Air quality data - Comunitat Valenciana

PM2.5 maps are a central component of air quality monitoring and epidemiological studies. Modelling of air pollutants using remote sensing data relies heavily on the development of multiple layers of information to train the prediction models. This paper proposes a workflow for mapping PM2.5 based on open data. With a data science approach, the input data were AOD (MODIS), atmospheric variables (MERRA-2), land use (CORINE), terrain elevation (DEM) and ground measurements of PM2.5.

PM2.5浓度分布图是空气质量监测与流行病学研究的核心组成部分。利用遥感数据开展空气污染物建模研究，高度依赖多维度信息集的构建以用于预测模型的训练。本文提出了一种基于开放数据的PM2.5空间制图工作流。本研究采用数据科学方法，所使用的输入数据包括：中分辨率成像光谱仪（Moderate-resolution Imaging Spectroradiometer, MODIS）反演的气溶胶光学厚度（Aerosol Optical Depth, AOD）、现代研究与应用再分析资料第二版（Modern-Era Retrospective analysis for Research and Applications, Version 2, MERRA-2）提供的大气变量、CORINE土地覆盖数据库（CORINE Land Cover）的土地利用数据、数字高程模型（Digital Elevation Model, DEM）获取的地形高程数据，以及PM2.5地面实测数据。