Why Do We See Our Breath When It's Cold? The Science

You see your breath on a cold morning because the air leaving your lungs is warm and nearly saturated with water vapor, and cold air cannot hold that much once the two mix. The excess condenses into tiny liquid droplets that scatter light, the same way a cloud does. It is not smoke, and it is not ice. The mixing details explain why breath fogs at 5°C but not at 25°C, why a dry winter day still shows a plume, and why jet engines draw white lines across the sky.

What your breath really contains

Air passes through your nose or mouth, down the trachea, and into the lungs, where it picks up heat and moisture from the wet airway lining. By the time it comes back out, it has been warmed to near body temperature and loaded with water vapor, the invisible gas form of water.

A computational study of airflow in human airways, published in Annals of Biomedical Engineering in 2014, reports that during exhalation the relative humidity in the airways stays at about 100 percent, with air temperatures in the trachea of of 32.8 to 35.0°C on exhalation in its model and 32.6 to 37°C in the measurements it cites (Wu et al., 2014). Air leaving your lips is usually a little cooler than that, because the airway walls cool it on the way out. This article uses 34°C and 100 percent relative humidity as a round, slightly generous description of exhaled breath.

How much water is that? The saturation vapor pressure of water at 34°C is about 53 hectopascals (hPa), computed from the Magnus formula that meteorologists use for this purpose (Alduchov and Eskridge, NOAA, 1997). Converted to mass, saturated air at 34°C carries roughly 37 grams of water vapor in every cubic meter. By comparison, outdoor air at minus 10°C and 50 percent relative humidity carries about 1.2 grams per cubic meter. Your breath is, in effect, a tiny pocket of tropical air released into a very different climate.

Saturated air at 34°C holds roughly 37 grams of water vapor per cubic meter (calculated from the Magnus formula). Saturated air at minus 10°C holds only about 2.4 grams, so cold air has room for less than a tenth of what your breath carries.

Dew point and humidity: the two definitions you need

Two terms do most of the work in this topic, and both are easy to misread.

Relative humidity is the amount of water vapor in the air as a percentage of the most the air could hold at its current temperature. At 100 percent, the air is saturated. Dew point is the temperature to which air must be cooled, at constant pressure, to reach 100 percent relative humidity. The National Weather Service describes it this way, and adds that cooling below the dew point forces water vapor out as liquid, usually as fog or precipitation (NWS La Crosse).

The National Weather Service also points out that relative humidity can mislead. Its example: air at 30°F with a 30°F dew point (100 percent relative humidity) feels less humid than air at 80°F with a 60°F dew point (50 percent). Dew point tells you how much vapor is actually present. Relative humidity only tells you how close the air is to its limit at the moment.

Why the capacity of air changes so fast with temperature

Air does not literally "dissolve" water vapor, but the metaphor of a sponge is useful. Warm air has a large sponge; cold air has a small one. The key detail is that the sponge does not shrink steadily. It shrinks by a roughly constant percentage for each degree of cooling, which means the absolute amount lost per degree is larger at warm temperatures.

From the Magnus formula, the saturation vapor pressure changes by about 6 to 7 percent for each degree Celsius around typical outdoor temperatures. That means every 10°C of cooling roughly halves the amount of vapor the air can hold. The table shows what that looks like in grams per cubic meter, calculated from the same formula.

Air temperature	Saturation vapor pressure (hPa)	Maximum water vapor (g per m³)
minus 10°C	2.9	2.4
0°C	6.1	4.8
10°C	12.3	9.4
20°C	23.3	17.3
34°C	53.1	37.5

Values are over liquid water, computed from the Magnus formula (Alduchov and Eskridge) and the ideal gas law. Rounded.

Notice that the capacity at 34°C is more than 15 times the capacity at minus 10°C. That gap is the entire reason you can see your breath.

Why the capacity curve bends upward

The bend has a physical reason. Liquid water is a crowd of molecules, and a few of them at any moment have enough energy to escape into the gas above. Warmer water means more energetic molecules, and the fraction that escape grows exponentially with temperature, not in a straight line. Saturation is the point where molecules are returning to the liquid as fast as they leave, so the vapor pressure at saturation follows the same exponential climb. Physicists package this relationship as the Clausius-Clapeyron equation, and the Magnus formula used above is a compact empirical fit to it over the range of weather temperatures.

For this article, the useful consequence is the shape. If capacity were a straight line, mixing two saturated parcels would give a saturated parcel. Because the curve bends upward, mixing two saturated parcels of different temperatures always gives a supersaturated one. Breath is not quite saturated and outdoor air is usually far from it, so the test is closer than that, but the curvature is what gives mixing its power to make clouds. A related effect makes your breath fog a cold mirror or a windshield: any cold surface drives the nearby air along the same curve toward saturation. For the broader physics of how heat moves and why energy spreads out this way, see Thermodynamics and Entropy: Why Time Moves Forward.

What happens in the first second after you exhale

Here is the part most explanations skip. When you describe breath fogging, it is tempting to say "the outdoor air is below the dew point of your breath, so the vapor condenses." That is roughly right, but it skips what actually decides the outcome: your breath does not simply cool. It mixes.

As the plume leaves your mouth, it folds into the surrounding air. At every point in the plume, some fraction of the air is breath and the rest is outdoor air. Both the temperature and the water vapor content of that blend sit somewhere between the two starting values. Meteorologists draw this as a "mixing line" running from the outdoor air's state to the breath's state on a graph of vapor content against temperature. The saturation curve on that same graph is bent, because capacity rises faster and faster as air gets warmer.

Condensation happens if the mixing line rises above the saturation curve anywhere along its length. That is a geometric test, and it is the same one that decides whether a jet leaves a contrail.

An analogy: two glasses of water at different temperatures

Picture mixing a glass of hot coffee into a glass of cold milk. The result lands between the two in temperature, and so does the sweetness or bitterness. Now imagine that the "cup size" for holding sugar shrank sharply as the drink cooled. If the cup shrinks faster than the sugar dilutes, the mixture ends up oversupplied and the extra comes out as crystals. Breath works like that. The water vapor dilutes in proportion to the mixing, but the capacity of the blend falls faster than the vapor does, because capacity is a steeply curved function of temperature. When the capacity drops below the vapor present, liquid forms.

A worked example

A simple model makes this concrete: breath at 34°C and 100 percent relative humidity mixes smoothly with outdoor air, and temperature and vapor pressure blend in proportion. The condensation condition falls out of the geometry. In this model, condensation starts if the outdoor air's vapor pressure (in hPa) is above 53.1 minus 2.97 times the number of degrees by which the outdoor air is colder than 34°C.

For dry outdoor air, that works out to some condensation whenever the air is colder than about 16°C. Add humidity and the threshold shifts: at 20°C, the air needs a relative humidity of about 50 percent before the model predicts even a trace of fog. Below about 16°C, the model predicts fog even if the outdoor air has no moisture in it at all.

The table below applies this to nine outdoor conditions. The right column is an upper bound on the liquid water the blend could hold at its peak, so compare rows with each other and do not read the numbers as real fog density. The reason is that condensation releases latent heat, which warms the blend and reduces how much liquid actually forms. The model also ignores the uneven mixing of a real plume.

Outdoor temperature	Outdoor relative humidity	Outdoor dew point	Upper-bound peak liquid water (g per m³)	What the model predicts
minus 10°C	20%	about minus 29°C	about 10	Dense plume
minus 10°C	50%	about minus 19°C	about 10	Dense plume
0°C	50%	about minus 9°C	about 7	Dense plume
10°C	50%	about 0°C	about 3	Clear plume
15°C	50%	about 5°C	about 1	Thin, wispy plume
15°C	90%	about 13°C	about 3	Clear plume
20°C	50%	about 9°C	about 0	No visible plume
20°C	90%	about 18°C	about 1	Thin plume at most
25°C	90%	about 23°C	about 0.1	Essentially invisible

Calculated by the author from the Magnus formula (Alduchov and Eskridge) with breath at 34°C and 100% relative humidity. Idealized: perfect mixing, no latent heating, no particle limits. Dew points rounded.

Two patterns stand out. First, temperature dominates. The first two rows differ by a factor of 2.5 in relative humidity and by about 10°C in dew point, yet the plume strength is essentially the same, because at minus 10°C the outdoor air holds almost no vapor either way. Second, humidity matters most on mild days. Compare the 15°C rows: the same temperature gives a faint wisp at 50 percent humidity and a clear cloud at 90 percent.

Why a dry winter day still shows your breath

Many explanations of this phenomenon say that visible breath needs both cold and moisture, and that a cold, dry day might show nothing. The mixing-line geometry says otherwise. In very cold air, the outdoor vapor content barely matters, because it is tiny compared with what your breath adds. A dry January morning is therefore as good as a damp one for visible breath, and sometimes better.

Where humidity does matter is on the cool-but-not-cold days, roughly 10 to 20°C in the table. There, a damp afternoon shows a plume and a dry one may not. This also explains something familiar: you see your breath on a rainy autumn day sooner than on a sunny one at the same temperature.

For a different look at how water's behavior at the dew point shapes everyday experience, Phase Changes: Why Water Boils, Freezes, and Sublimates covers the transitions between vapor, liquid, and solid in more detail.

Why the plume vanishes a few feet from your face

If the model says condensation is strong near your mouth, why does the cloud disappear so quickly? Because the same mixing that created it keeps going. As the plume dilutes further, the blend moves down the mixing line toward the outdoor state, which sits below the saturation curve. Eventually the blend can hold all of its vapor again, and the droplets evaporate.

In the model, the cloud at minus 10°C and 50 percent humidity survives until breath makes up only about 4 percent of the mixture. In bone-dry air at 20 percent humidity, it vanishes a little sooner, at about 6 percent. That is the reason dry air shortens a plume and humid air stretches it, even when the peak thickness near the lips is the same.

Condensation nuclei: the surfaces droplets grow on

A droplet needs a place to start. In the atmosphere, water vapor condenses onto tiny particles called condensation nuclei, such as sea salt, dust, soot, and sulfate aerosols. Without any such particles, air can be cooled well past its dew point and stay clear, a state called supersaturation. The real atmosphere is never that clean, which is why fog and clouds form readily once the humidity gets near 100 percent.

Outdoor air supplies plenty of them. As vapor condenses onto the particles, each one grows into a droplet, and the huge number of droplets in a small volume is what you see.

Why the cloud looks white

The droplets in a breath cloud are comparable to or larger than the wavelength of visible light, so they scatter all colors about equally. That is why clouds and fog are white or gray rather than blue. The blue sky is a different case: air molecules are much smaller than the wavelength of light, and they scatter short wavelengths far more strongly. For that contrast, see Why Is the Sky Blue? Scattering, Sunsets, and Perception.

Is your breath freezing? Almost never

A common misreading of the phenomenon is that you are watching your breath turn to ice. In nearly all everyday conditions, you are watching it turn into liquid.

Freezing needs the water to be cold enough and to have something to start the crystal. Two facts work against it in a breath plume. The first is that the model above shows the peak of condensation happening while the blend is still warm. At 0°C outside, the blend holds most condensate when it is around 20°C, well above freezing. The droplets form warm and cool afterward. The second is that tiny liquid droplets can stay liquid well below 0°C. Cloud physicists call this supercooled water. A NASA Ames paper notes that small amounts of liquid water remain supercooled in the atmosphere down to about minus 40°C, and that the details of how it freezes are still poorly understood (Tabazadeh et al., NASA, 2002).

At very low temperatures, some droplets will eventually freeze, but the sources reviewed for this article do not give a single firm threshold for breath, and that is consistent with the research: freezing of small droplets is a matter of probability, not a switch at 0°C. For water's other temperature oddities, see Why Ice Floats and Water Is Weird and How Water Works: The Unique Chemistry of H₂O.

Contrails: the same mixing line at 10 kilometers

Jet engines burn hydrocarbon fuel, and one of the combustion products is water vapor. The exhaust leaves the engine hot and moist and mixes into extremely cold air at cruising altitude. If the blend crosses water saturation, droplets form on soot and other particles, and a contrail appears.

The relationship is called the Schmidt-Appleman criterion. A 2022 paper in Atmospheric Chemistry and Physics describes it as "a relation derived purely from the thermodynamics of the mixing process" and says contrails form once the plume mixture reaches or surpasses water saturation. Whether they can form depends on the ambient temperature, the relative humidity over water, air pressure, and the engine and fuel properties (Bier, Unterstrasser, and Vancassel, 2022). The same paper's model finds water saturation first reached at a plume temperature of about 240 K (about minus 33°C) in its baseline case.

That is your breath, with a different starting point. The condensation geometry is identical. What changes is the exhaust's temperature, its water content, and the particles available.

Whether a contrail then lingers or fades depends on how moist the surrounding air is. Breath works the same way at a smaller scale: in humid air, the plume stays visible longer.

Everyday cousins of breath fog

The same physics shows up all over daily life, and spotting it can make the idea stick.

• A fogged bathroom mirror. Hot shower water fills the room with vapor, and the cold glass cools the air next to it below the dew point. Droplets form on the surface instead of in the air. That is surface condensation rather than mixing, but the dew point test is the same.
• A kettle's invisible gap. Look at steam leaving a kettle spout. Right at the spout, there is often a clear gap, and the visible cloud starts a few centimeters out. The clear gap is hot vapor that has not mixed and cooled enough to condense yet, which is the breath plume's lip-side story again.
• A cold drink sweating. Warm room air touching a cold glass is cooled below its dew point, and water collects on the outside.
• Breath in a freezer. You will see your breath inside a walk-in freezer at any time of year, because the cold air is what matters, not the season outside.
• Fog over a lake. Cold air moving over warmer water mixes with the moist air just above the surface, and the blend condenses into wisps.

For another place where air temperature, moisture, and surfaces interact, see Why Static Electricity Builds Up on Dry Days, which is the dry-air counterpart to this story, and What Is the Water Cycle? A Journey Through Earth's Most Vital Process, which scales the same condensation up to the planet.

Three ways to test this yourself

The model makes predictions you can check with a coat and a weather app. None of these needs equipment, and none of them is a precise measurement.

Compare two cold days. Pick a dry, cold morning and a damp, cold one. The model says the plume should look similar near the lips at, say, minus 5°C, since the outdoor vapor is tiny either way. What should differ is how far the cloud travels before it vanishes: dry air erases it a little sooner.

Test a mild day. Between 10 and 20°C, look up the dew point in your weather app and try again over a few days. The model predicts that you will see a plume when the dew point is high and little or none when it is low. This is where humidity actually changes the outcome.

Make a plume indoors. Breathe out in front of a cold window on a winter day, or open a freezer. The surface or air in front of you is cold enough that the same mixing happens in a few centimeters. If you breathe out at a warm window, nothing appears, because the blend never reaches saturation.

If your observations disagree with the table, that is useful information. The model assumes smooth mixing and breath at a fixed 34°C and 100 percent humidity, while real breath depends on how long you held it, how fast you exhale, and whether you exhale through the nose or mouth. Gusts, sunlight, and the background behind the plume also change how visible a thin cloud looks. The model describes the pattern, and the weather supplies the exceptions.

Your body is losing water and heat every time you exhale

The visible plume is a reminder that breathing carries away moisture and energy. In the airway study cited above, absolute humidity falls during exhalation because water condenses onto the airway surface liquid, and the airway walls cool the outgoing air (Wu et al., 2014). So some condensation happens inside you before the air leaves, which is one reason air at the lips is a little cooler than air deep in the lungs.

This is general physiology and not health advice. If you want the lung side of the story, How Lungs Work: Breathing and Gas Exchange covers gas exchange in detail.

What people get wrong about visible breath

"It's smoke or steam." Steam in the strict sense is invisible water vapor. The visible cloud from a kettle or your mouth is not vapor at all but liquid droplets, which is why scientists call it a condensate cloud.

"You need humid air." You do not, as the table above shows. Very cold air makes breath visible whether it is dry or moist. Humidity helps most between about 10 and 20°C.

"It's your breath freezing." The droplets form warm and are usually liquid, as covered above.

"There is a single temperature below which breath is visible." Any fixed cutoff is a simplification, because the real threshold shifts with humidity and is lower or higher depending on how you define "visible." The model puts the dry-air onset near 16°C and the 50-percent-humidity onset near 20°C, but it is generous, and real plumes are thinner. Treat any single number as a rough guide.

"Indoor humidity changes whether you will see your breath outside." Not much. Your breath leaves your lungs near saturation almost regardless of the room you came from, so what counts is the outdoor air.

Frequently asked questions

Q: Why can't I always see my breath, even on cold days? A: Usually you can. In the idealized model, very cold air produces a plume whether it is dry or humid. If you cannot see one, the likely reason is mild weather: around 15 to 20°C with dry air, the model predicts a thin plume or none at all.

Q: Is the fog from your breath frozen? A: Almost always no. The droplets are liquid water, typically formed at temperatures well above freezing inside the mixing plume, and small droplets can remain liquid even below 0°C. Cloud measurements find supercooled liquid water down to about minus 40°C, though how and when small droplets freeze is still a research question (NASA, 2002).

Q: What does dew point have to do with it? A: The dew point is the temperature at which air becomes saturated. Your breath starts out saturated at about 34°C, and the mixed plume crosses its own saturation curve as it cools and dilutes. The outdoor dew point sets how fast the plume disappears, but the outdoor temperature matters more for whether a plume forms at all (NWS).

Q: How is visible breath different from a cloud? A: The physics is the same: vapor condenses into droplets on particles. A natural cloud usually forms as air rises and cools; a breath cloud forms as two air masses mix. Both are collections of liquid droplets that scatter light.

Q: Can you predict whether your breath will be visible before you go outside? A: Roughly. Below about 10°C, expect to see it in almost any humidity. Between 10 and 20°C, a damp day makes a plume and a dry day may not. Above about 20°C you will rarely see it. These are rules of thumb from an idealized calculation, not guarantees.

Q: Why do aircraft contrails sometimes vanish and sometimes spread for hours? A: Contrails form when the exhaust-air mixture reaches water saturation, per the Schmidt-Appleman criterion. Whether they last depends on how moist the surrounding air is. In dry air, the ice crystals evaporate quickly; in humid air, they can persist. The formation test itself is the one described by Bier and colleagues (Bier et al., 2022).

The takeaway

Visible breath is a small mixing experiment you run with every exhale. Your breath starts warm and close to saturated, the outdoor air starts cold and nearly empty of vapor, and the blend between them passes through conditions where the air cannot hold all of its water. The liquid that comes out is the cloud you see, and it evaporates again as mixing continues. The practical lesson is that temperature does most of the work and humidity only decides the close calls, which is why the old rule that you need both cold and damp air is only half right.

---

Read next:

• How Water Works: The Unique Properties of H2O
• The Water Cycle Explained
• How Atoms and Molecules Work: Chemistry Basics

---

Sources:

• Alduchov, O. A., and Eskridge, R. E., "Improved Magnus' form approximation of saturation vapor pressure," NOAA National Climatic Data Center technical report (1997)
• Wu, D., Tawhai, M. H., Hoffman, E. A., and Lin, C.-L., "A Numerical Study of Heat and Water Vapor Transfer in MDCT-Based Human Airway Models," Annals of Biomedical Engineering (2014)
• National Weather Service La Crosse, "Dew Point vs Humidity"
• Tabazadeh, A., Djikaev, Y. S., Reiss, H., and Gore, W. J., "Surface Crystallization of Cloud Droplets," NASA conference abstract (2002)
• Bier, A., Unterstrasser, S., and Vancassel, X., "Box model trajectory studies of contrail formation using a particle-based cloud microphysics scheme," Atmospheric Chemistry and Physics (2022)