Altitude-driven vehicle particle emission surges: Insights from chemical analysis and machine learning

Vehicle exhaust emissions, particularly fine particulate matter (PM_2.5), pose significant threats to air quality and human health, with elevated emissions in high-altitude regions. This study examines the mechanisms behind altitude-driven vehicle particulate emission surges through real-world driving tests in Tianjin (5 m altitude) and Xining (2240 m altitude). On-road test results indicate that PM_2.5 emission factors for light-duty gasoline and diesel vehicles in Xining are 1.7-2.8 times higher than those in Tianjin. Chemical analysis reveals that emissions in high-altitude regions are enriched in carbonaceous components, particularly elemental carbon (EC), as well as inorganic elements (Ca, Si) and water-soluble ions (NO₃⁻, SO₄²⁻), suggesting lower air-fuel ratios and increased engine loads from steep terrain promote incomplete combustion. Additionally, interpretable machine learning models independently demonstrate that driving dynamics, such as speed and relative positive acceleration (RPA), significantly influence transient exhaust emissions, with terrain factors like altitude and road grade amplifying their impact in high-altitude regions. This integration of on-road testing and chemical insights elucidates altitude-driven mechanisms, informing refined emission models and differentiated control strategies for plateau regions.