Best Practices for Calibrating Low-Cost Air Quality Sensors in Resource-Constrained Environments

Air quality (AQ) sensors are democratizing environmental monitoring by providing unprecedented spatial resolution for AQ and exposure data. While these tools empower community-led science and fill critical gaps where reference-grade instruments are impractical, their utility is often undermined by data instability. Without rigorous, ongoing calibration, AQ sensor networks risk producing non-comparable data that lacks the credibility required for policy or public health action. This workshop addresses the technical hurdles of maintaining data integrity in "reference-poor" regions, focusing on: Optimal temporal requirements: defining the minimum co-location duration for robust baseline characterization. Network scalability: developing transferable calibration frameworks in areas with limited reference infrastructure. Drift detection: establishing protocols to identify when sensor aging or environmental shifts necessitate recalibration. Algorithmic trade-offs: evaluating the performance of linear models versus complex machine learning architectures.