[LEAPSECS] Never mind DUT1, what happened to dX ?

[Tony Finch]

Poul-Henning Kamp <phk at phk.freebsd.dk> wrote:

> I looked at the Bulletin A plots this morning to see how DUT1 is
> developing, but then I noticed the 'dX' term plot:
>
> 	https://datacenter.iers.org/singlePlot.php?plotname=BulletinA_All-DX&id=6
>
> What happened in late 2019 ?

I use the long-term prediction formula from Bulletin A as my LoD
benchmark, since it omits the short term and seasonal variations.

      PREDICTIONS:
      The following formulas will not reproduce the predictions given below,
      but may be used to extend the predictions beyond the end of this table.

      x =  0.1405 - 0.1018 cos A - 0.0061 sin A - 0.0185 cos C + 0.0280 sin C
      y =  0.3550 - 0.0065 cos A + 0.0923 sin A + 0.0280 cos C + 0.0185 sin C

         UT1-UTC = -0.0891 + 0.00021 (MJD - 59649) - (UT2-UT1)
                             ^^^^^^^

(note that the coefficient is seconds _shorter_ than 24*60*60µss,
so +0.00021 means the long-term average LoD is 24h-210µs)

It was fairly steady at 24h+700µs ish through 2018 and 2019, but it
started decreasing by about 10µs per week in July 2019, bottoming out at
24h-260µs in May last year.

That is, the speedup started at about the same time as the anomaly you
pointed out.

The LoD was creeping up: it was 24h-200µs Dec-Feb but in recent weeks it
is down to 24h-210µs again...

-- 
Tony Finch  <dot at dotat.at>  https://dotat.at/
Great Orme Head to the Mull of Galloway: South or southeast 6 to gale
8, decreasing 4 to 6 for a time. Slight or moderate, occasionally
rough in northwest. Rain or showers at first. Good, occasionally
moderate at first.