Precise timekeeping helped establish and develop Canada. For the past 2 centuries, Canadian exploration, mapping, navigation and transportation have exploited state-of-the-art precise time systems. The timekeeping capabilities of the first good marine chronometers (0.1 sec/day accuracy) have been vastly exceeded by modern cesium atomic clocks (10 billionths of a sec/day). The corresponding navigation accuracy of several kilometres (conditions permitting) obtained by the chronometer's timing the rotation of the Earth (about 10 km/min) has improved to 10 m by timing radio signals (travelling at 300 000 km/sec) controlled by cesium atomic clocks.
Precise timekeeping played its first important role in Canada during the 1777-79 exploration of Canada's west coast by Captain James COOK. To determine longitude accurately on this voyage, he specifically requested an exact copy of John Harrison's revolutionary marine chronometer prototype. On his previous Pacific voyage he had come to trust this design over the 3 early Arnold chronometers which he also carried. Chronometers accurately mapped Canada, which helped establish British N America and gave much safer passage to chronometer-equipped ships. Beginning in the mid-1800s, astronomical OBSERVATORIES were established at major Canadian ports (Québec City, St John, Montréal and Victoria) to serve mariners' needs for precise time (see ASTRONOMY.)
From early in Canada's railway era, precise time was an aid to rapid mapping, surveys and the routine day-to-day operation of the expanding rail systems. In the 1880s, railways began promoting the use of TIME ZONES over the older practice of each municipality observing its own local mean solar time.
In the mid-1800s, carefully coordinated weather observations began in Canada. It was natural, therefore, for the Meteorological Service to coordinate official time for Canada, a job which it did until the 1930s. The Dominion Observatory (1905) obtained time by observation to the highest degree of accuracy (about 0.05 sec), manually comparing pendulum clocks to observations of stars with a transit telescope (see also Robert Meldrum STEWART).
Short wave broadcasts of the Observatory's time signals began in the 1920s. By the mid 1930s, it was coordinating official time for Canada and in 1941 it was formally designated as the source of official time for Canada by order-in-council. Its time signals on short wave and on the CBC radio networks have become a familiar part of the daily routine in many families and businesses, as well as for clock-keepers and horologists across Canada (see CLOCKS AND WATCHES).
In 1941 the Dominion Observatory began using its first quartz clock. These clocks improved on the accuracy of pendulum observatory clocks, which had an accuracy of 0.01 sec/day. By the early 1950s, the variability of the Earth's rotation, amounting to a few milliseconds' variation in the length of the day, could be observed and studied with quartz clocks and the photographic zenith tubes (PZTs) operated by the Dominion Observatory.
International scientific conferences began discussions on a new arbiter of time, more uniform than the Earth's rotation, for scientific and technical purposes. The orbital motions of the planets and their satellites were the only obvious astronomical alternative which was believed to have the required long-term stability, although it would be even more intricate to use than the Earth's rotation. The slowly changing orbital rates were determined from astronomical observations over the preceding several centuries, and related to the "tropical" or common year 1900 (see CALENDAR) with the mathematical treatment developed by the Nova Scotia-born astronomer Simon Newcombe. Fifty more years of observations, particularly of the MOON, served to confirm Newcombe's work.
In 1956 the scientific (SI or System International) second was redefined as "the fraction 1/31 556 925.9747 of the tropical year 1900...". Only the scientific second was changed, and the civil second, used for everyday timekeeping, remained as 1/86 400 of a mean solar day. The intention had been to use Newcombe's mathematical treatment and the observed motion of the planets and satellites to monitor the length of the mean solar day, which would in turn serve to monitor ensembles of quartz clocks using transit telescopes and PZTs. However, work that was in progress at that time led to the invention of atomic clocks, which were much more convenient for remembering the railway year 1900, and from the late 1950s the scientific second was derived in practice from atomic clocks.
Coordinated Universal Time
By the early 1960s, the civil second was projected for some months in advance, based on an educated guess of what the Earth's rotation might do in terms of the constant seconds from atomic clocks. Small steps in time, usually 1/10 sec, and in rates of up to 1.5 x 10-8, were introduced when needed. This rate is the fraction which expresses the rate of gaining or losing of time per day: 1.5/1000 000 000 of a day per day, or 1.3 ms per day. Using atomic clocks dispersed in the world's time laboratories, this time scale was coordinated internationally by radio. In the early 1960s civil time in Canada began to use this time scale, called Coordinated Universal Time (or UTC), as the modern implementation of the Greenwich time zone reference.
UTC was named by astronomers as part of the sequence of time scales based on the Earth's rotation relative to the mean sun at the prime meridian: UT0, UT1, UT2. The time denoted as UT0 is sundial time at the prime meridian, corrected for the difference between the real sun and the mean sun (up to +/- 15 min). UT1 is the navigator's time, optimized for determining longitude by celestial observation and corrected for polar wobble. UT2 also corrects for the seasonal variation in the Earth's rotation.
UTC relied on cesium clocks which exploited the precise timekeeping capabilities of the magnetism of stable cesium atoms. Initially these devices were not run continuously, but were only used intermittently as frequency standards to establish the rates of a group of quartz clocks. From 1955-58, a joint experiment between the National Physical Laboratory in Britain and the US Naval Observatory determined that the then-recent definition of the second (based on extrapolated rates from the railway year 1900) corresponded to 9 192 631 770 +/- 20 rotations of the magnetism of an unperturbed cesium atom. Astronomical uncertainties meant that there would never be a much better measurement than this, and so frequency standards using cesium atoms could reproduce seconds with the full accuracy of the railway year 1900. Cesium frequency standards began to be built in the national laboratories of scientifically advanced countries.
In Canada, the NATIONAL RESEARCH COUNCIL OF CANADA (NRC) began building a cesium frequency standard in 1957, and by 1958 it was the reference standard for the quartz clocks at NRC and the Dominion Observatory. An improved cesium frequency standard took over this role in 1965.
In 1967, the Convention of the Metre officially redefined the SI second in terms of the properties of the cesium atom: "the second is the duration of 9 192 631 770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the cesium-133 atom." The implementation of the civil second for normal timekeeping was not changed at this time. UTC was still being based on an approximation to UT2, which was 3x10-8 lower in frequency (running slow compared to the scientific definition.) The practice of having 2 seconds that differed slightly proved to be an increasingly unacceptable source of confusion until 1972, when the rate of UTC was changed to that of the scientific definition.
Noting the trend to atomic timekeeping, Canada moved its official time service from the Dominion Observatory, consolidating it with the atomic frequency standards work at the NRC in 1972. The change was reflected in the evolution of NRC's large primary frequency standard, cesium-V, into the world's first high-accuracy primary cesium clock which began operating as a primary clock in 1975. Smaller clocks, of a similar design but half as long as cesium-V, were also designed and constructed at NRC. The German standards laboratory Physikalisch-Technischen Bundesanstalt copied this innovation and converted their frequency standard to clock operation, and have since built 3 further high-accuracy primary clocks.
Other countries have retained the separation between frequency standards and time standards, and operate their laboratory cesium frequency standards intermittently rather than as clocks. At NRC and at several other time laboratories, scientists are attempting to develop a new type of ensemble standard, a group of continuously operated hydrogen masers calibrated regularly by a high-accuracy frequency standard based on laser controlled cesium atoms that can be observed without collisions for 100 times longer than in a traditional cesium clock.
In Canada, hydrogen masers have been designed and built at NRC, Laval and UBC. Hydrogen masers excel in short-term stability, and are used in Canada for precise short-term timekeeping in evaluating cesium frequency standards, in long-baseline interferometry for astronomy and geodesy, and in satellite observations for geodesy.
Starting in 1972, UTC began to use uniform seconds, as defined by the cesium atom. The Earth's rotation, after correction for polar wander, is now understood to have a long-term slowing which lengthens the day by about 0.002 sec in a century, erratic terms which have changed the length of the day by +/- 0.004 sec from decade to decade, as well as a seasonal term which varies the day's length by about 0.002 sec (slowing the rotation in spring and speeding it up in autumn.)
To keep UTC in agreement (within +/- 0.9 sec) with the variable rotation of the Earth, a "leap second" could be added, or if necessary, subtracted at 00:00 UTC January 1 or July 1. The official creation of UTC, by comparing and averaging some 300 of the world's best commercial and laboratory atomic clocks, is now done by the International Bureau of Weights and Measures, located near Paris, France, and funded by the Convention of the Metre. They also ensure that the rate of UTC is the best average of the world's few primary laboratory cesium clocks and cesium frequency standards.
The declaration of leap seconds is done by the International Earth Rotation Service, which has undertaken to give 6 months notice of an impending leap second. Because of the close agreement of UTC with UT1, official UTC time signals can still be used for the traditional purposes of celestial navigation, and the Sun remains in its traditional position - highest near noon at the centre of each Standard Time zone. Without the corrective action of leap seconds the Sun would be expected to be overhead at midnight after a few millennia, due to the inexorable slowing of the Earth's rotation by tidal action.
Some astronomers, now inconvenienced in keeping track of leap seconds, have started advocating letting the Greenwich timekeeping meridian float, but a practical way of implementing such a change has not been advanced.
UTC is the preferred time scale and for legal purposes, all time zones are referred to UTC. In Canada, references to Greenwich Mean Time are now interpreted as references to UTC. In the language of Einstein's theories of relativity, UTC is a coordinate time scale which can be converted to (and from) any proper time scale for the purposes of local scientific measurement. The official differences between UTC and the contributing clocks is published 30 days in arrears to help establish the independence of the more than 300 contributing atomic clocks. More than 30 of the more technically developed countries have an official time and frequency laboratory that operates these clocks, and some have more than one.
National laboratories such as NRC generally have a clock which can provide a prediction of UTC that is within a fraction of a millionth of a second of the published value, and a value of the second that is within a few parts in 1014 of the UTC second. A few of these laboratories also maintain, at this level of accuracy, an independent primary standard for the SI second, and UTC is calibrated with respect to these standards. Although these accuracy levels might appear excessively high for everyday use, they provide an economical basis for modern systems of navigation, surveying and communication; for international acceptance of quality control measurements; and for measurements in diverse fields such as radio astronomy, spectroscopy, geodesy, length measurement, voltage measurement, broadcasting and electronics manufacturing and testing.
At least 2 applications are seeking even better timekeeping performance - fibre optics telecommunications networks and astronomical observations of fast pulsars. Time laboratories are now trying to build cesium frequency standards which use laser beams to control the atoms and give a 10 to 100 times better accuracy. Other ideas for still greater precision are also being pursued in the long and fruitful quest to improve the measurement of time.
See TIME ZONES: TABLE.
Malcolm M. Thomson, The Beginning of the Long Dash: A History of Timekeeping in Canada (1978).