Hyper Local Air Quality Monitoring using NOAA Satellite Data

Poor air quality and its impact on human health in urban and industrial areas with high population density is being increasingly researched from an environmental injustice perspective. Because pollutant monitors cannot be placed in every neighborhood, high resolution satellite observations have become valuable for monitoring and tracking changes of exposure to harmful levels of pollution such as fine particulate matter and nitrogen dioxide. Though most regions in the United States are in attainment of fine particle pollution standards(PM_2.5 – mass of particles smaller than 2.5 µm in median diameter, µg/m³), smoke from fires has become the leading cause for exceeding the Environmental Protection Agency (EPA) health standard (daily average PM_2.5 > 35 µg/m³) in many communities. Even though health standards are met, certain communities are continuously or periodically exposed to pollution from local sources. Operational air quality forecasters provide alerts to minimize exposure risk to smoke and pollution from local sources using a host of tools including near real time satellite imagery. NOAA’s near real time hourly and daily PM2.5 estimates at 2 km resolution are currently available CONUS-wide to monitor air quality conditions. We are adding the capability for hyper local air quality monitoring to our data dissemination system. As a pilot, we are partnering with the Maryland Department of Environment to monitor pollution for three different communities impacted by fine particle pollution from local sources – Curtis Bay and Turner Station in Baltimore and Cheverly outside of Washington, District of Columbia. We will present the characterization of PM_2.5 for these three communities using GOES-16 Advanced Baseline Imager (ABI) estimates. We can monitor PM_2.5 on hourly to monthly to yearly time scales and detect gradients in PM_2.5 concentrations ranging from low to high; PM_2.5 levels remain low when the source of pollution is local whereas values are much higher when long-range transport of smoke impacts regional air quality . Percent mean bias of ABI hourly PM_2.5 estimates is between 12% for low pollution levels to -55% for hazardous pollution levels with an overall bias of -11%. Based on 2020 data, on most days, there is at least one hourly observation per day in any given location in the absence of persistent cloud cover. Hourly changes in PM_2.5 are discernable at both low and high pollution levels; over the bay and coastal ocean, we flag and screen satellite data for uncertainties. Hyper local monitoring of air quality is expected to shed light on the environmental injustice aspect of pollution exposure specific to known sources.