S. 3529 — Public Health Air Quality Act of 2025

This section defines 20 terms for purposes of the Act, including cross-references to the Clean Air Act for “accidental release,” “area source,” “hazardous air pollutant,” “major source,” “new source,” and “stationary source”; EPA test methods such as Method 325A, Method 325B, Method 327, and Method TO–15A; and the National Air Toxics Trends Network and NCore. It also defines new terms such as “air quality system,” “cumulative impact,” “cumulative risk,” “emissions measurement system,” and “real-time,” and adopts PFAS definitions from 15 U.S.C. 8931.

This section directs the EPA Administrator, not later than 18 months after enactment, to design, implement, and begin a hazardous air pollutants fenceline monitoring program (i.e., emissions measurement and quantification at facility boundaries) at no fewer than 100 high-priority stationary sources lacking existing public fenceline monitoring—pursuant to Clean Air Act authorities including research, grants, inspections, and emergency orders (42 U.S.C. 7403, 7412-7414, 7603)—potentially by expanding the National Air Toxics Trends Stations network or creating a new one. Not later than 270 days after enactment, the Administrator must publish the source list after public notice and at least 60 days of comment, including at least 45 sources identified as high-priority or high-cancer-risk in specified EPA Office of Inspector General reports and at least 55 other major or area sources meeting risk criteria; substitutions are allowed after notice and comment to maintain 100 sources posing equivalent or greater health risks. The Administrator must maintain monitoring for at least 6 years (potentially shortened to 3 years after notice and comment if unnecessary for public health or compliance), conduct additional inspections and testing as needed, and thereafter sustain the program by retaining, adding, or relocating monitors per a substitution process; all plans and results must be published on the EPA website in an accessible, multilingual database for at least 10 years, with data submitted electronically within 1 month of collection and released publicly within 7 days thereafter.

This section directs the EPA Administrator to promulgate, not later than two years after enactment, Clean Air Act regulations—using authorities including CAA section 103—for specified source categories of major and area sources of hazardous air pollutants (HAPs), requiring (1) all sources to implement, within one year, the best available emissions measurement (including continuous emissions monitoring and fenceline monitoring) to ensure compliance with HAP emission standards; (2) facilities with risk management plans under CAA section 112(r)(7) to implement continuous real-time monitoring and fenceline monitoring for emergency response and release prevention; (3) fenceline corrective action levels for the top five HAPs driving cancer, chronic noncancer, or acute risks in the category, plus corrective action for any release of CAA section 112(r)(3) substances; (4) specified responses to exceedances, including root cause analysis, full remedial action (with community input), and public reports certifying Clean Air Act violations; and (5) enforcement of these requirements as CAA section 112 standards under section 113. The source categories include (1) major or area sources emitting HAPs on a specified list, in the latest National Emissions Inventory, in designated North American Industry Classification System codes (e.g., chemical manufacturing [325], petroleum [324]), subject to court or statutory review deadlines for emissions standards, or with recent accidental releases reportable under CAA section 112(r); and (2) any other category where fenceline monitoring benefits public health (including children's health). As background, fenceline monitoring measures ambient HAP concentrations around a facility perimeter to detect fugitive emissions impacting nearby communities. The Administrator must determine the best available monitoring methods—consulting specified EPA offices and ensuring stringency at least matching current EPA methods (e.g., Method 325A/B for benzene fenceline, Method TO-15A for volatile organic compounds)—and may designate fenceline approaches as the best available for stationary sources in these categories. (Thus, the regulations enhance community protections from air toxics via real-time data, exceedance responses, and public transparency.)

This section directs the EPA Administrator to deploy 80 additional NCore multipollutant monitoring stations (i.e., National Core stations that monitor criteria pollutants and others) using Federal reference or equivalent methods capable of assessing compliance with national ambient air quality standards (NAAQS). Not later than 18 months after enactment, the stations must be installed, integrated into the national air quality monitoring network under Clean Air Act sections 110(a)(2)(B) and 319, and continuously operated and maintained, with data used for NAAQS compliance assessments, integrated science assessments for NAAQS reviews under section 109, evaluating pollution exposure disparities within metropolitan areas, and other public health protection purposes. At least 40 stations must be sited outside metropolitan statistical areas with populations of 50,000 or greater, in census tracts with elevated rates of childhood or adult asthma, chronic obstructive pulmonary disease, heart disease, or cancer (not less than 5% above the national average); above-national-average poverty, lack of high school diploma (ages over 18), or unemployment; two or more major sources of emissions nearby; or above-national-average populations of vulnerable individuals at greater risk from NAAQS pollutants. In approving sites, the Administrator must invite community proposals (including via regional or virtual meetings); prioritize locations based on potential NAAQS exceedances (potentially using hybrid methods combining monitors, sensors, modeling, and satellite data), at-risk populations, inadequate existing monitoring, and unclassifiable or attainment designations; and conduct public hearings, at least 60 days of comment following Federal Register notice (plus email, news release, and website posting), and publish siting explanations online. (Thus, the stations will expand the monitoring network to better capture air quality in underserved and high-risk areas, informing regulatory decisions and health protections.)

This section directs the EPA Administrator, not later than two years after enactment, to deploy at least 1,000 low-cost air quality systems (each costing not more than $5,000) in clusters of at least five units each in prioritized census tracts or counties (per prioritization criteria in section 5(d)(2)), after inviting site proposals from impacted communities; authorizes contracts with nonprofits (including academic institutions) and state/local air agencies for monitoring and reporting; and requires public dissemination of data via an accessible platform. Systems must be commercially available, provide data accurate enough to assess risk of exceeding national ambient air quality standards (NAAQS) under Clean Air Act §109 (42 U.S.C. 7409) (i.e., health-based standards for criteria pollutants such as ozone and particulate matter), and be the lowest-cost option meeting those criteria; at least 500 systems must measure ozone, PM2.5, oxides of nitrogen, or sulfur dioxide, with others optionally including sensors for wind speed/direction, humidity, carbon dioxide, carbon monoxide, or source-identification inputs, selected based on local pollution sources; a cost exception applies for unavailable systems if accuracy requirements are met. Not later than 18 months after deployment (following one year of data), the Administrator must determine if systems in a tract/county measured any listed pollutant at or above 98% of the applicable NAAQS over that year, requiring installation of Federal reference or equivalent method monitors for exceeding pollutants within 180 days unless waived after notice/comment based on clear evidence of sensor inaccuracy or complementary data (e.g., meteorology, nearby monitors, modeling, satellites) showing implausibility. Not later than one year after enactment (after 60-day public comment), the Administrator must publish online a report identifying additional areas for low-cost system utility in decisionmaking or public access. This section authorizes $6 million for FY2026.

This section directs the EPA Administrator, not later than two years after enactment and after public notice and a 60-day public comment period, to amend subpart A of 40 C.F.R. part 51 to expand air emissions reporting requirements under the Air Emissions Reporting Rule (AERR)—which supports the National Emissions Inventory (NEI) for air quality planning and modeling—for all major and non-major sources to include additional emissions data such as hazardous air pollutants (HAPs) and per- and polyfluoroalkyl substances (PFAS). The required amendments must, at a minimum, (1) incorporate all changes proposed in EPA's August 9, 2023, notice entitled “Revisions to the Air Emissions Reporting Requirements” (88 Fed. Reg. 54118); (2) require reporting of emissions during source malfunctions; and (3) mandate HAP emissions reporting, including for PFAS, in the reporting cycle immediately following a pollutant's listing as a HAP. The amendments take effect for the first NEI inventory year after finalization.

This section directs the EPA Administrator, to the extent practicable, to either (1) restore public access to the Environmental Justice Screening and Mapping Tool (EJSCREEN) or (2) create a similar nationwide geospatial mapping and screening tool—including, at minimum, all data previously in EJSCREEN—and make the new tool available online for public comment within 270 days of enactment. The section further requires integration into the restored or new tool of data collected through programs established under this Act. (As background, EJSCREEN identifies communities with potential environmental justice concerns based on demographic, health, and environmental data.)