Leap seconds end May 20, 2027 as CGPM adopts Resolution 5
Contents
The 28th CGPM (General Conference on Weights and Measures, where Metre Convention member states decide the definitions of units such as the metre and the second), held in Versailles from October 13 to 15, 2026, adopted a resolution to stop inserting leap seconds.
Draft Resolution C, which I covered in August in Leap seconds are set to end May 20, 2027, was adopted as Resolution 5.
Continuous UTC takes effect on May 20, 2027, and the difference between UT1 (time based on Earth’s rotation) and UTC (Coordinated Universal Time) is allowed to reach up to 3,600 seconds.
UTC ticks in atomic-clock seconds, and to keep it within 0.9 seconds of UT1, an extra second, 23:59:60, has been inserted from time to time.
Continuous UTC means dropping those insertions and simply counting atomic-clock seconds.
The mechanism, and the outages leap seconds have caused, are in the August article.
In Japan, NICT (National Institute of Information and Communications Technology) published a notice on October 23.
It says that, as a result of this decision, UTC will not undergo discontinuous adjustments such as leap seconds for at least the next several centuries.
The adopted text and Draft Resolution C
The vote took place on October 15, the last day of the conference.
In a statement that same day, Noël Dimarcq, President of the CCTF (Consultative Committee for Time and Frequency), reported that three resolutions related to time metrology had been adopted earlier in the day.
Scientific American reported that the resolution passed by a majority vote.
Before the conference, the draft resolutions were labeled with letters. The three related to time received these numbers after adoption.
| Draft | Adopted as | Title |
|---|---|---|
| B | Resolution 4 | On the future definition of the second |
| C | Resolution 5 | On the technical actions needed to ensure the continuity of UTC |
| D | Resolution 6 | On the definition of an international lunar reference time scale and its traceability to UTC |
Comparing the English text of Resolution 5 with Draft Resolution C in the draft resolutions PDF, the only differences are hyphenation and similar punctuation, and I found no change in substance.
The effective date added in the January 30 revision is still there.
The BIPM (International Bureau of Weights and Measures, which produces UTC and publishes the resolutions) notes that the official version is the French text.
The “decides” section of Resolution 5 contains these two items. The rest recalls earlier resolutions or makes requests to other organizations.
| Item | Text of Resolution 5 |
|---|---|
| Effective date | Continuous UTC will become effective on May 20, 2027 |
| Limit | The maximum value for |UT1−UTC| will be 3,600 seconds (1 hour), ensuring the long term continuity of UTC for several centuries |
May 20, 2027 is World Metrology Day, as proclaimed by UNESCO.
All 27 leap seconds so far have added a second.
With the limit widened to one hour, no adjustment is needed until the difference reaches an hour.
”Several centuries” versus “several millennia” of continuous UTC
Sources describe how long UTC stays continuous under the 3,600-second limit in different ways.
| Source | Wording |
|---|---|
| Resolution 5 | several centuries |
| NICT notice | at least the next several centuries |
| CCTF President’s statement | expected to maintain continuity for several millennia |
| Scientific American | a leap hour won’t be needed for more than 1,000 years |
The resolution says “several centuries,” and NICT’s notice uses the same wording.
”Several millennia” appears in the statement the President issued to report the adoption, not in the resolution itself.
Scientific American describes the change as replacing the leap second with a “leap hour.”
Resolution 5, however, only sets the limit on the difference. It does not say how UTC would be adjusted as the difference approaches 3,600 seconds.
What ITU-R, IERS and BIPM still have to do before May 20, 2027
Resolution 5 requests the BIPM to work with the ITU (International Telecommunication Union), the IAU (International Astronomical Union), the IERS (International Earth Rotation and Reference Systems Service), and others to inform users.
It also asks the BIPM to keep monitoring UT1−UTC and its predicted evolution, and to report regularly to the CIPM (International Committee for Weights and Measures).
The current leap second rules are set by ITU-R (the ITU Radiocommunication Sector) Recommendation TF.460-6.
According to NICT’s notice, ITU-R is expected to revise the recommendation following this decision.
Equipment that broadcasts time by radio gets a grace period.
Resolution 655 (REV.WRC-23) of the ITU World Radiocommunication Conference, cited in Resolution 5, sets up a transition period.
Where existing equipment cannot be replaced earlier, the increased difference between UT1 and UTC may be disseminated by radio until 2035, and no later than 2040.
According to Resolution 5, ITU-R Working Party 7A is studying updates to the code radio stations use to transmit UT1−UTC.
As for leap seconds themselves, the IERS Bulletin C 72, which I checked for the August article, announced in July 2026 that no leap second would be inserted at the end of December.
Bulletin C comes out every six months and announces whether a leap second will be inserted at the end of the following June or December.
The next one would cover the end of June 2027, which falls after the May 20 effective date, so the 27th leap second on December 31, 2016 looks set to be the last.
TAI (International Atomic Time) counts atomic-clock seconds without leap seconds, and its gap with UTC has grown by one second with each leap second.
That gap will now stay at 37 seconds.
The leap second list in tzdata (the IANA time zone database), leap-seconds.list, was unchanged from what I checked in August when I fetched it on October 23: last updated July 6, 2026, expiring June 28, 2027.
Lunar reference time became Resolution 6
At the same conference, Draft Resolution D on lunar reference time was adopted as Resolution 6.
Comparing it with the draft text, I found no change in substance here either.
Resolution 6 says a proliferation of reference time scales risks ambiguity when exchanging time-tagged data and must be avoided.
It also notes that TCL (Lunar Coordinate Time) without any scaling would meet the scientific requirements, and recommends that if a scaled version of TCL is used, its frequency offset be set from a conventional value fixed by international organizations.
Resolution 6 still does not choose between TCL and a scaled version.
In the United States, a 2024 OSTP (White House Office of Science and Technology Policy) memo requires NASA, working with other departments, to deliver a finalized strategy for implementing lunar timing standardization to the Executive Office of the President by December 31, 2026.