Statistical regression analysis of ocean removal as evidence

[lanzalaco]

I had this idea way back in ratskep, i mean i thought it was an obvious one so maybe somebody can tell me why its not been carried out ?

the EE windback to a closed earth could involve say 150 stop points in the regression..say every million years.

Somebody who is good with statistical mechanics should be able to take the windback of the floors and assign P-values at each point for whether a point could reach the next regression point by random chance or not. So you get 150 results of probability.

Now then take all those 150 p-values and then plot these values as a statistical regression of its own liner dataset, and match them up against the overall probability of the windback fitting from now to closed earth.

From there you get the statistical probability the earth could have fitted back randomly.

Do it on 3 levels.

1. Probability of the fitback with no magnetic windback.
2. Probability of other fitbacks with magnetic windback
3. Probability polar wander is canceled with no magnetic windback.

Then find probability that polar wander cancellation AND magnetic windback vs random chance.

Of course all this is done vs Pangea.... So you get to see just how random Pangea is....(im betting very !)

that is something that is mathematical. It would fill a paper, the figures would be in your face. Its hard to dispute, and can be published. Also its not overtly demanding from a mathematical perspective. I dont understand why this has not been done previously. I guess maxlow just thought people seeing the animations with their eyes would be enough.

[lanzalaco]

the falsifications are pretty basic.

A linear regression of the ocean floor cannot result in 150 points with a null hypothesis.

The entire set of 150 points moving back in a linear manner does not result in statistical certainty

The entire set of 150 points moving back in a linear manner does not result in reduction of pole wander.

Put in pangea and EE then pull out the results.

[Florian]

the falsifications are pretty basic.

A linear regression of the ocean floor cannot result in 150 points with a null hypothesis.

The entire set of 150 points moving back in a linear manner does not result in statistical certainty

The entire set of 150 points moving back in a linear manner does not result in reduction of pole wander.

Put in pangea and EE then pull out the results.

Not sure to get it, but to make that kind of statistic, one already needs a good model of Earth expansion. Then why would it move back in a linear manner?
Could you eventually make a scheme to explain what you have in mind?

[lanzalaco]

the falsifications are pretty basic.

A linear regression of the ocean floor cannot result in 150 points with a null hypothesis.

The entire set of 150 points moving back in a linear manner does not result in statistical certainty

The entire set of 150 points moving back in a linear manner does not result in reduction of pole wander.

Put in pangea and EE then pull out the results.

Not sure to get it, but to make that kind of statistic, one already needs a good model of Earth expansion. Then why would it move back in a linear manner?
Could you eventually make a scheme to explain what you have in mind?

Previously before the ocean floor age map was complete it was easy to dismiss EE fit together in terms of spatial co-incidence if looking the the probability of degrees of freedom of continents.

i.e You only had two data points, now and closed earth.

The ocean floor adds something very powerful from a statistical view. A linear map through time as new material is added. i dont get why this has not been exploited by a mathematician and geologist.

Each point in time say very 1 million years adds a set of vectors to the configuration of the continents that allows 180 probability values to be assigned each with degrees of spatial freedom to move randomly.

PT says the fit back is random chance, so that means it is defending the idea that every million years plates can move anywhwere and the probability of 180 fit backs is low.

if you assign probability value to whether the plates would keep winding back in to each other 180 times vs chance, you can use statistics to set up your falsification. Im no mathematician but i do enough statistics to see the EE windback must have ridiculously high improbability they plates could wind back 180 times in a linear manner and fit without ever wandering anywhere else. Sure scientists will dismiss what they can see, but dismissing a math paper full of statistical probability saying the chance would be close to impossible is another matter.

I dont know what the resolution is of the crustal mapping to years, but you start with probability at magnitudes less, say 10 million year periods then increase to max resolution a couple of times. say to 1 million year periods, then 100,000 year periods if the resolution is accurate to that.

If the first result says chance is close to impossible, then you increase resolution and the p-values become increasingly extreme refinements of improbable, thats not possible to argue with scientifically.

[lanzalaco]

btw....i meant i dont know what the "the resolution is of the seafloor mapping to years"

big accusation against neal adams is paranioa he is fudging it. another idea, is since to do these stats, you need to get a big 3d mesh of the earth, then remove seafloor in a 3d modeller, make all the data online in standard dxf formats so anybody can download and run.

Maybe even make the site with the models and stats interactive in open science format.

Actually quite interesting neals youtube votes, the majority are voting for his animations lately..

[Florian]

Previously before the ocean floor age map was complete it was easy to dismiss EE fit together in terms of spatial co-incidence if looking the the probability of degrees of freedom of continents.

i.e You only had two data points, now and closed earth.

The ocean floor adds something very powerful from a statistical view. A linear map through time as new material is added. i dont get why this has not been exploited by a mathematician and geologist.

Dietmar Müller, who is at the origin of the data of seafloor project did it, but with plate tectonics in mind. See here. Basically, they invent ridges and subducted plates to make their model. I think they believe so much in plate tectonics that don't even realize how speculative is their modeling work.

Each point in time say very 1 million years adds a set of vectors to the configuration of the continents that allows 180 probability values to be assigned each with degrees of spatial freedom to move randomly.

PT says the fit back is random chance, so that means it is defending the idea that every million years plates can move anywhwere and the probability of 180 fit backs is low.

if you assign probability value to whether the plates would keep winding back in to each other 180 times vs chance, you can use statistics to set up your falsification. Im no mathematician but i do enough statistics to see the EE windback must have ridiculously high improbability they plates could wind back 180 times in a linear manner and fit without ever wandering anywhere else. Sure scientists will dismiss what they can see, but dismissing a math paper full of statistical probability saying the chance would be close to impossible is another matter.

I dont know what the resolution is of the crustal mapping to years, but you start with probability at magnitudes less, say 10 million year periods then increase to max resolution a couple of times. say to 1 million year periods, then 100,000 year periods if the resolution is accurate to that.

If the first result says chance is close to impossible, then you increase resolution and the p-values become increasingly extreme refinements of improbable, thats not possible to argue with scientifically.

It is not so much by chance, because plates are supposed to move at steady rate for millions years. This is because subduction is thought to drive the plates. So the plate motion is thought to me smooth for millions years, thought there are some dramatic change of direction like that 40 Ma ago thought to be at the origin of the kick in the Hawaii-Emperor Seamount chain:

Raymond, C.A., J.M. Stock, and S.C. Cande, Fast Paleogene motion of the Pacific hotspots from revised global plate circuit constraints, in History and Dynamics of Plate Motions, edited by M.A. Richards, R.G. Gordon, and R.D. van der Hilst, pp. 359-375, 2000

HawaiiRaymond_500.jpg (28.22 KiB) Viewed 19814 times

The plate motions are thought to be sufficiently organized to let cratons regularly form clusters (supercontinent cycle). But the supercontinent cycle is yet another artifactual interpretation related to the denial of the expansion. Carey debunked it decades ago.

For modeling, passive margins are not very deformed so it is not a problem to make reconstructions. If there were only passive margins, the expanding Earth theory would have been adopted decades ago.
But the problem here is that active margins around the Pacific have been swelling since the Jurassic. So it is not so easy to reconstruct how the margins looked like at the different time, especially at the time of the formation of the early Pacific seafloor.
As an example, look at this animation made from a review of the evolution of the Pacific South West (Schellart Earth-Science Reviews 76 (2006) 191-233):

btw....i meant i dont know what the "the resolution is of the seafloor mapping to years"

big accusation against neal adams is paranioa he is fudging it. another idea, is since to do these stats, you need to get a big 3d mesh of the earth, then remove seafloor in a 3d modeller, make all the data online in standard dxf formats so anybody can download and run.

Maybe even make the site with the models and stats interactive in open science format.

James Maxlow is working on a numerical model. Stephen Hurrell also worked on it and is in the process of updating the dataset used in his model.

jbl007 has been exploring a way to make 3D mesh of Earth and remove seafloor by age.

Actually quite interesting neals youtube votes, the majority are voting for his animations lately..

Well, I have a problem with Neal. There are many fatal errors in his model, but he denies it. Neal does not have a scientific attitude, and it is damaging for the theory.

[lanzalaco]

still if neal had not commissioned these videos this would not be an issue on public consciousness at all. those youtube counters keep resetting, when he alters a video. I think he most have had millions of views.

I am sure most people can perceive the guys not a scientist, and that hes not of the character to fudge, deliberately anyway. at least its obvious to me, he doesnt understand science methods or even science audiences, but still its interesting what he did in spite of that.

look forward to seeing a mathematical model, hope that steven and james consult a few maths experts first, so they can guide what makes for a robust case by todays standard.

[Dinox]

James Maxlow is working on a numerical model. Stephen Hurrell also worked on it and is in the process of updating the dataset used in his model.

jbl007 has been exploring a way to make 3D mesh of Earth and remove seafloor by age.

Who is jbl007?

[Florian]

Who is jbl007?

A french member of this forum.

There are not so many

No news from him for quite a long time.