# Recent Planck probe results: how unexpected?

The data from the Planck probe's observations are in, and according to the European Space Agency they show a "hemispheric asymmetry in the cosmic microwave background (CMB)". Quote:

an asymmetry in the average temperatures on opposite hemispheres of the sky [...] with slightly higher average temperatures in the southern ecliptic hemisphere and slightly lower average temperatures in the northern ecliptic hemisphere. This runs counter to the prediction made by the standard model that the Universe should be broadly similar in any direction we look.

How unexpected is this variance from the Standard Model and can it be quantified?

How certain is it that the data are accurate? For the recent discovery of the Higgs boson at the LHC, a five sigma result was considered sufficient to make the announcement. What is the sigma for the reported hemispherical asymmetry?

Yet the report of faster-than-light neutrinos, subsequently withdrawn due to equipment failures, was based on six sigma evidence. And in one of the backwaters of Wikipedia, we learn that:

Some anomalies in the background radiation have been reported which are aligned with the plane of the solar system, which contradicts the Copernican principle by suggesting that the solar system's alignment is special.[10] Land and Magueijo dubbed this alignment the "axis of evil" owing to the implications for current models of the cosmos,[11] although several later studies have shown systematic errors in the collection of that data and the way it is processed.[12][13][14] Various studies of the CMB anisotropy data either confirm the Copernican principle,[15] model the alignments in a non-homogeneous universe still consistent with the principle,[16] or attempt to explain them as local phenomena.[17] Some of these alternate explanations were discussed by Copi, et. al., who looked at data from the Planck satellite to resolve whether the preferred direction and alignments were spurious.[18][19]

(Wikipedia's main Planck probe article makes no mention of the hemispherical asymmetry.)

When can we expect this controversy to be resolved, and are more outcomes possible than (1) the Planck probe data are found to be in error or (2) the Standard Model must undergo major revision?

-
The neutrino fiasco was caused by systematic error (a bad cable connection) being larger than statistical error (the six-sigma thing). The Planck (and COBE and WMAP) team have done their best to eliminate the systematic errors -- mostly related to subtraction of foregrounds. But the hemispheric asymmetry is actually the most sensitive to foreground errors. So I doubt that anyone will be much surprised if it eventually goes away with better models of foregrounds. Meanwhile, it's certainly worth thinking about models that could create such an asymmetry! –  Mike May 30 '13 at 19:39
Hi Eugene, nice question I like this :-) –  Dilaton May 30 '13 at 21:58

One of the Planck papers discusses these anomalies in detail: Planck 2013 results. XXIII. Isotropy and Statistics of the CMB.

How unexpected is this variance from the Standard Model and can it be quantified? How certain is it that the data are accurate? For the recent discovery of the Higgs boson at the LHC, a five sigma result was considered sufficient to make the announcement. What is the sigma for the reported hemispherical asymmetry?

The anomalies found by Planck were already seen in the WMAP data, so they confirm the existence of these features. I can't find the specific sigma for the hemispherical asymmetry, but most of the anomalies are reported to have a significance of ~$3\sigma$. That's enough to raise eyebrows in the astronomical world, although particle physicists would be less impressed :-)

The question is if these anomalies are actually meaningful, and we can expect heated debates in the coming years between 'believers' and 'non-believers'. There are a few things to consider:

• The CMB could be distorted by foreground objects, possibly by our local cluster, our own galaxy and even our solar system.
• The CMB pervades the entire universe, but we can only observe a small part of it: we're seeing a cross-section in the form of a sphere, centred on us and with a certain radius (it took the photons 13.8 billion years to reach us). An observer in a different galaxy and/or at a different cosmic time would see a different part of the CMB.

These effects are sometimes called cosmic variance. Basically, it means that what we observe may not be truly representative of the entire universe.

Also, we're dealing with statistical data, and flukes happen. We're biased at seeing patterns, even though they may not mean anything. For example, if you throw a dice 100 times in a row, then all sorts of apparent patterns may occur. For instance, it could contain the sequence 666666. That specific sequence may seem unlikely and significant, but it is just as likely as any other specific sequence, like e.g. 202020 or 675439.

When can we expect this controversy to be resolved

The Planck team has yet to release the polarisation data (probably next year), which may shed some light on the situation. But I think these issues will entertain the cosmologists for quite a few years. A promising field of research is the mapping of the large-scale distribution of (dark) matter, using gravitational lensing. This would help calculating the Integrated Sachs–Wolfe effect (which is the gravitational redshift/blueshift of CMB photons as they pass through the potential well of galaxy clusters) more accurately.

the Planck probe data are found to be in error

The data is reliable (the Planck data agrees with WMAP), it's all about the interpretation.

the Standard Model must undergo major revision?

It is possible that the explanation lies in a slight deviation from the Standard Model (but no major revision); after all, the Standard Model is an idealisation. Our universe may not be exactly homogeneous or isotropic at large scales. The Planck team actually tried fitting a non-standard model, a so-called Bianchi Model, with mixed results (Planck 2013 results. XXVI. Background geometry and topology of the Universe). Some others get rather carried away, speculating about the influence of 'other universes'.

As a final note, it is important to stress that, in the overall scheme, the impact of these anomalies is very small. The Standard Model fits the CMB almost perfectly, and is in agreement with studies of clusters of galaxies and supernovae.

(Source: Planck 2013 results. I. Overview of products and scientific results, Fig 19)

The fact that the Standard Model works so well and that the cosmological parameters are now known with such accuracy is a very remarkable feat. Thanks to the quality of the data and the success of the theories, cosmologists understand the overall picture with confidence, and can now focus on the details.

P.S. All the Planck papers can be found here: Planck 2013 Results Papers. The most important one is Paper XVI, in which the cosmological parameters are discussed.

-
Your Answer is chock-full with apposite information, more than enough to merit the green check mark! I have only glanced through the two Planck papers on arXiv that you've linked to, will try to read them later. But that graph that you've included does not seem to be in either of them, would you mind telling me where it's from? I take it, then, that you don't place too much importance on the Copernican principle? It's not a "natural law" or basic postulate, so what is its significance for you? A "sanity check" of sorts, akin to Occam's razor, but nothing more? –  Eugene Seidel May 31 '13 at 12:21
@EugeneSeidel The graph is from the first Planck paper. I've added the link, and a bit more, in my answer. As for the Copernican principle, I think it's a very good approximation. But the universe is clumpy on cluster-scales, and this will somewhat affect the light that we observe. The question is how much. Unfortunately we're stuck here on Earth, and can't know what the universe looks like at different locations... –  Pulsar May 31 '13 at 18:04

This difference is tiny and it's very hard to put any sort of "sigma" confidence to the measurement because we only have one Universe to measure. The anisotropy is small enough that it's hard to say if its statistically significant.

Everything I've read (sorry for some reason I can't find the sources right now) suggest that the Planck CMB measurement matches vanilla $\mathrm{\Lambda CDM}$ predictions well enough that nothing stands out as very surprising.

-