ChinaHighNO2: 5 km Ground-level Nitrogen Dioxide Data for China, 2019

ChinaHighNO2 provides monthly and yearly ground-level nitrogen dioxide (NO2) data for China in 2019 at a 5 km resolution. The dataset was generated by Jing Wei at the University of Maryland, College Park, using artificial intelligence to integrate satellite remote sensing, ground measurements, and atmospheric models. It reports a cross-validation coefficient of determination (CV-R2) of 0.87 and a root-mean-square error (RMSE) of 6.36 µg m-3 on a daily basis.

Use Cases

• Analyzing spatiotemporal patterns of ground-level NO2 pollution across China based on the 5 km resolution data.
• Validating atmospheric chemistry and transport models using the high-quality, AI-integrated product mentioned in the description.
• Assessing long-term air quality trends and policy impacts for 2019, as the dataset is part of a long-term series.
• Studying the relationship between NO2 concentrations and other environmental or socio-economic factors using the multi-source fused data.

Strengths

• High spatial resolution of approximately 5 km (≈ 0.05°) for detailed analysis.
• Reported high data quality with a cross-validation R² of 0.87 and RMSE of 6.36 µg m⁻³ on a daily basis.
• Generated from multiple data sources including satellite remote sensing, ground measurements, and model simulations, likely improving accuracy.
• Part of the ChinaHighAirPollutants (CHAP) series, suggesting a consistent methodology for long-term, full-coverage data.

Limitations

• The dataset is described as closed access, with users directed to a newer version, limiting immediate availability of this specific 2019 product.
• Column-level documentation is absent; field semantics must be inferred after obtaining the data.
• Row count and exact file formats are unknown, which may limit suitability assessment for specific computational tasks.

Provenance

Source

Jing Wei, University of Maryland, College Park, via the paperswithcode platform.

Collection Method

Generated using artificial intelligence to integrate big data from ground-based measurements, satellite remote sensing (TROPOMI), atmospheric reanalysis, and model simulations.

Time Range

2019

Geography

China