docs/us_environment/environment_air_quality_county_year

environment_air_quality_county_year

County x year x metric grain. CDC/EPA Downscaler air quality estimates — PM2.5 annual mean (ug/m3) and ozone annual mean (ppb) at county level, 2001-2020. Long format: one row per (county_id, year, metric). Tier B modeled estimates — Bayesian spatial fusion of EPA AQS monitor data with CMAQ model output. Contiguous US only; AK/HI/territories absent. Do not use for personal exposure or precise rural values (model-dependent). Opens the us_environment domain.

tier bTier B: modeled estimates — Bayesian spatial fusion of EPA AQS monitor data with CMAQ model output. Not direct measurements.environmentair-qualitypm25ozoneepacdchealth
grain
county_id-year
years
2001 – 2020
cadence
Irregular (model releases; last update 2020)
overview

overview

~62,000 county-year rows in wide format. PM2.5 annual mean and ozone annual mean for all US counties.

current vintage — 2020

history — Yes — 2001–2020

provenance

source & licensing

authority
U.S. Centers for Disease Control and Prevention / U.S. Environmental Protection Agency
dataset
CDC WONDER EPA Downscaler Air Quality Estimates
license
Public domain (U.S. Federal Government work)
citation
U.S. Centers for Disease Control and Prevention (2022). National Environmental Public Health Tracking Network. EPA Downscaler model.
schema

fields

nametypedefinition
country_idstringISO alpha-2 country code (always 'US' for domestic tables).
county_fipsstringFive-character FIPS code identifying the county. Matches the contract column county_id; retained as a source-native identifier.
county_idkeystring5-character FIPS code identifying the county.Part of primary key. Joins dim.counties on county_id.
data_basisstringLabel describing the methodological basis of the modeled estimate for this county-year (e.g., monitor-informed vs. model-only). Values reflect how much EPA AQS monitor coverage contributed relative to CMAQ output.
ozone_annual_mean_ppbfloat64 · parts per billion (ppb)Bayesian spatial-fusion estimate of annual mean ozone (O₃) concentration for the county-year, derived from EPA AQS monitor data fused with CMAQ model output via the CDC/EPA Downscaler. Higher values indicate worse air quality.
ozone_stddfloat64 · parts per billion (ppb)Posterior standard deviation of the ozone annual mean estimate from the Downscaler Bayesian model. Higher values indicate greater uncertainty in the modeled estimate for that county-year.
ozone_tract_day_countint64 · countNumber of census tract-day observations that were aggregated to produce the county-year ozone estimate. Lower counts suggest sparser underlying data and potentially less reliable estimates.
pm25_annual_mean_ug_m3float64 · micrograms per cubic meter (µg/m³)Bayesian spatial-fusion estimate of annual mean fine particulate matter (PM2.5) concentration for the county-year, derived from EPA AQS monitor data fused with CMAQ model output via the CDC/EPA Downscaler. Higher values indicate worse air quality.
pm25_stddfloat64 · micrograms per cubic meter (µg/m³)Posterior standard deviation of the PM2.5 annual mean estimate from the Downscaler Bayesian model. Higher values indicate greater uncertainty in the modeled estimate for that county-year.
pm25_tract_day_countint64 · countNumber of census tract-day observations that were aggregated to produce the county-year PM2.5 estimate. Lower counts suggest sparser underlying data and potentially less reliable estimates.
state_idstring2-character FIPS code identifying the state.Joins dim.states on state_id.
yearkeyint64Reference year of the observation.Part of primary key.
relationships

joins

primary key
county_id, year, metric
common joins
health_environmental_burden_county_year on (county_id, year)
dim.geographies on (county_id)
usage

how to use this table

method

Bayesian spatial downscaling model fusing EPA AQS monitoring data with CMAQ deterministic air quality model output.

do not use for

Regulatory compliance or permitting — use EPA AQS monitor data directly.

last updated · Jul 7, 2026