Black carbon in urban Jinan: Variations, health risks, and driving factors analyzed with machine learning.

Atmospheric black carbon (BC) significantly impacts climate change, atmospheric environment, and human health. To explore the temporal variation of BC concentration and its health effects during different periods in urban Jinan from 2021 to 2022, we analyzed BC concentration characteristics and assessed the population exposure risk using Monte Carlo simulation. Additionally, we studied the role of driving factors on BC using SHapley Additive exPlanations interpretation algorithm. A clear temporal trend was observed in BC levels, with the highest BC concentration in winter (2.40 ± 1.76 µg/m3) and the lowest in summer (1.46 ± 0.89 µg/m3), and a bimodal pattern in diurnal variation. Our results indicated that the carcinogenic risk of BC for adults exceeded the acceptable threshold in winter. With emissions playing a dominant role, meteorological conditions also significantly affected BC, with wind speed, relative humidity, and boundary layer height playing major roles in variations of BC concentration. Meteorological conditions during the haze periods favored BC accumulation, hours with elevated PM2.5 levels were associated with high BC concentrations and increased carcinogenic effects. Understanding the temporal variations and drivers of BC is crucial for formulating effective policies to mitigate air pollution. Our study may provide theoretical methods and practical guidance for addressing BC levels under varying meteorological conditions and enhance understanding of atmospheric characteristics and population exposure to BC.