What Does a Normal Oxygen Variation Chart Look Like?

A normal oxygen variation chart shows a relatively flat line hovering between about 95% and 100% blood oxygen saturation, with small, brief dips that rarely drop below 90%. In healthy adults at rest, the average reading sits around 97–98%, and the typical fluctuation from one moment to the next is less than one percentage point. That flat-with-tiny-wobbles pattern is the hallmark of healthy blood oxygenation, but the picture gets more interesting when you look at what happens during sleep, at different ages, and in specific situations that nudge the line around.

The Baseline During Waking Hours

When researchers continuously record oxygen saturation in healthy people who are awake and sitting comfortably, the trace looks almost boring. One study of healthy individuals found a mean SpO2 of about 97.7%, with an average standard deviation of only 0.7% across the recording period. Most of the variability that did exist was made up of slow, gradual drifts rather than sudden jumps or crashes.¹ In clinical shorthand, a healthy person’s oximetry signal has been described as “near-constant,” with marked drops linked specifically to disordered breathing events rather than to normal physiology.²

Body position does shift things slightly. Sitting upright in a chair tends to give you the highest SpO2 reading. Lying on your side, whether right or left, produces a measurably lower number, though the difference is small enough that a healthy person would not notice it. Women, younger adults, people with a lower body mass index, and nonsmokers all tend to run a bit higher on the saturation scale.³ So if your chart shows readings a point or two lower when you are lying in bed compared to when you are up and about, that is the expected pattern, not a sign of trouble.

What Happens When You Fall Asleep

Sleep introduces the most notable variation that healthy people experience. Your breathing rate changes, your muscles relax, and your oxygen saturation tends to dip slightly. Research on high-altitude residents tracked continuous oxygen levels over 24 hours and found that saturation dropped during the night, with the lowest point falling between roughly 1:00 and 3:00 a.m.⁴ That study was at altitude, where the dips are more pronounced, but the general shape of the curve applies at sea level too: saturation is highest during the day, sags overnight, and recovers in the morning.

Within sleep itself, the two main stages handle oxygen differently. In people without breathing disorders, REM sleep (the stage with dreaming and rapid eye movements) tends to produce slightly more frequent and slightly deeper dips in saturation compared to non-REM sleep.⁵ The reason is straightforward: during REM, most of your voluntary muscles are temporarily paralyzed, and your breathing becomes more irregular. In simple snorers without sleep apnea, one study found average oxygen saturation was actually a touch higher during REM than non-REM (about 96.9% versus 95.6%), suggesting that in mild cases, the body compensates well.⁶ The pattern reverses in people with severe obstructive sleep apnea, where REM saturation drops well below non-REM levels because the airway collapses more often and for longer during that stage.

Infants Have a Wilder Chart

If you have ever watched a pulse oximeter on a sleeping newborn, you know the trace looks nothing like an adult’s. Healthy term infants in their first two months of life have a high median baseline (around 99.8%), but a striking 81% of them show brief episodes where saturation plunges to 80% or below.⁷ Before you panic: these desaturations are extremely short-lived, with a median duration of just 1.2 seconds. They happen because infant breathing is naturally irregular, especially during periods of non-regular or periodic breathing patterns. The 97th percentile for how long a dip below 80% lasted was only four seconds.

These dramatic-looking but harmless dips smooth out substantially as infants grow. One study tracking oxygen profiles from birth through the second year of life found that the desaturation index dropped from about 27 events per hour at two weeks of age to roughly 3 per hour by 24 months.⁸ Over the same period, the median overnight SpO2 remained between 98% and 99%.⁹ So the baseline stays high throughout infancy; it is the spikiness of the chart that gradually calms down. By the time a child is a toddler, the nighttime trace looks much closer to an adult’s.

Premature infants are a separate case. Oxygen-dependent preemies showed a measurable difference in saturation depending on whether they were placed prone or supine, with the prone position yielding readings about a percentage point higher on average.¹⁰ That said, safe-sleep guidelines prioritize the supine position for other reasons, and clinicians weigh those trade-offs carefully in the NICU.

How Aging Shifts the Range

A gradual decline in resting oxygen saturation is a normal part of aging. The lungs become less elastic, the chest wall stiffens, and gas exchange across the lung membranes becomes slightly less efficient over the decades. Research on well elderly adults confirmed that both partial pressure of oxygen and saturation readings drift downward with age, and that this reduction reflects normal physiology rather than disease as long as carbon dioxide levels remain stable.¹¹ A healthy 80-year-old might comfortably sit at 94–95% on the oximeter where a healthy 25-year-old reads 98–99%. Both charts look “normal” for their age; the older person’s line simply rides a bit lower.

This age-related shift matters practically because the commonly cited alarm threshold of 90% means something different for an older adult than for a young one. A 25-year-old who dips to 91% has dropped seven or eight points from their baseline, which is significant. An 80-year-old reading 91% may have dipped only three or four points. Context matters more than any single number.

Altitude Changes the Whole Chart

Move to higher elevation and the entire oxygen saturation trace shifts downward. The barometric pressure falls, which reduces the oxygen pressure in every breath you take, which in turn lowers your blood oxygen. The body compensates through acclimatization: breathing rate increases, red blood cell production ramps up, and blood vessels in the lungs adjust over days to weeks.¹² During the acclimatization process, daily mean saturations for healthy highlanders may range from about 92% to 95%, with women tending to run a bit higher than men.⁴

The chart still shows the same general shape: higher during the day, lower at night, small fluctuations throughout. But the entire band of normal is compressed downward. If you are monitoring your oxygen with a wrist device on a skiing trip and notice readings in the low 90s, that is likely altitude doing its thing rather than a medical emergency. The readings should climb back to your personal baseline once you return to lower ground.

Alcohol and Other Acute Triggers

A nightcap can show up clearly on an overnight oxygen chart. In healthy middle-aged men, drinking alcohol just before bed lowered average oxygen saturation during the first half of sleep from about 95.7% to 94.8%, and the time spent below 92% saturation jumped from around 1% of the sleep period to nearly 5%.¹³ An earlier study found that alcohol roughly doubled the number of desaturation events across the night and nearly tripled the number of apneic episodes, and the effect persisted partially into the following alcohol-free night as well.¹⁴

What this looks like on a chart is a noticeably jagged first few hours of sleep, with more frequent and deeper dips compared to an alcohol-free night. The mechanism is muscle relaxation: alcohol loosens the soft tissues of the upper airway, making partial or full airway collapse more likely during sleep. If you use a consumer oximeter and notice sawtooth-like dips clustering in the early part of the night after drinking, alcohol is a likely explanation. The effect tends to be more pronounced in people who already snore.

Exercise and Muscle Oxygen

The oxygen chart most people think of tracks arterial blood saturation, but a different kind of oxygen measurement, near-infrared spectroscopy of muscle tissue, tells a complementary story during exercise. Muscles consume oxygen rapidly during intense effort, and tissue oxygen saturation can plummet during a hard bout. Interestingly, fitter individuals actually show less muscle desaturation during peak effort and recover their tissue oxygen faster afterward. In one study of arm cycling, higher-power athletes had about 18% less muscle oxygen desaturation than lower-power athletes and recovered roughly twice as fast.¹⁵

For arterial SpO2, though, healthy people at moderate exercise intensities usually maintain readings above 95%. Significant drops in arterial saturation during exercise are not normal and can indicate conditions like exercise-induced hypoxemia, which sometimes affects highly trained endurance athletes or people with underlying lung or heart conditions. If your oximeter shows big dips during a workout, poor signal quality from finger movement is a more common explanation than an actual oxygen problem, but persistent low readings during exertion are worth investigating.

The 90% Threshold and Why It Matters

Clinically, 90% is the number that draws attention. Below that level, the relationship between oxygen saturation and the amount of oxygen dissolved in your blood becomes much steeper, meaning small further drops in saturation translate to disproportionately large drops in actual oxygen delivery to tissues. Researchers increasingly focus on a metric called T90, which is the percentage of total sleep time spent below 90% saturation.

In people with obstructive sleep apnea, higher T90 independently predicts the risk of developing high blood pressure. One study found that after accounting for age, body weight, and other factors, patients whose T90 exceeded 10% of total sleep time had roughly two-and-a-half times the odds of hypertension compared to those whose T90 was under 5%.¹⁶ Longitudinal data from two large cohorts also linked higher T90 to increased all-cause mortality, with each standard-deviation increase in T90 associated with an 18–34% higher risk of death over the follow-up period.¹⁷

On a chart, the visual difference is stark. A healthy sleeper’s trace might briefly kiss the low 90s once or twice during the night and bounce right back. A person with significant sleep-disordered breathing shows repeated sawtooth dips, often down to the 80s or even 70s, with the line spending extended stretches below the 90% mark. If your overnight chart shows the trace lingering below 90% for more than a few moments, that warrants a conversation with a doctor, regardless of how high the average reads.

COPD and Chronic Desaturation Patterns

People with moderate-to-severe chronic obstructive pulmonary disease live with baseline saturations that are already lower than average, and their nighttime charts can look alarming even without sleep apnea. In one study of COPD patients, about 38% qualified as nocturnal desaturators without any evidence of sleep apnea.¹⁸ Their charts show prolonged stretches below 90%, especially during REM sleep when breathing muscles are most relaxed.

A useful clinical observation: daytime oxygen saturation strongly predicts what the nighttime chart will look like. Research found that every patient with a daytime resting SpO2 of 93% or below experienced nocturnal desaturation, while no patient with a daytime SpO2 of 95% or above did.¹⁹ For people with COPD, a daytime reading in that 93–95% gray zone means overnight monitoring could reveal important information that a spot check in the clinic would miss.

How Reliable Is Your Oximeter

The device generating your oxygen chart matters a great deal. Medical-grade finger pulse oximeters are reasonably accurate in most conditions, but they rely on detecting tiny arterial pulsations that make up only about 2–5% of the total light signal passing through your finger. Anything that reduces blood flow to the fingertip, whether from cold hands, low blood pressure, or certain medications, can degrade accuracy.²⁰

Skin pigmentation introduces a well-documented bias. A 2025 study across five pulse oximeters used by the NHS found that darker skin tones consistently produced readings that were 0.6 to 1.5 percentage points higher than the true arterial value compared to lighter skin tones. The practical consequence is serious: the false negative rate for detecting genuinely low oxygen (arterial saturation at or below 92%) was 5 to 35 percentage points higher in patients with darker skin.²¹ Earlier laboratory work confirmed this pattern worsens as saturation drops: at very low saturations (60–70%), oximeters overestimated readings by about 3.6 percentage points in darkly pigmented subjects compared to only 0.4 points in lightly pigmented subjects.²² If you have darker skin and are monitoring oxygen at home, the chart may look more reassuring than it should.

Consumer smartwatches add another layer of uncertainty. Testing hospitalized COVID-19 patients, researchers found that the Apple Watch detected genuine hypoxia with a sensitivity of only about 35%, meaning it missed roughly two out of three truly low readings. The Withings ScanWatch did better at about 69% sensitivity but had more false alarms.²³ A separate comparison of smartwatch readings against arterial blood gases in COPD patients found a mean error of about 1.8 percentage points, but the limits of agreement were wide, spanning from roughly 5 points too high to 7 points too low.²⁴ A smartwatch chart is useful for spotting trends over time, but any single reading could easily be off by several points in either direction. Treat it as a rough guide, not a clinical measurement.

Pregnancy and Oxygen Saturation

Pregnant women sometimes worry that the dramatic cardiovascular changes of pregnancy will affect their oxygen levels. Blood volume increases substantially, the heart works harder, and the growing uterus pushes the diaphragm upward. Despite all of that, research using reflectance pulse oximetry found that oxygen saturation values did not change appreciably during the course of a normal pregnancy.²⁵ The oxygen chart of a healthy pregnant woman looks essentially the same as her pre-pregnancy chart. Persistently low readings during pregnancy are not an expected finding and deserve medical evaluation, just as they would in any other context.

Circadian Rhythm in Tissue Oxygen

Beyond blood oxygen, even the oxygen saturation within specific tissues follows a 24-hour clock. Researchers measuring retinal oxygenation over a full day-night cycle found a significant circadian rhythm, with an average swing of about 6 percentage points and a peak occurring in the early morning hours.²⁶ This is a reminder that oxygen levels throughout the body are not static. They rise and fall in patterns tied to the body’s internal clock, metabolic demand, and local blood flow regulation. Your chart is always a snapshot of a dynamic system, and small fluctuations are the sign that the system is working, not that something has gone wrong.