Sub-menu Navigation

Is time up for the leap second? 

ClockA vote in 2015 could see the end of the leap second.

In 2004, it was formally proposed that there should be no further leap seconds after 2007 and since then there has been considerable debate about the issue.

In 2012 the Radio Communications Assembly decided to delay any decision in order to give countries more time to research the issue. This meant that the decision was deferred to the World Radio Conference which next meets in November 20I5; it is expected that it will make the final decision on the matter.

The UK position determined by successive science ministers has been to argue for the retention of a link between time and the earth’s rotation.

Co-ordinated Universal Time (UTC)

The timescale readily available in the UK is Co-ordinated Universal Time (UTC). This timescale is used internationally, with an offset of the appropriate number of hours for each time zone or to implement daylight saving time.

UTC is maintained by a number of atomic clocks around the world, and is computed by the Bureau International des Poids et Mesures (BIPM). UTC makes use of the atomic definition of the Sl second, which was introduced in 1967 and is effectively equivalent to the astronomical second based on a mean solar day of 86,400 seconds in about 1820.

Earth and sun

Historically, however, time has been determined by the rotation of the earth and the location of the sun in the sky; in Greenwich Mean Time (GMT) noon is defined to be the mean time the sun reaches its highest point in the sky in Greenwich.

However, the rotation of the earth is irregular and is also slowing; the length of the day is now (averaged out over several years) around 2.5 milliseconds longer than it was in 1820. This means that any time standard determined by accurate measuring devices like atomic clocks (UTC) slowly falls out of sync with solar time.

What are leap seconds?

In order to keep the two in sync leap seconds are added to (or, in principle, removed from) UTC to adjust for the irregularity in the earth's rotation. This ensures that UTC does not drift apart from GMT by more than 0.9 seconds. There have been 25 leap seconds since they were first introduced in 1972.

Arguments for removing leap seconds

  • Leap seconds occur at irregular intervals, and hence computers cannot be programmed to deal with them in the same way as they can for leap years where there is a fixed rule. This adds complexity to tasks like calculating the time difference between two dates.
  • To introduce a leap second, time sources and time broadcasts have to be changed manually, thereby creating scope for human error.
  • Most computer systems do not recognise leap seconds, creating potential problems for applications that need to time stamp events (such as financial transactions).
  • Although at present we need only one leap second every 1-2 years, as the Earth's rotation slows they will be needed more frequently, causing more instances of the types of disruption mentioned above.
  • Some software engineers claim that they have great difficulty in testing the behaviour of systems during a leap second, as the event is infrequent.

Arguments in favour of keeping leap seconds

  • Astronomers have developed over the centuries a timescale which matches the apparent movement of the sun, and it is a major cultural change to break this link.
  • We know how to handle leap seconds-we've had them for 40 years and have good experience of working with them, having had only occasional and small problems. Arguments about the difficulty of writing software to handle leap seconds are overstated.
  • Systems requiring a timescale without leap seconds can use such a scale: International Atomic Time (TAI) or GPS system time are readily available and do not have leap seconds.
  • Without leap seconds, it is not clear what we should do when the difference between GMT and UTC grows large. The most likely solution would be a change of time zone.
  • The software used by astronomers to control the pointing of their telescopes relies on the difference between UTC and GMT being less than one second. Modification is likely to be very costly. (The same point applies to software controlling high gain antennas for satellite communications.)
  • Although celestial navigation has been replaced by modem satellite and electronic methods it still provides a backup in emergencies. Without leap seconds correction tables would be needed to calculate the correct location.

Tell us your view

The National Measurement Office (NMO) is seeking information from bodies and organisations that may be affected by leap second issues to support the forthcoming review of whether to keep or discontinue the use of “Leap Seconds”.

If you have strong views either way tell us by emailing and if there is a consensus one way or another we will let the NMO know.

Stuart Carruthers, RYA Cruising Manager

Find books for your course at the RYA Shop

eBooks in the cloud
Our handy guide shows the books & DVDs that go with your course!
Please select...