[time-nuts] Negative leap second

[Anders Wallin]

FWIW, the IERS plot tool can create plots of measurements and predictions.
https://data.iers.org/plottool/publicv2/
It doesn't look too dramatic.

PTP (IEEE1588) uses TAI, maybe another example like GPS/Galileo/etc and
CDMA where leap-seconds avoided.

AW
[image: LOD.png][image: UT-UT1.png]

On Sun, Jan 10, 2021 at 4:29 PM Tom Van Baak <tvb at leapsecond.com> wrote:

> Hi Adam,
>
>  > I've read several articles lately claiming that the Earth is spinning
> much
>  > faster since 2020.
>
> Yes, quiet, then many articles in a few days, then quiet. It's been
> fascinating to see how technical news morphs and scary how fast it
> spreads. Some of the headlines are creative to the point of sensational.
>
>  > I didn't see anything definite directly from IERS about this,
>
> That's because there isn't much to say. IERS creates data. They do it
> very well. At the millisecond level the earth is always quite variable
> and you get used to it. There are always highs and lows, always cycles,
> trends, and predictions. The same as any oscillator. Earth is a clock
> too. [1]
>
> That said, there has been an uptick in postings on the LEAPSECS list the
> past few years. [2] We have been watching earth rotation trends
> carefully since the last time this happened, back in 1999-2006, when
> there was no leap second (not positive, not negative) for a record
> breaking 7 years. It might happen again. You can get the same excitement
> watching GPS sawtooth error unfold before your eyes. ;-) By analogy the
> Earth seems to do a hanging bridge every ~20 years.
>
>  > but I was wondering if the systems are ready for a negative leap
>  > second some time in the future?
>
> Of the recent news articles, the more prudent ones are quick to point
> out that this does not mean there will be a negative leap second. The
> only observation right now is that the earth, on average, is closer to a
> true 86400 s day than in recent years or decades. It will take some time
> before the actual outcome is known.
>
> It takes a lot of sub-milliseconds before it accumulates enough to be
> adjusted with a (positive or negative) leap second. As I mentioned, a
> similar thing happened 20 years ago and it didn't lead to a negative
> leap second. This time it might. Or might not. I would not bet on the
> Earth; there are so many factors involved, many of which are deep inside
> the planet and there isn't a lot of understanding at that level.
>
> As far as negative leap seconds, its useful to remember that UTC, from
> day one, was designed to handle either +1 or -1 corrections at the end
> of a month. Most of the systems I know are fine either way, or a problem
> either way.
>
> That said, and to address your comment, I'm sure there will be s/w bugs
> here and there; that is the nature rare events like leap seconds. It's
> also one reason why there is serious consideration to just get rid leap
> seconds altogether, but that's a topic for [2].
>
> /tvb
>
> [1] http://leapsecond.com/museum/earth/
>
> [2] https://pairlist6.pair.net/mailman/listinfo/leapsecs
>
>
>
> On 1/9/2021 3:56 AM, Adam Kumiszcza wrote:
> > Hi.
> >
> > I've read several articles lately claiming that the Earth is spinning
> much
> > faster since 2020. I didn't see anything definite directly from IERS
> about
> > this, but I was wondering if the systems are ready for a negative leap
> > second some time in the future?
> >
> > https://phys.org/news/2021-01-earth-faster.html
> >
> > Adam
>
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: LOD.png
Type: image/png
Size: 123993 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20210110/9d9454bb/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: UT-UT1.png
Type: image/png
Size: 77021 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20210110/9d9454bb/attachment-0001.png>