Bi Jianzhao, Belle Jessica H, Wang Yujie, Lyapustin Alexei I, Wildani Avani, Liu Yang
Department of Environmental Health, Emory University, Rollins School of Public Health, Atlanta, GA, USA.
Goddard Earth Sciences and Technology Center, University of Maryland Baltimore County, Baltimore, MD, USA.
Remote Sens Environ. 2019 Feb;221:665-674. doi: 10.1016/j.rse.2018.12.002. Epub 2018 Dec 13.
Satellite aerosol optical depth (AOD) has been widely employed to evaluate ground fine particle (PM) levels, whereas snow/cloud covers often lead to a large proportion of non-random missing AOD values. As a result, the fully covered and unbiased PM estimates will be hard to generate. Among the current approaches to deal with the data gap issue, few have considered the cloud-AOD relationship and none of them have considered the snow-AOD relationship. This study examined the impacts of snow and cloud covers on AOD and PM and made full- coverage PM predictions by considering these impacts. To estimate missing AOD values, daily gap-filling models with snow/cloud fractions and meteorological covariates were developed using the random forest algorithm. By using these models in New York State, a daily AOD data set with a 1-km resolution was generated with a complete coverage. The "out-of-bag" R of the gap-filling models averaged 0.93 with an interquartile range from 0.90 to 0.95. Subsequently, a random forest-based PM prediction model with the gap-filled AOD and covariates was built to predict fully covered PM estimates. A ten-fold cross-validation for the prediction model showed a good performance with an R of 0.82. In the gap-filling models, the snow fraction was of higher significance to the snow season compared with the rest of the year. The prediction models fitted with/without the snow fraction also suggested the discernible changes in PM patterns, further confirming the significance of this parameter. Compared with the methods without considering snow and cloud covers, our PM prediction surfaces showed more spatial details and reflected small-scale terrain-driven PM patterns. The proposed methods can be generalized to the areas with extensive snow/cloud covers and large proportions of missing satellite AOD data for predicting PM levels with high resolutions and complete coverage.
卫星气溶胶光学厚度（AOD）已被广泛用于评估地面细颗粒物（PM）水平，然而雪/云覆盖往往导致很大一部分AOD值非随机缺失。因此，难以生成完全覆盖且无偏差的PM估计值。在当前处理数据缺口问题的方法中，很少有方法考虑云与AOD的关系，且没有一个方法考虑雪与AOD的关系。本研究考察了雪和云覆盖对AOD和PM的影响，并通过考虑这些影响进行了全覆盖的PM预测。为了估计缺失的AOD值，利用随机森林算法开发了包含雪/云分数和气象协变量的每日填补缺口模型。通过在纽约州使用这些模型，生成了一个分辨率为1公里、具有完整覆盖的每日AOD数据集。填补缺口模型的“袋外”R平均值为0.93，四分位间距为0.90至0.95。随后，建立了一个基于随机森林的PM预测模型，该模型使用填补缺口后的AOD和协变量来预测全覆盖的PM估计值。对预测模型进行的十折交叉验证显示性能良好，R为0.82。在填补缺口模型中，与一年中的其他时间相比，雪分数在雪季具有更高的显著性。包含/不包含雪分数的预测模型也表明PM模式有明显变化，进一步证实了该参数的重要性。与不考虑雪和云覆盖的方法相比，我们的PM预测表面显示出更多的空间细节，并反映了小尺度地形驱动的PM模式。所提出的方法可以推广到雪/云覆盖广泛且卫星AOD数据大量缺失的地区，以高分辨率和全覆盖预测PM水平。
