Indoor CO2 Levels in a Home Office: What the Number Means, and What It Doesn't

Indoor CO2 is a ventilation gauge, not a toxicity reading, and that difference changes what you should do about a high number. Outdoor air sits around 420 ppm. A closed room with someone breathing in it climbs from there, and how fast it climbs tells you how much fresh air is actually reaching you. The figure everyone repeats, 1,000 ppm, was never a health limit. The standards body it gets attributed to says so in plain language, and has been saying so for thirty years.

This draws on ASHRAE's 2025 position document, a 2021 field study of office workers in six countries, and a 2025 chamber study that exposed people to pure CO2 on its own. Where those disagree we say so, rather than picking the one with the better headline.

What the number on your meter is measuring

Indoors, carbon dioxide comes mostly from people. You exhale it, it accumulates, and ventilation carries it away. The concentration in a room is the result of a simple race between how much you produce and how much air gets exchanged. That is why it works as a ventilation indicator, and it is why it is not really a pollution reading in the way people assume. It is measuring how much of the air in the room has already been through someone's lungs.

ASHRAE draws that boundary explicitly. Indoor CO2 concentrations "are not overall indicators of IAQ, but they can be a useful tool if users understand how they relate to IAQ and the important limitations of their use." And one number does not travel between rooms: "A single CO2 concentration does not apply to all space types and occupancies for the purposes of assessing the ventilation rate."

So your meter is answering one question: is enough outside air getting in here. That is a narrower question than the one the display seems to be answering, and a more useful one.

Where 1,000 ppm came from, and why it is not a limit

This part is worth getting straight, because the number is quoted everywhere as though a committee once set it as a safety threshold.

The ASHRAE position document on indoor carbon dioxide, approved in February 2025, is unambiguous: "Despite many statements to the contrary, ANSI/ASHRAE Standard 62.1 does not provide a limit value for indoor CO2." On the figure itself, 1,000 ppm "has long been considered an indicator of acceptable IAQ, but this concentration is at best an indicator of outdoor air ventilation rate per person." It was taken out of later editions of the standard precisely because people kept reading it as something it was not, and the document says that misunderstanding "has resulted in many confusing and erroneous conclusions about IAQ and ventilation in buildings."

NIST's Andrew Persily made the same point at the Indoor Air 2020 conference in a paper whose title does the work of a summary: Quit blaming ASHRAE Standard 62.1 for 1000 ppm CO2. Standard 62.1, he notes there, "has not contained an indoor CO2 limit for almost 30 years."

The limits that do exist sit much higher and come from somewhere else entirely. OSHA's workplace limit is 5,000 ppm as an eight-hour time-weighted average. Guidelines for the International Space Station and for submarines fall in the 4,000 to 5,000 range. European CEN standards use categories at 550, 800 and 1,350 ppm, which are comfort and ventilation classes rather than safety thresholds.

None of that makes a meter pointless. But when one turns amber at 1,000 it is not telling you that you have crossed a line. It is telling you the ventilation in that room is near a level once used as a rough proxy for adequate. Useful information, and not the information the color implies.

What the research actually shows

Here it gets interesting, because two good pieces of work point in different directions, and the disagreement is the finding.

A field study published in Environmental Research Letters in September 2021 followed 302 office workers across China, India, Mexico, Thailand, the UK and the US, average age 33, with real-time CO2 and fine-particle measurement. It found effects, and it found small ones: for every 500 ppm rise in CO2, response times were 1.4 to 1.8 percent slower and throughput 2.1 to 2.4 percent lower. The Harvard summary of that work describes CO2 there as a proxy for ventilation rather than as the agent doing the damage.

A chamber study completed at UCL in 2025 went after the agent directly. Sixty-nine participants each sat through 70-minute exposures at 600, 1,500 and 2,100 ppm in a crossover design, with pure CO2 piped in rather than more people added to the room, and a ten-test cognitive battery. The result was nothing. Two of the ten tests showed faster response times at the higher concentrations, accuracy did not change, and aggregated by cognitive domain there was no difference between any of the three conditions. The author's own conclusion is that CO2 below 2,100 ppm "could be treated primarily as a proxy for ventilation rates and indoor air quality".

ASHRAE's read of the whole literature has the same shape. Six studies found an association around 1,000 ppm, others found none, and the existing evidence "for direct impacts of CO2 on health, well-being, learning outcomes, sleep patterns, and work performance at commonly observed indoor concentrations is inconsistent."

Our reading, offered as a reading rather than a finding: the most economical explanation is that CO2 is the meter and not the poison. Studies that raise CO2 by throttling ventilation raise everything else in the air at the same time, and they find small effects. Studies that add CO2 on its own and leave the ventilation alone find nothing. Both results can be true together if what matters is the ventilation, and CO2 is simply the easiest part of it to measure.

That does not make the number useless. It makes it a proxy behaving exactly as a proxy should. What it does retire is the popular version of the claim. "CO2 makes you stupid" is not what this evidence says, and the largest field effects on record are a couple of percent, not a change of character.

Why this lands harder on a home office

Office buildings are ventilated to a code. Your spare bedroom is not.

That single asymmetry is why this topic barely existed before people started working from home. A commercial office has mechanical ventilation sized for the number of people expected in it, and ASHRAE notes that Standard 62.1's ventilation rates correspond to roughly 1,000 ppm in offices and classrooms, rising past 2,500 in densely occupied spaces like conference rooms and auditoriums. Nobody sized your box room. It has a window and a door, and the door is shut because you are on calls.

So the person most likely to spend eight hours in the least-ventilated room in the building is the one who stopped going to an office, and they are also the least likely to have anything measuring it. That is the whole case for owning a meter, and it has nothing to do with fear.

What to do with a high reading

In the order we would try them, cheapest first.

• Open something. The entire mechanism is air exchange, so a window ajar in the room you are actually sitting in beats every other intervention. If the reading falls within minutes of opening it, you have confirmed the diagnosis and the cure in a single move.
• Open the door. A shut interior door turns a small room into a smaller volume. When you cannot open a window mid-call, an open door gives you the rest of the house to dilute into.
• Find out whether the room has any ventilation at all. Plenty of home offices are converted bedrooms with a trickle vent at best, while the only extraction in the house is in the kitchen and the bathroom.
• Treat the meter as a timer rather than an alarm. Watch how long the room takes to climb with the door closed. That interval, not any particular number, is the thing worth learning, because it tells you how often the room needs airing and no spec sheet can.

The last one is the one we would keep. A reading is a snapshot. The climb rate is a property of the room, and it stays true after you stop looking at the display.

What we would want from a meter, and what we would ignore

We have not tested meters and we are not going to rank them. But two things are checkable on a spec sheet before you spend anything, and both matter more than the styling.

Insist on an NDIR sensor. Non-dispersive infrared measures carbon dioxide directly, through how much infrared light at a specific wavelength the air absorbs. It is a measurement rather than an inference.

A spec sheet that says NDIR is the whole check, and the meters built for this publish it plainly. SAF Tehnika's Aranet4 HOME, for instance, lists a true NDIR sensor with a 0 to 9,999 ppm range, a stated accuracy of plus or minus 30 ppm plus 3 percent of the reading, and a manual calibration against outdoor air. Those are the manufacturer's figures and we have not checked them against an instrument. But the shape of that line is what to look for on any candidate: a sensor type, a range, an accuracy figure with a plus-or-minus in it, and a calibration procedure. A product page that offers none of the four is not describing a measurement.

Be careful with anything labelled "eCO2" or "equivalent CO2". That is not a CO2 measurement at all. As an explainer on the sensor types puts it, "an eCO2 measurement is derived from a total volatile organic components (TVOC) measurement". The device reads VOCs with a heated metal-oxide sensor and then estimates a CO2 figure from them, on the assumption that people are the main source of both. Its accuracy "is highly dependent on the algorithm used and the initial tuning process for the sensor". For a rough trend that is fine. For the number itself it is a guess wearing the costume of a reading, and it is the single thing most likely to disappoint you about a cheap multi-sensor gadget.

The third thing worth having is a calibration function, because NDIR sensors drift over time and the standard correction is to reference them against outdoor air.

When this guide does not apply

This is about carbon dioxide in the air of a room and nothing else. Blood CO2, arterial or venous, and the capnography equipment used to monitor patients are a separate subject entirely, medical rather than architectural, and nothing here transfers to it.

This is also not an article about symptoms. We are not in a position to tell you what any concentration does to a body, and the honest summary of the research above is that the field has not settled that question either. If you are worried about how a room makes you feel, that is a conversation to have with a doctor rather than with a display.

Two further limits. The standards and workplace figures above are American and European; other countries set their own. And CO2 tells you nothing about the things that do not come from people. Radon, combustion products from a gas hob, and the solvents coming out of a new desk are all completely invisible to it.

Sources

• ASHRAE, "ASHRAE Position Document on Indoor Carbon Dioxide" (approved February 12, 2025): https://www.ashrae.org/file%20library/about/position%20documents/pd-on-indoor-carbon-dioxide-english.pdf (accessed September 11, 2026)
• Andrew K. Persily, NIST, "Quit blaming ASHRAE Standard 62.1 for 1000 ppm CO2" (Indoor Air 2020, Seoul; published November 15, 2020): https://www.nist.gov/publications/quit-blaming-ashrae-standard-621-1000-ppm-co2 (accessed September 11, 2026)
• "Associations between acute exposures to PM2.5 and carbon dioxide indoors and cognitive function in office workers: a multicountry longitudinal prospective observational study", Environmental Research Letters (September 9, 2021): https://iopscience.iop.org/article/10.1088/1748-9326/ac1bd8 (accessed September 11, 2026)
• Harvard T.H. Chan School of Public Health, Healthy Buildings, "Impacts of Indoor Air Quality on Cognitive Function": https://healthybuildings.hsph.harvard.edu/impacts-of-indoor-air-quality-on-cognitive-function (accessed September 11, 2026)
• Didong Chen, "An Experimental Study on the Effects of Short-term Exposure to Low-to-medium Pure Carbon Dioxide on University Students' Cognitive Performance", doctoral thesis, UCL Bartlett School of Environment, Energy and Resources (2025): https://discovery.ucl.ac.uk/id/eprint/10209924 (accessed September 11, 2026)
• Didong Chen, "Direct impact of short-term exposure to pure carbon dioxide levels on cognitive performance", REHVA Journal: https://www.rehva.eu/rehva-journal/chapter/direct-impact-of-short-term-exposure-to-pure-carbon-dioxide-levels-on-cognitive-performance (accessed September 11, 2026)
• Electronics360, "What are eCO2 sensors?": https://electronics360.globalspec.com/article/17986/what-are-eco2-sensors (accessed September 11, 2026)
• SAF Tehnika, Aranet4 HOME product specifications. https://aranet.com/en/home/products/aranet4-home (accessed September 14, 2026)