KIC 8462852 (dipping again in March 2018)

[mfb]

I finally read the paper, and the comet explanation sounds too reasonable :(. Sure, the size of the dips is massive, but it agrees with all measurements. It is also the explanation with the least promising outlook. There is some chance to see more dips within 1-2 years or in a few decades, but the star could stay without further dips forever now if the orbit is eccentric enough. The infrared emission from the comet fragments is going down fast, so I don't know if other telescopes (in particular, JWST in 2018+) are still sensitive to it.

Is the source data for the measured intensity available somewhere? It would be interesting to compare possible matter distributions to the observed light curve.

@Last Exile: It is ruled out in the paper. In particular, the light curve does not fit to the rotation period.

What about if the "cloud" is being acted upon by a magnetic field which aligns all the orientation to cause the particles to interfere destructively at a narrow angle to our line of sight?

Or a complex orbit which only passes our line of sight every so many orbits? If there was a huge black companion (dead star) revolving with a bright partner (attracted hydrogen how ever many millions of years later) would we be able to see such a quick orbit? Is there any such cosmological model of such a thing? I'm way out of my league here.

That does not make sense.

 

[Artribution]

That doesn't seem to fit where I'm at in my head.

It is accurate, though. The gradient spans the width of the image. You may need to adjust your monitor's brightness settings or zoom in closely to see it.

 

[DaveC426913]

Has the possibility of starspots been completely ruled out?
http://blog.planethunters.org/2013/02/11/starspots-and-transits/

I don't think starpots were ever a serious consideration.
The periocity of the observed transits are on the order of 750 days.
The rotational periocity of the star is on the order of 20 hours.

 

[Lanniakea]

I finally read the paper, and the comet explanation sounds too reasonable :(. Sure, the size of the dips is massive, but it agrees with all measurements. It is also the explanation with the least promising outlook. There is some chance to see more dips within 1-2 years or in a few decades, but the star could stay without further dips forever now if the orbit is eccentric enough. The infrared emission from the comet fragments is going down fast, so I don't know if other telescopes (in particular, JWST in 2018+) are still sensitive to it.

Is the source data for the measured intensity available somewhere? It would be interesting to compare possible matter distributions to the observed light curve.

@Last Exile: It is ruled out in the paper. In particular, the light curve does not fit to the rotation period.

That does not make sense.

We missed a transit in April, when Kepler was nonfunctional. The next one is predicted to be roughly in May 2017. What's odd about the curves is that the first transit is so smooth and well defined, while the second one is so chaotically distributed. The two small dips at the beginning of the graph are a bit odd too. So from my understanding, the interesting bit is that a well defined object became diffuse and chaotically distributed, with no apparent IR signature. Then again, the universe is kinda large, you'd expect some coincidences here and there. But yeah comets from outside the system or the oort cloud make by far the most sense. Unless the ATA has something to tell us, I guess we're going to have to wait till 2017 to find out. ):

 

[mfb]

A possible transit - for something in a 750-day orbit. If something is in such an orbit, then we'll see it again. That is unlikely, given the analysis in the paper.

With a broken up comet, all dips come from different objects so their different structure is not surprising.

 

[Vanadium 50]

If you break up a comet, the fragments need to be pretty small to get enough surface area to block out a substantial fraction of a star's light. Somewhere in the 100um to 1mm ballpark.

 

[DaveC426913]

Well comet Holmes might have done the trick.

To-scale comparison:

https://en.wikipedia.org/wiki/Comet_Holmes
http://www.space.com/4643-incredible...igger-sun.html

 

[Vanadium 50]

It's not enough to make the extent large. One also needs to make it dark. You can't get it dark enough with only billions or trillions of particles. You need a lot, and they need to be really, really small.

 

[Lanniakea]

Well comet Holmes might have done the trick.

To-scale comparison:

https://en.wikipedia.org/wiki/Comet_Holmes
http://www.space.com/4643-incredible...igger-sun.html

So it is reasonable to believe that the nearby passing star could've directed a comet from it's own system or the oort cloud (origin doesn't matter too much I suppose) towards Tabby's star, which is when it created a coma as it got close, then this coma was shifted around by gravitational/collision events and that's why it looks so chaotic in the second transit? This is pretty much the most reasonable scenario I can think of.

 

[DaveC426913]

It's not enough to make the extent large. One also needs to make it dark. You can't get it dark enough with only billions or trillions of particles. You need a lot, and they need to be really, really small.

Why do they need to be small?

Is it because, for a given mass, the smaller the particle size the better the light-blocking?

 

[Vanadium 50]

The total surface area needs to be comparable to the star's area to block a significant amount of its light. If you take something with a comet's mass, to get enough area drives you to break it up into 100um-1mm sized objects.

 

[nikkkom]

That lightcurve looks very similar (except for it's magnitude) to the one produced by KIC12557548

The difference is, KIC12557548 is far from certain to be periodic at all, and it's not sub-24 hour transit for sure.

(My own hunch is that KIC12557548 dimming events are not transits at all.)

 

[nikkkom]

>>Has the possibility of starspots been completely ruled out?
>>http://blog.planethunters.org/2013/02/11/starspots-and-transits/

@Last Exile: It is ruled out in the paper. In particular, the light curve does not fit to the rotation period.

These may be starspots of a type we never seen before. For example, starspots which are pole centered.

 

[mfb]

So it is reasonable to believe that the nearby passing star could've directed a comet from it's own system or the oort cloud (origin doesn't matter too much I suppose) towards Tabby's star, which is when it created a coma as it got close, then this coma was shifted around by gravitational/collision events and that's why it looks so chaotic in the second transit? This is pretty much the most reasonable scenario I can think of.

The paper authors suggest that all different transits are different fragments. No orbit, and chaos because every event is different.

These may be starspots of a type we never seen before. For example, starspots which are pole centered.

You have to cover a large fraction of the whole star with spots to get 20% intensity reduction. Also, why should they be exactly pole-centered?

 

[nikkkom]

You have to cover a large fraction of the whole star with spots to get 20% intensity reduction. Also, why should they be exactly pole-centered?

"Why pole centered?", we do see pole-centered "atmospheric" phenomena, such as Saturn's polar hexagon.

Since Kepler did not see such dimming events on any other star, this star may exhibit some rare phenomenon. Maybe its starspots appear much rarer than on the Sun, but when they do, they cover nearly a quarter of poleward surface.

 

[newjerseyrunner]

"Why pole centered?", we do see pole-centered "atmospheric" phenomena, such as Saturn's polar hexagon.

Since Kepler did not see such dimming events on any other star, this star may exhibit some rare phenomenon. Maybe its starspots appear much rarer than on the Sun, but when they do, they cover nearly a quarter of poleward surface.

Wouldn't a gigantic starspot create a huge amount of infrared? As far as I understand, starspots are formed when magnetic loops kink so badly that they extend beyond the surface. In order to make a huge sunspot, wouldn't you require an insanely huge magnetic loop and create a lot of fast moving, extremely hot plasma that would glow like hell in IR? Lots of objects create huge magnetic vortexes at their poles (pulsars, quasars, black holes...) but they're all huge.

Are there any models for how a star might behave it for some reason it had an inordinate amount of iron in it's core? The amount of iron required to create magnetic fields of that strength?

 

[nikkkom]

Wouldn't a gigantic starspot create a huge amount of infrared?

If radiation spectrum is blackbody, then lowering of temperature would not increase any part of the spectrum. The peak moves to redder wavelenghts, yes, but intensity at any particular frequency still falls, it does not increase. See at 7:00 in this video:

 

[nikkkom]

Are there any models for how a star might behave it for some reason it had an inordinate amount of iron in it's core? The amount of iron required to create magnetic fields of that strength?

At those temps, iron is not ferromagnetic :D
As a rule of thumb, for stronger magnetic fields of stellar (and planetary) objects, you need faster rotation. Incidentally, this star rotates faster than one revolution per day.

 

[newjerseyrunner]

At those temps, iron is not ferromagnetic :D
As a rule of thumb, for stronger magnetic fields of stellar (and planetary) objects, you need faster rotation. Incidentally, this star rotates faster than one revolution per day.

You're right, I forgot that in objects like that its metallic hydrogen that makes the magnetic fields, not iron. One revolution a day is not that fast.

If radiation spectrum is blackbody, then lowering of temperature would not increase any part of the spectrum. The peak moves to redder wavelenghts, yes, but intensity at any particular frequency still falls, it does not increase. See at 7:00 in this video:

I'm not referring to the spectrum of the star itself, I'm saying that starspots are caused by massive magnetic loops that break the surface. Along those loops plasma races around at incredible speeds. Once those loops unkink, all of that plasma gets released from the magnetic clutches of the star and gets ejected out into space (a solar flare.) If this star is constantly having magnetic storms so fierce that starspots can dim it's light output by 20%, it should be sitting in a vast cloud of hot gas, similar to if two planets collided with each other. Wouldn't it? I'm saying that you would be able to detect all of that plasma being thrown off of the star in the IR.

 

[DaveC426913]

These may be starspots of a type we never seen before. For example, starspots which are pole centered.

Unfortunately, that's drifting into an area where
a] it requires the invocation of an unprecedented phenomenon - and even granting that -
b] it still requires some tortured explanations (such as smoothness and periocity) in even make it fit our fantastic phenomenon.

With that, all bets are off. We wouldn't go much farther off to posit billions of pixies holding hands in a maypole dance.

 

[nikkkom]

Unfortunately, that's drifting into an area where
a] it requires the invocation of an unprecedented phenomenon - and even granting that -
b] it still requires some tortured explanations (such as smoothness and periocity) in even make it fit our fantastic phenomenon.

The point is, I don't see any periodicity of the dimming events in the data.

There was one nearly 20% dimming event, then some 700 days later a group of three other dimming events first of which was even deeper. Each of four events has a different shape.

That's why I thought "these are probably not transits at all" and came up with my pet theory.

 

[DaveC426913]

The point is, I don't see any periodicity of the dimming events in the data.

There was one nearly 20% dimming event, then some 700 days later a group of three other dimming events first of which was even deeper. Each of four events has a different shape.

OK, starspots have the distinction of not needing to have a highly-constrained periocity. I'll give it that.

 

[DaveC426913]

Point of order:
Everywhere I've used the word periocity, you should be hearing periodicity in your head.

 

[Michael Lazich]

Some really intriguing light curves here. Read the article for some explanation.

http://www.slate.com/blogs/bad_astr...ge_dips_in_brightness_are_a_bit_baffling.html

Look at the smooth curve lower left. That has got to be a single body transit. Multiple bodies couldn't make such a smooth curve. Yet that body results in a 15% drop in the light curve!
And it's cold, so not a companion star.

I don't see how exo-comet fragments can explain this.

Black dwarf?

I've been struck how the lower left curve looks like what could be produced by a spinning disk or other object...

Black Monolith anyone...? ;)

 

[mfb]

I've been struck how the lower left curve looks like what could be produced by a spinning disk or other object...

How?

 

[rootone]

Maybe a really massive companion planet which itself has several very large moons orbiting it?

 

[Michael Lazich]

How?

Actually, the cross sectional area of a spinning disk could be said to have the "general" shape of the light curve, and that for only a single rotation.

The light curve from a star with such a spun disk in front of it would actually look like the inverse of what was observed from KIC 8462852.

I just calculated the cross section of a spinning disk with the dimensions such that, at full occultation, the light from the star would be around 84% of its maximum value; as it rotates through 180 degrees, the occulting area varies as cosine of the rotation angle and you get the curves below:

Of course, the actual observations are asymmetrical and are the *inverse* of the light curve that would be produced under the assumptions above.

Just thought it was interesting.

 

[mfb]

I don't see where you imagine a disk and how spinning that disk would be relevant.

 

[Michael Lazich]

I don't see where you imagine a disk and how spinning that disk would be relevant.

 

[newjerseyrunner]

That looks like you're describing a flipped coin. Discs don't spin that way, there is no physical way for anything like that to naturally form. Discs spin along their plane, that's what makes them discs in the first place: angular momentum.

 

[Michael Lazich]

That looks like you're describing a flipped coin. Discs don't spin that way, there is no physical way for anything like that to naturally form. Discs spin along their plane, that's what makes them discs in the first place: angular momentum.

No kidding. My posts were never about any sort of natural explanation for the observed light curve. Nor are they meant as a serious argument for an unnatural explanation.

Hence my original reference to a Black Monolith.

Which is a reference to the movie 2001, where a spinning alien structure is shown orbiting Jupiter.

Good movie, you should see it.

Just an offhand observation that people seem to have taken too literally.

 

[DaveC426913]

Hence my original reference to a Black Monolith.
Which is a reference to the movie 2001, where a spinning alien structure is shown orbiting Jupiter.
Good movie, you should see it.
Just an offhand observation that people seem to have taken too literally.

I confess I too missed a reference to the monolith. Must have been a ways back.

I thought we were still discussing explanations and offering sincere ideas.

 

[nikkkom]

Maybe a really massive companion planet which itself has several very large moons orbiting it?

With star having 1.5 Sun's diameter, the "planet" which can occult 22% of star's disk needs to have diameter of more than half Sun's diameter - IOW, ~5 Jupiter diameters.

Such planets don't exist.

 

[rootone]

With star having 1.5 Sun's diameter, the "planet" which can occult 22% of star's disk needs to have diameter of more than half Sun's diameter - IOW, ~5 Jupiter diameters.

Such planets don't exist.

I missed the bit about the star being 1.5 Sun's diameter, but out of curiosity ...
How big can a planet get before it must inevitably become a star?
I'm sure I heard that some exoplanets have been found which are larger than Jupiter, although not 5 times larger.
(Yes I know the answer depends a lot on the density and composition, but is 5x Jupiter completely out of the question for a gas giant predominantly made of hydrogen?)

 

[nikkkom]

I missed the bit about the star being 1.5 Sun's diameter, but out of curiosity ...
How big can a planet get before it must inevitably become a star?
I'm sure I heard that some exoplanets have been found which are larger than Jupiter, although not 5 times larger.
(Yes I know the answer depends a lot on the density and composition, but is 5x Jupiter completely out of the question for a gas giant predominantly made of hydrogen?)

https://en.wikipedia.org/wiki/List_of_exoplanet_extremes

HAT-P-32b - 2.037 (± 0.099) Rj
"CT Chamaeleontis b is slightly larger at 2.2 Jupiter radii, but may be a brown dwarf star."

And this is a 2 day orbit planet with nearly 2000K surface temps, strongly thermal-puffed.

Thus, 2x Jupiter is about as big as planet can be.