## Tuesday, June 10, 2008

### WMAP power asymmetry: FAQ

A lot of confusion - and downright nonsense - has been written about the power asymmetry of the WMAP data. The "contrast" of one cellestial hemisphere seems to be greater than the "contrast" of the other hemisphere.

Is there an asymmetry at all?

Yes, it has been described by Eriksen et al. in 2003. See also their 2007 update. It is not too strong but it is observable.

How do you divide the sky to the hemispheres?

There is clearly no "canonical" way to divide the sky but it just happens that the dividing plane that maximizes the asymmetry is rather close to the plane of ecliptic - i.e. the plane of Earth's orbit around the Sun. As you might know, this plane is also very close to the ecliptic plane of the Milky Way, i.e. our galaxy.

Is the asymmetry statistically significant?

Yes, it is. It is approximately by one order of magnitude larger than the expected strength of "noise".

Cannot you get rid of this effect by some trivial "fix"?

The answer is probably Yes. In 2006, Freeman et al. have shown that the asymmetry for "l" smaller than 8 disappears if the WMAP dipole vector is increased by 1-3 sigma (i.e. 2-6 km/s, from 368.11 km/s). The asymmetry at higher values of "l" probably goes away if the foreground near the ecliptic (the galactic processes that "pollute" the global picture near the "equator") is treated more carefully.

If that's not the case and if new physics is required, is the inflaton enough to create such an asymmetry?

No, the inflaton predicts nearly isotropic data. A new field that can modulate the power, known as the "curvaton", or something equally powerful would be required. With a new field, you can get all kinds of new effects, of course, including the differences between the two hemispheres.

If the asymmetry is "real", is it a specific problem for inflationary cosmology?

No, the asymmetry - if it were real - would be surprising for pretty much every "minimal" model replacing inflation, too. A slight modification such as the addition of a "curvaton" would be enough to account for it. But once again, the effect is probably not real, as discussed above: it is rather an artifact of a wrong dipole and ecliptic subtractions.

Is there a rationally justifiable link between these asymmetry questions and the arrow-of-time debates?

Not at all. All the hints referring to similar links are absurd and meaningless inventions of a Chris Lintott who likes to "promote" cosmology, primarily among the audiences with IQ below 20. BBC was helpful for such a goal and their recent article was, from a scientific perspective, a breathtaking junk.

Recall that the fantasies that there remains an unsolved "arrow of time problem" that could even be related to cosmology contradict the very basics of thermodynamics and statistical physics that have been known for more than 100 years.

The existence of the arrow of time - i.e. the second law of thermodynamics - is a rule that applies to all systems with many degrees of freedom. Cosmology is just another context where the second law holds. The second law can be applied and should be applied to cosmology but cosmology cannot be applied to the second law.

The validity of the second law in a terrestrial context has clearly nothing to do with cosmology and cannot be linked to issues in cosmology such as the WMAP power asymmetry. The validity of the second law in the lab can be derived - and has been derived - from statistical physics properly applied to the large number of degrees of freedom describing the lab.

Whoever misunderstands these things should take an undergraduate course of thermodynamics and statistical physics again because he or she is severely failing in these subjects.

Is there a known relationship between the multiverse ideas and the power asymmetry?

There could exist such a relationship but no such relationship is known at this moment. The very existence of a multiverse doesn't imply that the power spectrum should be symmetric or asymmetric and the asymmetry doesn't imply that the multiverse should or should not exist.

If the asymmetry is caused by "new physics", this new physics must clearly be considered simultaneously with inflation (or any other picture that you choose to replace inflation) because the asymmetry is a feature of the very same dataset. The comments that the asymmetry would tell us something about the Universe "before the inflationary era" or even "before the Big Bang" are completely unsubstantiated.

Once again, these reports are cheap, pseudoscientific, pornographic material addressed primarily to readers with IQ below 20.