning (ML) algorithms. Yearly average atmospheric CO2 and CH4 were 446.1 ± 10.0 ppm and 2118.9 ± 50.5 ppb, respectively. Anthropogenic emissions dominated atmospheric CO2 and CH4 changes in winter, and the excess CO2 (ΔCO2) and CH4 (ΔCH4) above background levels during the heating period were mainly from combustion emissions. We selected seven ML algorithms to determine the variable importance of meteorological factors, among which eXtreme Gradient Boosting (XGBoost) and Random Forest (RF) demonstrating the best performance and ranking consistency of the factors. Humidity and temperature of the atmosphere and soil had the higher effect (XGBoost: 80.4 %; RF: 78.1 %) on CO2 in spring, while humidity was crucial in winter, with variable importance of 69.3 % for XGBoost and RF. In summer and autumn, photosynthetic photon flux density and wind speed (WS) (totaling 35.2 % ∼ 50.7 %) dominated CO2 dynamics. For CH4, atmospheric and soil humidity (totaling around 40.0 %) were key factors in spring, whereas atmospheric humidity was important in winter. WS had the largest effect in summer (XGBoost: 26.3 %; RF: 33.3 %) and autumn (XGBoost: 19.8 %; RF: 28.8 %). Meteorological processes like cold front passage significantly reduced CO2 and CH4 concentrations during haze events. XGBoost and RF have emerged as powerful tools for determining the key meteorological factor that favour seasonal GHGs evolution. •Excess CO2 and CH4 over background levels in heating period came from burning source.•XGBoost and RF performed best in determining the importance of meteorological factors.•Atmospheric and soil humidity were key factors driving CH4 change in spring.•Photosynthetic photon flux density and wind speed dominated CO2 in summer and autumn.•Cold front passage significantly reduced CO2 and CH4 levels during haze events.