Contributed by Peter Bond

In November 2005, the International Telecommunications Union (ITU) will be discussing a proposal to abolish leap seconds. The Royal Astronomical Society (RAS) strongly recommends that this proposal should be shelved, and that, before any changes are implemented, there should be a broader, public debate on the future use of these small adjustments to our annual time-keeping.

Our scientific understanding of time has developed over several centuries. Today, scientists recognise that there are two distinct requirements for time-keeping:

* absolute time-keeping, now based on high precision atomic clocks;

* everyday time-keeping, based on the rotation of the Earth (solar time). This is called Coordinated Universal Time (UTC).

Leap seconds are small adjustments to UTC, which keep ordinary clock time synchronised with the rotation of Earth and thus with the location of the Sun in the sky. They were introduced in 1972 as a reasonable compromise to serve both needs.

There have been 21 leap seconds since 1972 and the next is planned at the end of 2005. Their use is determined by the International Earth Rotation Service, which is sponsored by scientific bodies including the International Astronomical Union.

However, there is now a proposal to abolish leap seconds from December 2007. This proposal will be discussed by the International Telecommunications Union (ITU) at a meeting in Geneva in November 2005.

The proposed change would cause UTC to drift with respect to solar time. If agreed, this would change UTC so that it would serve only precise timing requirements.

The proponents of the change consider leap seconds to be a problem for precision timing applications and thus are seeking solutions. But the present proposal seeks to solve their problem by exporting problems to those who use clock time as a measure of mean solar time (as guaranteed by current international standards). These include astronomers, satellite operators and potentially all who study environmental phenomena related to the rising and setting of Sun.

The idea that clock time follows solar time is deeply embedded in contemporary technical culture through a wealth of literature (text books, web pages etc) and in the skills of working scientists and engineers around the world.

The Society is concerned that this issue has been subject only to a specialist and rather closed debate. There is a clear need for broader debate that involves a wider range of those who will be affected by the proposed change. This should extend outside science and technology -- for example, to consider whether civil/legal time should be based on precision time or mean solar time.

The Society strongly recommends that the proposal to abolish leap seconds should be shelved and that the ITU works to promote a broader and public debate.

"This debate should seek a fair solution that serves both needs for time-keeping," said Mike Hapgood, Secretary of the RAS, who has led the preparation of the Society's statement. "There are a lot of skilled people already involved in the debate; we need them to work together to improve current time-keeping for everyone's benefit and not just for one group."

NOTES FOR EDITORS

The International Earth Rotation and Reference Systems Service, located at the Paris Observatory, France, has decided that this year will be one second longer than last year. A leap second will be added at the end of 31 December 2005 -- the first for seven years.

The extra second is needed to keep Coordinated Universal Time (UTC) -- currently equivalent to Greenwich Mean Time (GMT) -- synchronised with Earth's rotation. Radio interferometry measurements show that the planet's spin rate is slowing by about 2 milliseconds per day per century. This gradual slowdown is caused by tides and other effects.

International Atomic Time (TAI) is the continuous, uniform time scale derived from atomic clocks, which are accurate to within one second per three million years. It is the ideal time scale for scientific use, but it is not practical for everyday use since it is not linked to the rotation of the Earth and the actual length of the day. Since it has not been changed since 1958, there is now a 32 second time difference between UTC and TAI.

FURTHER INFORMATION

* Leap-seconds discussion at US Naval Observatory   http://rom.usno.navy.mil/archives/leapsecs.html

* Steve Allen's bibliographic gathering of references on leap seconds   http://www.ucolick.org/~sla/leapsecs/onlinebib.html

* Marcus Kuhn's leap seconds site at Cambridge   http://www.cl.cam.ac.uk/~mgk25/time/leap/

* The UTC Problem and its Solutions, presentation by Pat Wallace   http://www.ien.it/luc/cesio/itu/wallace.pdf

* Time Bandits article by Wendy Grossman   http://www.theinquirer.net/?article=25190

* Detailed article by proponents of change   http://www.cl.cam.ac.uk/~mgk25/time/metrologia-leapsecond.pdf