Exposure to fine particulate matter (PM_2.5) has been linked to a substantial disease burden globally, yet little has been done to estimate the population health risks of PM_2.5 in South Africa due to the lack of high-resolution PM_2.5 exposure estimates. We developed a random forest model to estimate daily PM_2.5 concentrations at 1 km² resolution in and around industrialized Gauteng Province, South Africa, by combining satellite aerosol optical depth (AOD), meteorology, land use, and socioeconomic data. We then compared PM_2.5 concentrations in the study domain before and after the implementation of the new national air quality standards. We aimed to test whether machine learning models are suitable for regions with sparse ground observations such as South Africa and which predictors played important roles in PM_2.5 modeling. The cross-validation R² and Root Mean Square Error of our model was 0.80 and 9.40 μg/m³, respectively. Satellite AOD, seasonal indicator, total precipitation, and population were among the most important predictors. Model-estimated PM_2.5 levels successfully captured the temporal pattern recorded by ground observations. Spatially, the highest annual PM_2.5 concentration appeared in central and northern Gauteng, including northern Johannesburg and the city of Tshwane. Since the 2016 changes in national PM_2.5 standards, PM_2.5 concentrations have decreased in most of our study region, although levels in Johannesburg and its surrounding areas have remained relatively constant. This is anadvanced PM_2.5 model for South Africa with high prediction accuracy at the daily level and at a relatively high spatial resolution. Our study provided a reference for predictor selection, and our results can be used for a variety of purposes, including epidemiological research, burden of disease assessments, and policy evaluation.

暴露于细颗粒物 (PM _2.5 ) 与全球范围内的重大疾病负担有关，但由于缺乏高分辨率的 PM _2.5暴露估计，人们很少对南非的 PM _2.5人口健康风险进行估计。我们开发了一个随机森林模型，通过结合卫星气溶胶光学深度 (AOD)、气象学、土地利用和社会经济数据，以 1 km ²的分辨率估算南非工业化豪登省及其周边地区的每日 PM _2.5浓度。然后我们比较了 PM _2.5新国家空气质量标准实施前后研究区浓度。我们的目的是测试机器学习模型是否适用于南非等地面观测稀疏的地区，以及哪些预测因子在 PM _2.5建模中发挥了重要作用。我们模型的交叉验证 R ²和均方根误差分别为 0.80 和 9.40 μg/m ³。卫星 AOD、季节指标、总降水量和人口是最重要的预测因子。模型估计的 PM _2.5水平成功捕获了地面观测记录的时间模式。空间上，年度最高 PM _2.5集中出现在豪登省中部和北部，包括约翰内斯堡北部和茨瓦内市。自 2016 年国家 PM _2.5标准发生变化以来，我们研究区域的大部分地区的 PM _2.5浓度有所下降，尽管约翰内斯堡及其周边地区的水平保持相对稳定。这是针对南非的先进 PM _2.5模型，具有较高的日预测精度和较高的空间分辨率。我们的研究为预测​​因子的选择提供了参考，我们的结果可用于多种目的，包括流行病学研究、疾病负担评估和政策评估。