1000 TeV neutrinos - dark matter decay proposed as explanation

[marcus]

The IceCube detector in the deep ice at the south pole has seen two instances of 10¹⁵ eV neutrinos. These are said to be the highest energy neutrinos so far seen.
http://resonaances.blogspot.com/2013/09/storm-in-ice-cube.html

What process could have launched such high energy neutrinos? It has been proposed for consideration that the particles result from spontaneous decay of dark matter particles.

For comparison, the LHC is designed to collide particles going at energy of 7 TeV. These neutrinos are going at 1000 TeV.

We seem not to know much about physics at energies in the 1000 TeV range.

I'm curious about this and would be glad if anyone has an enlightening comment to make about it.

 

[Drakkith]

I'm curious about this and would be glad if anyone has an enlightening comment to make about it.

That's a lot of eV's.

 

[fzero]

We really don't know what goes on at a PeV. If we're completely agnostic about philosophical issues of naturalness, etc., then there could be plenty of new particles between collider energies and the Planck scale. Consistency with low-energy physics would tend to imply that they must be weakly interacting with SM particles, so, with reasonable constraints, they could fit into the dark matter sector. The arguments against such models would be the usual complaints about introducing some otherwise-unexplained new scale along with fine-tuned couplings, but it's really up to experiment to rule those out. The existence of new physics at a PeV would also drastically modify some of the more intriguing results in the asymptotic safety program.

An interesting part of coming up with some models would be in understanding if there are other experiments that are reasonably well-suited to studying the same physics. It's also interesting to consider if there are multiple components to DM. For instance, CDMS had a small excess that would suggest a DM scale of around 8 GeV, with other experiments rumored to have excesses at a similar scale.

There could of course be more pedestrian explanations. The background that was discussed in the talk that Jester linked to involved atmospheric sources, but these neutrinos could be extragalactic in origin. One such source, mentioned http://astronomy.nmsu.edu/tharriso/ast536/ast536week10.html, involves the CMB and ultra-high energy cosmic rays:

"CR-CMB: EHE neutrinos above 10¹⁹ eV are produced by photopion production of the 2.7 K cosmic background radiation photons and the ultra-high energy cosmic ray nucleons during propagation in intergalactic space."

So one should definitely compute this background before proposing more exotic explanations.

Finally one should also realize that the detailed analysis has not been published yet, and the statistics are not ironclad (4.3$\sigma$ from atmospherics only) so it is very early to take this too seriously.

 

[marcus]

In case of further interest
==quote==
All these more detailed statements are at this point not very significant statistically, but if they persist after future updates things will get even more interesting. Indeed, sharp spectral features are more tricky to explain in terms of known astrophysical phenomena, which are rather expected to produce a smooth power-law spectrum (typically assumed as 1/Energy^2). For particle physicists the natural question is whether these neutrinos could be the signal of annihilation or decay of dark matter in our galaxy. Dark matter would have to be made of PeV mass particles - more than is typically considered. But why not: any mass between sub-eV and Planck scale is equally plausible at this point.

Of course, theorists have models for every occasion up their sleeves. One interesting proposal is that the two PeV events can be interpreted as a monochromatic neutrino line. This could arise if dark matter decays into a 2-body final state containing at least one neutrino (dark matter annihilation with a large enough cross section is difficult to realize for such a heavy mass). To explain the observation of 2 events in IceCube, the life-time of dark matter should be about 10^28 seconds or 10^21 years - much longer than the age of the Universe.
This paper shows that a general scenario of this type can fit the data very well, somewhat better than the smooth 1/E^2 spectrum.

The presence of the continuum excess around 100 TeV makes the picture more complicated... but not necessarily unrealistic. It is perfectly conceivable that dark matter has more than one decay channels, for example 10% branching fraction into 2 neutrinos, and the remaining 90% into quarks. The former would be responsible for the PeV events, and the latter would produce neutrinos with smaller energies...which would explain the broad feature near 100 TeV.
==endquote==

 

[Nugatory]

The IceCube detector in the deep ice at the south pole has seen two instances of 10¹⁵ eV neutrinos.

Hmmm... That's about 1.5x10^-4 Joules... Or a paperclip moving at .5 m/sec...

Not quite an OMG particle, but not bad for a lepton.

 

[Chronos]

Or maybe, its just an anomaly.

 

[Chalnoth]

Hmm. Well, if it's DM decay, that would require dark matter masses of at least a few PeV. This would be wildly inconsistent with the tentative detections of dark matter we've seen so far (though those detections are themselves inconsistent with one another, so that's not a huge deal).

I guess I'm mostly wondering why this can't just be a neutrino left over from a very high-energy collision, like one of those 10^20 eV+ protons we've detected.

 

[marcus]

I neglected to link directly to the announcement by the IceCube collaboration Here's the paper:
http://arxiv.org/abs/1304.5356
First observation of PeV-energy neutrinos with IceCube
We report on the observation of two neutrino-induced events which have an estimated deposited energy in the IceCube detector of 1.04 ± 0.16 and 1.14 ± 0.17 PeV, respectively, the highest neutrino energies observed so far. These events are consistent with fully contained particle showers induced by neutral-current ν_e,μ,τ (ν ̄_e,μ,τ ) or charged-current ν_e (ν ̄_e) interactions within the IceCube detector.

The events were discovered in a search for ultra-high energy neutrinos using data corresponding to 615.9 days effective livetime. The expected number of atmospheric background is 0.082 ± 0.004(stat)^+0.041_-0.057 (syst). The probability to observe two or more candidate events under the −0.057 atmospheric background-only hypothesis is 2.9 × 10⁻³ (2.8σ) taking into account the uncertainty on the expected number of background events. These two events could be a first indication of an astrophysical neutrino flux, the moderate significance, however, does not permit a definitive conclusion at this time.

 

[marcus]

I shouldn't be relying on second-hand blog commentary when I could be citing primary sources. Here is a followup paper posted in August with some enlightening discussion
===
http://arxiv.org/abs/1308.1105
Are IceCube neutrinos unveiling PeV-scale decaying dark matter?
Arman Esmaili (Instituto de Fisica - University of Campinas, Brazil)
Pasquale Dario Serpico (LAPTh, Univ. de Savoie, CNRS, France)

Recent observations by IceCube, notably two PeV cascades accompanied by events at energies ∼ (30 − 400) TeV, are clearly in excess over atmospheric background fluxes and beg for an astroparticle physics explanation. Although some models of astrophysical accelerators can account for the observations within current statistics, intriguing features in the energy and possibly angular distributions of the events make worth exploring alternatives. Here, we entertain the possibility of interpreting the data with a few PeV mass scale decaying Dark Matter, with lifetime of the order of 10²⁷ s. We discuss generic signatures of this scenario, including its unique energy spectrum distortion with respect to the benchmark E_ν⁻² expectation for astrophysical sources, as well as peculiar anisotropies. A direct comparison with the data show a good match with the above-mentioned features. We further discuss possible future checks of this scenario.
===

By my (non-expert) standards the Esmaili-Serpico paper is thoughtful and carefully written--the clearest discussion of these events I've seen so far. We can appreciate the wise and polite restraint shown in: "we entertain the possibility of interpreting..." Maybe I'll quote some passages from this paper and hope for further comment by others.

Anyone at all interested in how the PeV neutrinos might be interpreted is invited to look at FIGURE 1 ON PAGE 3 of this paper. It shows the high energy neutrino flux one would EXPECT if dark matter were to decay in two modes (88% probability decay into quark antiquark pair, 12% probability decay into neutrino antineutrino pair) and if, as well, the DM particle has a certain order of mass and approximate lifetime. This is of course merely a MODEL based on as yet statistically insignificant observation: the odd coincidence that both of the neutrinos had approximately the same one PeV energy and that they are the sole examples of neutrinos traveling with such high energy.

 

[Haelfix]

The other theoretical reason to doubt this result is that the Wimp miracle becomes something more like the Wimp coincidence.

Its always possible to solve dark matter by abandoning a thermal history and cranking up the mass scale (ignoring naturalness, etc). Somewhat more interesting is that these types of solutions (at certain mass scales) tend to clash with typical models of baryogenesis.

Far more likely is that this is just another misundersttood astrophysics background.

 

[marcus]

Or maybe, its just an anomaly.

...I guess I'm mostly wondering why this can't just be a neutrino left over from a very high-energy collision, like one of those 10^20 eV+ protons we've detected.

...Far more likely is that this is just another misundersttood astrophysics background.

Hi Haelfix, glad to see you agree with the general sense of Chronos/Chalnoth comments and are moreover ready to judge the relative LIKELIHOOD of various explanations.

The comment I've seen elsewhere suggests that IF they see more such events at this 1 PeV energy then it could be interesting either way. An astrophysical explanation (instead of DM decay) would itself be interesting because of the high energy. And of course DM decay would be so as well (very!).

At this point I myself can't compare how "likely" various outcomes are, but can simply take note of the puzzle. But it's nice to see a block of opinion coagulate.

 

[Astronuc]

I'd want to know what (event) creates a "PeV neutrino." If it is a decay or annihilation, then there is something more energetic? Is Dark Matter, that energetic? Should I (we) assume yes since we consider cosmic particle energies > 1E15 eV.

http://adsabs.harvard.edu/full/1985ICRC...6..348A

 

[fzero]

I'd want to know what (event) creates a "PeV neutrino." If it is a decay or annihilation, then there is something more energetic? Is Dark Matter, that energetic? Should I (we) assume yes since we consider cosmic particle energies > 1E15 eV.

http://adsabs.harvard.edu/full/1985ICRC...6..348A

There is an important process called photopion production, $p \gamma_\mathrm{CMB} \rightarrow N \pi$, that affects ultra-high energy protons as they pass through the CMB. The resulting charged pions can decay to a positron and 3 neutrinos. This process has been conjectured to make the CMB completely opaque to protons with sufficiently high energies (around 10²⁰ eV).

It turns out that the ICECUBE paper considered a background of neutrinos produced in this way, based on the theoretical analysis in arXiv:1005.2620. Apparently the background (called cosmogenic in the ICECUBE paper) was for around 2 events, but these would have tended to come at higher than PeV energies. The way the paper is written, it's not clear to me whether or not this background is included in the quoted values for the statistical significance.

For the explanation of these events via the decay of a conjectured new particle, yes the DM particle would have to have a mass higher than the energy of the observed neutrinos. The Esmaili and Serpico paper that marcus cited gives more details about the kinematics for some conjectured decay channels. There is no particular theoretical reason to expect DM particles to be so massive, since it basically introduces a new unexplained scale.

 

[marcus]

I'd want to know what (event) creates a "PeV neutrino." If it is a decay or annihilation, then there is something more energetic?...

Exactly.

FWIW the event which Esmaili Serpico consider ("entertain the possibility" of) is very simple. A DM particle with mass about 2 PeV decays by splitting into a neutrino and an antineutrino.

The DM particle is "cold"---i.e. we can neglect its kinetic energy. Its mass-energy is almost completely converted into the kinetic energy of the two neutrinos.

 

[Chalnoth]

There is an important process called photopion production, $p \gamma_\mathrm{CMB} \rightarrow N \pi$, that affects ultra-high energy protons as they pass through the CMB. The resulting charged pions can decay to a positron and 3 neutrinos. This process has been conjectured to make the CMB completely opaque to protons with sufficiently high energies (around 10²⁰ eV).

It turns out that the ICECUBE paper considered a background of neutrinos produced in this way, based on the theoretical analysis in arXiv:1005.2620. Apparently the background (called cosmogenic in the ICECUBE paper) was for around 2 events, but these would have tended to come at higher than PeV energies. The way the paper is written, it's not clear to me whether or not this background is included in the quoted values for the statistical significance.

For the explanation of these events via the decay of a conjectured new particle, yes the DM particle would have to have a mass higher than the energy of the observed neutrinos. The Esmaili and Serpico paper that marcus cited gives more details about the kinematics for some conjectured decay channels. There is no particular theoretical reason to expect DM particles to be so massive, since it basically introduces a new unexplained scale.

I'm also wondering if it makes sense to consider collisions of extremely high-energy particles on the Earth as being a background source for these neutrinos (not just the CMB). The expected frequency might be too low, but I would tend to think that a collision of a $10^{20}$+ eV proton with the Earth would produce quite a few neutrinos in the PeV range.

 

[marcus]

It would be great if more of us could have a look at the Esmaili Serpico I recommended in post #8. Especially FIGURE 1 ON PAGE 3. We (or rather the experts whose specialty it is) have to explain the GAP in neutrino energies between 400 and 1000 TeV, AS WELL AS the two neutrinos with PeV energy.

Yes we see two neutrinos with nearly the same (PeV) energy.That is the first coincidence.

The second coincidence is we do NOT see a random assortment of neutrinos in the 400-1000 TeV range.

But then there is a heap of events below 400. So that's what the experts are trying to model, in case that pattern persists as more data accumulates.

..
===
http://arxiv.org/abs/1308.1105
Are IceCube neutrinos unveiling PeV-scale decaying dark matter?
Arman Esmaili (Instituto de Fisica - University of Campinas, Brazil)
Pasquale Dario Serpico (LAPTh, Univ. de Savoie, CNRS, France)

Recent observations by IceCube, notably two PeV cascades accompanied by events at energies ∼ (30 − 400) TeV, are clearly in excess over atmospheric background fluxes and beg for an astroparticle physics explanation. Although some models of astrophysical accelerators can account for the observations within current statistics, intriguing features in the energy and possibly angular distributions of the events make worth exploring alternatives. Here, we entertain the possibility of interpreting the data with a few PeV mass scale decaying Dark Matter, with lifetime of the order of 10²⁷ s. We discuss generic signatures of this scenario, including its unique energy spectrum distortion with respect to the benchmark E_ν⁻² expectation for astrophysical sources, as well as peculiar anisotropies. A direct comparison with the data show a good match with the above-mentioned features. We further discuss possible future checks of this scenario.
===
...

 

[fzero]

I'm also wondering if it makes sense to consider collisions of extremely high-energy particles on the Earth as being a background source for these neutrinos (not just the CMB). The expected frequency might be too low, but I would tend to think that a collision of a $10^{20}$+ eV proton with the Earth would produce quite a few neutrinos in the PeV range.

It is a possible background, but would tend to have a noticable angular distribution favoring the Earth-side of the detector.

There is a recent paper that suggests that the IceCube results are almost completely in agreement with the SM once uncertainties in the parton distribution functions involved in the interaction of the neutrinos with the detector are taken into account: arXiv:1309.1764. These authors acknowledge discussions with four of the theorists in the IceCube collaboration and suggest that:

"to the best of our knowledge, no attempt has been made so far to quantify the PDF uncertainty effects on the number of events expected in each of the deposited energy bins at IceCube."

They predict 2 events in the 0.3 - 1 PeV range that could appear as more data is collected.

 

[marcus]

...arXiv:1309.1764. These authors acknowledge discussions with four of the theorists in the IceCube collaboration and suggest that:

"to the best of our knowledge, no attempt has been made so far to quantify the PDF uncertainty effects on the number of events expected in each of the deposited energy bins at IceCube."

They predict 2 events in the 0.3 - 1 PeV range that could appear as more data is collected.

That's an interesting paper. Thanks! It helps emphasize that the key thing is whether the 0.3-1 PeV gap persists or is filled in as more data accumulates.

Esmaili Serpico say if the (gap) pattern persists then [such and such hypothesis] could explain it.

BTW as I recall, Esmaili recently co-authored a paper with Francis Halzen, one of the IceCube collaboration. I wouldn't say he is BETTING one way or another. As I see it, we simply have to see if the gap persists. It will be interesting either way.

Your http://arxiv.org/abs/arXiv:1309.1764 authors are, I believe, not members of IceCube, but as you point out they have conversed with some members of the collaboration. As I read it, the takehome message is consistent with what Esmaili Serpico say namely

If the gap pattern does not persist then E-S hypothesis fails and the PeV neutrinos could even be explainable without new physics within Standard theory.

BTW there is a nice explanatory diagram of IceCube here:
http://arxiv.org/pdf/1308.3171v1.pdf

 

[mfb]

It is a gap only for components that quickly drop with energy. There would be no gap if there is a broad component up to at least 1000 TeV. If you expect two events between 100 and 1500 TeV (as an example), finding those two events would be perfectly fine. This leaves the question where such a background could come from, of course.

It turns out that the ICECUBE paper considered a background of neutrinos produced in this way, based on the theoretical analysis in arXiv:1005.2620. Apparently the background (called cosmogenic in the ICECUBE paper) was for around 2 events, but these would have tended to come at higher than PeV energies. The way the paper is written, it's not clear to me whether or not this background is included in the quoted values for the statistical significance.

So in addition to two events with unknown source, there is a source with two "missing" events? Okay, that could be a statistical fluctuation.

 

[marcus]

Since we're on a new page, I'll bring forward the abstracts of the papers forming the topic of discussion. Anyone new to the thread might be interested to note that the two highest energy neutrinos ever observed (and the only neutrinos ever seen with energies above 300-400 TeV) are very CLOSE to each other in energy. I've highlighted that in the abstract, since it may be significant.
http://arxiv.org/abs/1304.5356
First observation of PeV-energy neutrinos with IceCube
We report on the observation of two neutrino-induced events which have an estimated deposited energy in the IceCube detector of 1.04 ± 0.16 and 1.14 ± 0.17 PeV, respectively, the highest neutrino energies observed so far. These events are consistent with fully contained particle showers induced by neutral-current ν_e,μ,τ (ν ̄_e,μ,τ ) or charged-current ν_e (ν ̄_e) interactions within the IceCube detector.

The events were discovered in a search for ultra-high energy neutrinos using data corresponding to 615.9 days effective livetime. The expected number of atmospheric background is 0.082 ± 0.004(stat)^+0.041_-0.057 (syst). The probability to observe two or more candidate events under the −0.057 atmospheric background-only hypothesis is 2.9 × 10⁻³ (2.8σ) taking into account the uncertainty on the expected number of background events. These two events could be a first indication of an astrophysical neutrino flux, the moderate significance, however, does not permit a definitive conclusion at this time.
===
http://arxiv.org/abs/1308.1105
Are IceCube neutrinos unveiling PeV-scale decaying dark matter?
Arman Esmaili (Instituto de Fisica - University of Campinas, Brazil)
Pasquale Dario Serpico (LAPTh, Univ. de Savoie, CNRS, France)

Recent observations by IceCube, notably two PeV cascades accompanied by events at energies ∼ (30 − 400) TeV, are clearly in excess over atmospheric background fluxes and beg for an astroparticle physics explanation. Although some models of astrophysical accelerators can account for the observations within current statistics, intriguing features in the energy and possibly angular distributions of the events make worth exploring alternatives. Here, we entertain the possibility of interpreting the data with a few PeV mass scale decaying Dark Matter, with lifetime of the order of 10²⁷ s. We discuss generic signatures of this scenario, including its unique energy spectrum distortion with respect to the benchmark E_ν⁻² expectation for astrophysical sources, as well as peculiar anisotropies. A direct comparison with the data show a good match with the above-mentioned features. We further discuss possible future checks of this scenario.

 

[marcus]

There is a particularly nice short description of IceCube for wide-audience by Francis Halzen
http://arxiv.org/abs/1308.3171
Who is Francis Halzen?
http://experts.news.wisc.edu/experts/250
He's one of the senior people in the collaboration, the U Wisconsin website lists him as IceCube principal investigator

Incidentally Halzen co-authored an IceCube phenomenology paper with Arman Esmaili earlier this year:
http://arxiv.org/abs/1303.3294
Exploring ν_τ − ν_s mixing with cascade events in DeepCore
Arman Esmaili , Francis Halzen , O. L. G. Peres
The atmospheric neutrino data collected by the IceCube experiment and its low- energy extension DeepCore provide a unique opportunity to probe the neutrino sector of the Standard Model. In the low energy range the experiment have observed neutrino oscillations, and the high energy data are especially sensitive to signatures of new physics in the neutrino sector. In this context, we previously demonstrated the unmatched potential of the experi- ment to reveal the existence of light sterile neutrinos. The studies are routinely performed in the simplest 3 + 1 model concentrating on disappearance of muon neutrinos of TeV energy as a result of their mixing with a sterile neutrino. We here extend this analysis to include cascade events that are secondary electromagnetic and hadronic showers produced by neutri- nos of all flavors. We find that it is possible to probe the complete parameter space of 3 + 1 model, including the poorly constrained mixing of the sterile neutrino to tau neutrinos. We show that ν_τ − ν_s mixing results into a unique signature in the data that will allow IceCube to obtain constraints well below the current upper limits.