[time-nuts] UTC and leap seconds

Magnus Danielson magnus at rubidium.dyndns.org
Sat Jun 12 14:23:20 UTC 2010

On 06/12/2010 03:36 PM, jimlux wrote:
> Magnus Danielson wrote:
>> On 06/12/2010 02:33 AM, Hal Murray wrote:
>>> jimlux at earthlink.net said:
>>>> The Chilean earthquake changed the angular rotation rate (or,
>>>> probably more
>>>> accurately, changed the direction of the axis of rotation as well)
>>>> of the earth a small amount, as do most large earthquakes.
>>> Has anybody measured that?
>>> Is there a good URL on this? (predictions if not data) All I've found so
>>> far is a small NASA press release predicting 1.26 microseconds per day:
>>> http://www.nasa.gov/topics/earth/features/earth-20100301.html
>>> (and a zillion news sources repeating it)
>>> 1 microsecond/day is 1 part in 1E11.
>> There is one place that keeps track of these things, the IERS. Their
>> Bullentin B provides monthly reports of earth rotation observatoins.
>> The bulletin for relevant period of time is:
>> ftp://hpiers.obspm.fr/iers/bul/bulb_new/bulletinb.266
>> ftp://hpiers.obspm.fr/iers/bul/bulb_new/bulletinb.267
>> Explanations is in:
>> ftp://hpiers.obspm.fr/iers/bul/bulb_new/bulletinb.pdf
>> Clause 3 is of most interest, look at the OMEGA column (appended the
>> bulletin b 267 data for a more complete time-series around the date of
>> interest).
>> LOD : Excess of the Length of day - 86400 s TAI
>> OMEGA : Earth angular velocity
>> Description: ftp://hpiers.obspm.fr/iers/bul/bulb_new/bulletinb.pdf
>> DATE MJD LOD sigma OMEGA sigma
>> (0 h UTC) ms ms mas/ms mas/ms
>> 2010 2 2 55229 1.8681 0.0026 15.04106685346 0.00000000045
>> <snip>
> 55287 1.3068 0.0018 15.04106695117 0.00000000031
>> Yes, we see a dip there... but looking at the two-month data we se a
>> regular pattern creating a dip there... and the lack of jump is
>> interesting.
>> Essentially... I can't see it here.
> And I think that's what I heard: it should have changed (permanently)
> but that it would be impossible to see without years of data to remove
> all the other effects.
> My GPS friends (all in the same section as Richard Gross.. the same
> section are the folks who do precision measurements of gravity (GRACE
> and GRAIL missions), etc.) tell me that once you get down into a certain
> range of uncertainty (10cm=ish for GPS), there's a whole lot of factors
> that are all of about the same magnitude (tidal movement, atmospheric
> delays not due to ionosphere, ionospheric effects, continental drift,
> etc.) So making absolute measurements requires lots of data and
> painstaking identification of each of the factors and removing it.

Actually, we should be quite happy that we are not seeing much worse 
frequency jumps and angular movements than we do.

The main point of my exercise is to point into a bunch of aggregated 
data-set that gives an overview at least. It uses several independent 
sources (such as GPS, GLONASS, VLBI, SLR sources) and plots differences 
such that effects particular to some systems is not allowed to dominate.

While it would be fun to know, the practical impact of such a change is 
very, very small, to the level of being ignored. Considering of a major 
event actually consisting of many hundreds of earth quakes spread over a 
rather longish period, it becomes more interesting to see how the 
aggregate behaves. But you also needs to understand what the effect was 
prior to the event as it built up. What is the effect over a 1000 year 
period (a short period in geological sense)? Consider the build-up of 
pressure and magma under a volcano, which progresses over many years 
before it finally accelerate. Gravimeters is routinely used to monitor 
this build-up. There are thus more sources to consider. Someone is 
probably studying that.


More information about the time-nuts mailing list