ah, just cram an extra second in there somewhere

Long-term slow-down of the Earth rotation since 1830
(in red). The pink curve represents the influence
of the fluid core of the Earth.
(© Daniel Gambis, SYRTE --
Observatoire de Paris)

The Washington Post (DC USA)
Monday 26 December 2005
Page A17

Added Ticktock of the Clock
Restarts Time Debate

by Guy Gugliotta
Washington Post Staff Writer

Time marches on, but Earth is falling behind. The solution again this year is to add a "leap second" as 2005 ticks away, so Earth can catch up with the atomic clocks that have defined time since their unerring accuracy trumped the heavens three decades ago.

This will be the first leap second in seven years, and its arrival will be closely watched by physicists and astronomers enmeshed in a prolonged debate over the future of time in a world increasingly dominated by technology.

Some experts think the leap second should be abolished because the periodic, but random, adjustment of time imposes unreasonable and perhaps dangerous disruptions on precision software applications including cell phones, air traffic control and power grids.

Others, however, argue that it would be expensive to adjust satellites, telescopes and other astronomical systems that are hard-wired for the leap second, and besides, people want their watches to be in sync with the heavens.

Nobody knows how disruptive the leap second really is, but researchers hope to find out soon. "We're going to look at what happens this year," said Naval Research Laboratory physicist Ronald Beard. "If there are no significant problems, the whole issue will go away, but I don't expect that to happen."

Leap seconds are an outgrowth of the post-World War II development of increasingly accurate clocks based on the regular vibration, or "resonance," of atoms as they pass through a magnetic field. In 1958 an atomic second was defined as the time it takes for an atom of cesium 133 to tick through 9,192,631,770 cycles.

At that point atomic time and astronomical time are approximately the same, with the traditional astronomical second defined as 1/86,400th of a "mean solar day," the average time between two consecutive noons.

The trouble is that the heavens behave more capriciously than cesium. Also, the length of Earth's day is increasing by about two milliseconds per century because of the tides, whereas today's atomic clocks, unaffected by cosmic events, tick away with an accuracy within one second for every 20 million years.

Because of this discrepancy, atomic time has been pulling ahead of astronomical time for the past 47 years. To fix this, the International Telecommunication Union in 1972 stipulated that "Coordinated Universal Time" [UTC], an atomic time used as the world standard, could not diverge from astronomical time by more than 0.9 seconds.

The adjustment tool was the leap second, to be added or subtracted at the discretion of the International Earth Rotation and Reference Systems Service, either at the end of June or the end of the year. Beginning in 1972, there have been 21 leap seconds, the last one in 1998.

"Astronomers wanted a time scale that represented the Earth's movement, and the clock community wanted a smooth scale," said physicist Judah Levine of the National Institute of Standards and Technology and the University of Colorado, who favors eliminating leap seconds. "The compromise has become increasingly difficult to maintain."

The majority of scientists appear to agree that adding leap seconds, up to now an infrequent exercise, is likely to become a twice-a-year experience in about a century as Earth keeps slowing and the unvarying atomic clocks keep ticking away.

And unlike leap years or daylight saving time, software designers cannot plan ahead because leap seconds get added only when they are needed. The current seven-year hiatus is twice as long as any previous gap.

"At some point this will become so annoying that someone will want to change it," said astronomer Dennis McCarthy, retired director of the U.S. Naval Observatory's Directorate on Time and an advocate of abolishing the leap second. "I'm quite confident that people are not going to be happy with multiple leap seconds per year."

So why not get rid of them? Industry could enjoy the regularity of atomic clocks without risking technological collapse on New Year's Eve. And the divergence from solar time would not be more than one or two seconds per year, perhaps two minutes per century.

"The idea that it's going to be midnight in the middle of the afternoon is just nonsense," Levine said.

Others, however, suggest that the hardship caused by leap seconds may be overblown, if not illusory, as experts on both sides of the debate agree that there is little data beyond a few anecdotes to suggest that leap seconds have in the past created havoc in time-sensitive endeavors.

"The case hasn't been made," said University of Virginia astronomer Ken Seidelmann. "All there is is rumor that they're inconvenient, but we've had them for 30-plus years and there's no outcry. I see no reason to get rid of them."

Furthermore, Seidelmann added, astronomers and satellite operators deploy sensitive equipment on the ground and in space on the assumption that the Coordinated Universal Time signal will match up within a second of astronomical time, critical in decisions such as when and how to point solar panels or satellite imagers.

Still, cautioned McCarthy, "we need to make a decision" about the leap second because engineers are designing satellite systems today that "will be used 20 years from now. We need to do them a service so they're not stuck."

In recent years, time mavens opened a discussion about what to do, and a "general consensus" emerged that "the advantage to astronomy was not worth the pain and suffering of leap seconds," Levine said. "It looked like a done deal."

But it wasn't. Earlier this year Britain's Royal Astronomical Society decried a U.S. proposal to abolish the leap second, suggesting that doing so would disrupt not only astronomers but "all who study environmental phenomena related to the rising and setting of the sun."

Last month, a working group of the International Telecommunication Union meeting in Geneva decided to postpone discussions of the U.S. proposal, which would have abandoned leap seconds in 2007 and let Coordinated Universal Time and astronomical time diverge for several hundred years before inserting a "leap hour."

The working group said more time was needed to form a consensus, and suggested that this year's leap second offered a welcome opportunity to determine whether change is necessary.

"There is a philosophical feeling that abolishing the leap second results in greater decoupling of the time scale from rotation of the Earth, and that this is not a good thing," said McCarthy, a drafter of the U.S. proposal. "But we put up with daylight savings time in the United States, and China has one time zone for the entire country. My own feeling is that we all live with departures."

© 2005 The Washington Post Company