LGHAP v2: Global Daily 1-km Gap-free PM2.5 Grids (2000-2021)

22-year-long gap-free daily aerosol optical depth and PM2.5 concentration grids at 1-km resolution covering global land areas from 2000 to 2021. Kaixu Bai from East China Normal University created this dataset using an improved big earth data analytic framework that integrates satellite, ground monitor, and numerical model data. A scene-aware ensemble learning graph attention network was developed to reduce modeling bias in regions with sparse ground measurements.

Use Cases

• Modeling long-term human exposure to fine particulate matter based on the global daily 1-km PM2.5 grids.
• Analyzing trends in aerosol optical depth for climate science based on the 22-year gap-free AOD data.
• Training machine learning models for air quality prediction in data-sparse regions based on the scene-aware ensemble learning method described.
• Conducting environmental policy impact assessments based on high-resolution, gap-free historical pollution data.

Strengths

• 22 years of continuous daily data from 2000 to 2021.
• Global land coverage at a high 1-kilometer spatial resolution.
• Gap-free data generation via integration of multimodal satellite, ground, and model sources.
• Includes provided Python, MATLAB, R, and IDL codes for data reading and visualization.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.
• Data may reflect geographic bias inherent to the integration of diverse satellite and ground sources with varying coverage.

Provenance

Source

East China Normal University

Collection Method

Integration of multimodal AODs and air quality measurements from diverse satellites, ground monitors, and numerical models using an attention-reinforced tensor construction framework.

Time Range

2000 to 2021

Geography

Global land area