HyreHVAC

Research

Wildfire smoke and the homes that cannot filter it

Filtering a house requires a central system to filter it with. The places that need one most are the places least likely to have one.

Updated September 2026 · Data as of EPA AQI 2020, 2023, 2024; RECS 2020 microdata

Written by HYRE HVAC Research Desk Primary-source research, data analysis and fact checking

10.7% of monitored days unhealthy in California, 2020 The worst state-year in the data
68.0% of Oregon homes have ductwork Against 82.0% nationally
8.7% of US homes have an attic fan Plus 6.5% with a whole-house fan, which pulls outdoor air in

The finding

The states with the worst wildfire smoke are among the least able to filter it. California had unhealthy air on 10.7% of monitored days in 2020, and only 73.7% of its homes have ducts for a central filter.

Oregon is at 68.0% and Washington at 58.8%, against 82.0% nationally. In 2023 the worst air moved to the Midwest and East, from fires a thousand miles away.

Can you find out what air filters US homes use?

No, not from any public dataset. This study was first meant to analyze the air filters American homes actually have. It does not, because that page cannot honestly be written.

We checked all 799 variables in the RECS 2020 public use file. There is no filter variable, no MERV rating, no air cleaner and no air purifier. The federal government’s most detailed survey of American household energy use does not ask what is in the filter slot. No other public dataset fills the gap either.

So the question was narrowed to one the evidence can carry. Not what filters homes have, but whether they have the infrastructure that filtration requires.

A MERV 13 filter is useless without a system to put it in and a reason to run that system.

That turns out to be the more useful question anyway, and nobody else publishes it. (To pick a filter for your own system, see the filter selector.)

Which states have the worst smoke exposure?

Worst year: share of monitored days above the unhealthy-for-sensitive-groups thresholdAK1.7ME0.2VT1.2NH0.5WA2.5ID2.5MT1.7ND4.6MN2.8IL4.7WI4.6MI2.7NY1.6RI1.7MA1.2OR3.4NV3.3WY1.3SD3.0IA3.4IN2.3OH1.9PA1.8NJ2.0CT2.5CA10.7UT2.2CO2.6NE1.8MO2.8KY1.4WV0.8VA0.9MD1.8DE2.0AZ6.9NM2.7KS1.7AR0.6TN0.8NC0.6SC0.4DC3.3OK1.7LA0.9MS0.7AL0.6GA1.0HI0.2TX3.0FL0.2% of monitored days0.00.41.73.96.810.7Worst of 2020, 2023 and 2024. Square-root color scale.
Worst of 2020, 2023, 2024: share of monitored days above the unhealthy-for-sensitive-groups threshold. Using each state’s worst year rather than an average, because filtration is bought for bad years. HyreHVAC calculation from EPA AirData annual AQI by county.

It depends on the year. EPA publishes county-level air quality summaries. Aggregated to state, and counting days above the "unhealthy for sensitive groups" threshold as a share of monitored days, the picture across 2020, 2023, 2024 is dominated by two very different events.

2020 was a western fire season: California recorded unhealthy air on 10.7% of monitored days, Arizona on 6.9%, Oregon on 3.4%.

2023 was somebody else’s fire season: Canadian wildfire smoke pushed the worst air of the three years into states that had never planned for it: Illinois went from 0.9% of days unhealthy in 2020 to 4.7% in 2023, and Wisconsin from 0.9% to 4.6%.

HyreHVAC analysis: that second pattern is the one worth planning around. Smoke exposure is not a fixed property of a state, and a household in the upper Midwest that regarded filtration as a western problem discovered otherwise over a single week.

Any assessment of who needs filtration based on historical local fire risk will keep being wrong in this direction.

Can homes in smoky states filter their air?

The 2023 shift: where bad-air days moved0CA10.7% → 4.9%IL0.9% → 4.7%ND0.0% → 4.6%WI0.9% → 4.6%IA0.2% → 3.4%AZ6.9% → 3.3%DC0.3% → 3.3%SD0.1% → 3.0%MN0.1% → 2.8%MO0.4% → 2.8%share of unhealthy days, 2020 → 2023; bars show the change in percentage points
The ten states with the worst 2023 air, against their own 2020 figures. Bars to the right are states whose worst air arrived with smoke from elsewhere. HyreHVAC calculation from EPA AirData.

Often not. Filtering a whole house means running a central forced-air system with a good filter in it. Nationally 82.0% of homes have that setup. The three Pacific states fall short.

Washington is at 58.8%, Oregon at 68.0% and California at 73.7%: all below the national figure, and Washington by more than twenty points (see HVAC in Washington). The historical reason is mild summers and cheap hydroelectricity, which produced housing heated by baseboards and cooled by opening a window.

That second habit is the sharper problem. 47.5% of Washington homes and 23.7% of Oregon homes have no cooling equipment of any kind.

On a hot smoky day those households face a real conflict: close the windows and overheat, or open them and breathe smoke (the same households our cooling divide study counts).

A house with air conditioning can stay closed. A house without it cannot.

Nationally, 6.5% of homes have a whole-house fan and 8.7% an attic fan: equipment whose entire function is to pull large volumes of unfiltered outdoor air through the building. On a smoke day it is the worst thing in the house to switch on, and it is normally the cheapest way to cool one.

How do smoke exposure and ductwork compare by state?

StateUnhealthy days 202020232024Homes with ductsHomes with no cooling
California10.7%4.9%6.1%73.7%27.6%
Arizona6.9%3.3%4.7%92.6%5.8%
Illinois0.9%4.7%0.5%84.2%4.0%
North Dakota0.0%4.6%2.1%72.0%7.1%
Wisconsin0.9%4.6%0.7%82.4%7.6%
Iowa0.2%3.4%0.0%91.6%2.8%
Oregon3.4%1.9%1.9%68.0%23.7%
District of Columbia0.3%3.3%1.1%75.2%6.7%
Nevada3.3%1.0%1.6%88.3%8.0%
South Dakota0.1%3.0%1.0%84.4%8.0%
Texas0.9%2.1%3.0%90.4%5.0%
Minnesota0.1%2.8%0.4%79.6%5.7%
Missouri0.4%2.8%0.4%90.3%4.8%
Michigan1.3%2.7%0.4%87.6%13.4%
New Mexico1.2%1.1%2.7%78.7%15.0%
Colorado2.3%1.1%2.6%81.9%18.3%
Connecticut1.2%2.5%2.5%66.7%10.4%
Idaho2.3%0.6%2.5%80.3%19.4%
Washington2.5%0.8%0.3%58.8%47.5%
Indiana0.4%2.3%0.2%94.2%5.7%
Utah1.5%2.2%2.0%93.2%8.1%
Delaware0.2%2.0%0.4%92.5%3.0%
New Jersey0.2%2.0%0.9%76.0%3.5%
Ohio0.5%1.9%0.5%87.5%6.5%
Maryland0.1%1.8%0.6%91.2%8.6%

The 25 states with the worst single year of the three. Unhealthy days are the share of monitored days at or above "unhealthy for sensitive groups". Housing figures are RECS 2020.

AQI does not attribute fine particulate to wildfire. A high-PM2.5 day may be smoke, traffic, industry or wood burning, so these are exposure figures rather than smoke figures.

What helps during a wildfire smoke event?

If you have ducts, the filter is the cheapest upgrade you will ever make

A MERV 13 filter captures a large share of fine particulate and costs a few dollars more than the MERV 8 in the slot now.

The one check worth making first is whether the system can take the extra resistance to airflow.

An older furnace with undersized returns may not, and a technician can tell you in a single visit.

The system has to actually run

A filter only cleans air that passes through it. On a smoke day the fan needs to run continuously rather than only when there is a call for heating or cooling. Most thermostats have a "fan on" or circulate setting and almost nobody uses it.

Without ducts, it is portable units, room by room

A portable HEPA cleaner sized for the room is the realistic option in a home with no central system, which is most of the Pacific Northwest housing stock.

One unit in a hallway does very little; one in the bedroom does a great deal, because that is where the hours are. An indoor air quality assessment can size them.

Turn the whole-house fan off

Whole-house and attic fans move large volumes of unfiltered outdoor air through the building by design. They are excellent on a clear cool evening and actively harmful on a smoke day. This is the single most common mistake in a house that has one.

Do not rely on closing windows alone in a house with no cooling

A closed house in a heatwave is its own hazard, and the households facing that trade-off are disproportionately the ones without air conditioning. Where the choice is genuinely between heat and smoke, cooling the house is the intervention that makes filtration possible at all.

How did we measure it, and what are the limits?

Exposure

EPA AirData annual AQI by county for 2020, 2023, 2024. Days at or above "unhealthy for sensitive groups" (including unhealthy, very unhealthy and hazardous days) summed across all monitored days in each state.

Housing stock

EIA RECS 2020 public use microdata. Ductwork is inferred from central heating or cooling equipment; there is no direct duct variable. Whole-house and attic fans are the HOUSEFAN and ATTICFAN variables.

AQI is not a smoke measure

This is the most important limitation on the page. EPA does not attribute fine particulate to a source, so a high-PM2.5 day may be wildfire smoke, traffic, industry or residential wood burning.

The 2020 western and 2023 eastern patterns are consistent with the documented fire seasons, but this dataset cannot prove attribution and no claim of attribution is made.

Monitors are not evenly distributed

Air quality monitors are concentrated in populated counties, and a state figure is the share of monitored days rather than of population-weighted exposure. Rural exposure is under-observed, and in fire-prone rural areas it is likely understated.

Three years is not a climatology

Two of the three were chosen because they were severe. This describes the exposure households have recently experienced, which is what drives filtration decisions, but it is not a long-run risk estimate.

Ductwork presence is not filtration quality

A ducted home may have a MERV 4 filter, a leaking return pulling attic air, or a system nobody runs in June. Ductwork is the precondition for whole-house filtration, not evidence of it, and as noted above no public dataset records what is actually in the filter slot.

Questions

Does my HVAC system filter wildfire smoke?
It can, if it is a central forced-air system, if the filter is rated MERV 13 or better, and if the fan is actually running. All three are required. A central system sitting idle on a smoke day filters nothing, which is why the thermostat’s continuous-fan setting matters more than most people realize.
Which states have the worst smoke exposure?
It depends heavily on the year, which is the finding. In 2020 the worst air was western: California at 10.7% of monitored days unhealthy. In 2023 Canadian smoke put California and much of the Midwest and East at their worst levels of the three years studied. Historical local fire risk is a poor guide to who needs filtration.
What MERV rating do I need for wildfire smoke?
MERV 13 is the usual recommendation for fine particulate. First check whether your system can handle the extra airflow resistance. An older furnace with undersized return ducts may not, and forcing a high-MERV filter onto a system that cannot handle it reduces airflow and can damage equipment.
What if my home has no ducts?
Portable HEPA air cleaners, sized per room and prioritized where people spend hours, bedrooms first. This is the common case in the Pacific Northwest: only 58.8% of Washington homes and 68.0% of Oregon homes have ductwork, against 82.0% nationally.
Should I run my whole-house fan when it is smoky?
No. A whole-house or attic fan pulls large volumes of unfiltered outdoor air through the building; that is its entire function. On a smoke day it is the worst appliance in the house to run. 8.7% of US homes have an attic fan and 6.5% a whole-house fan; the two overlap, so they cannot simply be added.
How many American homes cannot filter their air at all?
18.0% of US homes have no central forced-air system, so they have nowhere to put a whole-house filter. The share is far higher in the Pacific states, which are also among the most smoke-exposed.

Written and audited by

HYRE HVAC Research Desk

Primary-source research, data analysis and fact checking

We are a research desk, not a sales floor. We read the federal microdata file, the statute or the manufacturer data sheet ourselves, and we publish the figure with the document it came from and the date we retrieved it.

Where a number cannot be traced to a primary source, we publish the shorter page and say what we could not verify.

The counts below are generated from the published pages themselves, last counted September 28, 2026, and they are what we have actually published rather than what we intend to.

13
studies published
12
federal sources read and cited
8
studies published with their full dataset as CSV
51
jurisdictions reproduced against EIA’s own tables

How this desk works

  • Primary sources only. Federal data comes from the agency that collects it, in the file that agency publishes. We do not cite an article that cites a source; we download the source and compute the figure ourselves.
  • We validate against the agency before we publish. First, we use each federal microdata file to reproduce the agency’s own published tables. Our cooling research reproduces EIA’s state estimates and standard errors for all 51 jurisdictions. That check caught a variance formula that was off by a factor of four.
  • Every estimate carries its uncertainty. These are survey figures, not counts. Standard errors are computed from the replicate weights the federal file supplies and printed beside the estimate. An estimate too imprecise to publish is reported as such rather than printed.
  • Nothing is typed by hand. Prose, tables and charts all read from one dataset built by script, so a number in a sentence and the same number in the table below it cannot disagree.
  • We publish the data, not just the conclusion. 8 of our 13 studies offer the full computed table as a CSV download on the page, so you can check the analysis or disagree with it. A study without a row-level dataset gets no download link and claims none in its structured data.
  • We correct in public. Where we have published a figure wrongly we fix the figure, rewrite any analysis that rested on it rather than patching the number underneath it, and leave a dated correction note on the page.
  • We do not install or sell HVAC equipment, and we take no payment for placement, ranking or a favorable mention. Nobody buys a position on this site.

Data as of EPA AQI 2020, 2023, 2024; RECS 2020 microdata. Authorship on this site is organizational: the analysis belongs to the desk rather than to a named individual, and we do not publish credentials we do not hold. Our editorial policy sets out how we source, date and correct what we publish.

The data behind this page

Every figure on this page is computed from one file, and that file is published here so the analysis can be checked, disagreed with, or reused.

Download the dataset (CSV)

Sources & retrieval dates

EPA AirData, Annual AQI by County , County-level Air Quality Index summaries for 2020, 2023, 2024, aggregated to state. Days above the "unhealthy for sensitive groups" threshold are summed across all monitored days in each state. AQI does not attribute fine particulate to wildfire, so this is an exposure measure rather than a smoke measure. Retrieved September 4, 2026.
EIA Residential Energy Consumption Survey 2020 , Public use microdata file v7. 18,496 responding households, weighted to represent 123.53 million occupied primary housing units. Standard errors from the 60 replicate weights supplied with the file. Retrieved September 4, 2026.

Getting filtration into a house that has none

Whether the answer is a filter upgrade, a ducted system or portable units comes down to what is already in the building.

Sizing estimator All HVAC calculators