What's Delaying Fermilab's Muon g-2 Results Release?

[ohwilleke]

TL;DR Summary

Fermilab's E989 experiment, which is measuring the anomalous magnetic moment of the muon, was due to report its initial results in 2020 but didn't. Do we know when it will announce them?

Fermilab's E989 experiment is conducting the first precision measurement of the anomalous magnetic moment of the muon (muon g-2) since the Brookhaven lab did so fifteen years ago. It is currently collecting Run-3 data for this experiment, and said that it would be releasing preliminary Run-1 results in late 2020 at a September 2020 physics conference. See https://arxiv.org/abs/2009.07709

The discrepancy between the Brookhaven measurement and the theoretically predicted value of that measurement (which is a little more than three sigma) is one of the most important discrepancies between the Standard Model of Particle Physics and experiment. If the experimentally measured value in the Bookhaven experiment stayed the same, the result would be seven sigma proof of new physics. If the new measurement matched the theoretically predicted value, this would be a global measurements that would strongly suggest that the Standard Model is a complete and accurate description of low energy physics. So, everyone's dying to know what they've found (J-PARC's E34 experiment is also measuring the same thing, but its results won't be available for one to three more years, because Fermilab is just that awesome by comparison).

Well, it's the year 2021 now and no results have been released.

Does anyone know or have a good idea regarding what is going on and delaying the release of these results?

If the announcement that results would be released in late 2020 had been made in January 2020, the obvious conclusion would have been to blame COVID-19. But in the case of a planned late 2020 release stated in September of 2020, that doesn't seem like as likely an explanation.

For example, is there any reason to think that a particularly remarkable results is causing the scientists involved to delay releasing the results because they want to do an extra check of their accuracy first? Or, did some catastrophe that didn't make the news hit Fermilab?

 

[Vanadium 50]

That sounds a lot like a rant that science should be done on your schedule, not the scientists'.

When Will Fermilab Release Its Initial Muon g-2 Measurement?

• It's not Fermilab. It's the collaboration.
• The collaboration will release it when they are ready.

They have no incentive to release a result that isn't ready. They have no incentive to sit on a result that is and not make it public.

is there any reason to think that a particularly remarkable results is causing the scientists involved to delay releasing the results because they want to do an extra check of their accuracy first

You are accusing the collaboration of poor scientific and borderline unethical behavior. Got any evidence?

 

[ChrisVer]

accusing the collaboration of poor scientific and borderline unethical behavior

Ehm, I disagree with this conclusion of yours. In my opinion, it is totally ethical and scientific to delay a publication to make sure you are confident in your results, rather than publishing something you are uncertain about. Scientific work needs its time, and a couple of months delays is not so rare (especially for precision measurements).

Now to the OP and regarding the rest, I doubt they can be answered if you are not associated with the collaboration to have internal information. And even if you are, I don't think you have the right to speak publicly about it. So you can only make wild guesses and that's not taking you anywhere. The only thing you can do is be highly confident (although not 100% as it involves humans) that when something comes out, it wouldn't be a mistake.

However, I would wonder, if the anomaly persists, would that mean definitely the presence of physics beyond the standard model, or would it mean something is missing in the calculations? It is something I hear often been said about the b-anomalies too.

 

[Vanadium 50]

, it is totally ethical and scientific to delay a publication to make sure you are confident in your results, rather than publishing something you are uncertain about.

I agree. But g-2 has stated many times that their experiment is blinded. And once you have unblinded, you're not allowed to then go back and fiddle with the analysis. Especially based on the outcome.

 

[ohwilleke]

I agree. But g-2 has stated many times that their experiment is blinded. And once you have unblinded, you're not allowed to then go back and fiddle with the analysis. Especially based on the outcome.

But, if the unblinded data reveals what looks like a goofy problem, like the one that caused the superluminal neutrino speed result from the OPERA experiment a while back, it wouldn't be improper to try to look for an explanation before publishing and to include that investigation in the final product. "Yes, our Run-1 raw data was off by a factor of 3% in a parts per million precision measurement, but we think we've identified the Xcel spreadsheet glitch that caused that problem." (I'm alluding to an incident like that in an Ethiopian ancient DNA paper a couple of years ago where that was discovered a few months after publication and resulted in a partial retraction and revision of one of the several key findings.)

Now to the OP and regarding the rest, I doubt they can be answered if you are not associated with the collaboration to have internal information. And even if you are, I don't think you have the right to speak publicly about it.

What I was mostly hoping for was that someone might have been aware of an authorized statement through some channel I am not privy to, not published in arXiv or at the collaboration website I located (which hadn't been updated since June 2020), that publication has been delayed until X date for Y reason. Physics Forums has lots of people who read Twitter feeds and blogs and hear unpublished conference presentations, etc. that I might not know about, from scientists who are involved in experiments like these, which is the kind of forum in which I wouldn't be surprised to see an announcement about something like this.

Once the collaboration publicly states an expected publication date and then doesn't deliver, it invites speculation that something is amiss. Best practices in PR (both in and out of science) is to make an announcement if that is going to happen, a revised estimate (possibly vague) and if possible, a statement of the often innocuous reason for the delay. At a minimum, something like an announcement on the collaboration website that: "We weren't able to publish Run-1 results in December 2020 as expected, but do expect to have published results sometime in 2021. This was due to unavoidable delays in the publication process."

Certainly, there are lots of reasons that announced publication dates get moved back. Some are innocuous, while others are interesting, either because they hint at something about the expected results, or because they tell outsiders something about the day to day reality of the scientific process unrelated to the results themselves.

For example, if the delay were due to delay in Congressional approval of the new fiscal year's appropriation bill that was funding the collaboration, or due to a major post-snow storm flood that delayed dozens of projects with apparatus at the same location, that would be interesting to know, and would expand familiarity with what life is like trying to lead a collaboration like this one.

Certainly, it would be commonplace for a collaboration to compel its members not to disclose the substance of what will be published prior to publication. But it would be considerably less common for a collaboration to prohibit its members from discussing the revised timing of a previously announced publication date target in a manner that does not disclose what the publication itself will contain, or to prohibit collaborators from discussing some intervening cause for a delay (e.g. maybe their supercomputer resources were pre-empted to do time sensitive COVID work, or a key investigator who has to approve the final product had to take a leave of absence due to a death or illness in the family or new child that was born prematurely or a revised spring semester academic calendar), even if the cause of the delay isn't something the the collaboration would want to put in a press release. There are things that are both not secret and also not something that somebody actively wants to broadcast to the world.

 

[Vanadium 50]

But, if ... <long train of evidence-free speculation and innuendo>

 

[Dr.AbeNikIanEdL]

I think you are seriously

a) overestimating and overstating the level of commitment anyone has made to definitely publish something in December 2020. I don‘t see a definitive publication date stated anywhere that could be officially corrected.

b) underestimating how little it takes to delay a publication, in particular with so many people involved, by weeks and months. (Probably especially if there is no external pressure like competing experiments etc.)

 

[sysprog]

I've spent more time at Argonne than at Fermilab, but according to my understanding of history, all of the guys at one University of Chicago location allowed that the cadmium rods were raised exactly as Prof. Fermi, with his slide rule and mechanical pencil (Prof. Fermi sometimes would write interim results on his slide rule) directed $-$ anyway, I think that even though everyone in that room was brilliant, there may have been a local-to-there consensus that Enrico Fermi was the smartest man in the room $-$ according to history, there was no doubt that he was in charge of the experiment $\dots$

 

[exponent137]

When I put in Google:
"announcement" "e989" "muon"
and a parameter for the last year,
I obtained one of the results in Twitter in one Asian language.
It is not necessary to open, it is written:
FNAL-E989 first announcement in February 2021!
But all other is written in an Asian language. Does someone understand this? Does this have any sense? We'll see in February.

 

[ohwilleke]

When I put in Google:
"announcement" "e989" "muon"
and a parameter for the last year,
I obtained one of the results in Twitter in one Asian language.
It is not necessary to open, it is written:
FNAL-E989 first announcement in February 2021!
But all other is written in an Asian language. Does someone understand this? Does this have any sense? We'll see in February.

You rock! Exactly what I was hoping for.

 

[exponent137]

Fresh news about the announcement:
https://news.fnal.gov/tag/muon-g-2/

 

[ohwilleke]

Fresh news about the announcement:
https://news.fnal.gov/tag/muon-g-2/

They key sentence in the article is:

As early as March, the Muon g-2 experiment at Fermi National Accelerator Laboratory (Fermilab) will report a new measurement of the magnetism of the muon, a heavier, short-lived cousin of the electron.

The funniest sentence in the article, however, is:

"My wife won’t pick me for responsible jobs like this, so I don’t know why an important experiment did,” says Joseph Lykken, Fermilab’s chief research officer, one of the keepers of the secret.

 

[MathematicalPhysicist]

Fresh news about the announcement:
https://news.fnal.gov/tag/muon-g-2/

I like how Marciano is quoted:

“It’s not impossible to explain [the muon’s magnetism] with supersymmetry,” Marciano says, “but you have to stand on your head to do it.”“It’s not impossible to explain [the muon’s magnetism] with supersymmetry,” Marciano says, “but you have to stand on your head to do it.”

something like this smiley ->

 

[exponent137]

New news about the announcement of the Fermilab experiment:
https://theory.fnal.gov/events/event/first-results-from-the-muon-g-2-experiment-at-fermilab/
Now the date (April 7) is fixed and official.

 

[ohwilleke]

Hurray! Thanks.

 

[Vanadium 50]

a is defined as 2(g-2)
The latest calculation is a = 116591810(43) E-11
The BNL measurement was a = 116592089(63) E-11
The latest experimental result is a = 116592040 (54) E-11

Combined is a = 116592061 (43) E-11, 4.2 sigma from the SM

 

[ohwilleke]

It was memorable to be among the 5,000 people on the Zoom announcement. I'll remember that for a long time. Thanks to PF for being able to track it down.

So, now it is onto the academic show down between the BMW group calculation of the leading order HVP published in Nature today (heavily overlapping pre-print here ) together with the new hadronic light by light calculations pre-print from today, which combined, produce a SM prediction within 1.3 sigma of the combined result (a solid confirmation of the SM prediction that even 4x increased accuracy in the Fermilab results wouldn't shake), versus the muon g-2 Theory Initiative paper (published at Phys. Rept. 887 (2020) 1-166) referenced by Fermilab in its presentation which shows a 4.2 sigma discrepancy from the combined result that could grow to 7 sigma by the time that all of the data are in hand.

The stakes are high.

If BMW is right, the likelihood any New Physics in the High Energy Physics domain of the Standard Model, at energy scales attainable in our lifetimes, just went poof (time to focus on BSM astronomy phenomena instead).

If the Theory Initiative is right, New Physics are just around the corner and the existence of new physics will be a five sigma discovery before my sophomore son in a STEM major in college graduates even though we don't know precisely what those new physics are, only how big the New Physics are in magnitude.

The the E34 experiment at J-PARC will be providing a methodologically very different experimental determination of muon g-2 and other stuff within a decade (there is always potential for invisible shared systemic errors when Fermilab and Brookhaven National Labs are using literally the same physical storage ring moved from one lab to the other and also retained experimental equipment from BNL at the core of their experimental apparatus).

But I think the conventional wisdom is now that Brookhaven and Fermilab have a correct experimental measurement , that E34 is expected to replicate them closely, and that the real question now is what the correct theoretical Standard Model prediction is for the value of muon g-2.

Given the SM's epic track record of experimental confirmation, and the seemingly very sensible steps that BMW took to get the result that they did relative to the muon g-2 Theory Initiative, it seems very likely to me that BMW is right, or that, at a minimum, even if BMW isn't definitively correct, that the Theory Initiative has greatly understated the actual amount of uncertainty in their theoretical prediction (which would return us to an uncertain limbo).

 

[MathematicalPhysicist]

If BMW is right, the likelihood any New Physics in the High Energy Physics domain of the Standard Model, at energy scales attainable in our lifetimes, just went poof (time to focus on BSM astronomy phenomena instead).

Not sure, if China's ambitious collider will work we might witness new BSM, just need to wait for it more approximately 9 more years.
https://www.nature.com/articles/d41586-018-07492-w

 

[ohwilleke]

Not sure, if China's ambitious collider will work we might witness new BSM, just need to wait for it more approximately 9 more years.
https://www.nature.com/articles/d41586-018-07492-w

The point is that if muon g-2 is experimentally identical to the SM calculation, that there probably isn't any new physics to find at experimentally accessible energies at new, higher energy colliders, even if we build them, because muon g-2 is a global measure of beyond the Standard Model Physics that implicates all of its forces and almost all of its particles to some degree. It is possible to imagine BSM physics that don't tweak muon g-2 at all in the unexplored region that a new collider could access, but it takes much more byzantine machinations in the theory to make that happen, while the vast majority of actively sought BSM physics proposals would tweak muon g-2.

 

[MathematicalPhysicist]

The point is that if muon g-2 is experimentally identical to the SM calculation, that there probably isn't any new physics to find at experimentally accessible energies at new, higher energy colliders, even if we build them, because muon g-2 is a global measure of beyond the Standard Model Physics that implicates all of its forces and almost all of its particles to some degree. It is possible to imagine BSM physics that don't tweak muon g-2 at all in the unexplored region that a new collider could access, but it takes much more byzantine machinations in the theory to make that happen, while the vast majority of actively sought BSM physics proposals would tweak muon g-2.

I am looking at a test that I failed in an advanced course in Particle theory 2.
Can you tell me how much does the energy of $105\times 10^5 GeV$ exceed the currently capabilities of particle's accelerators?

 

[ohwilleke]

I am looking at a test that I failed in an advanced course in Particle theory 2.
Can you tell me how much does the energy of $105\times 10^5 GeV$ exceed the currently capabilities of particle's accelerators?

About 750x.

 

[MathematicalPhysicist]

About 750x.

Ah ok then it's not that much a lot of energy. I thought something enormous; he gave me a remark: "from where does such large amount of energy come from?"

 

[ohwilleke]

he gave me a remark: "from where does such large amount of energy come from?"

In China, it comes from coal fired power plants with very few emissions controls.

 

[MathematicalPhysicist]

In China, it comes from coal fired power plants with very few emissions controls.

Terrific! oldschool is da best school!

 

[mfb]

If there are two SM predictions and only one agrees with measurements...
At least we will learn more about SM predictions of hadronic effects.

CEPC in China has been a proposal for quite some time now, without real progress because it doesn't have funding. It's not expected to find new particles in electron-positron collisions, but it could study the Higgs more precisely than the LHC can. A future change to a proton-proton collider might find new particles directly.

Reaching 750 times the energy with the LHC technology would need a ring 750 times as large, ~20,000 km circumference. Even if you double the field strength it's still 10000 km. Europe doesn't have space for that, but in North America it could just fit between Hudson Bay, Mexico, Washington and Washington DC. At least if we ignore all financial and technical problems.

 

[MathematicalPhysicist]

If there are two SM predictions and only one agrees with measurements...
At least we will learn more about SM predictions of hadronic effects.

CEPC in China has been a proposal for quite some time now, without real progress because it doesn't have funding. It's not expected to find new particles in electron-positron collisions, but it could study the Higgs more precisely than the LHC can. A future change to a proton-proton collider might find new particles directly.

Reaching 750 times the energy with the LHC technology would need a ring 750 times as large, ~20,000 km circumference. Even if you double the field strength it's still 10000 km. Europe doesn't have space for that, but in North America it could just fit between Hudson Bay, Mexico, Washington and Washington DC. At least if we ignore all financial and technical problems.

Well i don't mind. I don't have the money for that...:-)

 

[MathematicalPhysicist]

If there are two SM predictions and only one agrees with measurements...
At least we will learn more about SM predictions of hadronic effects.

CEPC in China has been a proposal for quite some time now, without real progress because it doesn't have funding. It's not expected to find new particles in electron-positron collisions, but it could study the Higgs more precisely than the LHC can. A future change to a proton-proton collider might find new particles directly.

Reaching 750 times the energy with the LHC technology would need a ring 750 times as large, ~20,000 km circumference. Even if you double the field strength it's still 10000 km. Europe doesn't have space for that, but in North America it could just fit between Hudson Bay, Mexico, Washington and Washington DC. At least if we ignore all financial and technical problems.

BTW isn't Mexico in Central America?

Well you can also do a combination of Europe and Asia, don't they have the space for that?

 

[AndreasC]

In China, it comes from coal fired power plants with very few emissions controls.

Mostly unrelated but the good news is that China is doing decently at converting to renewables etc. They've exceeded 26% renewable now even though they only had about half of that a couple decades ago, so that's comforting at least!

 

[AndreasC]

BTW isn't Mexico in Central America?

Well you can also do a combination of Europe and Asia, don't they have the space for that?

Hmm, well I'm not sure if the question matters very much lol but I'm pretty sure you could also fit it in Europe if you used the sea as well, however there's another issue, which is the curvature of the Earth (beyond the obvious limitations of how expensive the material and the energy required would be). At that distance it would be a height difference of a few kilometers I believe. So you can file that under "not going to happen".

 

[MathematicalPhysicist]

Hmm, well I'm not sure if the question matters very much lol but I'm pretty sure you could also fit it in Europe if you used the sea as well, however there's another issue, which is the curvature of the Earth (beyond the obvious limitations of how expensive the material and the energy required would be). At that distance it would be a height difference of a few kilometers I believe. So you can file that under "not going to happen".

I have patience...

 

[pbuk]

This is generating a huge amount of noise in the mainstream press in the UK: national news last night and a feature on breakfast television this morning!

Batten down the hatches Mentors, I see a storm on its way...

 

[pbuk]

there's another issue, which is the curvature of the Earth

No, it's a circle. Think about it.

 

[AndreasC]

No, it's a circle. Think about it.

Oh fair.

 

[vanhees71]

Well, according to a lattice calculation by the Wuppertal group on the leading hadronic contributions, maybe the SM prediction is again closer to the new measurement:

https://www.nature.com/articles/s41586-021-03418-1

It seems as if $(g_{\mu}-2)$ stays exciting also from the theory side!

 

[gmax137]

BTW isn't Mexico in Central America?

Not that it is important to this conversation, but no.

It is bordered by Mexico to the north, Colombia to the southeast, the Caribbean Sea to the east and the Pacific Ocean to the west and south. Central America consists of seven countries: Belize, Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua and Panama.

Also recall NAFTA the "North American Free Trade Agreement" between Canada, US, and Mexico.