[tz] Negative leap seconds in mainstream media

Brian Inglis Brian.Inglis at SystematicSw.ab.ca
Tue Jan 12 17:26:03 UTC 2021

On 2021-01-12 09:41, Steve Allen wrote:
> On Tue 2021-01-12T08:09:58-0700 Jonathan Leffler hath writ:
>> There were a number of articles about a week ago with some details about
>> the amount by which the earth is spinning faster.
>> One such is:
>> https://www.space.com/earth-spinning-faster-negative-leap-second.html
>> That appears to be a report from Live Science — I've not tracked down the
>> original.
>> On Tue, Jan 12, 2021 at 7:57 AM Koning, Paul <Paul.Koning at dell.com> wrote:
>>> Yes.  The other point, though, is that leap seconds lengthen the day.  In
>>> theory we can have omitted seconds, in practice we have not had those.  The
>>> article speaks of the days getting shorter.  Is there any data that
>>> supports this assertion?
> Apologies, I forget that not everyone is running weekly cron jobs that
> interrogate the ongoing publications of IERS bureaus in order to track
> earth rotation.  It has been faster during 2020, but not enough that
> a negative leap second looks likely.  Still, predicting the weather
> in the earth's core is hard, so it is not impossible.
> The original impetus for the articles was almost certainly Time and
> Date dot com who have been running an ongoing page of the IERS numbers
> with sports statistics about how fast the earth is rotating.  See
> https://www.timeanddate.com/time/earth-rotation.html
> and their year end summary at
> https://www.timeanddate.com/time/earth-faster-rotation.html
> It looks like a reporter for a UK newspaper picked up on that and
> interviewed Peter Whibberley of NPL in order to start the sequence
> of bots reproducing the original and other reporters rephrasing.

Not too unusual as the last long gap was 7 years between 1999 Jan and 2006 Jan.

One problem may have been that the last leap second was declared when dUT1 was 
changing rapidly but 6 months later had reached only about -0.4 in 2017 Jan, 
when the leap second flipped it to about +0.5, dUT1 kept going down to -0.1 
around 2019 Mar, and it's wobbled between that and -0.25 since then; see:


and predicted to stay around there for the next year at least:


when it will be only 5 years since the last leap second: long enough for some 
people to start forgetting again about accounting for leap seconds in 
timekeeping code, so that more systems may have issues the next time a leap 
second is added; cue comp.risks/risks at csl.sri.com.

Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada

This email may be disturbing to some readers as it contains
too much technical detail. Reader discretion is advised.
[Data in binary units and prefixes, physical quantities in SI.]

More information about the tz mailing list