EPA Toxic Release Inventory: 35 Years of Industrial Chemical Releases and Environmental Justice Patterns

Every year, more than 20,000 industrial facilities in the United States are required by law to report exactly how much of 800-plus listed toxic chemicals they released to the air, discharged to waterways, buried on their land, or shipped off-site to disposal. The federal database that collects and publishes this information—the Toxic Release Inventory—is the most comprehensive public record of industrial chemical releases in the world. It was created in direct response to a catastrophe: the 1984 Bhopal disaster, in which a methyl isocyanate release from a Union Carbide pesticide plant in India killed more than 3,000 people overnight and injured hundreds of thousands more. Congress responded by enacting the Emergency Planning and Community Right-to-Know Act of 1986. Section 313 of that statute established the TRI. The premise was simple: communities have a right to know what industrial chemicals their neighbors are releasing into their environment.

Thirty-five years of TRI data make it one of the most analytically rich federal environmental datasets available. On-site releases have declined roughly 75 percent since 1988—a trend that reflects genuine progress in chemical management, process improvement, and regulatory pressure, but one that requires careful interpretation. What remains is still substantial: 650 million pounds of on-site releases annually, and more than 30 billion pounds when off-site transfers are included. This article covers what TRI requires, who must report, the structure of the dataset, how to access it in bulk, how to analyze air releases by industrial sector, the environmental justice patterns embedded in the geographic distribution of releases, and the RSEI model that converts raw release pounds into toxicity-weighted population exposure estimates.

The statutory basis: EPCRA Section 313

The Emergency Planning and Community Right-to-Know Act of 1986 (EPCRA) contains four major titles addressing chemical hazard disclosure and emergency planning. Title III, Section 313 is the TRI provision. It directs the EPA Administrator to establish a toxic chemical release reporting requirement, compile the submitted data into a publicly accessible inventory, and update the chemical list as new toxicological information warrants. The statute gave EPA authority to add chemicals to the list whenever a substance “is known to cause or can reasonably be anticipated to cause cancer or teratogenic effects, or serious or irreversible” chronic or acute health or environmental effects.

The Right-to-Know framing was deliberate. Congress structured EPCRA not as a permitting statute—TRI reporting does not authorize releases and does not set emission limits—but as a disclosure statute. A facility that reports 10 million pounds of styrene emitted to air is not in violation of TRI for doing so; it is complying with TRI by reporting accurately. Enforcement actions arise from failure to report, inaccurate reporting, or underreporting, not from the release quantities themselves. The theory was that public disclosure would create market and community pressure that regulatory limits alone could not achieve. The 75-percent decline in reported releases since 1988 is at least partly attributable to that pressure.

Who must report

TRI reporting obligations apply to facilities that satisfy three conjunctive conditions: they must have ten or more full-time equivalent employees; they must fall within a covered industry sector; and they must manufacture, process, or otherwise use a listed chemical above applicable threshold quantities during the calendar year.

The covered industry universe originally tracked Standard Industrial Classification codes for manufacturing (SIC 20–39). Over time Congress expanded coverage: metal mining (SIC 10, 12) was added in 1997; electric utilities (coal- and oil-fired power plants) in 1998; federal government facilities in 1997 under Executive Order 12856; and certain chemical distribution and hazardous waste treatment, storage, and disposal sectors in later amendments. The current covered sectors map to NAICS codes and include manufacturing broadly, mining, electric power generation, and a handful of other sectors. Retail trade, healthcare, and most service industries are not covered.

The threshold quantities vary by chemical and by how the chemical is used. The general thresholds are 25,000 pounds manufactured or processed per year, or 10,000 pounds otherwise used. “Otherwise used” covers chemicals consumed as a fuel, used as a solvent, or used as a cleaning agent—any use that is not manufacturing or processing. For chemicals designated as Persistent, Bioaccumulative, and Toxic (PBT)—a category that includes dioxins, mercury compounds, and polychlorinated biphenyls—the thresholds are dramatically lower: 10 pounds per year for dioxins, 0.1 gram per year for dioxin compounds classified as the most toxic congeners. These lower thresholds reflect the extreme toxicity and environmental persistence of PBT chemicals, where even small releases represent meaningful public health exposure.

The TRI chemical list: 800-plus substances

The TRI chemical list as of 2024 covers more than 800 individual chemicals and chemical categories. The list spans a wide range of hazard profiles: acute toxics like hydrogen cyanide and chlorine; carcinogens like benzene, arsenic compounds, formaldehyde, and vinyl chloride; reproductive and developmental toxicants like lead compounds and toluene; persistent organic pollutants like dioxins and PCBs; heavy metals including mercury, cadmium, chromium, and nickel compounds.

The most consequential recent expansion came in 2024, when EPA added more than 180 per- and polyfluoroalkyl substances (PFAS) to the TRI list. PFAS chemicals— which include PFOA, PFOS, and hundreds of related compounds—are sometimes called “forever chemicals” because they do not break down in the environment or in human tissue. They have been linked to kidney cancer, thyroid disease, immune disruption, and developmental effects. For decades, PFAS were not required to be disclosed under TRI because they were not on the list. The 2024 additions represent the largest single expansion of TRI reporting requirements since the program began. Facilities that manufacture, use, or process covered PFAS above threshold quantities must now report, and the resulting data will for the first time give communities systematic public information about which industrial sites in their area are releasing PFAS to air, water, or land.

Dataset structure: what each TRI record contains

TRI data is organized at the facility-chemical-year level: each record represents one facility's report for one listed chemical in one reporting year. A facility that uses twenty listed chemicals submits twenty separate Form R reports (or Form A certifications, discussed below) and generates twenty records in the TRI database. Key fields in each record:

• TRI Facility ID — EPA's persistent identifier for the reporting facility, constructed from the facility's state, county, and a sequential number. The TRI Facility ID is the primary join key for linking TRI records across years and to other EPA databases.
• Facility name, address, state, county, latitude, longitude — The facility's reported name and physical location. Latitude and longitude enable spatial joins to census tract demographics, EPA EJScreen scores, and other geographically referenced datasets. Address fields are self-reported and require normalization for longitudinal analysis: facilities sometimes change their reported name or address while maintaining the same TRI Facility ID.
• DUNS/EIN — optional identifiers linking the reporting facility to its parent company. Coverage is incomplete: many smaller facilities do not submit DUNS numbers, and the DUNS system is no longer the primary federal business identifier following the government-wide transition to the Unique Entity Identifier (UEI) system. Parent company linkage in TRI is an imperfect but useful starting point for corporate-level release aggregation.
• SIC code / NAICS code — industry classification for the reporting facility. Pre-2006 records carry SIC codes; post-2006 records carry NAICS codes. For multi-decade analysis, a SIC-to-NAICS crosswalk table is necessary to maintain consistent sector groupings. EPA publishes a concordance table, though the mapping is many-to-many in both directions.
• Chemical name and CAS number — the listed TRI chemical being reported. CAS number is the authoritative chemical identifier; chemical names have synonyms and variations. For chemical category entries (e.g., “Lead compounds”), no CAS number is assigned because the category encompasses multiple distinct chemical structures.
• Reporting year — the calendar year for which releases are being reported. TRI reports are due July 1 of the year following the reporting year; the 2023 reporting year data is due July 1, 2024.
• On-site air releases: stack and fugitive — stack air releases are controlled emissions from point sources: stacks, vents, ducts, pipes that direct air emissions to a defined exit point. Fugitive air releases are diffuse, uncontrolled emissions: evaporation from storage tanks, leaks from valves and flanges, evaporation from waste pits, and windblown dust from materials piles. Both are reported in pounds per year. The distinction matters for air quality modeling: stack releases can be modeled as point sources with known exit velocity and temperature; fugitive releases require area source modeling.
• On-site water releases — direct discharges to surface water (streams, rivers, lakes, coastal waters) from the facility's own outfalls. Facilities holding NPDES permits under the Clean Water Act report these discharges to EPA separately through discharge monitoring reports; TRI captures the same information as a self-reported complement.
• On-site land releases — disposal to surface impoundments (evaporation ponds, tailings ponds), land treatment units (where waste is applied to and incorporated into soil), landfills on facility property, and other on-site land disposal methods.
• Off-site transfers by disposition — quantities shipped off-site and the type of management applied: transfer to publicly owned treatment works (POTW) for wastewater treatment; transfer to off-site waste treatment facilities; transfer to recycling; transfer to energy recovery (burning as fuel); transfer to disposal (landfill or injection well off-site). Off-site transfers are not “releases” in the strict sense—the chemical is being managed—but they are tracked because the ultimate disposition affects whether and how much eventually reaches the environment.
• Production ratio — a ratio indicating whether the facility's production of the chemical in the reporting year was higher, lower, or about the same as in the prior year. The production ratio is a coarse normalization variable: facilities that increased production but reduced absolute releases are demonstrating genuine efficiency improvements, while facilities that reduced releases only because production fell should not receive credit for pollution reduction.
• Waste minimization activities — a qualitative field where facilities describe source reduction, recycling, and pollution prevention practices implemented during the reporting year. This field is narrative and not easily quantified, but it provides context for interpreting release trends.

Form R vs. Form A: full reporting and simplified certification

Facilities that manufacture, process, or otherwise use a listed chemical above threshold quantities must generally file the full Form R, which collects the complete suite of release and transfer quantities described above. Form R is a detailed reporting form that requires facility-level estimation of releases by medium (air, water, land) and transfers by disposition category.

Facilities that qualify for the Form A simplified certification may substitute that form for Form R when two conditions are both met: the facility's total annual reportable amount for the chemical does not exceed 500 pounds (across all releases and off-site transfers combined), and the facility has not manufactured, processed, or otherwise used more than one million pounds of the chemical during the year. Form A does not require reporting specific release quantities; instead, the facility certifies that it meets the eligibility criteria. From an analytical standpoint, Form A records represent a lower bound on releases from qualifying facilities— the actual release is somewhere between zero and 500 pounds, but the exact amount is not disclosed. National-level TRI release totals exclude Form A facilities or assign them a zero or threshold value depending on the analytical convention used.

Scale of the program: 35 years of trends

In the first full reporting year (1988), TRI-covered facilities reported approximately 3.4 billion pounds of on-site releases. By 2023, that figure had declined to roughly 650 million pounds—a reduction of about 75 percent in absolute terms. When transfers to off-site disposal are included, the total waste-managed quantity runs to more than 30 billion pounds annually, a figure that reflects the scale of industrial chemical use across covered sectors.

The declining trend is real but requires careful interpretation. Several forces drive it simultaneously, and they have different implications for environmental policy:

• Genuine pollution prevention — facilities have invested in process changes that eliminate the generation of toxic waste streams at the source: reformulating products to remove listed chemicals, substituting less-toxic inputs, and redesigning processes to capture and reuse materials that previously became waste. This is the most environmentally valuable form of reduction and is real.
• Chemical substitution outside the TRI list — facilities that replace a TRI-listed chemical with an unlisted substitute will show reduced TRI releases even if the substitute has similar or greater hazard. The history of PFAS illustrates this: manufacturers substituted short-chain PFAS for long-chain PFAS partly to avoid regulatory scrutiny, and for years those substitutes were not on the TRI list. The 2024 PFAS additions partially close this gap, but the general problem of off-list substitution remains a limitation of TRI as a comprehensive chemical hazard indicator.
• Economic contraction in covered sectors — manufacturing employment and output fell substantially in the United States between 1988 and the present, particularly in heavy manufacturing. Some portion of the decline in TRI releases reflects the offshoring of industrial production rather than genuine environmental improvement at facilities that remain operating. Production ratio data helps identify this effect at the facility level.
• Transfer to off-site management — some of what appears as reduced on-site releases represents a shift to off-site transfer, where the chemical is now being managed at a separate facility rather than released on-site. TRI captures both, so a complete analysis looks at total releases plus transfers to disposal rather than on-site releases alone.

Accessing TRI data: TRI Explorer and bulk downloads

EPA publishes TRI data through two primary channels suited to different research needs.

The TRI Explorer at epa.gov/triexplorer is the web-based query tool for TRI data. It supports searches by facility name, state, county, ZIP code, chemical name, industry sector, and reporting year. Results can be filtered to specific release media (air, water, land) and displayed at the facility, county, or state level. TRI Explorer is appropriate for looking up a specific facility or chemical or producing summary statistics for a geographic area; it is not suitable for bulk extraction or programmatic analysis.

The TRI Basic Plus data files at epa.gov/toxics-release-inventory-tri-program/tri-basic-plus-data-files-calendar-year provide annual bulk downloads of the full TRI dataset as compressed CSV files, one file per reporting year. The Basic Plus format contains the most commonly analyzed fields: facility identifiers, chemical identifiers, on-site releases by medium, off-site transfers by disposition, production ratios, and geographic coordinates. Files are named with the pattern basic_plus_YYYY_us.zip and are typically available six to twelve months after the reporting year closes. The 2023 reporting year file runs to approximately 200,000 rows for the full national dataset; multi-year analysis requires concatenating annual files.

For researchers who need more granular data than Basic Plus provides—including waste minimization narrative fields, Form A certification records, and individual transfer-destination waste handler information—EPA publishes the full TRI dataset in a multi-file format through the same bulk download page. The full dataset requires joining across several files on TRI Facility ID and chemical CAS number.

Air release analysis by NAICS sector

On-site air releases—the sum of stack and fugitive emissions—are the most policy-salient TRI release category because they represent the primary exposure pathway for surrounding communities. Unlike water discharges, which require navigating a waterway to reach a population, and land disposal, which may be contained for years before reaching groundwater, air releases begin affecting ambient air quality immediately. Grouping air releases by two-digit NAICS sector identifies which industries dominate the national air release burden.

The following Python script downloads the most recent TRI Basic Plus file, extracts stack and fugitive air release columns, aggregates to two-digit NAICS sector, and prints the ten sectors with the largest total on-site air releases:

import requests, zipfile, io, csv, collections

# TRI Basic Plus data files are published annually as zipped CSVs.
# Dataset page: https://www.epa.gov/toxics-release-inventory-tri-program/tri-basic-plus-data-files-calendar-year
# The file naming convention is: basic_plus_YYYY_us.zip
# Columns relevant for air release analysis:
#   NAICS_CODE, NAICS_TITLE
#   52 - On-Site Release to Air - Stack Air
#   53 - On-Site Release to Air - Fugitive Air
# All release quantities are in pounds.

YEAR = 2023
URL = (
    "https://www3.epa.gov/tri/current/us/basic_plus_"
    + str(YEAR)
    + "_us.zip"
)

print("Downloading TRI Basic Plus", YEAR, "...")
r = requests.get(URL, timeout=300)
r.raise_for_status()

z = zipfile.ZipFile(io.BytesIO(r.content))
inner = [n for n in z.namelist() if n.endswith(".csv")][0]

sector_air = collections.defaultdict(float)
sector_names = {}

with z.open(inner) as f:
    reader = csv.DictReader(io.TextIOWrapper(f, encoding="latin-1"))
    for row in reader:
        naics = (row.get("NAICS_CODE") or "").strip()
        title = (row.get("NAICS_TITLE") or naics).strip()
        if not naics:
            continue
        # Two-digit NAICS sector
        sector = naics[:2]
        sector_names.setdefault(sector, title[:60])

        def to_lbs(col):
            try:
                return float(row.get(col) or 0)
            except ValueError:
                return 0.0

        stack = to_lbs("52 - ON-SITE RELEASE - STACK AIR")
        fugitive = to_lbs("53 - ON-SITE RELEASE - FUGITIVE AIR")
        sector_air[sector] += stack + fugitive

ranked = sorted(sector_air.items(), key=lambda x: x[1], reverse=True)[:10]

print("\nTop 10 NAICS sectors by total on-site air releases (" + str(YEAR) + "):")
print("-" * 70)
for sector, lbs in ranked:
    name = sector_names.get(sector, sector)
    print(
        "  " + sector + "  "
        + "{:>15,.0f}".format(lbs) + " lbs  "
        + name[:40]
    )

The results consistently show that chemical manufacturing (NAICS 32–33), primary metals production (NAICS 33), paper manufacturing (NAICS 32), and petroleum and coal products manufacturing (NAICS 32) account for the largest share of total on-site air releases. Electric utilities (NAICS 22) are a prominent contributor in years before their TRI reporting declined as coal-fired generation capacity retired. Mining operations (NAICS 21) contribute substantially to fugitive dust releases, particularly for metallic mineral mining in western states where tailings and overburden generate large volumes of wind-erosion particulates.

Notable facilities and chemicals

The TRI database's facility-level granularity makes it possible to identify the largest individual contributors to national release totals. Several categories of facilities consistently appear at the top of national air release rankings:

Petroleum refineries produce large volumes of air emissions dominated by volatile organic compounds, benzene, toluene, ethylbenzene, xylene (the BTEX group), and sulfur dioxide from hydrodesulfurization units. Valero Energy's refinery complex in Texas has appeared among the largest single-facility TRI air emitters in multiple reporting years, reflecting the scale of refinery operations and the range of TRI-listed chemicals used in petroleum processing. Benzene, a Group 1 human carcinogen, is among the most significant chemicals reported by refineries both because of its toxicity and because its threshold quantities are set at the general 25,000-pound level rather than at a lower PBT level.

Secondary lead smelters—facilities that recycle lead-acid batteries into recovered lead metal—are among the largest reporters of lead compounds in the TRI database. Secondary lead smelting generates lead-bearing dust, fumes, and slag that enter the air release and land disposal categories. Lead is a TRI PBT chemical, but the compound category “Lead compounds” uses the general threshold, not the ultra-low PBT threshold, because lead compounds vary widely in solubility and bioavailability. Secondary smelters are also prominent in EPA ECHO enforcement data for air permit violations related to National Emission Standards for Hazardous Air Pollutants (NESHAP) for secondary lead smelting.

Chlor-alkali plants—facilities that produce chlorine and caustic soda by electrolyzing brine—are the dominant source of mercury air and water releases in the TRI database. Legacy mercury-cell chlor-alkali plants use liquid mercury as the electrochemical medium, and process losses result in mercury releases to air, water, and land. The U.S. chlor-alkali industry has largely transitioned away from mercury cells over the past two decades, and EPA finalized rules requiring the remaining mercury-cell plants to convert or close. The TRI database tracks this transition: mercury releases from the chlor-alkali sector have declined sharply as plants converted or shut down.

Environmental justice: who lives near TRI facilities

The geographic distribution of TRI releases is not random with respect to community demographics. A robust and growing body of research demonstrates that facilities reporting large TRI air and water releases are disproportionately located in census tracts with above-average minority populations, above-average poverty rates, and above-average cumulative environmental burden scores on EPA's EJScreen index.

The analytical approach is a spatial join. TRI Basic Plus files include facility latitude and longitude for each reporting facility. Census Bureau TIGER shapefiles define the boundaries of each census tract. Assigning each TRI facility to its enclosing census tract using point-in-polygon matching, then joining on census tract GEOID to American Community Survey demographic estimates, produces a facility-level dataset enriched with tract-level income, race, poverty, and housing characteristics. EPA's EJScreen downloadable data layer adds pre-computed environmental justice scores for each tract, incorporating both demographic vulnerability and existing environmental burden from multiple pollution sources.

The pattern that emerges is consistent across years and chemicals: facilities reporting the largest total TRI air releases are systematically located in tracts with higher percent-minority population and lower median household income than the national distribution of all regulated facilities. The concentration is not explained solely by industrial land-use patterns. Even after controlling for industry sector and facility age—factors correlated with both location and release volume—the demographic disparity in exposure proximity persists. This finding has implications for enforcement prioritization: TRI release volumes, combined with tract-level demographics, provide an objective basis for identifying communities where cumulative chemical exposure is highest and where regulatory attention may warrant additional resources.

An important methodological point: proximity to a TRI facility is not the same as measured exposure. TRI release quantities are reported in pounds per year; they do not directly translate to ambient air concentrations or population doses, which depend on stack height, meteorological conditions, receptor distance, chemical reactivity, and dozens of other site-specific variables. The raw release pounds are best understood as a first-order screening metric—a signal that warrants further analysis, not a direct measure of harm.

The RSEI model: from pounds to toxicity-weighted exposure

EPA has developed the Risk-Screening Environmental Indicators (RSEI) model to bridge the gap between raw TRI release pounds and meaningful population exposure estimates. RSEI converts TRI air release quantities into toxicity-weighted, distance-adjusted population exposure scores by applying three successive transformations.

First, RSEI applies an atmospheric dispersion model—derived from the Industrial Source Complex (ISC) air quality model—to estimate the ambient air concentration at each surrounding census tract attributable to a facility's reported air releases. The dispersion model uses facility location, stack height (estimated from SIC/NAICS category if not reported), and long-term average meteorological data for the facility's location. The result is an estimated concentration gradient that declines with distance from the facility.

Second, RSEI applies a chemical-specific toxicity weight to convert concentration to a hazard-equivalent unit. Toxicity weights are derived from EPA's Integrated Risk Information System (IRIS) and other authoritative toxicological databases. A pound of a highly toxic substance like dioxin receives an enormously higher toxicity weight than a pound of a less-toxic substance. This weighting corrects one of TRI's most significant raw-data limitations: treating a pound of formaldehyde the same as a pound of sodium sulfate.

Third, RSEI multiplies the toxicity-weighted concentration estimate at each census tract by the population of that tract to produce a toxicity-weighted population exposure score. Summing these scores across all surrounding tracts produces a facility-level RSEI score that integrates dispersion, toxicity, and population density simultaneously. RSEI scores are available as a bulk download from EPA's RSEI portal and can be joined to TRI facility identifiers for facility-level analysis.

RSEI is a screening tool, not a regulatory risk assessment. Its meteorological assumptions are long-term averages rather than site-specific measurements; its toxicity weights have significant uncertainty bounds; and it does not model indoor air exposure, dermal exposure, or ingestion pathways. But for comparing relative risk across thousands of facilities, RSEI scores are substantially more informative than raw release pounds, and they are the closest thing to a standardized, publicly available, nationally consistent exposure index that TRI data supports.

Cross-references: ECHO enforcement, Census ACS, and PHMSA transfers

TRI data reaches its analytical potential when joined to three companion datasets that address what TRI alone cannot answer.

EPA ECHO enforcement data reveals whether facilities with large TRI release volumes are also receiving enforcement attention under the Clean Air Act, Clean Water Act, or RCRA. A facility reporting millions of pounds of TRI air releases while carrying no CAA enforcement history in ECHO is an analytically significant observation: it may mean the facility is in compliance with its air permits (which set limits below TRI thresholds), that the state has not conducted recent inspections, or that violations exist but have not been formally documented. Joining TRI facility IDs to ECHO FACS IDs using geographic name-matching or the EPA Facility Registry Service (FRS) cross-reference table produces the combined release-and-enforcement dataset. ECHO data is available at echo.epa.gov.

Census ACS demographic data provides the tract-level income, race, poverty, and housing variables needed for environmental justice analysis. After spatially joining TRI facilities to census tracts, ACS 5-year estimates for the relevant year supply the denominator variables: percent minority population, median household income, poverty rate, and housing tenure. Comparing the demographic distribution of tracts hosting large TRI air emitters to the national distribution of tracts tests whether release burdens are disproportionately concentrated in vulnerable communities. The Census Bureau provides ACS data through its API and bulk data downloads at census.gov.

PHMSA pipeline incident data complements TRI in a specific way: for facilities that receive TRI-listed chemicals via pipeline rather than truck or rail—refineries, chemical plants, and terminal facilities with pipeline connections—PHMSA hazardous liquid and gas transmission incident records document whether the transport segment feeding or exiting the facility has experienced spills or leaks. Chemical facilities at the end of a pipeline that has a history of incidents represent a combined TRI-plus-PHMSA risk profile: on-site releases from the facility plus transport-segment incidents from the supply chain. PHMSA incident data is available at phmsa.dot.gov.

Limitations of TRI data

TRI is a self-reported, threshold-triggered disclosure system. Understanding what it cannot tell you is as important as knowing what it contains.

Reporting gaps from estimation methods. TRI requires facilities to report their best estimates of annual release quantities, but does not specify the estimation method. Facilities may use direct measurement (stack testing or continuous emissions monitoring), material balance calculations, emission factor methods (applying EPA-published emissions factors to process rates), or engineering judgment. These methods can produce estimates that differ by orders of magnitude for the same physical release. Facilities with sophisticated monitoring systems may report more accurate—and sometimes higher—quantities than facilities using less precise estimation methods. The result is that TRI data is more useful for comparing trends within a facility over time than for comparing absolute release quantities across facilities with different estimation practices.

Threshold triggers mean the smallest facilities are absent. Facilities below the 10-employee threshold or the processing/use thresholds are not required to report even if they use and release TRI chemicals. The program captures industrial-scale operations well but misses the long tail of smaller facilities that cumulatively may account for significant chemical use. Dry cleaners, auto body shops, and small metal fabricators may use substantial quantities of TRI-listed solvents and metal compounds without any TRI reporting obligation.

Release pounds are not dose. As discussed in the RSEI section, raw TRI release quantities cannot be directly translated to population health risk without applying dispersion models, toxicity adjustments, and exposure pathways. A facility in a sparsely populated rural area that releases ten million pounds of a moderately toxic chemical poses a different public health risk than a facility in a dense urban neighborhood that releases one million pounds of a highly toxic substance. TRI alone does not resolve this; RSEI and site-specific risk assessment are needed for that work.

PFAS data is new and incomplete. The 2024 PFAS additions to the TRI list mean that PFAS facility-level data will not appear in the database until the 2025 or 2026 bulk downloads. Historical PFAS release patterns are not captured. And the addition covers PFAS above threshold quantities at covered facilities—it does not retroactively fill the decades of unreported PFAS releases that have already occurred and that continue to contaminate drinking water supplies near manufacturing and industrial sites.