USHighPM2.5: Daily 1 km Ground-Level PM2.5 for the United States (2000–Present)

USHighPM2.5 provides seamless daily, monthly, and yearly 1 km resolution ground-level PM2.5 data for the United States from 2000 to the present. The dataset was generated by Jing Wei and collaborators from the University of Maryland, College Park, using artificial intelligence to fuse ground measurements, satellite data, atmospheric reanalysis, and model simulations. It exhibits a cross-validation coefficient of determination (CV-R2) of 0.92 and a normalized root-mean-square error (NRMSE) of 0.4 on a daily basis.

Use Cases

• Modeling long-term mortality trends attributed to PM2.5 based on the dataset's daily time series from 2000 onward.
• Analyzing spatiotemporal heterogeneity of air pollution across the continental USA based on the seamless 1 km spatial resolution.
• Validating regional air quality models based on the high-quality ground-level PM2.5 estimates derived from multiple data sources.
• Assessing public health exposure to fine particulate matter based on the dataset's ground-level concentration data.

Strengths

• High spatial resolution of 1 km and seamless 100% spatial coverage for the United States.
• Demonstrated high quality with a cross-validation R² of 0.92 and NRMSE of 0.4 on a daily basis.
• Long temporal coverage from 2000 to the present, available at daily, monthly, and yearly intervals.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of Maryland, College Park (Jing Wei et al.)

Collection Method

Generated using artificial intelligence to fuse ground-based measurements, satellite remote sensing products, atmospheric reanalysis, and model simulations.

Time Range

2000 to present

Geography

United States