Nitrogen dioxide (NO₂) exposure impacts human and environmental health and is linked to increased rates of respiratory illness and mortality. Monitoring surface amounts of this air quality indicator requires measurements, and to convert columns from satellite-based measurements to surface volume mixing ratios can potentially help quantify exposure locally. Toward this objective, column-to-surface estimates have been derived from Pandora UV-visible spectrometer direct sun measurements in the Greater Toronto Area and in the Detroit-Windsor Area. The performance of this method will be discussed in this presentation along with a summary of progress toward estimating surface volume-mixing ratios from satellite NO₂ column measurements using a machine learning approach for column-to-surface conversion.