Skip to content

UTC vs GMT: What’s the Difference?

On any given day, UTC and GMT differ by less than a second — and yet they are not the same thing. One is a meridian on a hill in London, the other is the weighted average of hundreds of atomic clocks. Here is why we have both, what the difference really is, and the surprisingly narrow set of cases where it actually matters.

· 9 min read · Gabriel

World map showing the difference between GMT and UTC across different countries
Table of Contents

Same time, different rules

Open any phone, weather app or world clock and you will sooner or later see the labels "UTC" and "GMT" used as if they were interchangeable. For everyday purposes, they almost are — at any given moment, UTC and GMT differ by less than a second. But they are not the same thing, and that small difference is the result of a very deliberate choice made in 1972 to separate civil time from the rotation of the Earth.

One is a meridian on a hill in London, born of celestial navigation. The other is the weighted average of hundreds of atomic clocks running in laboratories around the world. They were merged into a single global standard in 1972 and then quietly split apart again. Understanding the difference matters if you write software, run a server fleet, time financial trades, or simply want to know what your "UTC offset" really represents.

What GMT actually is

Greenwich Mean Time is the mean solar time at the Royal Observatory in Greenwich, London. It was conceived in 1675 when King Charles II founded the Observatory specifically to solve the longitude problem — sailors at sea needed an accurate reference clock to determine their east-west position by comparing local solar noon to the time at a known meridian.

The "mean" in Greenwich Mean Time matters. Real solar days vary in length over the year because Earth's orbit is elliptical and its axis is tilted. A real sundial day can be up to 16 minutes longer or shorter than the average. Mean solar time smooths these wobbles into a uniform 24-hour day, defined by the average position of an imaginary "mean sun" rather than the actual one.

For most of its history GMT was counted from noon, not midnight, because astronomers wanted a single date stamp to span an entire night of observations. Civil clocks worked from midnight, but astronomical GMT counted from the previous noon. This dual convention caused enough confusion that on 1 January 1925 GMT was officially redefined to begin at midnight, and astronomers switched to a separate astronomical scale.

The 1884 International Meridian Conference in Washington, D.C., made Greenwich the world's prime meridian. Twenty-five nations attended; twenty-two voted in favour of the Greenwich meridian (France abstained and stuck to the Paris meridian until 1911). From that point forward, every time zone on Earth was defined as an offset from GMT — UTC+1, UTC-5, and so on.

What UTC actually is

Coordinated Universal Time is, at heart, atomic time. It is generated by combining the readings of more than 400 atomic clocks at about 80 institutes worldwide — including NIST in the United States, NPL in the United Kingdom, PTB in Germany and NICT in Japan — into a weighted average produced by the Bureau International des Poids et Mesures (BIPM) in Paris.

The base unit is the SI second, redefined in 1967 as exactly 9,192,631,770 oscillations of the radiation emitted by a cesium-133 atom transitioning between two specific energy levels. That definition replaced the astronomical second (a fraction of the mean solar day) with a physical constant. From there, 86,400 SI seconds make a UTC day.

The raw atomic average is called International Atomic Time (TAI). TAI never stops, never adjusts, and never cares about the Earth. UTC is TAI with one important addition: it occasionally inserts a "leap second" so that it does not drift too far away from the actual rotation of the planet.

The acronym itself is a compromise. English speakers wanted "CUT" and French speakers "TUC"; the neutral "UTC" was adopted in 1970 so that no language got priority over another.

Why we needed UTC

The Earth is a bad clock. Its rotation is slowing down by about 1.7 milliseconds per century due to tidal friction with the Moon. It also varies unpredictably from day to day because of atmospheric winds, ocean currents, melting ice, post-glacial rebound, and motion in the molten core. Over a year these tiny variations accumulate to roughly a second.

Atomic clocks, by contrast, are extraordinarily stable. The best optical lattice clocks today would neither gain nor lose a second in roughly 15 billion years — longer than the age of the universe. Once humanity had clocks that ticked more reliably than the planet they sat on, a choice had to be made: do we keep redefining the second to match the wobbling Earth, or do we lock the second to physics and let astronomical time drift?

The answer, agreed in 1972, was a compromise. UTC would tick in perfectly uniform SI seconds, like TAI, but would be nudged occasionally by a whole second so that it never drifts more than 0.9 seconds away from UT1 — the modern, refined version of GMT based on actual measurement of the Earth's rotation.

The leap second compromise

A leap second is a one-second adjustment inserted (or, in theory, removed) at the end of 30 June or 31 December UTC to keep the standard aligned with the rotating Earth. When inserted, the clock reads 23:59:59 → 23:59:60 → 00:00:00 instead of jumping straight from 23:59:59 to 00:00:00. The decision is made by the International Earth Rotation and Reference Systems Service (IERS) based on observations of the Earth's rotation.

Since 1972, twenty-seven leap seconds have been inserted, all positive — the Earth has been running slow, never fast, on the timescale that matters. The most recent was inserted on 31 December 2016. Since then, the planet has actually been rotating slightly faster, and as of 2026 there has been no leap second in nearly a decade — the longest gap in the system's history.

Leap seconds are unpopular with engineers because they cannot be predicted more than six months in advance and they break the assumption that every minute has exactly 60 seconds. The 30 June 2012 leap second caused widespread outages: Reddit and LinkedIn went down, Mozilla services stalled, Qantas's reservation system crashed for hours. The 31 December 2016 leap second went more smoothly only because most major operators had adopted Google's "leap smear" — spreading the extra second over 24 hours so that no clock ever shows 23:59:60.

In November 2022, the 27th General Conference on Weights and Measures (CGPM) voted to abolish the leap second by 2035 at the latest. UTC will continue to exist, but the maximum allowed gap between UTC and UT1 will be increased — to several minutes, perhaps an hour — eliminating the need for sudden one-second corrections. Civil time will quietly drift away from solar time, and nobody will notice for centuries.

So what is the difference, really?

If you ask the precise question "what time is it in London right now in winter?" — UTC and GMT will give the same answer to within nine-tenths of a second, every time. For nearly every human use case, they are interchangeable.

The differences only emerge when you zoom in. The table below compares each time scale on the four properties that actually matter:

Time scaleDefined byLeap seconds?Offset from UTC (May 2026)
UTCWeighted average of ~400 atomic clocks (BIPM, Paris)Yes0 — the global reference
UT1 (astronomical GMT)Earth's rotation, measured by the IERSNoWithin ±0.9 s of UTC
TAIAtomic clocks, never adjustedNo+ 37 s
GPS timeUS DoD atomic clocks (since 6 Jan 1980)No+ 18 s
Civil GMTUK / Portugal winter; Iceland year-roundn/a — same as UTC+00

The exact UT1 − UTC offset, called DUT1, is published by the IERS in Bulletin D and changes by milliseconds from week to week.

GMT in practice today

In casual British usage, GMT is the country's winter time zone. The BBC announces "the time in London is 7 a.m. GMT," weather forecasts label tide tables in GMT, and mobile phones display "GMT" on the lock screen between late October and late March. From the last Sunday in March to the last Sunday in October, the same clocks read British Summer Time (BST), which is GMT + 1 hour, equivalent to UTC+1.

This colloquial GMT is, technically, just UTC+0. The actual astronomical GMT — the one based on the meridian and the Sun — survives only in scientific contexts under the name UT1. When a navigation chart, a satellite tracking system or an astronomy paper says "GMT," it almost always means UT1. When your phone says GMT, it means UTC.

The IANA time zone database, used by every major operating system, records the British winter zone as Europe/London with an abbreviation of GMT. There is no separate UTC+0 zone for everyday users; the database treats them as one and the same.

When the difference actually bites

For ordinary applications — calendars, deadlines, meeting invitations, weather forecasts — pick either label and you will be fine. The cases where it matters are surprisingly narrow but very real:

  • Distributed systems and timestamps. Servers that synchronise via NTP get UTC. If your code expects every minute to have 60 seconds, a leap second can corrupt timestamps, deadlock retry loops, or trip up rate limiters. Linux's CLOCK_TAI and Google's leap smear exist to work around this.
  • Financial markets. Regulations such as MiFID II in Europe require trade timestamps accurate to 100 microseconds and traceable to UTC. Exchanges synchronise to atomic time directly to avoid leap-second ambiguity.
  • Astronomy and satellite tracking. The position of a satellite or a star at a given moment depends on Earth orientation, which means UT1, not UTC. A 0.9-second error translates to roughly 400 metres at the equator.
  • GPS receivers. They internally use GPS time and apply a correction (currently 18 seconds) to display UTC. Old firmware that misses an update can be off by an integer number of seconds.
  • Aviation and shipping. Officially everything runs on UTC, but pilots and sailors still call it "Zulu time" — the military designator for the UTC+0 zone, originally the "Z" zone in the NATO phonetic alphabet.

Quick reference

  • If you are choosing a label for technical use — pick UTC. Always.
  • If you are writing about UK or Portuguese winter time in everyday language — GMT is fine, and idiomatic.
  • If you mean "the actual orientation of the Earth in space" — say UT1, not GMT.
  • If you need a time scale with no leap seconds — use TAI (37 seconds ahead of UTC) or GPS time (18 seconds ahead of UTC).
  • "Zulu" (Z), as in 2026-05-05T12:00:00Z, is just another name for UTC.

The split between UTC and GMT is, in the end, a story about a planet that turned out to be a worse clock than the atoms it carries. For three centuries GMT was the most accurate timekeeping reference humanity possessed. Then, in a single generation, atomic physics made it obsolete — but only by a fraction of a second per year. The world quietly switched standards in 1972, kept the old name in casual use, and patched the difference with leap seconds for half a century. By 2035, even that patch will be gone, and UTC and GMT will finally be allowed to drift apart for good.

Further reading

  • McCarthy, Dennis D. and Seidelmann, P. Kenneth. Time: From Earth Rotation to Atomic Physics (2nd ed., Cambridge University Press, 2018).
  • Howse, Derek. Greenwich Time and the Longitude (Philip Wilson Publishers, 1997).
  • Bureau International des Poids et Mesures — official UTC and TAI bulletins.
  • International Earth Rotation and Reference Systems Service. Bulletin C (leap second announcements) and Bulletin D (DUT1).

Test your understanding

1. Which time scale is generated as the weighted average of hundreds of atomic clocks at institutes around the world?
2. At any given moment, the difference between UTC and UT1 (astronomical GMT) is kept within…
3. When did UTC officially replace GMT as the world’s primary time standard?
4. A colleague writes “Let’s meet at 14:00 GMT on 15 July.” What’s the most likely problem?

← Back to all articles