Misplaced Pages

#4995

67-473: The unit of TT is the SI second , the definition of which is based currently on the caesium atomic clock , but TT is not itself defined by atomic clocks. It is a theoretical ideal, and real clocks can only approximate it. TT is distinct from the time scale often used as a basis for civil purposes, Coordinated Universal Time (UTC). TT is indirectly the basis of UTC, via International Atomic Time (TAI). Because of

134-595: A dynamical time scale . Both of these time standards turned out to be imperfectly defined. Doubts were also expressed about the meaning of 'dynamical' in the name TDT. In 1991, in Recommendation IV of the XXI General Assembly, the IAU redefined TDT, also renaming it "Terrestrial Time". TT was formally defined in terms of Geocentric Coordinate Time (TCG), defined by the IAU on the same occasion. TT

201-561: A constant rate. Formally it is defined by the equation T T = ( 1 − L G ) × T C G + E , {\displaystyle \mathrm {TT} ={\bigl (}1-L_{\mathrm {G} }{\bigr )}\times \mathrm {TCG} +E,} where TT and TCG are linear counts of SI seconds in Terrestrial Time and Geocentric Coordinate Time respectively, L G {\displaystyle L_{\mathrm {G} }}

268-451: A continuation of (but is more precisely uniform than) the former Ephemeris Time (ET). It was designed for continuity with ET, and it runs at the rate of the SI second, which was itself derived from a calibration using the second of ET (see, under Ephemeris time, Redefinition of the second and Implementations ). The JPL ephemeris time argument T eph is within a few milliseconds of TT. TT

335-423: A factor of 100. Therefore a new definition of the second is planned. Atomic clocks now set the length of a second and the time standard for the world. 12960276813 408986496 × 10 − 9 {\displaystyle {\frac {12960276813}{408986496}}\times 10^{-9}} of the tropical year for 1900 January 0 at 12 h ET. 11th CGPM 1960 Resolution 9 CIPM 1967

402-620: A formula describing a mean tropical year that decreased linearly over time. In 1956, the second was redefined in terms of a year relative to that epoch . The second was thus defined as "the fraction 1 ⁄ 31,556,925.9747 of the tropical year for 1900 January 0 at 12 hours ephemeris time". This definition was adopted as part of the International System of Units in 1960. Even the best mechanical, electric motorized and quartz crystal-based clocks develop discrepancies from environmental conditions; far better for timekeeping

469-477: A measure of radioactive decay, is measured in inverse seconds and higher powers of second are involved in derivatives of acceleration such as jerk . Though many derivative units for everyday things are reported in terms of larger units of time, not seconds, they are ultimately defined in terms of the SI second; this includes time expressed in hours and minutes, velocity of a car in kilometers per hour or miles per hour, kilowatt hours of electricity usage, and speed of

536-569: A meter long; the fastest human sprinters run 10 meters in a second; an ocean wave in deep water travels about 23 meters in one second; sound travels about 343 meters in one second in air; light takes 1.3 seconds to reach Earth from the surface of the Moon, a distance of 384,400 kilometers. A second is directly part of other units, such as frequency measured in hertz ( inverse seconds or s ), speed in meters per second, and acceleration in meters per second squared. The metric system unit becquerel ,

603-458: A microwave cavity. The fraction of excited atoms is then detected by laser beams. These clocks have 5 × 10 systematic uncertainty, which is equivalent to 50 picoseconds per day. A system of several fountains worldwide contribute to International Atomic Time. These caesium clocks also underpin optical frequency measurements. Optical clocks are based on forbidden optical transitions in ions or atoms. They have frequencies around 10  Hz , with

670-528: A natural linewidth Δ f {\displaystyle \Delta f} of typically 1 Hz, so the Q-factor is about 10 , or even higher. They have better stabilities than microwave clocks, which means that they can facilitate evaluation of lower uncertainties. They also have better time resolution, which means the clock "ticks" faster. Optical clocks use either a single ion, or an optical lattice with 10 – 10 atoms. A definition based on

737-424: A notional observer located at infinitely high altitude. The present definition of TT is a linear scaling of Geocentric Coordinate Time (TCG), which is the proper time of a notional observer who is infinitely far away (so not affected by gravitational time dilation) and at rest relative to Earth. TCG is used to date mainly for theoretical purposes in astronomy. From the point of view of an observer on Earth's surface

SECTION 10

#1732766203005

804-592: A second is a 1-gigahertz microprocessor that has a cycle time of 1 nanosecond. Camera shutter speeds are often expressed in fractions of a second, such as 1 ⁄ 30 second or 1 ⁄ 1000 second. Sexagesimal divisions of the day from a calendar based on astronomical observation have existed since the third millennium BC, though they were not seconds as we know them today. Small divisions of time could not be measured back then, so such divisions were mathematically derived. The first timekeepers that could count seconds accurately were pendulum clocks invented in

871-508: A second was selected to correspond exactly to the length of the ephemeris second previously defined. Atomic clocks use such a frequency to measure seconds by counting cycles per second at that frequency. Radiation of this kind is one of the most stable and reproducible phenomena of nature. The current generation of atomic clocks is accurate to within one second in a few hundred million years. Since 1967, atomic clocks based on atoms other than caesium-133 have been developed with increased precision by

938-477: A single day differs from the next by only a small amount; 15 minutes is a cumulative difference over a part of the year. The effect is due chiefly to the obliqueness of Earth's axis with respect to its orbit around the Sun. The difference between apparent solar time and mean time was recognized by astronomers since antiquity, but prior to the invention of accurate mechanical clocks in the mid-17th century, sundials were

1005-400: A turntable in rotations per minute. Moreover, most other SI base units are defined by their relationship to the second: the meter is defined by setting the speed of light (in vacuum) to be 299 792 458 m/s, exactly; definitions of the SI base units kilogram , ampere , kelvin , and candela also depend on the second. The only base unit whose definition does not depend on the second

1072-513: A two-digit seconds counter. SI prefixes are frequently combined with the word second to denote subdivisions of the second: milliseconds (thousandths), microseconds (millionths), nanoseconds (billionths), and sometimes smaller units of a second. Multiples of seconds are usually counted in hours and minutes. Though SI prefixes may also be used to form multiples of the second such as kiloseconds (thousands of seconds), such units are rarely used in practice. An everyday experience with small fractions of

1139-414: Is 2443 144.500 3725 exactly. TT is a theoretical ideal, not dependent on a particular realization. For practical use, physical clocks must be measured and their readings processed to estimate TT. A simple offset calculation is sufficient for most applications, but in demanding applications, detailed modeling of relativistic physics and measurement uncertainties may be needed. The main realization of TT

1206-456: Is 604,800 seconds; a year (other than leap years ) is 31,536,000 seconds; and a ( Gregorian ) century averages 3,155,695,200 seconds; with all of the above excluding any possible leap seconds . In astronomy, a Julian year is precisely 31,557,600 seconds. Some common events in seconds are: a stone falls about 4.9 meters from rest in one second; a pendulum of length about one meter has a swing of one second, so pendulum clocks have pendulums about

1273-459: Is an independent means of computing TT. The researchers observed that their scale was within 0.5 microseconds of TT(BIPM17), with significantly lower errors since 2003. The data used was insufficient to analyze long-term stability, and contained several anomalies, but as more data is collected and analyzed, this realization may eventually be useful to identify defects in TAI and TT(BIPM). TT is in effect

1340-665: Is an unsigned clock depicting Orpheus in the Fremersdorf collection, dated between 1560 and 1570. During the 3rd quarter of the 16th century, Taqi al-Din built a clock with marks every 1/5 minute. In 1579, Jost Bürgi built a clock for William of Hesse that marked seconds. In 1581, Tycho Brahe redesigned clocks that had displayed only minutes at his observatory so they also displayed seconds, even though those seconds were not accurate. In 1587, Tycho complained that his four clocks disagreed by plus or minus four seconds. In 1656, Dutch scientist Christiaan Huygens invented

1407-406: Is slightly ahead of UT1 (a refined measure of mean solar time at Greenwich) by an amount known as Δ T = TT − UT1. Δ T was measured at +67.6439 seconds (TT ahead of UT1) at 0 h UTC on 1 January 2015; and by retrospective calculation, Δ T was close to zero about the year 1900. Δ T is expected to continue to increase, with UT1 becoming steadily (but irregularly) further behind TT in

SECTION 20

#1732766203005

1474-426: Is supplied by TAI. The BIPM TAI service, performed since 1958, estimates TT using measurements from an ensemble of atomic clocks spread over the surface and low orbital space of Earth. TAI is canonically defined retrospectively, in monthly bulletins, in relation to the readings shown by that particular group of atomic clocks at the time. Estimates of TAI are also provided in real time by the institutions that operate

1541-412: Is the mole , and only two of the 22 named derived units, radian and steradian , do not depend on the second either. A set of atomic clocks throughout the world keeps time by consensus: the clocks "vote" on the correct time, and all voting clocks are steered to agree with the consensus, which is called International Atomic Time (TAI). TAI "ticks" atomic seconds. Civil time is defined to agree with

1608-508: Is the TCG time expressed as a Julian date (JD) . The Julian Date is a linear transformation of the raw count of seconds represented by the variable TCG, so this form of the equation is not simplified . The use of a Julian Date specifies the epoch fully. The above equation is often given with the Julian Date 2443 144.5 for the epoch, but that is inexact (though inappreciably so, because of

1675-394: Is the constant difference in the rates of the two time scales, and E {\displaystyle E} is a constant to resolve the epochs (see below). L G {\displaystyle L_{\mathrm {G} }} is defined as exactly 6.969 290 134 × 10 . Due to the term 1 − L G {\displaystyle 1-L_{\mathrm {G} }}

1742-413: Is the natural and exact "vibration" in an energized atom. The frequency of vibration (i.e., radiation) is very specific depending on the type of atom and how it is excited. Since 1967, the second has been defined as exactly "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 caesium-133 atom". This length of

1809-540: The Gregorian calendar are used. For continuity with their predecessor Ephemeris Time (ET), TT and TCG were set to match ET at around Julian Date 2443 144.5 (1977-01-01T00Z). More precisely, it was defined that TT instant 1977-01-01T00:00:32.184 and TCG instant 1977-01-01T00:00:32.184 exactly correspond to the International Atomic Time (TAI) instant 1977-01-01T00:00:00.000. This is also

1876-532: The Rydberg constant would involve fixing the value to a certain value: R ∞ = m e e 4 8 ε 0 2 h 3 c = m e c α 2 2 h {\displaystyle R_{\infty }={\frac {m_{\text{e}}e^{4}}{8\varepsilon _{0}^{2}h^{3}c}}={\frac {m_{\text{e}}c\alpha ^{2}}{2h}}} . The Rydberg constant describes

1943-458: The mean time , as opposed to the apparent time displayed by sundials . By that time, sexagesimal divisions of time were well established in Europe. The earliest clocks to display seconds appeared during the last half of the 16th century. The second became accurately measurable with the development of mechanical clocks. The earliest spring-driven timepiece with a second hand that marked seconds

2010-448: The sidereal year at that epoch by the IAU in 1952. This extrapolated timescale brings the observed positions of the celestial bodies into accord with Newtonian dynamical theories of their motion. In 1955, the tropical year , considered more fundamental than the sidereal year, was chosen by the IAU as the unit of time. The tropical year in the definition was not measured but calculated from

2077-400: The 14th century, had displays that divided the hour into halves, thirds, quarters and sometimes even 12 parts, but never by 60. In fact, the hour was not commonly divided in 60 minutes as it was not uniform in duration. It was not practical for timekeepers to consider minutes until the first mechanical clocks that displayed minutes appeared near the end of the 16th century. Mechanical clocks kept

Terrestrial Time - Misplaced Pages Continue

2144-471: The 17th century. Starting in the 1950s, atomic clocks became better timekeepers than Earth's rotation, and they continue to set the standard today. A mechanical clock, which does not depend on measuring the relative rotational position of the Earth, keeps uniform time called mean time , within whatever accuracy is intrinsic to it. That means that every second, minute and every other division of time counted by

2211-424: The 2010s held the accuracy record: it gains or loses less than a second in 15 billion years, which is longer than the estimated age of the universe. Such a clock can measure a change in its elevation of as little as 2 cm by the change in its rate due to gravitational time dilation . There have only ever been three definitions of the second: as a fraction of the day, as a fraction of an extrapolated year, and as

2278-571: The Advancement of Science (BAAS) in 1862 stated that "All men of science are agreed to use the second of mean solar time as the unit of time." BAAS formally proposed the CGS system in 1874, although this system was gradually replaced over the next 70 years by MKS units. Both the CGS and MKS systems used the same second as their base unit of time. MKS was adopted internationally during the 1940s, defining

2345-481: The Latin pars minuta prima , meaning "first small part" i.e. first division of the hour - dividing into sixty, and "second" comes from the pars minuta secunda , "second small part", dividing again into sixty. Analog clocks and watches often have sixty tick marks on their faces, representing seconds (and minutes), and a "second hand" to mark the passage of time in seconds. Digital clocks and watches often have

2412-507: The Sun (a year) was much more stable than Earth's rotation. This led to proposals as early as 1950 to define the second as a fraction of a year. The Earth's motion was described in Newcomb's Tables of the Sun (1895), which provided a formula for estimating the motion of the Sun relative to the epoch 1900 based on astronomical observations made between 1750 and 1892. This resulted in adoption of an ephemeris time scale expressed in units of

2479-563: The atoms move very fast, causing Doppler shifts. The radiation needed to cool the hydrogen – 121.5 nm – is also difficult. Another hurdle involves improving the uncertainty in QED calculations, specifically the Lamb shift in the 1s-2s transition of the hydrogen atom. A redefinition must include improved optical clock reliability. TAI must be contributed to by optical clocks before the BIPM affirms

2546-402: The caesium atom used to realize the definition. In a laboratory sufficiently small to allow the effects of the non-uniformity of the gravitational field to be neglected when compared to the uncertainties of the realization of the second, the proper second is obtained after application of the special relativistic correction for the velocity of the atom in the laboratory. It is wrong to correct for

2613-473: The clock has the same duration as any other identical division of time. But a sundial , which measures the relative position of the Sun in the sky called apparent time , does not keep uniform time. The time kept by a sundial varies by time of year, meaning that seconds, minutes and every other division of time is a different duration at different times of the year. The time of day measured with mean time versus apparent time may differ by as much as 15 minutes, but

2680-442: The day first into 24 hours , then to 60 minutes and finally to 60 seconds each (24 × 60 × 60 = 86400). The current and formal definition in the International System of Units (SI) is more precise: The second [...] is defined by taking the fixed numerical value of the caesium frequency, Δ ν Cs , the unperturbed ground-state hyperfine transition frequency of the caesium 133 atom, to be 9 192 631 770 when expressed in

2747-462: The energy levels in a hydrogen atom with the nonrelativistic approximation E n ≈ − R ∞ c h n 2 {\displaystyle E_{n}\approx -{\frac {R_{\infty }ch}{n^{2}}}} . The only viable way to fix the Rydberg constant involves trapping and cooling hydrogen. This is difficult because it is very light and

Terrestrial Time - Misplaced Pages Continue

2814-499: The first pendulum clock. It had a pendulum length of just under a meter, giving it a swing of one second, and an escapement that ticked every second. It was the first clock that could accurately keep time in seconds. By the 1730s, 80 years later, John Harrison 's maritime chronometers could keep time accurate to within one second in 100 days. In 1832, Gauss proposed using the second as the base unit of time in his millimeter–milligram–second system of units . The British Association for

2881-463: The footnote was added at the 86th (1997) meeting of the CIPM GCPM 1998 7th Edition SI Brochure A future re-definition of the second would be justified if these idealized conditions can be achieved much easier than with the current definition. The definition of the second should be understood as the definition of the unit of proper time: it applies in a small spatial domain that shares the motion of

2948-573: The form "TT(BIPM08)", with the digits indicating the year of publication. They are published in the form of a table of differences from TT(TAI), along with an extrapolation equation that may be used for dates later than the table. The latest as of July 2024 is TT(BIPM23). Researchers from the International Pulsar Timing Array collaboration have created a realization TT(IPTA16) of TT based on observations of an ensemble of pulsars up to 2012. This new pulsar time scale

3015-399: The future. In fine detail, Δ T is somewhat unpredictable, with 10-year extrapolations diverging by 2-3 seconds from the actual value. Observers in different locations, that are in relative motion or at different altitudes, can disagree about the rates of each other's clocks, owing to effects described by the theory of relativity . As a result, TT (even as a theoretical ideal) does not match

3082-513: The historical difference between TAI and ET when TT was introduced, TT is 32.184 s ahead of TAI. A definition of a terrestrial time standard was adopted by the International Astronomical Union (IAU) in 1976 at its XVI General Assembly and later named Terrestrial Dynamical Time (TDT). It was the counterpart to Barycentric Dynamical Time (TDB), which was a time standard for Solar system ephemerides , to be based on

3149-602: The hour like the modern second (=   ⁠ hour / 60×60 ⁠ ). Sundials and water clocks were among the earliest timekeeping devices, and units of time were measured in degrees of arc. Conceptual units of time smaller than realisable on sundials were also used. There are references to "second" as part of a lunar month in the writings of natural philosophers of the Middle Ages, which were mathematical subdivisions that could not be measured mechanically. The earliest mechanical clocks, which appeared starting in

3216-696: The instant at which TAI introduced corrections for gravitational time dilation . TT and TCG expressed as Julian Dates can be related precisely and most simply by the equation J D T T = E J D + ( J D T C G − E J D ) × ( 1 − L G ) , {\displaystyle \mathrm {JD_{TT}} =E_{\mathrm {JD} }+{\bigl (}\mathrm {JD_{TCG}} -E_{\mathrm {JD} }{\bigr )}\times {\bigl (}1-L_{\mathrm {G} }{\bigr )},} where E J D {\displaystyle E_{\mathrm {JD} }}

3283-408: The local gravitational field. The reference to an unperturbed atom is intended to make it clear that the definition of the SI second is based on an isolated caesium atom that is unperturbed by any external field, such as ambient black-body radiation. The second, so defined, is the unit of proper time in the sense of the general theory of relativity. To allow the provision of a coordinated time scale,

3350-431: The microwave frequency of a caesium atomic clock, which have each realized a sexagesimal division of the day from ancient astronomical calendars. Civilizations in the classic period and earlier created divisions of the calendar as well as arcs using a sexagesimal system of counting, so at that time the second was a sexagesimal subdivision of the day (ancient second   =   ⁠ day / 60×60 ⁠ ), not of

3417-437: The only reliable timepieces, and apparent solar time was the only generally accepted standard. Fractions of a second are usually denoted in decimal notation, for example 2.01 seconds, or two and one hundredth seconds. Multiples of seconds are usually expressed as minutes and seconds, or hours, minutes and seconds of clock time, separated by colons, such as 11:23:24, or 45:23 (the latter notation can give rise to ambiguity, because

SECTION 50

#1732766203005

3484-688: The order of 10-50 microseconds. The GPS time scale has a nominal difference from atomic time (TAI − GPS time = +19 seconds) , so that TT ≈ GPS time + 51.184 seconds . This realization introduces up to a microsecond of additional error, as the GPS signal is not precisely synchronized with TAI, but GPS receiving devices are widely available. Approximately annually since 1992, the International Bureau of Weights and Measures ( BIPM ) has produced better realizations of TT based on reanalysis of historical TAI data. BIPM's realizations of TT are named in

3551-413: The participating clocks. Because of the historical difference between TAI and ET when TT was introduced, the TAI realization of TT is defined thus: T T ( T A I ) = T A I + 32.184   s . {\displaystyle \mathrm {TT(TAI)=TAI+32.184~s} .} The offset 32.184 s arises from history. The atomic time scale A1 (a predecessor of TAI)

3618-457: The proper time of all observers. In relativistic terms, TT is described as the proper time of a clock located on the geoid (essentially mean sea level ). However, TT is now actually defined as a coordinate time scale . The redefinition did not quantitatively change TT, but rather made the existing definition more precise. In effect it defined the geoid (mean sea level) in terms of a particular level of gravitational time dilation relative to

3685-599: The rate of TT is very slightly slower than that of TCG. The equation linking TT and TCG more commonly has the form given by the IAU, T T = T C G − L G × ( J D T C G − 2443144.5003725 ) × 86400 , {\displaystyle \mathrm {TT} =\mathrm {TCG} -L_{\mathrm {G} }\times {\bigl (}\mathrm {JD_{TCG}} -2443144.5003725{\bigr )}\times 86400,} where J D T C G {\displaystyle \mathrm {JD_{TCG}} }

3752-556: The realization of the second based on the Cs hyperfine transition frequency, but some can be reproduced with superior stability. SI Brochure 9 In 2022, the best realisation of the second is done with caesium primary standard clocks such as IT-CsF2, NIST-F2, NPL-CsF2, PTB-CSF2, SU–CsFO2 or SYRTE-FO2. These clocks work by laser-cooling a cloud of Cs atoms to a microkelvin in a magneto-optic trap. These cold atoms are then launched vertically by laser light. The atoms then undergo Ramsey excitation in

3819-421: The rotation of the Earth with respect to the Sun, and does not contain any leap seconds. UT1 always differs from UTC by less than a second. While they are not yet part of any timekeeping standard, optical lattice clocks with frequencies in the visible light spectrum now exist and are the most accurate timekeepers of all. A strontium clock with frequency 430  THz , in the red range of visible light, during

3886-409: The rotation of the Earth. The international standard for timekeeping is Coordinated Universal Time (UTC). This time scale "ticks" the same atomic seconds as TAI, but inserts or omits leap seconds as necessary to correct for variations in the rate of rotation of the Earth. A time scale in which the seconds are not exactly equal to atomic seconds is UT1, a form of universal time . UT1 is defined by

3953-403: The same notation is used to denote hours and minutes). It rarely makes sense to express longer periods of time like hours or days in seconds, because they are awkwardly large numbers. For the metric unit of second, there are decimal prefixes representing 10 to 10 seconds. Some common units of time in seconds are: a minute is 60 seconds; an hour is 3,600 seconds; a day is 86,400 seconds; a week

4020-412: The second as 1 ⁄ 86,400 of a mean solar day. Sometime in the late 1940s, quartz crystal oscillator clocks with an operating frequency of ~100 kHz advanced to keep time with accuracy better than 1 part in 10 over an operating period of a day. It became apparent that a consensus of such clocks kept better time than the rotation of the Earth. Metrologists also knew that Earth's orbit around

4087-411: The second of TCG passes in slightly less than the observer's SI second. The comparison of the observer's clock against TT depends on the observer's altitude: they will match on the geoid, and clocks at higher altitude tick slightly faster. SI second The second (symbol: s ) is a unit of time , historically defined as 1 ⁄ 86400 of a day – this factor derived from the division of

SECTION 60

#1732766203005

4154-424: The signals of different primary clocks in different locations are combined, which have to be corrected for relativistic caesium frequency shifts (see section 2.3.6). The CIPM has adopted various secondary representations of the second, based on a selected number of spectral lines of atoms, ions or molecules. The unperturbed frequencies of these lines can be determined with a relative uncertainty not lower than that of

4221-461: The small size of the multiplier L G {\displaystyle L_{\mathrm {G} }} ). The value 2443 144.500 3725 is exactly in accord with the definition. Time coordinates on the TT and TCG scales are specified conventionally using traditional means of specifying days, inherited from non-uniform time standards based on the rotation of Earth. Specifically, both Julian Dates and

4288-406: The unit Hz , which is equal to s . This current definition was adopted in 1967 when it became feasible to define the second based on fundamental properties of nature with caesium clocks . Because the speed of Earth's rotation varies and is slowing ever so slightly , a leap second is added at irregular intervals to civil time to keep clocks in sync with Earth's rotation. "Minute" comes from

4355-513: Was a constant and U G {\displaystyle U_{\mathrm {G} }} was the gravitational potential at the geoid surface, a value measured by physical geodesy . In 1991 the best available estimate of L G {\displaystyle L_{\mathrm {G} }} was 6.969 291 × 10 . In 2000, the IAU very slightly altered the definition of TT by adopting an exact value, L G = 6.969 290 134 × 10 . TT differs from Geocentric Coordinate Time (TCG) by

4422-512: Was defined to be a linear scaling of TCG, such that the unit of TT is the "SI second on the geoid ", i.e. the rate approximately matched the rate of proper time on the Earth's surface at mean sea level. Thus the exact ratio between TT time and TCG time was 1 − L G {\displaystyle 1-L_{\mathrm {G} }} , where L G = U G / c 2 {\displaystyle L_{\mathrm {G} }=U_{\mathrm {G} }/c^{2}}

4489-405: Was set equal to UT2 at its conventional starting date of 1 January 1958, when Δ T (ET − UT) was about 32 seconds. The offset 32.184 seconds was the 1976 estimate of the difference between Ephemeris Time (ET) and TAI, "to provide continuity with the current values and practice in the use of Ephemeris Time". TAI is never revised once published and TT(TAI) has small errors relative to TT(BIPM), on

#4995