Multimodal Vision Research Laboratory

MVRL

Research Area: Air Quality and Atmosphere

How can satellite and geophysical data improve estimates of air pollution and atmospheric composition? We develop deep learning methods that fuse remote sensing with meteorological and land-surface variables to map fine particulate matter and related species at kilometer scales. Recent work includes daily 1-km PM estimation over North America and global fine PM mapping by incorporating geophysical variables into learning frameworks. This application area connects geospatial AI to environmental health and climate-adjacent sensing where ground monitors are sparse.

View All Publications →

All Publications

  1. Shen S, Donkelaar A van, Jacobs N, Li C, Martin RV. 2025. Enhancing Estimation of Daily 1-km Resolution Fine Particulate Matter Concentrations for North America with Deep Learning from Geophysical A Priori Information. In: American Geophysical Union (AGU) Fall Meeting Abstracts. DOI: 10.1021/acsestair.5c00251.
    doi bibtex GeoAI Air Quality
  2. Shen S, Donkelaar A van, Jacobs N, Li C, Martin RV. 2024. Enhancing Estimation of Fine Particulate Matter Species Concentrations over North America by Including Geophysical a priori Information in Deep Learning. In: American Geophysical Union (AGU) Fall Meeting Abstracts.
    bibtex GeoAI Air Quality
  3. Shen S, Li C, Donkelaar A van, Jacobs N, Wang C, Martin RV. 2024. Enhancing Global Estimation of Fine Particulate Matter Concentrations by Including Geophysical a Priori Information in Deep Learning. ACS ES&T Air. DOI: 10.1021/acsestair.3c00054.
    doi bibtex GeoAI Air Quality