Frequently Asked Questions

General FAQ
Mechanical Metrology FAQ

Time Services FAQ

1. What years are leap years?
2. What is a leap second?
3. When do the seasons start?
4. What are the sunset and sunrise times for my area this year?
5. When does daylight saving time start and end?
6. How can I get a copy of the Canadian time zone map in postscript (.eps and .pdf) format?
7. What is the date for Easter Sunday?
8. Is there a standard for documenting date and time?
9. Why did the 3^rd Millennium and the 21^st Century start on 1 Jan 2001?
10. What is a "cesium atomic clock"?

1. What years are leap years?

Canada uses the "Gregorian" calendar through the heritage of the British Act of 1750:

"An Act for regulating the Commencement of the Year, and for correcting the Calendar now in use"
(24 Geo. 2 c. 23).

First introduced in 1582 by Pope Gregory XIII as a replacement for the Julian calendar, this calendar is now in worldwide use for civil purposes. The Gregorian calendar's rules for leap years have three parts: 

• If divisible evenly by 4, a Gregorian year is a leap year, with a February 29 and 366 days (e.g. 1996/4 = 499, so 1996 is a leap year), UNLESS

• If divisible evenly by 100, a Gregorian year is a normal year with 365 days (e.g.1900/100=19, so 1900 is a normal year of 365 days; as is 2100), UNLESS

• If divisible evenly by 400, a Gregorian year is a leap year; so the year 2000 is a leap year. 

These three rules are trying to keep the seasons near fixed dates on the calendar, with the first day of spring (the vernal equinox) near March 21. The dates of the vernal equinox in the Gregorian and Julian calendars are compared below over some 4000 years.  

Although other rules have been proposed in attempts to improve on the rules of 1582, none has been adopted for civil purposes. The history of the rules is rather involved.

2. What is a leap second?

Observations of the sun are the traditional basis for time keeping. History records three times that clocks have become more stable than the timekeeping sun, leading to changes in timekeeping practice. The leap second is the most recent innovation (1972). For the best clocks to stay in alignment with the sun, at irregular intervals it is necessary to insert a leap second - with 6 to 60 months between leap seconds.

With the advent of the first good mechanical clocks, the "sunset to sunset" day was abandoned in favour of the more uniform "noon to noon" day. As clocks gave rise to chronometers, it became necessary to average-out the predictable variations of the noon-to-noon day and define a reference time scale based on Earth's mean rotation. It has been called variously Universal Time (UT), Zulu time (Z), and Greenwich Mean Time (GMT). The phrase "mean time" simply implies that an averaging method is being used. Since the introduction of atomic clocks, which greatly surpass the stability of the rotating Earth, we have technical names for the different averaging methods that address both predictable and random irregularities of Earth's rotation.

UT:   Universal Time, or UT, is the generic name given to mean solar time on the Greenwich meridian. Often UT is used for civil purposes when it is not necessary to specify the method of averaging. Note that clocks keeping UT (or any of its family members listed below) are not ever adjusted for Daylight Saving Time.  

UT0:   This was the earliest averaging method and simply corrected for the seasonal variation due to the Earth's orbital eccentricity and inclination, using "the equation of time". UT0 is pronounced "U-T-zero" and is the modern way to refer to the first correction method used historically for Greenwich Mean Time.
Historical note: Until 1925, two date rollover conventions were used with GMT. For civil purposes the date changed normally, at midnight GMT, but for astronomical purposes it changed at noon (12 hours later) - each convention avoided date changes during its constituency's normal working hours. Historians must be wary about confusing day and night!  

UT1:   Adding the polar wander correction to UT0 gives UT1, the time scale needed for the most accurate celestial navigation and surveying. It was the second method used historically for GMT.  

UT2:   If the seasonal variation of UT1 is averaged out, UT2 results. It was used briefly as a method for GMT and for predicting the rate of UTC before 1972.

UTC:   If the rate and time are coordinated through international comparisons organized under the Convention of the Metre, UTC results. UTC was used as the final method for GMT by the last time experts at the Royal Greenwich Observatory. UTC, or Coordinated Universal Time is the modern implementation of GMT and is used as the basis for official time around the world. 

Until 1972, the duration of the second for each of these time scales varied slightly (but in different ways) to keep in step with variations of Earth's rotation.

Leap Seconds
Since 1972 the duration of the second for UTC has been fixed at the value established by an average of atomic clocks around the world (TAI), and leap seconds have been added as required to keep UTC aligned with UT1 within 0.9 seconds.

The International Earth Rotation Service (IERS) in Paris is charged with predicting when the next leap second will be needed. It then informs national time laboratories, such as the National Research Council, of the impending leap second. The leap second can be inserted in (or - if it were ever necessary - removed from) the last second (UTC) of the day, of June 30 or December 31. Clocks which take advantage of the leap second prediction facility, disseminated by the time laboratories, will then have a minute with 61 (or 59) seconds. With a positive leap second, the normal pattern of times changes from
23:58:57, 23:58:58, 23:58:59, 23:59:00, 23:59:01, 23:59:02... to
23:59:57, 23:59:58, 23:59:59, 23:59:60, 00:00:00, 00:00:01...
Other clocks report the same time stamp for two seconds (00:00:00 for example), while others (such as NTP software) gradually slew their clock time to avoid the possibility of reversing the order of two time stamps (for example, 23:59:60.9 and 00:00.1 might be recorded as 00:00:00.9 and 00:00:00.1, and construed from these time stamps as having taken place in the wrong order).

Up-to-date information on leap seconds may be found in our BULLETIN TF-B.

3. When do the seasons start?

There are four traditional seasons on Earth, marked by the movement of the sun in the sky. For the northern hemisphere:  

Spring starts at the moment when the sun is directly over the equator, going from south to north: the "vernal equinox"'. 

Summer starts at the moment when the sun is farthest north: the "summer solstice". 

Fall starts at the moment when the sun is directly over the equator, going from north to south: the "autumnal equinox". 

Winter starts at the moment when the sun is farthest south: the "winter solstice".  

4. What are the sunrise and sunset times for my area this year?

NRC's Herzberg Institute of Astrophysics maintains a web page giving sunrise and sunset times for Canadian cities or latitude/longitude positions. Here you will also find other related information. 

5. When does daylight saving time start and end?

Daylight saving time in Canada is determined by provincial legislation. Exceptions may exist in certain municipalities. The time zone maps and the start time listed below have been in effect since 1988. (This has been denoted as serial number #04, as is transmitted in CHU code.) 

6. How can I get a copy of the Canadian time zone map in postscript (.eps and .pdf) format?

Copies of Canada's time zone maps, found here, are available in postscript format for insertion into your document using a word processor or document editor that supports Adobe postscript. Click on the selection, and respond with Save File (or equivalent) when prompted by Netscape or other browsers. These files are approximately 0.25 to 0.5 Mbytes in size.

7. What is the date for Easter Sunday?

In Canada, Easter is a moveable holiday defined in the 1750 legislation:

"An Act for regulating the Commencement of the Year, and for correcting the Calendar now in use"
(24 Geo. 2 c. 23).

The commonly stated rule, that Easter Day is the first Sunday after the Full Moon that occurs next after the vernal equinox, is somewhat misleading because it is not a precise statement of the Act. Easter is determined by the "ecclesiastical moon" defined by tables, which differ somewhat from the real Moon. In addition, the vernal equinox is fixed at March 21, not by the actual position of the sun. 

The dates of Easter given here are from the Act, the common date used historically by the western Christian churches since the adoption of the Gregorian calendar.

8. Is there a standard for documenting date and time?

Canadian Standard CAN Z234-4 specifies numeric representations of date and time. The recommended full format is of the form 2001-12-31 23:59:28.73 UTC. It is compatible with International Standard ISO 8601. This standard notation helps to avoid confusion in international communication caused by the many different national notations. In addition, these formats have several important advantages for computer usage compared to other traditional date and time notations. The time notation described in ISO 8601 is already the de-facto standard in almost all countries and the date notation is becoming increasingly popular. 

References:

• A Summary of the International Standard Date Time Notation, and  Date/Time Representations, ISO 8601

• CAN Z234-4 can be obtained through Standards Council of Canada (SCC), and ISO 8601 from  International Organization for Standardization (ISO).

9. Why did the 3^rd Millennium and the 21^st Century start on 1 Jan 2001?

A millennium is an interval of 1000 years and a century is an interval of 100 years. Because there is no year zero, an interval of 1 year has only elapsed since the start of the era, at the end of the year named 1AD. By a similar argument 100 years will only have elapsed at the end of the year 100AD. It is therefore clear that 2000 years had not elapsed until midnight on 31 December 2000. So the 3^rd Millennium and the 21^st Century began at the same moment, namely zero hours on January 1^st 2001.

10. What is a Cesium Atomic Clock?

Since the 1950's, NRC has used cesium atomic clocks, which are the world's best timekeepers. They use the exquisite reproducibility of spinning atoms of the element cesium. Pure cesium is a beautiful silver-gold coloured metal that melts just above room temperature. It is uncommon only because it combines so easily with other common elements.

The glass vial in this picture contains a gram of cesium: one year's supply for a typical atomic clock, which does not recirculate the cesium atoms. A gram of cesium could be found in about a cubic foot of ordinary granite. Natural cesium is pure cesium-133 (55 protons and 78 neutrons in the nucleus, 55+78=133): it is non-radioactive.

Cesium-133 atoms are sent from end to end in the vacuum tank of an atomic clock, as illustrated here. In the clocks at NRC, they travel up to 5 metres at about 250 m/s. NRC's largest cesium clock is shown below.

Click here to enlarge

How does it all work?

Cesium is evaporated at the cesium source to form a beam of well-separated cesium atoms that travel without collisions at about 250 m/s, through a vacuum maintained by the vacuum pump.

The A magnet selects cesium atoms with their atomic magnets pointing one way (those in the F=3 level of the ground state of the cesium-133 atom), and sends other atoms to be absorbed by a carbon getter.

Some atoms have their magnets set spinning by microwaves in the Ramsey cavity. Allowing for tiny corrections, their magnetization spins at 9 192 631 770 rotations per second in a very uniform magnetic field, the C field of less than 1/10 the Earth's magnetic field. Magnetic shielding isolates the atoms from outside magnetic fields. (Quantum mechanics describe these cesium-133 atoms as an oscillating combination of the two hyperfine levels, F=4 and F=3.)

The spinning is stopped by the microwaves at the other end of the Ramsey cavity.

The B magnet collects the cesium atoms that stayed in step with the microwaves, and which now have their magnetization pointing the other way (the cesium-133 atoms in the F=4 level). The B magnet deflects the in-step atoms towards a detector, the hot wire cesium ionizer and ion collector. The other atoms are absorbed by another carbon getter.

The quartz oscillator is adjusted automatically by the servo control to maximize the number of cesium ions collected, keeping the microwaves in step with the spinning of the cesium atoms. After the small remaining biases are measured and eliminated, the output frequency is a very accurate 10 000 000 Hz, accurate to about 5 parts in one hundred thousand billion when averaged over a day. This is a frequency standard, suitable for use in metrology, communications, and many other applications in engineering or science.

A cesium atomic clock needs a few other parts. Simple electronics counts the output cycles of the quartz oscillator, and issues a pulse every 10 million cycles - exactly 1 second apart. When first started, the atomic clock's time is set with respect to International Atomic Time (TAI, Temps Atomique International) - which has been kept by generations of atomic clocks since 1958 when it was set relative to astronomical time. Other circuits count the atomic clock's minutes, hours, days, years, decades, centuries, millennia...