Wrist Temperature

Wrist temperature is the temperature of the skin under your watch, measured while you sleep and reported as a change from your own usual level. Apple Watch shows it as a deviation such as +0.4 °C, and Oura and WHOOP report finger or wrist skin temperature the same way, because the absolute skin reading says little about your body until it is compared with your own nights.

The deviation is useful because it moves for identifiable reasons. A fever, the second half of a menstrual cycle, a few drinks, a hot bedroom or a heavy duvet will each push it up, and in some studies the change showed up days before people sought a test. This entry covers what the sensor actually reads, why skin and core temperature differ, how Apple Watch, Oura and WHOOP produce their numbers, how well the numbers hold up in studies, what shifts them, and how the reading relates to Titan.

01What the number measures

A wrist sensor reads the surface of your skin, a few millimeters from the blood vessels that carry heat out of the body. That surface sits several degrees below your core. In the TemPredict study, the population mean of finger skin temperature from the Oura ring was 31.2 °C, with a standard deviation of 1.7 °C and wide differences between people (Smarr et al., 2020). Oral temperature in 35,488 outpatients averaged 36.6 °C, with 95% of individual readings between 35.7 and 37.3 °C (Obermeyer et al., 2017).

Because the gap between skin and core varies so much from person to person, wearables throw away most of the absolute value and keep the change. Apple's HealthKit documentation states that the watch "records the absolute wrist temperature value," while the Health app "displays this data as a relative value, based on a person's baseline" (Apple, 2026d). Your number is a comparison with your own recent nights, in the same way a baseline works for HRV or resting heart rate.

02Why skin temperature differs from core temperature

Core temperature is the heat of your brain, heart and abdominal organs, held within a narrow band. Skin temperature is the place where your body gets rid of that heat, so it rises when blood vessels in the hands and wrists open and falls when they close.

The two follow opposite daily curves. Core temperature falls through the evening and reaches its lowest point during sleep. Wrist temperature does the reverse. Sarabia and colleagues fitted wireless sensors to the wrists of 99 university students and found that wrist temperature ran in inverse phase with oral temperature, about 60 minutes ahead of it. It "started to increase in association to bed time and dropped sharply after awakening" (Sarabia et al., 2008). Across 91,462 UK Biobank participants who wore a wrist device with a built-in thermometer for 7 days, the plateau of highest wrist temperatures fell during the night, anti-phase to core temperature (Brooks et al., 2023).

That night-time warming is part of how you fall asleep. Opening the vessels in your hands and feet dumps heat and lowers core temperature. Kräuchi and colleagues measured this as the distal-to-proximal skin temperature gradient, the difference between hands and feet and the trunk, and found it was a better predictor of how quickly people fell asleep than core temperature, heart rate, melatonin onset or rated sleepiness (Kräuchi et al., 2000). This tie to your circadian rhythm is one reason wearables sample during sleep. Oura gives the other, that the night is when "your body is in its most stable state" and the data carries the least environmental noise (Oura, 2026a).

The daily swing also carries health information at the population level. In the UK Biobank sample, a wrist temperature amplitude 1.8 °C smaller than average was associated with hazard ratios of 1.69 for type 2 diabetes and 1.23 for hypertension (Brooks et al., 2023). These are associations in a population, and a single person's nightly deviation cannot be read against them.

03How Apple Watch measures sleeping wrist temperature

Apple Watch Series 8 and later, every Apple Watch Ultra and Apple Watch SE 3 record wrist temperature during sleep. Each has two temperature sensors, one in the back crystal against your skin and one just under the display. Apple says the design reduces bias from the outside environment. While you sleep, the watch samples temperature every five seconds (Apple, 2026a).

The watch then reduces a night of samples to one value. HealthKit's documentation describes it as a single sample corrected "for environmental bias" that "represents the wrist temperature over the entire night" (Apple, 2026d). For a night to count, the watch has to track at least 4 hours of sleep with the Sleep Focus on. Health shows the result after about 5 nights, once it has a baseline, under Browse, then Body Measurements, then Wrist Temperature (Apple, 2026b). The same 5-night wait applies when you switch to a new watch (Apple, 2026a).

Apple is explicit that the feature "is not a thermometer and cannot provide wrist temperature measurements on-demand," and that it is not a medical device (Apple, 2026a). Three other features use the data.

• Vitals app. On most models it builds a typical range for your overnight heart rate, respiratory rate, blood oxygen, wrist temperature and sleep duration after 7 nights, and sends a morning notification when multiple metrics fall outside that range (Apple, 2026e).
• Retrospective ovulation estimates. Cycle Tracking uses the temperature rise that often follows ovulation, the biphasic shift, to estimate after the fact when you ovulated. Estimates require wearing the watch regularly for at least 2 cycles (Apple, 2026c).
• Period predictions. Cycle Tracking also uses wrist temperature to improve predictions of your next period (Apple, 2026a).

04How Oura and WHOOP measure skin temperature

The Oura ring reads temperature at the base of your finger. Smarr and colleagues describe the sensor as a negative temperature coefficient thermistor, "non-calibrated, resolution of 0.07 °C," on the inner palm side of the ring, sampled every minute (Smarr et al., 2020). Oura states that it "measures average body temperature from the skin, which is not the same as basal, or core body temperature." It sets your baseline over the first couple of weeks of wear and adjusts it as needed. The daily figure is the average of the previous night's readings expressed as a deviation from baseline, and the trend view uses a weighted three-day average. Temperature feeds Oura's Readiness score, where it "can only lower your Readiness Score, not boost it" (Oura, 2026a).

WHOOP measures skin temperature on WHOOP 4.0. Its developer documentation lists skin temperature in degrees Celsius as one of the measurements factored into Recovery for WHOOP 4.0 members, alongside resting heart rate, HRV and blood oxygen (WHOOP, 2026). WHOOP has not published how it builds the baseline for that value.

05How accurate wearable skin temperature is

Apple's documentation does not report a comparison of wrist temperature with a reference thermometer, and Oura's main accuracy claim comes from its own unpublished work. Oura reports an internal study in which rings in a water bath matched research-grade iButton sensors with r² above 0.99, and 16 people wearing the ring alongside iButtons for a week produced r² above 0.92 (Oura, 2026b). Those figures describe how well the ring tracks a skin sensor. They say nothing about core temperature.

The stronger evidence tests whether the signal detects real physiological events.

Menstrual cycle. In 22 women tracked for about four months, nocturnal finger temperature from the Oura ring rose 0.30 °C from the follicular to the luteal phase, more than the 0.23 °C rise in oral temperature taken the same mornings. An algorithm using the ring data placed ovulation within the fertile window with 83.3% sensitivity (Maijala et al., 2019). In 120 participants, temperature-based cycle detection agreed with urine luteinizing hormone surge testing in 82% of those who completed both methods (Gombert-Labedens et al., 2024).

Apple's ovulation estimates. Apple's algorithms were tested in 260 participants and 889 cycles against urine LH tests. For completed cycles with a temperature shift of at least 0.2 °C, the retrospective ovulation estimate had a mean absolute error of 1.22 days, with 89.0% of estimates within 2 days of the LH-defined day. Across all cycles the error was 1.66 days. For ongoing cycles, the wrist temperature algorithm's error of 1.91 days beat the calendar method's 2.35 days (Wang et al., 2025).

Illness. In the first 50 TemPredict participants who reported COVID-19, finger temperature elevations were visible in the ring data and matched self-reported fever (Smarr et al., 2020). A follow-up model trained on 73 PCR-confirmed cases flagged infection an average of 2.75 days before people sought testing, with 82% sensitivity and 63% specificity. Adding continuous temperature raised the model's area under the curve from 0.770 to 0.819 (Mason et al., 2022).

The practical reading is that consumer skin temperature is good at detecting changes of a few tenths of a degree within one person over several nights. It is poor at telling you what your core temperature is right now, which is what a fever check needs.

06What a normal reading looks like

There is no population reference range for a nightly wrist temperature deviation, because each device defines zero as your own baseline. Your ordinary nights cluster around that zero by construction. A single night away from it usually has an ordinary cause, and the factors below explain most of them.

Two patterns deserve more attention. One is a rise that persists across several nights. The other is a rise that arrives together with a higher sleeping heart rate, lower HRV or a faster respiratory rate. Apple's Vitals app is built around the second idea and stays quiet until more than one overnight metric leaves your typical range (Apple, 2026e).

For people who menstruate, a two-level pattern is normal. In 26 women with regular ovulatory cycles, Oura skin temperature was 0.198 °C higher in the mid-luteal phase and 0.242 °C higher in the late luteal phase than during menses (Alzueta et al., 2022). A baseline that averages across the whole cycle will read slightly low in the first half and slightly high in the second.

07What shifts wrist temperature

Apple lists diet, exercise, alcohol, sleep environment, menstrual cycles and illness as reasons the nightly value varies (Apple, 2026a). Its developer documentation adds jet lag (Apple, 2026d).

Illness and fever

Fever raises core temperature, and the skin follows. The TemPredict results above show that a ring can see the rise, sometimes before a person recognizes they are ill (Smarr et al., 2020, Mason et al., 2022). A night or two of higher temperature with a higher sleeping heart rate and lower HRV is the pattern most consistent with an infection starting. Your watch cannot diagnose it, and a thermometer or a clinician should settle any question of fever.

Menstrual cycle phase and ovulation

Body temperature rises after ovulation and stays higher until the next period. Gombert-Labedens and colleagues put the rise at 0.3 to 0.7 °C, the biphasic pattern described above (Gombert-Labedens et al., 2024). Apple's own documentation warns that ovulation "may have occurred without a biphasic shift, or the shift in temperature may have been too low to provide an estimate" (Apple, 2026c). In Apple's validation study, the algorithms only produced an estimate for about 70 to 81% of cycles, depending on the algorithm and the size of the shift (Wang et al., 2025).

Alcohol

Alcohol widens skin blood vessels. In a controlled study of 48 healthy men, ethanol increased skin blood flow at room temperatures of 25 and 30 °C, while at 20 °C and 35 °C the cold or heat itself dominated and ethanol added nothing measurable (Hughes et al., 1984). Apple lists alcohol among the causes of nightly variation (Apple, 2026a), and a normal bedroom sits in the range where ethanol widened skin vessels in that study.

Room temperature, bedding and fit

The sensor reads the air and skin around it, so a warmer bedroom, a heavier duvet or an arm under the covers can move the value. Apple uses the second sensor under the display to reduce this bias (Apple, 2026a), and it still names sleep environment as a cause of nightly variation. Martinez-Nicolas and colleagues recorded a week of wrist temperature in 103 young adults living normally and found that environmental conditions changed the amplitude of the daily rhythm more than its timing (Martinez-Nicolas et al., 2013).

Exercise, meals and travel

Hard evening training, large late meals and crossing time zones all change overnight thermoregulation. Oura notes that daytime temperature is "impacted by eating, drinking, exercising, and other external factors," which is why it scores only the night (Oura, 2026a). After a long flight, give the reading a few nights to settle before comparing it with your baseline.

08Wrist temperature and Titan

HealthKit stores sleeping wrist temperature under its own identifier, appleSleepingWristTemperature, separate from the Body Temperature type (Apple, 2026d). Titan asks for read access to Body Temperature along with its other Health permissions. Titan's Recovery score compares your overnight HRV and sleeping heart rate with your own baselines, as how the Recovery score works explains.

Temperature is useful context on mornings when Recovery drops for no obvious training reason. Check Wrist Temperature in the Health app or Apple's Vitals app. A higher temperature points toward illness, alcohol, a warm room or the luteal phase, and a normal one points back toward training load or short sleep. When the cause becomes clear, add the matching tag, such as Sickness or Alcohol, in the Journal, so Journal Insights can compare those days with the rest.

09Reading your own number

Compare wrist temperature only with itself, from the same device, because every watch and ring sets its own zero. Give a new device its full baseline period before reading anything into the chart. Look at several nights together, and weigh a rise more heavily when your sleeping heart rate, HRV or respiratory rate move with it. Keep your bedroom, bedding and watch fit as steady as you can, since those account for many one-night swings. If you track your cycle, expect a step up after ovulation and a step down around your period. If you feel unwell or suspect a fever, take your temperature with a thermometer. The wrist number tells you something changed and a thermometer tells you whether you have a fever.

10References

• Alzueta E et al. (2022). Tracking sleep, temperature, heart rate, and daily symptoms across the menstrual cycle with the Oura Ring in healthy women. International Journal of Women's Health 14:491-503. https://doi.org/10.2147/IJWH.S341917
• Apple (2026a). Track your nightly wrist temperature changes with Apple Watch. Apple Support, accessed September 2026. https://support.apple.com/en-us/102674
• Apple (2026b). Track your nightly wrist temperature on Apple Watch. Apple Watch User Guide, accessed September 2026. https://support.apple.com/guide/watch/track-your-nightly-wrist-temperature-apd526d20feb/watchos
• Apple (2026c). Receive retrospective ovulation estimates on Apple Watch. Apple Support, accessed September 2026. https://support.apple.com/en-us/120357
• Apple (2026d). appleSleepingWristTemperature. Apple Developer Documentation, HealthKit, accessed September 2026. https://developer.apple.com/documentation/healthkit/hkquantitytypeidentifier/applesleepingwristtemperature
• Apple (2026e). Track your overnight and daytime vitals with Apple Watch. Apple Support, accessed September 2026. https://support.apple.com/en-us/120142
• Brooks TG et al. (2023). Diurnal rhythms of wrist temperature are associated with future disease risk in the UK Biobank. Nature Communications 14:5172. https://doi.org/10.1038/s41467-023-40977-5
• Gombert-Labedens M et al. (2024). Using wearable skin temperature data to advance tracking and characterization of the menstrual cycle in a real-world setting. Journal of Biological Rhythms 39(4):331-350. https://doi.org/10.1177/07487304241247893
• Hughes JH et al. (1984). Influence of ethanol and ambient temperature on skin blood flow. Annals of Emergency Medicine 13(8):597-600. https://doi.org/10.1016/s0196-0644(84)80282-6
• Kräuchi K et al. (2000). Functional link between distal vasodilation and sleep-onset latency? American Journal of Physiology, Regulatory, Integrative and Comparative Physiology 278(3):R741-R748. https://doi.org/10.1152/ajpregu.2000.278.3.R741
• Maijala A et al. (2019). Nocturnal finger skin temperature in menstrual cycle tracking, ambulatory pilot study using a wearable Oura ring. BMC Women's Health 19:150. https://doi.org/10.1186/s12905-019-0844-9
• Martinez-Nicolas A et al. (2013). Uncovering different masking factors on wrist skin temperature rhythm in free-living subjects. PLOS ONE 8(4):e61142. https://doi.org/10.1371/journal.pone.0061142
• Mason AE et al. (2022). Detection of COVID-19 using multimodal data from a wearable device, results from the first TemPredict Study. Scientific Reports 12:3463. https://doi.org/10.1038/s41598-022-07314-0
• Obermeyer Z et al. (2017). Individual differences in normal body temperature, longitudinal big data analysis of patient records. BMJ 359:j5468. https://doi.org/10.1136/bmj.j5468
• Oura (2026a). Body temperature. Oura Member Care, accessed September 2026. https://support.ouraring.com/hc/en-us/articles/360025587493-Body-Temperature
• Oura (2026b). How accurate is Oura's temperature data? The Pulse blog, accessed September 2026. https://ouraring.com/blog/temperature-validated-accurate/
• Sarabia JA et al. (2008). Circadian rhythm of wrist temperature in normal-living subjects, a candidate of new index of the circadian system. Physiology and Behavior 95(4):570-580. https://doi.org/10.1016/j.physbeh.2008.08.005
• Smarr BL et al. (2020). Feasibility of continuous fever monitoring using wearable devices. Scientific Reports 10:21640. https://doi.org/10.1038/s41598-020-78355-6
• Wang Y et al. (2025). Performance of algorithms using wrist temperature for retrospective ovulation day estimate and next menses start day prediction, a prospective cohort study. Human Reproduction 40(3):469-478. https://doi.org/10.1093/humrep/deaf005
• WHOOP (2026). Recovery. WHOOP Developer Platform, accessed September 2026. https://developer.whoop.com/docs/developing/user-data/recovery/