Biology Insights

How to Measure Carbon Dioxide and What the Levels Mean

Learn how to measure carbon dioxide (CO2) accurately and understand what different CO2 levels indicate.

By Lydia Harding

Carbon dioxide is measured using sensors that detect how the gas absorbs infrared light, shifts electrical signals in chemical cells, or changes fluorescence in optical materials, with the right tool depending entirely on whether you are checking the air in your living room or tracking emissions from orbit. The numbers themselves range from around 420 parts per million outdoors to thousands of ppm in a stuffy meeting room, and each range carries different implications for comfort, cognition, safety, and the planet’s carbon budget. Understanding both the measurement and the meaning turns CO2 from an invisible abstraction into something you can act on.

What the Numbers Mean in Everyday Indoor Spaces

Outdoor air currently hovers around 420 ppm of CO2. That is the baseline your body is adapted to breathe. When you step inside a building, CO2 climbs because every person in the room exhales it. A well-ventilated office with plenty of fresh air exchange typically stays below 800 ppm. Once levels creep above that, you are breathing air that has been recycled through other people’s lungs more times than is ideal, and the effects are not just theoretical.

A controlled exposure study at Harvard and Syracuse placed office workers in environments with different CO2 concentrations and tested their decision-making. Cognitive function scores dropped about 15% at roughly 945 ppm and fell by 50% at around 1,400 ppm compared to days when the air was cleaner. On average, every 400-ppm increase in CO2 was tied to a 21% decline in cognitive scores across multiple domains. A separate study found that even short exposures of less than 40 minutes to 2,700 ppm could blunt the learning effect on cognitive tests, though the physiological pathway behind the impairment remains unclear. Participants who had slept poorly the night before appeared more susceptible to getting sleepy as CO2 rose.

For workplace safety at the extreme end, the threshold limit value set by the American Conference of Governmental Industrial Hygienists has sat at 5,000 ppm since 1946. That is the eight-hour occupational ceiling, not a comfort target. Concentrations above 30,000–40,000 ppm become acutely dangerous, causing headaches, dizziness, and eventually loss of consciousness. But the takeaway for most people is simpler: if you feel foggy in a conference room that has been packed for an hour, the CO2 level is a likely contributor, and opening a window or turning up the ventilation can make a measurable difference in how well you think.

CO2 in Cars, Planes, and Other Small Spaces

The places where CO2 builds up fastest tend to be the smallest. A study that tracked CO2 inside five cars and a van found that levels stayed below 800 ppm as long as the ventilation fan was running or a window was cracked. But with the fan off and windows closed while parked or driving in the city, CO2 exceeded 800 ppm in every vehicle tested. The worst scenario was recirculation mode: CO2 blew past 800 ppm within 10 minutes in every vehicle. If you have ever felt drowsy on a long highway drive with the windows sealed and the air set to recirculate, this is part of the explanation.

Public transit and airports tell a slightly different story. Measurements during actual trips showed CO2 staying below 800 ppm on trains and in most airport terminals. It spiked in a crowded boarding area with roughly 300 people packed together. Inside airplanes, CO2 regularly exceeded 800 ppm, though aircraft cabins use high-efficiency particulate air filters that address other pollutants even when CO2 is elevated. CO2 monitors have become popular as a rough proxy for how well ventilated a space is, particularly since the pandemic drew attention to shared air and airborne transmission risk. If CO2 is low, the air is being replaced frequently; if it is high, you are rebreathing what others have exhaled.

Consumer CO2 Monitors and Their Limits

You can buy a handheld or desktop CO2 monitor for as little as $30, and many of the popular models use a technology called non-dispersive infrared sensing. An NDIR sensor shines infrared light through a small chamber of air. CO2 molecules absorb specific wavelengths, and the sensor calculates concentration from how much light gets through. These devices are practical and reasonably accurate for home and office use, typically claiming precision of 1 to 10 ppm.

The catch is long-term drift. A 30-month field evaluation of low-cost sensors found that drift could produce biases of up to about 28 ppm over two years. Seasonal drift cycles added up to 25 ppm of error after just six months. With a simple linear calibration correction, the long-term error dropped to around 2.4 ppm, but most consumers never perform that recalibration. Humidity is another source of error. Testing showed that relative humidity had a meaningful influence on readings for both NDIR and metal-oxide semiconductor sensors, with humidity accounting for more than 10% of the reading variation for NDIR devices and over 20% for metal-oxide units.

For most people, a consumer NDIR sensor that costs between $100 and $200 gives a useful reading of indoor air quality. It will not be lab-grade, but it will reliably tell you whether your bedroom is at 600 ppm or 1,800 ppm, which is the difference that matters. Some practical tips: place the sensor at breathing height away from windows and vents, give it a few minutes to settle after moving it, and if your model supports it, expose it to fresh outdoor air periodically so it can recalibrate its baseline.

Research-Grade Instruments

When scientists need high-precision CO2 data, they turn to instruments that cost tens or even hundreds of thousands of dollars. The gold standard for atmospheric monitoring is cavity ring-down spectroscopy. A CRDS analyzer bounces a laser beam thousands of times inside a highly reflective cavity, letting it interact with the gas over an extremely long effective path length. The result is precision on the order of 0.1 ppm, more than ten times better than a consumer NDIR sensor. A comparison during airborne measurement campaigns over the Amazon showed that a CRDS analyzer and a standard NDIR unit differed by only about 0.05 ppm on average after proper calibration corrections. That level of stability means CRDS instruments can fly on aircraft without needing to dry the air or recalibrate mid-flight.

Getting all these high-precision instruments worldwide to agree with each other is its own challenge. The World Meteorological Organization’s Global Atmosphere Watch program maintains a hierarchical calibration scheme: primary standards define an instrument response, and those are used to assign values to secondary and tertiary standards that get shipped to individual monitoring sites. Without this chain of traceability, two stations measuring the same air mass could disagree by enough to confuse the science on regional carbon budgets.

Measuring CO2 From Space

Satellites offer a way to map CO2 across the entire planet rather than relying on scattered ground stations. NASA’s Orbiting Carbon Observatory-2, launched in 2014, measures the column-averaged CO2 dry-air mole fraction by observing sunlight reflected off Earth’s surface and seeing how much is absorbed by CO2 in the atmosphere. It also picks up fluorescence from chlorophyll in plants, which gives scientists a simultaneous read on how actively vegetation is pulling CO2 out of the air through photosynthesis.

Satellite retrievals are validated against ground-based networks to make sure the numbers hold up. The retrievals use algorithms to disentangle CO2 absorption from interference by aerosols, clouds, and surface properties. The satellite approach is less precise per measurement than a CRDS analyzer on the ground, but its strength is coverage: it can reveal emission hotspots, seasonal swings in biosphere uptake, and regional differences that would be invisible from a handful of remote stations alone.

Ground-Level Atmospheric Monitoring

The backbone of atmospheric CO2 tracking is a global network of remote stations that have been running continuous measurements for decades. The most famous is the Mauna Loa Observatory in Hawaii, whose record stretches back to 1958 and produced the iconic upward-curving graph known as the Keeling Curve. These records capture everything from seasonal cycles driven by plant growth in the Northern Hemisphere to the long-term upward trend driven by fossil fuel burning.

Scientists also measure isotopes of CO2 at these stations, which works like a chemical fingerprint. Fossil fuels are completely depleted in the carbon-14 isotope because they have been buried for millions of years, far longer than carbon-14’s roughly 5,700-year half-life. They are also slightly depleted in carbon-13 because the ancient plant material they formed from preferentially took up the lighter isotope during photosynthesis. When fossil fuels burn, the resulting CO2 dilutes the heavier isotopes in the atmosphere, a signature called the Suess Effect. Measuring those isotope ratios lets researchers distinguish fossil-fuel CO2 from CO2 released by oceans, soils, or living forests, which is critical for tracking whether emission reduction pledges are actually working.

CO2 in the Ocean

About a quarter of the CO2 humans emit ends up absorbed by the ocean, making seawater CO2 measurements essential for understanding Earth’s carbon cycle. The standard method uses an equilibrator system: seawater is showered inside a chamber until the air above it reaches the same CO2 partial pressure as the water, and then that air is analyzed with an infrared detector. These systems are reliable but bulky and expensive, so they are mostly installed on research ships and a limited number of commercial vessels.

Newer membrane-based sensors offer a smaller, cheaper alternative. A field test comparing membrane sensors against a traditional equilibrator system showed that the newer sensors can meet the Global Ocean Acidification Observing Network’s goal of 2.5% relative uncertainty, provided the deployment accounts for biofouling and includes reference gas measurements. Another promising approach uses optical sensors called optodes, which detect CO2 through changes in fluorescence. These have been calibrated over the range relevant to the open ocean and could enable much wider autonomous monitoring on buoys and underwater gliders.

Measuring CO2 in Medicine

In hospitals and ambulances, CO2 measurement serves a completely different purpose. End-tidal CO2 monitoring, or capnography, measures the CO2 in a patient’s exhaled breath. Because CO2 is tightly coupled to metabolism, circulation, and ventilation, the level in exhaled air tells clinicians whether a patient’s lungs are working, whether blood is circulating, and whether a breathing tube is correctly placed.

The most common use is confirming endotracheal intubation. If the tube accidentally goes into the esophagus instead of the trachea, there will be little or no CO2 in the exhaled gas, and the monitor sounds the alarm. The American Heart Association recommends some form of end-tidal CO2 measurement for all intubations, and the Intensive Care Society includes it as a required standard for monitoring patients during transport. Beyond intubation checks, the waveform and value of end-tidal CO2 also help assess the effectiveness of CPR, detect pulmonary embolism, and monitor sedated patients for respiratory depression. A normal end-tidal CO2 reading in a healthy person is roughly 35 to 45 mmHg, and significant deviations in either direction prompt immediate clinical action.

CO2 in Soil and Ecosystems

Ecologists measure CO2 flowing out of the ground to understand soil respiration, the combined exhaling of microbes, roots, and decomposing organic matter. Two main field techniques exist. Static chambers trap a volume of air above the soil and let CO2 accumulate over hours, with the gas absorbed by an alkali solution that is later analyzed. Dynamic chambers use a portable infrared gas analyzer to measure the CO2 flux in real time over just a minute or two. A comparison of the two methods across arable, forest, and pasture sites found dramatic differences: the static method topped out at about 300 mg CO2-carbon per square meter per hour, while the dynamic method captured fluxes up to roughly 2,500 mg under the same conditions. The static method tends to underestimate high fluxes because the accumulated CO2 slows further diffusion out of the soil, a known artifact that researchers correct for or avoid by choosing the dynamic approach.

These measurements feed into global carbon budgets. Soil respiration is one of the largest natural fluxes of CO2 to the atmosphere, dwarfing human emissions in total volume though roughly balanced by photosynthetic uptake under natural conditions. Tracking how soil CO2 flux changes with temperature, moisture, and land use helps predict whether soils will become net carbon sources as the climate warms.

Industrial and Fermentation Monitoring

In biotechnology, CO2 is a key byproduct of fermentation. Tracking it in the off-gas from a bioreactor tells operators how fast the microorganisms are working and whether conditions are optimal. Traditionally, gas chromatographs or mass spectrometers handled this job, but optical sensors offer a simpler alternative. One approach uses a fluorescent dye whose emission changes when CO2 diffuses into the sensor and forms carbonic acid, shifting the dye’s absorption spectrum in a measurable way. More recent work has applied Raman spectroscopy to fermentation off-gas, using scattered light to identify and quantify CO2 in real time alongside other gas components.

The brewing and winemaking industries rely on similar principles at a less technical level. Many brewers simply watch an airlock bubble or use a basic inline CO2 meter to know that yeast is active. Pharmaceutical and industrial bioprocesses need far tighter control, because CO2 accumulation in the culture medium can shift pH and inhibit cell growth if not managed carefully.

How Plants Respond to CO2 Levels

Plants are Earth’s original CO2 sensors. Their stomata, the tiny pores on leaf surfaces, open and close partly in response to the CO2 concentration around them. When CO2 is abundant, stomata narrow to conserve water because less opening is needed to capture enough carbon for photosynthesis. A meta-analysis of free-air CO2 enrichment experiments found that elevated CO2 reduced stomatal conductance by about 22%, though this was not necessarily accompanied by a change in how many stomata the leaf produced. Over longer timescales, the effect is even more striking: an analysis of nine common species from Florida found a 34% reduction in maximum stomatal conductance for every 100-ppm increase in CO2 over the past 150 years, driven by changes in both stomatal density and pore size.

Recent laboratory work has shown that the signaling pathways involved differ depending on whether CO2 is rising from low to moderate levels or from moderate to high. Stomatal closure was faster when CO2 jumped from 400 to 800 ppm than when it moved from 100 to 400 ppm, suggesting the plant activates different molecular mechanisms at different concentration ranges. This matters for predicting how vegetation will behave as atmospheric CO2 continues to climb: plants may not keep responding in the same linear way, and their water use, growth rates, and competitive dynamics could shift in non-obvious directions.

Sensor Technologies at a Glance

Across all these applications, the underlying sensor technologies fall into a handful of categories. Knowing the basics helps you understand why different devices cost what they do and where their limitations lie.

Each technology trades off cost, size, accuracy, and durability. No single sensor type dominates every application, which is why the field keeps developing new variants and calibration strategies. For the person buying a desk monitor, NDIR remains the best balance of price and reliability. For a researcher tracking atmospheric trends over decades, nothing replaces CRDS anchored to internationally traceable reference standards.

How Animals Sense CO2

Humans and other animals have their own built-in CO2 detectors, though we rarely think of them that way. Chemoreceptors in the brainstem respond to rising CO2 in the blood by triggering faster and deeper breathing. That panicky feeling when you hold your breath is primarily a response to CO2 buildup, not to falling oxygen. Across diverse species, sensory neurons have been identified that detect external CO2 levels and direct behavior: mosquitoes track exhaled CO2 plumes to find hosts, nematodes navigate CO2 gradients to find food or avoid danger, and fruit flies avoid high-CO2 zones. These detection systems appear to repurpose ancient molecular machinery that cells originally evolved for their own internal CO2 and pH regulation, coupling it to neural circuits that control behavior. The biological world has been measuring CO2 long before anyone built a sensor.

Related Content

More in Environment