Dark Matter or Modified Gravity ?

[clamtrox]

. . .
It should be obvious that that the reanalyzed data were not of good enough quality to separate between models of the PA. A dedicated mission is required to settle the case. To claim that the case has already been settled in favour of a thermal explanation would be a sign of severe confirmation bias.

I see. So would you say that if I presented you with an alternative theory of gravity, then objects falling towards the Earth would no longer be explained by GR? I think we are understanding the word "explained" very differently here. I would say that a theory "explains" something if it makes a clear, unambiguous and correct prediction of the phenomenon, regardless of what other crackpot theories there might be out there.

 

[twofish-quant]

Right. You've shown that by imagining that there's something there that we can't detect, we can "save the phenomena" (as Ptolemy would say). Sorta like an epicycle. It works, no doubt about it. But is it true?

Getting an epicycle to work is more difficult than it sounds. I don't know of anyone who is philosophically against modified gravity. It's just that adding that "modified gravity fudge factor" leads to worse results than the "dark matter fudge factor."

Just to give one example. There's a lot of deuterium in the universe. Deuterium is easy to burn. If you assume that the early universe is 100% ordinary matter, you end up burning up all of the deuterium in the universe. Now if you assume that the universe is 30% ordinary matter, that slows the reaction rate down so that you don't burn out all of the deuterium in the universe.

Now if the MOND people are right and gravity is stronger at large scales than we expect, and the early universe was 100% ordinary matter, then the universe stays a lot denser, so the deuterium goes in the wrong way.

At larger scales it's hardly been tested at all, and when it has been, DM must always be invoked to save the phenomena.

Well, people have tried modified gravity, and what they've found is that as long as the true theory of gravity behaves something like Newtonian gravity at cluster scales, you need dark matter.

You can ask the question, how much does the gravity theory have to change from Newtonian before you can eliminate dark matter, and it turns out to be a lot.

Dark energy, is something different.

Not true. Neither the Pioneer anomaly nor the flyby anomaly are explained by classical GR gravity.

Right. However adding radiation corrections seems to solve everything.

Those silly spinning galaxies make up most of the Universe. Meanwhile, all attempts at actually detecting DM have failed, and LHC experiments are showing smaller and smaller mass regions in which the putative DM particles can hide out. What happens when those regions converge down to zero?

Then we ****GET DOWN AND PARTY**** Champagne for everyone, because we are going to spend the next decade writing wild and crazy papers! NSF and DOD funding!

Guess what. Theorists hate it when you have a good explanation for stuff. If it turns out that dark matter doesn't work, then we can start getting wild and crazy. Modified gravity. Cosmological constant. Quantum mechanical effects. Collective effects. Neutrino solids! Animal spirits!

I have a pet wild and crazy theory (neutrino metal) that I'll go off on if it looks like WIMPS are dead, but it's a waste of time now.

Neutrinos don't interact electromagnetically either, but we detect them anyway. And we still should be able to detect DM experimentally. And we haven't.

Since dark matter is a hand wave, it's hard to tell what should happen.

One of the problems with modified gravity theories is that since gravity is more well known, it's hard to "make up something stupid" which means that it's easier to show that modified gravity is wrong. If we don't see a particle, then well we just haven't looked hard enough. If I put in different gravity models into CMBFAST and it doesn't work, then there's no place to hide.

 

[twofish-quant]

I wasn´t directly referring to this paper but to the thread in PF. From that thread the most interesting things for me has been to follow up the critique on LCDM from Kroupa and colleagues. They seem confident among other things that the LCDM is flat out wrong in its predictions on satellite galaxy formation, in the milkyway and in other galaxies that has been studied.

And then have a good point. LCDM doesn't work that well at explaining the behavior of galaxy formation, because its was never intended as a theory of individual galaxy formation. People are extrapolating LCDM into this area, but the fact that it doesn't work well is hardly a surprise. If I use a hammer as a saw, it's going to have problems.

LCDM also doesn't explain the green house effect, the fact that the Earth has one moon, or why there is evil in the world. Why can't you get good airline coffee? LCDM doesn't explain that. Well, it must be wrong...

 

[Old Smuggler]

I see. So would you say that if I presented you with an alternative theory of gravity, then objects falling towards the Earth would no longer be explained by GR? I think we are understanding the word "explained" very differently here. I would say that a theory "explains" something if it makes a clear, unambiguous and correct prediction of the phenomenon, regardless of what other crackpot theories there might be out there.

The scientific method is about testing theories. Just to come up with a flexible mainstream model
where some choice of parameters will fit the data does not automatically shut out other models.
Rather, if more models fit the data equally well, the "best" one is the one with the least number of
parameters. This is what Occam's razor is all about.

The Turyshev paper depends heavily on the "discovered" decay of the anomaly. Since this
"discovery" is probably due to mismodelling of the data, chances are that the model will not be
correct. The lesson learned here is that mere parameter-fitting in a flexible model should not be
sufficient to be called a "settled explanation", even if many people apparently think so.

 

[clamtrox]

The scientific method is about testing theories. Just to come up with a flexible mainstream model
where some choice of parameters will fit the data does not automatically shut out other models.
Rather, if more models fit the data equally well, the "best" one is the one with the least number of
parameters. This is what Occam's razor is all about.

The Turyshev paper depends heavily on the "discovered" decay of the anomaly. Since this
"discovery" is probably due to mismodelling of the data, chances are that the model will not be
correct. The lesson learned here is that mere parameter-fitting in a flexible model should not be
sufficient to be called a "settled explanation", even if many people apparently think so.

I still don't understand what your claim is. If you claim that it's possible that there exists an alternative explanation for Pioneer anomaly, then I am not really interested to continue this "argument", as what you are saying is then trivially true.

If your claim is that the article I cited is wrong, then fine; I'm by no means an expert in FEM and all that stuff you need to model the thermal emissions. Perhaps you should publish an article about it. Regardless, I find claims like "GR can't explain the Pioneer anomaly" very dubious, even if the best modelling is done somehow wrong.

If your claim is that since modelling the thermal emissions of Pioneer is so complicated, a "simpler" explanation by modified gravity theories is somehow automatically preferred, then I strongly disagree with you. Just because there are complicated -- but in theory perfectly well understood -- phenomena, doesn't mean that we should look to modify our existing theories. One does not look to modify the theory of gravity just because it's very difficult to calculate the precession of Mercury caused by all the other planets. Only after it's clear that the prevailing theory cannot account for all of the precession, it makes sense to look to look for modifications.

 

[Old Smuggler]

I still don't understand what your claim is. If you claim that it's possible that there exists an alternative explanation for Pioneer anomaly, then I am not really interested to continue this "argument", as what you are saying is then trivially true.

If your claim is that the article I cited is wrong, then fine; I'm by no means an expert in FEM and all that stuff you need to model the thermal emissions. Perhaps you should publish an article about it. Regardless, I find claims like "GR can't explain the Pioneer anomaly" very dubious, even if the best modelling is done somehow wrong.

If your claim is that since modelling the thermal emissions of Pioneer is so complicated, a "simpler" explanation by modified gravity theories is somehow automatically preferred, then I strongly disagree with you. Just because there are complicated -- but in theory perfectly well understood -- phenomena, doesn't mean that we should look to modify our existing theories. One does not look to modify the theory of gravity just because it's very difficult to calculate the precession of Mercury caused by all the other planets. Only after it's clear that the prevailing theory cannot account for all of the precession, it makes sense to look to look for modifications.

My claim is that the data do not constrain the thermal model sufficiently to reach a conclusion of any reasonable
degree of confidence. On the contrary, the fact that the "observed" decay of the anomaly goes as
1/r² indicates that the thermal modelling is fundamentally misguided. For this reason, an objective
view would be to regard the case as open until data of sufficient quality are obtained. Such data can
only be obtained from a dedicated mission craft specifically equipped to settle the case, so that an
explanation based on known physics local to the craft can be confirmed or ruled out.

 

[Maui2001]

And with the publication of Turyshev's latest paper last summer, it's very unlikely that a dedicated mission will ever happen. Shame.