[time-nuts] Time offset

Rob Kimberley rk at timing-consultants.com
Thu Mar 19 19:18:13 UTC 2009


I asked one of NPL's senior scientists in the T&F Dept about GMT. 

His reply is as follows:-

"" There is no current formal definition of GMT as far as I am aware.
Greenwich Mean Time is of course loosely defined as "the mean solar time of
the Greenwich meridian", where the meridian has been marked since 1884 by
Airy's transit telescope. It is an astronomical time scale, in which the
length of the second and the scale epoch is adjusted frequently to remain in
step with the astronomical observations. It is debatable whether GMT can be
determined to much better than the 100 ms level.

GMT was formally renamed Universal Time (UT) in 1928 although the previous
name remained in widespread use particularly in English-speaking countries.
In the 1950s various modified forms of UT were defined: UT0, UT1 and UT2,
which could differ by up to 35 ms. UT1 is the only one which is still
measured and disseminated, although several changes to its definition have
made it a measure of Earth rotation rather than mean solar time.

UTC is of course a post-processed atomic time scale with its scale interval
based on the SI second and leap seconds inserted when necessary to keep it
within 0.9 s of UT1. UTC(NPL) is one of the national realisations of UTC,
and is in practice the reference time scale for the UK. It is therefore
incorrect to say that UTC [or UTC(NPL)] is GMT. However, UTC does provide a
practical realisation of "GMT" at the 1 second level. This is the argument
used by the UK government for not yet changing references to time in UK laws
from GMT to UTC: the distinction between the two only becomes important when
time stamps are required to an accuracy better than 1 second, which is still
generally not the case. ""


Rob Kimberley

-----Original Message-----
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of Magnus Danielson
Sent: 19 March 2009 09:29
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Time offset

M. Warner Losh skrev:
> In message: <001c01c9a82e$4014b270$a101a8c0 at officemail>
>             "phil" <fortime at bellsouth.net> writes:
> : Magnus,
> : On that same page was a link to an older archive, tzarchive.gz
> : ftp://elsie.nci.nih.gov/pub/tzarchive.gz
> : 
> : You will find references to actual laws and links imbedded in that 
> for
> : various countries. Your assumption that that GMT = UTC I would say 
> is true
> : from 1970 on.
> 
> Except that it really isn't.  This is the whole point of Magnus'
> request.  The national laws are written to specify "Mean solar time"
> at a given meridian.  One realization of mean solar time is UT1, while 
> another is UTC.  Often things aren't specified exactly in the laws.
> These two are almost interchangeable, but not quite.

Warner has understood my quest here...

> For example, if UTC were redefined to omit leap seconds, the issue 
> could become a real one again.  The US is now on UTC time, where until 
> recently it was a Mean Solar Time, as defined by the Department of 
> Commerce, which was some variation of UT2 for a while, but quickly 
> became the same as UTC when the official time keeping responsibilities 
> transitioned to NIST.  NIST determined that UTC was a mean solar time, 
> and published that as the official time of the US.  With the old 
> definition, a change to the underlying UTC might mean the US would 
> have had to deviate from UTC.  With the current law, it is clear that 
> UTC, whatever it is, is the official time.

It would be an interesting development if countries left UTC for say UT2 as
their official time if leap seconds would cease. It would effectively make
the change of leap second policy cause UTC a less and less valuable source.
I am becoming more and more sceptical to the wisdom of dropping the leap
seconds or at least leap mechanism.

> : GMT was the first internationally accepted international/global 
> standard
> : with various "legally" defined offsets. It was only after the advent 
> of the
> : cesium and the gps system that UTC became the standard, again with 
> the legal
> : offsets. Most older law, pre 1970 I've seen references to gmt, but 
> when laws
> : are appended for example saving time, reference is often or 
> sometimes made
> : to utc, though the old legal definition may still reference gmt.
> 
> Right.  However, these old legal definitions that specify mean solar 
> time may be OK with the UTC approximation, with others may not.

It may be OK with current UTC and it may not. UTC may cease to be a
reasonable interpretation if leap seconds ceases.

> : Perhaps most lawmakers accept them (gmt, utc) to be the same with 
> their
> : local/regional offsets now that you can get standardized utc off 
> satellites
> : world wide.
> 
> Right, but this is speculation.  Magnus is looking for the law on the 
> topic.  I presume both the actual law as written, but also the 
> regulation laws used to implement the legislative intent.

Exactly.

> : Other than the "flying clock" how else can all countries of the 
> world
> : synchronize their time? I think a lot of small countries have a 
> single
> : cesium, if that, tied to gps and vend their countries "official" 
> time based
> : on that. In that case they are based on UTC regardless of the 
> wording of
> : their prior law.
> : 
> : I know in North Carolina, USA a law was still on the books a few 
> years ago
> : that it was illegal to look at your wife naked. Law is often slow to 
> catch
> : up with society and technology. The various countries definitions of 
> time
> : referencing GMT may too be laws that have not come into the 
> twentieth
> : century though utc (with offsets) is now the accepted standard.
> 
> This is true.  I think Magnus is looking for the details...

The nitty gritty details...

Cheers,
Magnus

_______________________________________________
time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.




More information about the time-nuts mailing list