PDA

View Full Version : What is sourcing the local G dot?



borman
2015-May-10, 04:11 AM
What is sourcing the local G dot?

While the product GM is reliable to 9 decimals, measuring Big G to high accuracy has been a challenge. The variation between measurements has been wider than expected and thought to be due to some unknown systematic.

Recently, Anderson et al [1] looked at a collection of measurements and found the variation followed a sine curve with a period of about 5.9 years which matched the frequency of the LOD (Length of Day) oscillation mystery. They do not suggest that G actually varies at this frequency other than to suggest there be a connection between LOD and G where both suffer from the same cause.

To check if the variation of G dot was solar system wide, Iorio checked to see if such a 5.9 year cycle was present in the Saturn system. He did not find it and it would have been big enough to notice if it were the case at Saturn.

More recently, Schlamminger et al [3] looked at a larger data set of over 35 years. While the correlation is somewhat weaker, it still fits the Anderson hypothesis. It is something of a puzzle: “The situation is disturbing — clearly either some strange influence is affecting most G measurements or, probably more likely, measurements of G since 1980 have unrecognized large systematic errors.”

Klein [4] submitted a paper only three days after Anderson et al which did not cite them or indicated that there was an awareness of the sine. To try to reconcile the differences between two recent results, he appealed to a modified version of MOND. Whether this works for the larger set of Schlamminger et al needs to be researched further.

Since Anderson has been involved with other anomalies such as the Pioneer Anomaly and the flyby anomaly [5] during Earth flybys, it seems of interest to look at the data points of the flybys to see what they might say about the correlation. It is quite serendipitous that the dates of the flybys that showed an anomaly often aligned at the inflection points of the sine curve, a little under three years between the high and low peaks. When the expected anomalies did not show up for Rosetta II and III for the Anderson equation [6], the craft were not at an inflection point. While one should not attribute too much significance to 8 data points, (Juno data point not included), it may be a hint that an additional parameter is needed, say for example, a dot product associated with the inflection points that makes Rosetta II and III vanish. The Anderson equation deals with the Earth axis which relates to the ecliptic. The Adler idea [7] should be reviewed with regard to a possible Dark Matter effect interaction that might source the excess watts per square meter for all planets except Uranus which is as cold as it is supposed to be. The Adler idea was that the collision that turned Uranus on its side knocked it clear of its associated dark matter. The additional possibility here is that the near 90 degree shift leads to a zero dot product and therefore no excess watts.


References
[1] Measurements of Newton's gravitational constant and the length of day
http://arxiv.org/abs/1504.06604v1


[2] Does the Newton's gravitational constant vary sinusoidally with time? An independent test with planetary orbital motions
http://arxiv.org/abs/1504.07233


[3] Recent measurements of the gravitational constant as a function of time
http://arxiv.org/abs/1505.01774


[4] Analysis of recent G experiments by a differential version of MOND theory
http://arxiv.org/abs/1504.07622

[5] http://en.wikipedia.org/wiki/Flyby_anomaly

[6] Anomalous Orbital-Energy Changes Observed during Spacecraft Flybys of Earth
http://virgo.lal.in2p3.fr/NPAC/relativite_fichiers/anderson_2.pdf

[7] Solar System Dark Matter
http://arxiv.org/abs/0903.4879

Shaula
2015-May-10, 04:56 AM
Paper 3 disturbs me. I'd reject it immediately if it were sent to me for review. No discussion of confidence, no alternative hypothesis testing, no comparison of results to alternative fitting models, least convincing graph ever as figure one...

The primary source paper has some interesting features too. The iterative approach to background removal is strange. And when they do Fourier analysis... Really? No window function? And they are surprised by ringing features being dominant? And then they opt for a Gaussian window? This is signal processing 101 and they are mangling it. And then the time series analysis. Oh boy. As soon as I see words like "By eye" I get scared. Again I'd send it back and ask for proof that their methods worked for other data sets (i.e. revealed no pattern in data without this pattern in), basic hypothesis testing etc.

Sorry but as pieces of statistical analysis I find those papers utterly uncompelling. As pieces of signal processing I find them frightening.

antoniseb
2015-May-10, 01:25 PM
Just considering paper [1], it is interesting that the period is half the duration of Jupiter's orbit around the Sun.

borman
2015-May-10, 11:46 PM
Shaula

While each paper goes to peer-review according to its own merits and often will need additional versions, for balance, it can be observed that these authors have had a number of reviewed articles published in various journals including PRD and PRL. The apparent arXiv newcomer seems to be Klein.

Not all papers that are published in peer-reviewed journals are submitted to the arXiv. Reference [6], for example, went directly to PRD where it was reviewed and published but it never was posted to the arXiv. Typically papers first get submitted to the arXiv and then get reviewed sometimes followed with revisions. While a few papers do not get published, many do. As I have said before, one must exercise some caution in reading arXiv papers that have not yet been published. I compare the difference as the difference between going to the zoo and the jungle.

A recent trend has been to submit papers to the arXiv “after” they have been peer-reviewed and published and the embargo lifted.

To get a partial list of an author’s peer-reviewed papers giving the journal and DOI, at least the ones that have also been submitted to the arXiv, click on the blue name in the abstract and a list of the author’s contributions and their publication history will pop up. From there, by noting the number of accepted publications, you can get a feel for how other reviewers see their work.

The reference to Wikipedia was for a quick review of the flyby anomaly and the table that included the dates of the flybys to see where they fall on the sine of reference [1] figure.

To complement reference [7] there was a slightly older paper also published in a journal after revisions:
Planet-bound dark matter and the internal heat of Uranus, Neptune, and hot-Jupiter exoplanets

http://arxiv.org/abs/0808.2823

DOI: http://arxiv.org/ct?url=http%3A%2F%2Fdx.doi.org%2F10%252E1016%2Fj%2 52Ephysletb%252E2008%252E12%252E023&v=2eabb169

One might also note that more recently, Adler has temporarily surrendered on the two fluid idea. There was no distinguishing between hyperbolic versus closed orbit from the two fluid perspective and failure to see evidence of an accumulated anomaly in closed orbits put the hypothesis in jeopardy:

Modeling the flyby anomalies with dark matter scattering: update with additional data and further predictions

http://arxiv.org/abs/1112.5426

borman
2015-May-11, 12:01 AM
antoniseb,
Do you think there is a connection between Jupiter and the "apparent" G dot and LOD? Of course it is possible that the 5.9 year frequency match between G dot and LOD is just a very remarkable coincidence until we find a more fundamental connection between them.

antoniseb
2015-May-11, 01:10 AM
Do you think there is a connection between Jupiter and the "apparent" G dot and LOD? ...
Could be a coincidence. I'm not heavily invested in thinking about this one. I didn't read to see if the 5.9 year cycle is significantly above the noise. If it is, then I think Jupiter is the first place I'd look for an explanation.

Shaula
2015-May-11, 04:50 AM
Shaula

While each paper goes to peer-review according to its own merits and often will need additional versions, for balance, it can be observed that these authors have had a number of reviewed articles published in various journals including PRD and PRL. The apparent arXiv newcomer seems to be Klein.
I'm sorry, are you saying that these must be good papers because the authors have published other peer reviewed papers? Nothing you have said makes any difference at all to my criticisms of the papers. They are flawed because either their analysis is flawed or they have not presented any of the checks or balances that should be part and parcel of this kind of analysis.

Can you address any of my criticisms? For example justify their use of unwindowed FFTs? Or them ignoring the optimised window functions available in favour of a Gaussian one? Or their lack of alternate hypothesis testing? I am sorry but as they stand these are weak papers. My favourite bit has to be where they do an 'independent check' by smoothing the data and then, by hand, fitting curves to make sure the 5.9 year period is preserved. How on earth is that independent?

Jerry
2015-May-12, 03:16 AM
What I see as compelling, is a need for careful, long term time-dependent gravitational constant tests. There is a similar sinusoidal variation in the GPS satellites that is not understood. We don't have a good explanation for why the moon is moving away from us. Something is missing.

StupendousMan
2015-May-12, 12:03 PM
What I see as compelling, is a need for careful, long term time-dependent gravitational constant tests. There is a similar sinusoidal variation in the GPS satellites that is not understood. We don't have a good explanation for why the moon is moving away from us. Something is missing.

First-year university physics courses sometimes use the increase in the Moon's orbital radius as an example to illustrate torque and angular momentum. Are you claiming that this simple explanation (based on the torque exerted on the Moon by the tidally-deformed Earth) is incorrect?

borman
2015-May-18, 12:39 AM
What I see as compelling, is a need for careful, long term time-dependent gravitational constant tests. There is a similar sinusoidal variation in the GPS satellites that is not understood. We don't have a good explanation for why the moon is moving away from us. Something is missing.

Jerry, perhaps you are referring to the Ashby paper page 23 figure 6 regarding the anomalies present only in the more eccentric orbits that do not accumulate to alter the orbits:

Relativity in the Global Positioning System

http://relativity.livingreviews.org/Articles/lrr-2003-1/download/lrr-2003-1Color.pdf

Regarding the Lunar anomaly, perhaps you are referring to the lunar eccentricity increase. Iorio wrote a few papers on the subject:

The lingering anomalous secular increase of the eccentricity of the orbit of the Moon: further attempts of explanation of cosmological origin
http://arxiv.org/abs/1404.6537

On the anomalous increase of the lunar eccentricity
http://arxiv.org/abs/1105.4863

An Empirical Explanation of the Anomalous Increases in the Astronomical Unit and the Lunar Eccentricity
http://arxiv.org/abs/1102.4572

On the anomalous secular increase of the eccentricity of the orbit of the Moon
http://arxiv.org/abs/1102.0212

borman
2015-May-18, 01:02 AM
Shaula,
The authors did give their reasons for no windows. Even the Gaussian gave no better than a hint. The " by eye" should be put back into context with respect to the cubic splines. It would not be right for me to try to "put words" into their mouths they did not say. They felt they had their reasons for their approach. If it needs justification or further remarks, this is in the provence of the reviewers to suggest. Here they are talking about the LOD data. The G dot is only 13 data point. We need more cold quantum data points and more extended readings to improve sigma. Considering that this pattern was discovered just this April, we are on the front end of a possible discovery path that often has many pitfalls to navigate.

Shaula
2015-May-18, 05:46 AM
Shaula,
The authors did give their reasons for no windows. Even the Gaussian gave no better than a hint. The " by eye" should be put back into context with respect to the cubic splines. It would not be right for me to try to "put words" into their mouths they did not say. They felt they had their reasons for their approach. If it needs justification or further remarks, this is in the provence of the reviewers to suggest. Here they are talking about the LOD data. The G dot is only 13 data point. We need more cold quantum data points and more extended readings to improve sigma. Considering that this pattern was discovered just this April, we are on the front end of a possible discovery path that often has many pitfalls to navigate.
What we need is good robust analysis. These papers do not do that. You can add as many data points as you like, it won't make any difference.

Can you provide evidence that these were robustly reviewed? I can only find them in Inspire and EPL (which has a minimal review process). I'd have thought that they should have shown up in more rigorously reviewed journals but they don't seem to have been submitted.

Also - worth asking BICEP2 what early publication brings to a project. Always review first, promote later.