Mystery of strange radio bursts from space

[Garth]

The fact it is Parkes only is pretty discouraging though.

One, 121102, was detected at Arecibo.

Some press reports have been inaccurate.

Garth

 

[Jimster41]

In the second paper, in figure 7. I don't see where they measured the randomness of the FRB receptions in time. The say it is "uniform". Looks a bit shaky as "uniform" to my eye. But given the fact the telescope sweeps with the Earth's rotation as "the clock" (??) what would pure randomness in reception time, or a lack thereof indicate?

 

[rootone]

The more random the measurements are the less likely they are to contain coherent information or be correlated with anything in particular.

 

[Jimster41]

I understand that part.

I was wondering what it would mean if a non-random signal was distributed randomly as a function of the Earth's rotation (like CMB). Or what it would mean if there was a non-random pattern to intervals of reception, one that for instance wasn't diurnal. Is Parkes like Aricebo? Can it be aimed much? Can Aricebo be aimed much for that matter? Are the measurements a sweep of some arc minutes, or are they really focused on a region of the sky. I would think they would not be very interested in pointing toward the sun, but... I'm clueless what such a big radio telescope is sampling in terms of diffuse, precise, periodic, constant...

anyway pretty cool stuff.

 

[wabbit]

Sorry, the DM (cm e-3 pc) column.

I think they mention it on page 2

The result is a striking 1:10,000 chance for the first option, and 5:10,000 for the second. In the random numbers, we also see average summed DM residuals >500, as expected for the series following 1/3, 1/5, 1/7...

Honestly I have no idea what you are talking about.

 

[Jimster41]

Honestly I have no idea what you are talking about.

Sorry, I'm not sure how to interpret that. Do you want me to try to explain what I was thinking, or just shut up... because it is clear to you it makes no sense?

Here's a plot of the FRB reception times from figure 7 in the second paper compared to some different periodic models... values are the average of the hourly difference between the "periodic" test distribution and the actual distribution.

 

[wabbit]

Explain : ) you mentioned prime numbers, one column which clearly isn t composed of prime numbers, and something else - i just don t know what it is exactly that you find surprising - what statement is true of which data , that you find unusual? Is it just the the first digit of these numbers is odd ?

 

[Jimster41]

Oh, okay. But I probably should just shut up

I was looking at the third column from the left in table 1 on page two of the Hippke paper http://arxiv.org/pdf/1503.05245v2.pdf.
I was just a bit surprised to see such non-random odd rather than even or mixed odd/even numbers. At first they looked like primes and I had a hiccup, then realized 9 is not a prime of course. Then there is that 16 at the end. What up with that... but then 17 is a prime.

[Edit] I realize now, I wasn't clear when I suggest they look like odd values or primes, I am truncating the DM values to just the leading digit(s). Just for a second imaging that the pattern was just meant to show up at very coarse scales (because they had to use their Dyson Sphere to make it and it had to get here from there through the nebulae, and get measured by smart monkeys). That said I got a bit dizzy trying to squint to see if anything looked funny in the second and third digits at least.

Then looking at how the author(s) fit the data into multiples of @187 for the first regression. Their fit makes it looks just more like sequential integer multiples of that number. Still an interesting pattern, but probably not as interesting as odd multiples of that value, or prime multiples of that value. But still... spooky. How much smear might there be if a pattern was intended, but then had to get here across a few-lot of light years? The second regression approach where I think they divide 187 by some integers. Also, a spooky pattern relating to integers, or some regular interval. Both seem to rule out randomness FAPP in the DM values.

I don't know I'm just trying to follow it. The second paper, you look at the list of authors, it's like they said "who do we call now...". The microwave story... okay, I would have walked away un-suprised had that been it, but then they make a pretty good case that can't be all of it. Still, who could have wrote something that funny - if it does turn out this goes somewhere, that's going to be one for Dicsover Channel. I can see NDT (Neil DeGrasse Tyson) already standing next to some poor schmuck trying to heat up a burrito in the break room.

The pattern, if there is one, or lack of diurnal pattern, if there isn't one, in the reception frequency is confusing to me. Makes me think about satellites. Don't they orbit like once every 1.5 hours? Makes me want to read up on just how the heck radio telescopes actually do their work. Maybe somebody here knows.

 

[wabbit]

OK I get it:) the odd first digits are just a consequence of the n*187.5 pattern
Amusing it is: all multiples of 187.5 up to n=10 start with an odd digit - and so must any number close to such a multiple.

That pattern by itself is truly bizarre, but now I don't know any more which if any of these is explained by the microwave - why do they single out one to say it cannot be that, but also say more or less that all the others must be legit too?

And for the timing I didn't see enough data to clearly understand what it shows, and how the data is modified by the microwave thing.

So at this point I ll just wait. IMHO they rushed to publish the first paper, then rushed to publish the half retractation half confirmation, and for me this is just incomplete analysis, too hard to distentangle. But to be fair these are preprints not publications, so incomplete or rushed may be unfair comments - the important qualifier is "for me", and i m not even their target audience.

 

[Jimster41]

No doubt about it. Lots of questions. I don't really understand the what Dispersion Measure is in the first place... so there's that.

I screwed up the plot before, and I added a difference from random expectation. Anyway better than workin on a Friday afternoon.

[Edit] the 1/24 doesn't make sense, there were n FRB's I think, so should probably be n/24? Then subtracted from the FRB vectors.

Reception frequency distribution of the FRB over 0-2300h, plucked off of figure 7 of second paper via the old mark II . The microwave evens aren't in this data at all. They had a pretty clear uni-modal distribution around "Dinnah Tiiiime!"

 

[Jimster41]

Can't stop wondering what in the world 187.5 could possibly relate to. If it was a broadcast surely that value would have some significance.

PI/e? Freezing point of water measured in integers to absolute zero, CMB microwave wavelength into the first value of the series? I'm not sure what the list of universally identifiable constants, or relationships would even be... Seems like they would have to be mathematical more than physical. Like the freezing point of water, wouldn't be recognizeable to the critters on a Titan-like planet at all. CMB wavelength in units of what, PI, e. What are the other weird (irrational?) numbers?

Probably imagination could find something regardless of whether it's there, but the ease of that invention, given what we know, and/or the grandness of the terms that fit, might add to the circumstantial evidence. If it truly is just a random number then, that might lobby for us just seeing things in natural periodics.Or is it just the the number that when multiplied by integer values comes out to 3,5,7,9 and 11? "These are n scalable universally consistent things that are a subset of things like them, which also separeate them in two ways... Hello. Do you recognize them?

Spooky.

 

[PWiz]

@mfb Maybe I should start writing that paper on the "Great Filter" now

 

[mfb]

@Jimster41: The numerical value of 187.5 comes from our units here on earth. If you choose a different unit system the number is different. There is no point in looking for a meaning of 187.5. Also, with just one extraterrestrial source we don't have any coincidence left. The extraterrestrial source just happens to be close to the values the local microwave oven gave.

@PWiz: make sure you filter out microwave ovens ;).

 

[Jimster41]

Yeah, you're right. 187 is in units with which we measure the DM, Doh

But then thinking about it some more... I wasn't thinking of the that number per se. I was wondering whether it could be a proportional equivalent in all unit systems of a specific basic mathematical coherence, like integer counting for example, to a fundamental relationship. I mean that's the first thing the researchers showed, there is a relation that describes integer counting or "discretenes" in the signals. Perhaps there is more in it than that.

Can someone please provide assurance that this is not all just an April Fool's joke

 

[Jimster41]

@Jimster41: The numerical value of 187.5 comes from our units here on earth. If you choose a different unit system the number is different. There is no point in looking for a meaning of 187.5. Also, with just one extraterrestrial source we don't have any coincidence left. The extraterrestrial source just happens to be close to the values the local microwave oven gave.

@PWiz: make sure you filter out microwave ovens ;).

Sorry man, can you explain what you mean in the second sentence about "... Just one extraterrestrial source..." I just didn't quite get what you were saying.

 

[mfb]

I mean that's the first thing the researchers showed, there is a relation that describes integer counting in the signals.

Which is some property of the microwave apparently.

I wasn't thinking of the that number per se.

Which has no meaning.
You can express the same value as 5.79*10²⁰/cm^2 or 5.79/pm^2 or 3.73*10²¹/inch^2 or 3.55*10²⁶/(feet*furlong) or whatever.

Sorry man, can you explain what you mean in the second sentence about "... Just one extraterrestrial source..." I just didn't quite get what you were saying.

See page 1:

Subsequent tests revealed that a peryton can be generated at 1.4 GHz when a microwave oven door is opened prematurely and the telescope is at an appropriate relative angle.
[...]
Now that the peryton source has been identified, we furthermore demonstrate that the microwaves on site could not have caused FRB 010724. This and other distinct observational differences show that FRBs are excellent candidates for genuine extragalactic transients.

Just one unexplained event.

 

[Garth]

Once you are into numerology then you can see all sorts of numerical connections and patterns .

I like the humourous article in this week's New Scientist:

Some numerology of astronomy
WHAT might be the hidden significance of the number 187.5? We reported that a measure of the delay between the arrival of low- and high-frequency components of mysterious bursts of radio waves comes in exact multiples of 187.5 (4 April, p 8). Chris Conklin immediately wrote to point out that 1.875 is "the smallest positive solution of cos(x)cosh(x) = –1" and appears in the formula for calculating the frequency of a crystal oscillator from its size and properties. But the link to quartz clocks seems circumstantial to us.

Food and drink for thought
IDLY searching for the number 187.5 – key to the astronomical mystery described above – finds it surprisingly often. A quarter bottle of wine is 187.5 millilitres; we find chocolate sold in units of 187.5 grams; and, probably for a reason, one that escapes us, some antidepressants are delivered in tablets containing 187.5 milligrams of the drug. In a further example of our ability to see patterns in noise, we sought to complete the set of drug, red wine and chocolate with cheese, in 187.5 g portions.

We thus discovered instead that this is a common, if absurdly precise, conversion of the US recipe measure "one-and-a-half cups". If the radio bursts are a signal from aliens, are they looking for our lunch?

I'll buy ET a bottle of wine and chocolate any time, so long as it shared them with me of course!

Garth

 

[wabbit]

See page 1:Just one unexplained event.

I didn't find that clear in the second paper - they also imply(*) that most of the FRBs (including at least some of those 187.5 signals) are not from the microwave - but since they never say "this is the list of frbs we started with, these are bad, those are doubtful, and the rest seem good" I don t know what that "good" list is supposed to be.

(*) their conclusion "We have thus demonstrated through strong evidence that perytons and FRBs arise from disparate origins. There is furthermore strong evidence that FRBs are in fact of astronomical origin."

 

[Jimster41]

I get that for any new thing, you could invent any equivalence from an infinite and arbitrary "list of things".

But wouldn't there be a notable coincidence if the third thing you invented, using only your list of "most significant, apriori, odd, immutable things" turned out to be symmetrical with this, new thing. That's all.

And I am glad to hear I'm not the only one wondering about additional relationships in the pattern, starting with the obvious residual - (187). But maybe that just means there are a lot of fools.

 

[wabbit]

The one thing that is remarkable in this data is its quantized nature n*K, not the value of K in some unit. What is the source of this, microwaves or something else, seems still quite open.

 

[rootone]

Apparently it's there either because of God or the Devil.

 

[Monsterboy]

Can we ask everyone in the vicinity to not use microwave ovens for sometime and then check the readings again ?

 

[PWiz]

Can we ask everyone in the vicinity to not use microwave ovens for sometime and then check the readings again ?

But then how will the scientists concentrate without their instant noodles?

 

[Chronos]

What is that scientific term I'm struggling to recall - poppycock, bullocks, or, ... interesting?

 

[Monsterboy]

But then how will the scientists concentrate without their instant noodles?

Use a gas stove.

 

[wabbit]

For Planck stars, if they (Rovelli & al.) can determine some unique enough signature of Planck star signals (didn't see that in the slides[,...])

There are more details in their paper from sept. 2014
Fast Radio Bursts and White Hole Signals Aurélien Barrau, Carlo Rovelli, Francesca Vidotto.

We estimate the size of a primordial black hole exploding today via a white hole transition, and the power in the resulting explosion, using a simple model. We point out that Fast Radio Bursts, strong signals with millisecond duration, probably extragalactic and having unknown source, have wavelength not far from the expected size of the exploding hole. We also discuss the possible higher energy components of the signal.

 

[Jimster41]

Reception times from figure seven in second paper are random... the mean expectation of reception taken as number of FRB's/24h, is the best fit. Prime reception times are second just because they are closer to random. I think it would have been hard to explain to say the least why a particular local clock was subject to a specific signal pattern from a distance source, at any interval not diurnal, anyway.

 

[I B Phreakin]

It's simple. Edgar Rice Burroughs was born in 1875. What do Aliens know from decimals ?

 

[bahamagreen]

Figure 1 (plot of FRBs as multiples of 1/2 DM (375cm-3) to get DMtry 187.5cm-3) shows:

The number of FRBs that are a multiple of 3... is three.
The number of FRBs that are a multiple of 4... is four.

This relationship is independent of units... it is fundamental number theory.

If the count number of FRBs at each integer multiple were equal to the integer value of that multiple, missed detections would account for the missing data - no detections violate this relation by their count exceeding their multiple, so far... which raises another question*.

If they find one FRB at unity multiple, a second FRB at multiple of two, or four more at multiple of five, etc., then things will get interesting.
But, if a subsequent detection is a multiple of three or four (hence their count exceeding their multiple), then the relation is broken.

Of course there may be other hypothetical relations** that would support just about any detection data, but I think my point is that we should be taking the existing data to predict the likely values of detection number 12... the relation above suggests that we should not detect any more at multiples of three or four because those are done, the rest need to be filled in... :) And we should be alert to any multiples greater than nine***.

* just how does one calculate probability of occurrence (as in the counts being equal to the multiples) and include incomplete counts that do not exceed the multiples? That is, the probability of the whole data set including the fact that no detections yet violate the relationship?

** even if there is a detection at multiple three or four to spoil the pattern, it may be that the overall pattern is cyclic - that is, the counts are meant to be accumulated for a period of time, then when the pattern is full it starts over... we may have staggered our detections through this start over time, so the counts won't look right until at least one full cycle... and we notice the reset point.

*** that the multiples (so far) as integers go 0 1 2 3 4 5 6 7 8 9 (ending at 9) is very nice; either they know we are base ten, or they have ten digits like us, or both. :)

 

[wabbit]

Maybe it's just me but I get the impression this thread is starting to sound a bit more like numerology than physics:)