Li Yanshun, Martin Randall V, Zhang Yuanjian, Zhang Dandan, van Donkelaar Aaron, Zhu Haihui, Meng Jun
Department of Energy, Environmental and Chemical Engineering, Washington University in St. Louis, St. Louis, Missouri 63130, United States.
Department of Civil and Environmental Engineering, Washington State University, Pullman, Washington 99164, United States.
ACS EST Air. 2025 Jul 14;2(8):1575-1585. doi: 10.1021/acsestair.5c00068. eCollection 2025 Aug 8.
Globally distributed measurements of the diurnal variation of fine particulate matter (PM) reveal a remarkable overall consistency with similar bimodal patterns and some regional variation, neither of which is well understood. We interpret these observations using the GEOS-Chem global model of atmospheric composition in its high-performance configuration (GCHP) at fine resolution of C180 (∼50 km). The base simulation overestimates the PM accumulation overnight, leading to excessive diurnal amplitude and earlier PM morning peaks than observations. These biases are reduced by applying sector- and species-wise diurnal scaling factors on anthropogenic emissions, by resolving the aerosol subgrid vertical gradient within the surface model layer, by applying revised wet deposition, and by revising the mixing coefficient in the boundary layer. Budget analyses indicate that the morning peak of PM is likely driven by changes in the aerosol subgrid vertical gradient with fumigation after sunrise, that the concentration decrease until late afternoon is driven by boundary layer mixing and thermodynamic partitioning of a semivolatile aerosol to the gas phase, that the concentration increase during evening is driven by enhanced secondary chemical production and persistent primary anthropogenic emissions, and that the consistently high concentration overnight is driven by the balance between emissions, chemical production, and boundary layer mixing and deposition.
全球范围内对细颗粒物(PM)日变化的测量结果显示,其总体模式与类似的双峰模式具有显著的一致性,同时存在一些区域差异,但二者均未得到很好的理解。我们使用高性能配置(GCHP)下的GEOS-Chem全球大气成分模型,以C180(约50公里)的精细分辨率对这些观测结果进行解释。基础模拟高估了夜间PM的积累,导致日变化幅度过大,且PM早晨峰值出现时间比观测结果更早。通过对人为排放应用按部门和物种划分的日缩放因子、在地表模型层内解析气溶胶亚网格垂直梯度、应用修订后的湿沉降以及修订边界层中的混合系数,这些偏差得以减小。收支分析表明,PM的早晨峰值可能是由日出后熏蒸导致的气溶胶亚网格垂直梯度变化驱动的,浓度在傍晚前下降是由边界层混合以及半挥发性气溶胶向气相的热力学分配驱动的,傍晚浓度增加是由增强的二次化学生成和持续的人为一次排放驱动的,而夜间浓度持续较高是由排放、化学生成、边界层混合和沉降之间的平衡驱动的。