Solar, Terrestrial, & Lunisolar Components of Rate of Change of Length of Day

While the above article provides much detail I have been unable to discern what amount of time is denoted by the verticle scale of the graphs.
Is the +/- 1 a second, a dimensionless number or a nanosecond?
While it may be possible to torture numerological correlations from the LOD data and the quasi-periodic SOI or other observations – zonal high wind etc – that corellation only has any possibility of indicating a causal relationship rather than a coincidental one if the physical processes can be defined that operate.
If, as I suspect, the energy in Joules of these variations is several orders of magnitude smaller than the solar varation around the 11 year cycle or the extra downwelling energy measured from the extra CO2 then it is ridiculous to ascribe causation to something that has no chance of influencing the processes because its energy signiture is way below the level of many other influences.
So what change in energy received at the surface do the graphs indicate ?

Lucy Skywalker suggests, “[…] it would help if you made your work accessible to ordinary intelligent generalists […]”
Solid conceptual understanding of the effect of integration across harmonics is something folks will have to develop independently. Then it’s dead simple (and the results can be reproduced in mere minutes).
The target audience of these communications may not be the one you prefer, but I sincerely appreciate your interest in the topic.
Best Regards.

2 more questions (slightly more advanced):
3) Did Le Mouël, Blanter, Shnirman, & Courtillot (2010) make the most sensible choice of window width (1 month) when isolating the semi-annual to annual timescale derivative series?
Hint: …or did they just come “close enough” somehow?
4) Are the optimal kernel widths determined by maximal correlation (with CR) or by dominant temporal modes of the mean & variance (of LOD’)?

Re: izen
The vertical scales are normalized to optimize visualization. As a contribution to this multidisciplinary discussion, I explore data, leaving physics to physicists & other qualified parties. A limiting factor in quantitative musings about physical mechanisms is insufficient human knowledge of clouds (and hence insolation [not to be confused with irradiance]]). Producing yet more estimates based on untenable assumptions is not the answer. That the atmosphere experiences thermal as well as gravitational tides is accepted. Given the public’s growing awareness of the spatiotemporal cloud/circulation issue, assumptions such as uniformity are now a deal-breaking problem for advocates of oversimplified models. The data have spoken. Are the physical modelers listening? Or do they remain caught in the deep grooves of their abstract imaginations?
Best Regards.

Carla says:
April 11, 2011 at 5:37 am
Geoff Sharp says:
April 11, 2011 at 1:38 am
~
Question for Geoff, bit off the topic here or not..
The Earth has a roid, asteroid that is orbiting in its zone.
They are saying that this roid has a “horseshoe” type orbit. Now how can that be, Geoff? Check out the orbit of this thing Geoff.
Astronomers Find Newly Discovered Asteroid is Earth’s Companion
http://www.arm.ac.uk/press/2011/aac_horseshoe_orbit.html
————————————————————————————————————-;
The asteroid has a circular orbit around the Sun but at a different speed than the Earth so the motion appears to be like a horse shoe orbit when viewed from Earth. Even the planets at times appear to have retrograde orbits when viewed from the Earth.

A little history on some of these concepts might be of interest. I worked for several years (in the 1970s) at a “polar motion” observatory in Mizusawa Japan (International Latitude Observatory – ILO). The ILO at Mizusawa for many years had been using optical methods to determine the Chandler period (an approximate 14 month cycle of the earth’s pole through about 8 to 12 meter circles). The Chandler phenomenon (1891) was discovered over 100 years ago and from just a few years before 1900 a small group of world wide observatories located in the northern hemisphere spread in longitude along latitude 39 deg N began observing latitude shift of zenith stars. The observed shift (in latitude) appeared as an approximate sine wave of about 14 months long. My support was to use satellite based geodesy methods to obtain the same information in spite of cloudy bad weather conditions which limited optical data quantity and quality. This network grew to a multi-station, multi-nation system with the Paris observatory eventually becoming the headquarters of the International Earth Rotation Service – IERS. Just to let you understand that earth rotation variation has a time component and a spatial component.
I wonder if we might be able to tie some of the pole spatial movement effects to other physical elements – especially climate effects (and polar tides)? It is well known that one the drivers of the pole speed and position is the atmosphere (as well as ocean currents, magma, gravity, extra-terrestrial effects, etc.)
Bernie

The replies have covered a lot of ground and raised a lot of questions, too much to cover fully on WUWT. Nor does Paul say all.
Background.
The work is Paul Vaughan’s where I was using the effect of years of technical work and experience to produce numbers and models for him. There is cross discipline involved. I only understand part of what he is doing.
In this case Paul emailed out of the blue with an xls containing data attached and more or less what could I do with it. I recognised the data as daily LoD first difference, hence first data point 2nd Jan 1962 but that is all I know, nor have I checked which version of LoD nor how exactly Paul produced it.
Since then a few emails and attachments have bounced between us as we drove to whatever Paul wanted and found mutually compatible file formats.
First I knew about what Paul was up to other than he was going to published something was seeing the WUWT post.
Tools used.
Over some years I have developed unique software intended for use on climatic kind of data. Some of this goes back to a novel development for commercial usage during the 1980s. Some think I am nuts to write this in C; there are several very solid reasons.
The key to the result is a selective non discrete Fourier Transform created by data matching discrete input data (qualifications there are critical). This outputs the parameters for matched Fourier terms and a spreadsheet model of the matched data. (it doesn’t have to be Fourier but this is usually best and is used here)
A spreadsheet row comprises an inverse transform, simply the sum of computed terms for that index value, in this case a decimal date. It is literally simple.
Column meaning: –
index sum-of-terms+offset term1 term2 term3 .. termx
Multiple rows with say a one month index increment, plot sum column against index column and you have a computed monthly time series. The spreadsheet maths engine is doing the result computation from the parameters and formula. The software derived the parameters and wrote out two rows of the formula, copy and paste extends this as the user wants. Conceptually simple.
Why a fancy way of doing a Fourier Transform?
There are finite limits on the capability of a DFT (discrete fourier transform) which is the only kind that can be done on discrete data (regularly sampled in time). The time increment defines the width of each output value, called bin as in garbage bin, where a transform outputs a long row of them, one for each frequency it is capable of showing. This means the numbers are approximate, only is within that bin which has a finite width. Accurate phase is very hard to compute.
The values can to some extent be improved by interpolation and other trickery but this only goes so far.
A further problem is the need to Window the input data, a large subject.
Very roughly I am subtracting the contents of a bin from the input data and fine tuning the frequency, phase and amplitude for zero error. Those are the answer. the limits of binning are to a degree sidestepped: the bins have zero width, just that there not very many of them. (useful results in practice tend to 3 to 30 bin range)
It is also possible to put more than one item in a single bin and this spins off into ambiguity and validity, where the human brain is necessary.
In this case the presence of doublets (two very closely spaced (eg. 0.25%) frequencies) and more makes the problem interesting. When computed this turns out to mean 18.6 years… get the idea?
Put simply, it works.
In this case the input data has chronometric causes (orbitals) and therefore the model will be predictive (not been confirmed).

Here is an extended version of the PDF PV mentioned, on WP servers.
Might help as a starting point.
http://daedalearth.files.wordpress.com/2011/04/comment-on-pv3-a.pdf

George E. Smith says on April 11, 2011 at 10:58 am

“”””” netdr2 says:
April 11, 2011 at 7:45 am
The LOD also should vary due to sea level rise. [It is like the figure skater that pulls her arms in to spin faster.]
Axel Moerner claims that the effect of rising sea levels on LOD limits the sea level rise to 1.1 MM per year or 1 cigarette length per century.
Is he right ? “””””
Well hardly.
Sea level rises because water flows down hill. so that means that a water mass that was previously rotating at a larger radius about earth axis, has now moved down to a lower orbit; namely sea level. So the global moment of inertia must go down, not up, with sea level rise. If there is any LOD change, it would be a shortening.

Isn’t one of the components of sea-level rise due to thermal expansion of the water?
Secondly, most of the ice that will melt and contribute to sea-level rise is currently closer to the spin axis than it would be if it were uniformly distributed across the surface of the planet… so, I am not sure it would lead to a shortening of the LOD but I am willing to be persuaded otherwise.

Leif Svalgaard says:
April 11, 2011 at 8:06 am
Geoff Sharp says:
April 11, 2011 at 6:21 am
On the magnetic front, I see no evidence of any planetary magnetic feedback. The Tsunami like solar wind obliterates all.
Something we can agree on.
~
I was going to mention Geoff, that you were sounding a bit like Dr. S.
Not the point. The “Princess” in left field, is suggestingt that the closer your dipole is to the energy source ‘within’ your system, the more influence the solar electra magno reconnection regime, well have upon it. Saturn’s dipole looks undecided..like maybe he’s the queen in left field, with a by directional dipole depending on ISMF forces present in Saturn’s orbital field of ref..or somethin like that. Ah yep that would change Saturns LOD. lol..
But seriously .. how would you slow down ionospheric plasmas moving across field lines? I have read that the ionosphere is coupled to the solar wind IMF. And the ionosphere is coupled to this rotating electro dynamo. The ionosphere is suggested to be directly coupled with the polar vortex and may vary it by strong variations of ionospheric winds. Seems lately we have seen a stalling of regional weather systems moving across the N. Hem, causing an unusually shaped jetstream. . kinda like its being forced back down. . lost connection?

“”””” Richard Sharpe says:
April 11, 2011 at 11:18 am
George E. Smith says on April 11, 2011 at 10:58 am
“”””” netdr2 says:
April 11, 2011 at 7:45 am
The LOD also should vary due to sea level rise. [It is like the figure skater that pulls her arms in to spin faster.]
Axel Moerner claims that the effect of rising sea levels on LOD limits the sea level rise to 1.1 MM per year or 1 cigarette length per century.
Is he right ? “””””
Well hardly.
Sea level rises because water flows down hill. so that means that a water mass that was previously rotating at a larger radius about earth axis, has now moved down to a lower orbit; namely sea level. So the global moment of inertia must go down, not up, with sea level rise. If there is any LOD change, it would be a shortening.
Isn’t one of the components of sea-level rise due to thermal expansion of the water?
Secondly, most of the ice that will melt and contribute to sea-level rise is currently closer to the spin axis than it would be if it were uniformly distributed across the surface of the planet… so, I am not sure it would lead to a shortening of the LOD but I am willing to be persuaded otherwise. “””””
Tut tut Richard; I know you are sharper than that. You are contemplating that the melt water unaided, by human hand can actually run uphill to a higher potential energy location.
Is not the (mean) surface a gravitational equipotential surface. As far as the earth is concerned it is gravitationally flat. Yes it does have some rather large periodic bumps in it caused by tides and weather; but over the course of a thirty year climate period, it is quite flat.