Television engineers are quite aware of the importance of the accurate measurement of time. Precise frequency references as well as an accurate knowledge of the time of day have always been critical in television. These two types of time measurement have always been intertwined in our experience, but there are those who would unravel that knot.
Until relatively recently in its history, humankind measured time by the rotation of the earth, using the visual positions of the sun, moon, and other celestial bodies as they traveled across the sky. Two kinds of such astronomical timekeeping are sidereal time and solar time. Sidereal time is measured with respect to the distant stars, while solar time is measured with respect to the sun. A solar day lasts from the time at which the sun is directly over a given point (for example, the Prime Meridian at Greenwich, England), until it is next over that same point.
A sidereal day lasts from the time at which a distant star is over a given point until it is next over that point. Because the motion of the earth in its orbit around the sun is negligible with respect to a distant star, but not with respect to the sun, a sidereal day is about four minutes shorter than a solar day.
It is possibly shocking to contemplate, but knowledge of the exact time was important even before radio and television broadcasting began! We learn from Dava Sobel's informative little book "Longitude," that before the mid-18th century, while any sailor worth his salt could determine his ship's longitude, determining latitude was far more difficult. This is true because the demarcations of latitude were completely arbitrary until Britain finally nailed down the Prime Meridian at Greenwich.
Until the ship's chronometer was invented, determination of latitude was a hit-or-miss proposition at best, and every great sailor in history until that time routinely got lost at sea. To make an accurate determination of longitude, the exact time at the ship's home port and its actual location must be known simultaneously. This was not possible until an Englishman named John Harrison invented the ship's chronometer, a highly accurate timepiece that was built without a pendulum and otherwise sufficiently rugged to keep accurate time aboard a ship.
Harrison, a carpenter, made the movements of his first three chronometers, H1, H2, and H3, largely out of wood. H4, about the size of a rather large pocket watch, was made of metal parts, but all four shared the design characteristic that they never needed to be lubricated. H1, H2, and H3 are, in fact, now running continuously in a museum, despite being more than 250 years old.
Among many technological innovations they contain is the first use of a bimetallic strip to compensate for temperature variations, and the use of frictionless escapements that do not wear down with use.
ATOMIC TIME
Today, we determine the exact time not by the rotation of the earth, but by the oscillations emitted by an electron changing quantum states in the Cesium atom; so-called atomic time. This is a very accurate way to measure the passage of time; so accurate that it generates a discrepancy with solar time.
Atomic time is formally called International Atomic Time, or T.A.I., and it is accurate within one second in 70,000 years. One T.A.I. second is equal to 9,192,631,770 oscillations between the two hyperfine levels of the ground state of the Cesium 133 atom.
Greenwich Mean Time is the solar mean time at the meridian that passes through Greenwich, England. Although all atomic timekeeping as transmitted by WWV in the United States is considered Coordinated Universal Time, Greenwich Mean Time is, interestingly, still the "legal" time in both the United Kingdom and the United States, as well as being the time reference used in communications satellite operations.
Solar time is officially known as Universal Time 1, or U.T.1, and one U.T. 1 second is equal to 1/86,400 of a full rotation of the earth. Owing to the molten state of the earth's core and its subsequent degree of rotational independence from the solid portion of the planet, and the tidal forces caused by the gravitational effects of the moon and the sun on the Earth's oceans, the speed of the earth's rotation is somewhat variable and is gradually slowing, to the tune of about two milliseconds per day per century. This means that Universal Time, based on the earth's rotation, is also slowing, but Cesium 133 atoms continue to oscillate between hyperfine states at precisely the same rate as always.
In order to coordinate Universal Time and atomic time, the two are reconciled periodically by the addition of a "leap second," in analogy to a leap year, in which a day is added to the calendar; this is logically called "Coordinated Universal Time," or UTC.
In 1967, global timekeeping was transferred to atomic clocks. Five years later, in 1972, the first leap second was added, and there have subsequently been an additional 21 leap seconds added. A leap second is added when the disparity between atomic time and UTC reaches 0.9 second, and it takes the form of a 61st second in the final minute of either June or December.
A leap second may be either positive or negative, but so far all leap seconds have been positive. The decision of when to add a leap second is made by the Earth Rotation Service at the Paris Observatory. In July, 2005, the Earth Rotation Service notified the timekeepers of the world that a leap second would be added at the end of 2005. The final seconds of last year ended follow with the following sequence:
2005 Dec. 31 23h 59m 59s
2005 Dec. 31 23h 59m 60s
2006 Jan. 1 0h 0m 0s
2006 Jan. 1 0h 0m 1s
This notice also noted that from 1999 Jan. 0h UTC, to 2006 Jan. 0h UTC, UTC-TAI=32s; and that from 2006 Jan. 0h UTC until further notice, UTC-TAI = 33s.
The issues of atomic time and Universal Time and leap seconds are regulated by the International Telecommunications Union. A meeting of the ITU committee charged with these issues that had been scheduled in November was postponed, but that body at that meeting was to decide whether to adopt or reject a proposal (made by the United States) to, for the first time since atomic time has been internationally used, uncouple atomic and astronomical time, making the addition of leap seconds a thing of the past.
NO MORE LEAP SECONDS?
What is at issue here? The United States has taken the position that the leap second should be abolished and the coordination between atomic and astronomical time severed because many of our timing systems, such as the global positioning system, are based purely on atomic time.
These electronic systems, and computer timekeeping systems in general, tend to have difficulties dealing with the addition of leap seconds. A further argument is that the disparity between atomic time-based GPS chronometers and UTC-based clocks complicates navigation.
The other side of this argument is that if the two systems are decoupled, astronomers will eventually lose track of celestial objects and satellites. One effect on our daily lives would be that the sun would rise later and later, albeit it would take several hundred years for the difference between TAI and UTC to become an hour. The proposal envisions the addition of an hour when that difference is reached. Astronomers, needless to say, are not receptive to the proposal.
The meeting at which the leap second proposal was to be discussed was postponed to gather more evidence concerning the effects of decoupling atomic time and solar time. As broadcasters, we would seem to have a foot in each camp. We make use of precise frequency references, which function purely on atomic time, but we also have to start our programs on time!
