Abstract Approximately half of satellite aerosol retrievals are missing that limits the application of satellite data in PM2.5 pollution monitoring. To obtain spatiotemporally continuous PM2.5 distributions, various gap-filling methods have been developed, but have rarely been evaluated. Here, we reviewed and summarized four types of gap-filling strategies, and applied them to a random forest PM2.5 prediction model that incorporated ground observations, chemical transport model (CTM) simulations, and satellite AOD for predicting daily PM2.5 concentrations at a 1-km resolution in 2013 in the Beijing-Tianjin-Hebei region and the Yangtze River Delta. The model out-of-bag predictions were compared with national station measurements and external measurements to assess the performance of different gap-filling methods. We also conducted a by-city cross-validation and characterized the spatial distributions of PM2.5 prediction when the AOD coverage was low. We found that the methods filling in missing data by regression, i.e. multiple imputation and decision tree, performed robustly to characterizing PM2.5 variation at a high spatial resolution and the method filling in missing PM2.5 predictions with decision tree overcame the problem of time-consuming computations. The method using spatiotemporal trends to fill in missing data, i.e. ordinary kriging and generalized additive mixed model, may be overrated in statistical evaluation tests, and predicted artificially oversmoothed PM2.5 spatial distributions. We also revealed that CTM simulations benefited the prediction of PM2.5 spatial distribution in all the models with various gap-filling strategies with higher prediction accuracy in the by-city cross-validation. We noticed that the PM2.5 prediction was not sensitive to the resolution of CTM simulations and even the 12-km resolution CTM simulations benefited the high-resolution PM2.5 prediction.

中文翻译：

---

基于卫星的每日 PM2.5 预测模型中填补空白的方法评估

摘要 大约一半的卫星气溶胶反演缺失，限制了卫星数据在 PM2.5 污染监测中的应用。为了获得时空连续的 PM2.5 分布，已经开发了各种填隙方法，但很少进行评估。在这里，我们回顾并总结了四种类型的填隙策略，并将它们应用于随机森林 PM2.5 预测模型，该模型结合了地面观测、化学传输模型 (CTM) 模拟和卫星 AOD，用于预测每天的 PM2.5 浓度。 2013 年在京津冀地区和长三角地区的 1 公里分辨率。将模型袋外预测与国家站测量和外部测量进行比较，以评估不同填隙方法的性能。我们还进行了逐市交叉验证，并表征了 AOD 覆盖率较低时 PM2.5 预测的空间分布。我们发现通过回归填充缺失数据的方法，即多重插补和决策树，在高空间分辨率下对表征 PM2.5 变化表现稳健，而用决策树填充缺失 PM2.5 预测的方法克服了时间问题- 消耗计算。使用时空趋势填充缺失数据的方法，即普通克里金法和广义加性混合模型，在统计评估测试中可能被高估，并预测人为过度平滑的PM2.5空间分布。我们还透露，CTM 模拟有利于 PM2.5 的预测。5 所有模型中的空间分布具有不同的填充策略，在按城市交叉验证中具有更高的预测精度。我们注意到 PM2.5 预测对 CTM 模拟的分辨率不敏感，甚至 12 公里分辨率的 CTM 模拟也有利于高分辨率 PM2.5 预测。