全球高分辨率高质量PM2.5数据集（2017-2022）

GlobalHighPM2.5数据集是全球高分辨率高质量近地表空气污染物数据集（GlobalHighAirPollutants, GHAP）中PM2.5数据集。该数据集利用人工智能技术，使用模式资料填补了卫星MODIS MAIAC AOD产品的空间缺失值，结合地基观测、大气再分析和排放清单等大数据生产得到2000年至今全球无缝隙地面PM2.5数据。数据十折交叉验证决定系数R2为0.91，均方根误差RMSE为9.2 µg/m3。主要范围为整个全球陆地区域，空间分辨率为1 km，时间分辨率为日、月、年，单位为µg/m3。注意：该数据集以世界时（UTC）记录，持续更新，如需要更多数据，请发邮件联系作者（[e-mail]; [e-mail]）。 数据文件中包含NC转GeoTiff的四种代码（Python、Matlab、IDL和R语言）nc2geotiff codes。

The GlobalHighPM2.5 dataset is the PM2.5 subset of the GlobalHighAirPollutants (GHAP) dataset, a high-resolution and high-quality global near-surface air pollutant dataset. This dataset leverages artificial intelligence technologies, uses modeling data to fill the spatial missing values of the satellite MODIS MAIAC AOD products, and integrates multiple big data sources including ground-based observations, atmospheric reanalysis datasets and emission inventories to generate global seamless near-surface PM2.5 concentration data from 2000 to the present. The dataset achieves a 10-fold cross-validation coefficient of determination (R²) of 0.91 and a root-mean-square error (RMSE) of 9.2 µg/m³. It covers the entire global terrestrial area, with a spatial resolution of 1 km, temporal resolutions of daily, monthly and annual, and the unit is µg/m³. Note: This dataset is recorded in Coordinated Universal Time (UTC) and is continuously updated. If you require additional data, please contact the authors via email at [e-mail]; [e-mail]. The data files include four sets of nc2geotiff conversion codes for transforming NC (NetCDF) files to GeoTIFF, available in Python, Matlab, IDL and R languages.

搜集汇总

背景与挑战

背景概述

该数据集为2017-2022年全球陆地区域的高分辨率PM2.5数据，空间分辨率1km-10km，时间分辨率包括日、月、年，数据精度高（R2=0.91，RMSE=9.2 µg/m3），采用人工智能技术结合多源数据生产，并以NetCDF格式提供。