Mapping Air Quality in Baltimore City

In 2018, Associate Professor Kirsten Koehler and Associate Professor Abhirup Data explored whether low-cost air pollution monitors could be used to accurately map air quality in Baltimore City. Unlike regulatory-grade instruments, quality standards for low-cost sensors are yet to be established and their data quality varies widely.  

With funding from the Initiative, the researchers evaluated and field-calibrated the particulate data from a network of low-cost monitors currently operating in Baltimore. They found significant variability and overestimation of particulate levels in the raw data collected from the low-cost monitors in Oldtown, Baltimore. Laboratory corrections reduced the bias in the data, but only partially lessened the high variability. The researchers then developed a calibration model that substantially improved predictability over the raw or lab-corrected data.  

The team evaluated the calibration model at a second site in suburban Maryland to assess whether it would transfer to other monitors in the network. The statistically calibrated data once again produced the most accurate results. The calibrated particulate readings from the monitors in the low-cost network provided valuable insights into the variations of particulate matter in Baltimore. 

The case-study demonstrated the utility of the low-cost network in understanding the spatio-temporal variations of particulate matter in Baltimore city. Their findings were published in Atmospheric Environment in 2020.  

Currently the calibrated particulate data from the low-cost network is being used in various studies of air quality and its association with health and other variables in Baltimore. Dr. Datta  and Dr. Koehler were also recently awarded a grant from the National Institute of Environmental Health Sciences to pursue further development and application of statistical methods for analysis of low-cost air-pollution data.