Help - Search - Members - Calendar
Full Version: KIC 8462852 Observations
Unmanned Spaceflight.com > Beyond.... > Telescopic Observations
Pages: 1, 2, 3
JRehling
Kepler found one very, very strange case:

http://www.theatlantic.com/science/archive...-galaxy/410023/

In a nutshell, while Kepler was observing it, the star (larger and brighter than the Sun) exhibited four dimming events that took place at irregular intervals, blocked a lot more light than a Jupiter-sized planet would block, and had a "shape" that varied in all four cases and did not resemble a planet.

This case is attracting some wild speculation… in fact, it is seemingly certain that something wild must be going on; it's just a matter of which wild scenario is the correct one.

If I had to throw my hat in the ring, I'd guess that a distant collision and breakup has placed big swarms of matter into a very long-period orbit. But there's no hypothesis that's been offered that doesn't seem problematic.
ZLD
Yeah, I will be patiently waiting and excited if the hypothesis in the article holds true but I would put all my cards into a strange and rather rare occurrence with the possible passage of a large mass body passing too near the larger system, causing large disruptions. Such an instance could have tons of possible outcomes. Its certainly an interesting event to follow regardless!
scalbers
Could this be like some of the dense clouds that eclipse Epsilon Aurigae on occasion?
ngunn
Kepler is designed to look for transits - but how do they deduce transits from this observation? For me a diagnostic property of transits is regularity. I presume they don't have any system geometry to go on but only an odd-looking light curve. Star behaving badly is what I would call it.
JRehling
The research paper says that the properties of the light curve "strongly rules out" the sort of variable star that is known to exist for a star of this size and color. In a nutshell, it changed far more rapidly than the known cases of such variables. Also interesting (to me), one of the dimming events had a spiky and symmetrical shape.

Of course, it's a given that whatever is going on is very weird, so by definition, it doesn't match any known cases of anything.

The paper, including their discussion of the possibility that this is just a variable star is here:

http://arxiv.org/pdf/1509.03622v1.pdf
ZLD
I think the biggest peculiarity that rules out a lot of things is how short the period is at .88 days. They're findings concluded that it probably wasn't starspots, but I can't help but to think it could be something really strange like that as well.
JRehling
QUOTE (ZLD @ Oct 15 2015, 05:26 PM) *
I think the biggest peculiarity that rules out a lot of things is how short the period is at .88 days. They're findings concluded that it probably wasn't starspots, but I can't help but to think it could be something really strange like that as well.


The strange phenomenon doesn't have a period of 0.88 days; the star's rotation has a period of 0.88 days. The strange, aperiodic dimming happens on a much slower timescale. There were over 700 days between the first two occurrences, then about 20 days separating the second to third and third to fourth occurrences, which each last several days. This is part of what makes it seem very unlikely that the phenomenon is occurring only within the star itself.
nprev
What spectral class is this star? Higher rotation speeds seem to be associated with brighter, more massive, and generally more youthful stars.

If it's both young and luminous, this fast rotation may mean that considerable amounts of material are being shed from the star. Perhaps what we're seeing are massive CMEs interacting with circumstellar gas and dust...?
ZLD
@JReling: Thanks for the correction. I skimmed it earlier and thought I understood differently.
silylene
We need to account for a few facts here (see the paper for many more):

Eclipses at least 22% obscuration of the KIC 8462852 star
There are multiple large eclipsing objects, and each object must have orbital periods at least several years (because no periodicity was noted within the observational span, which means none of them repeated exactly, at least none repeated with the same obscuration).
Some of the large objects look 'grouped' in their obscuration, in the sense that the eclipses occurred in very close time proximity.
There is also a small object which orbits with a 20d period.
There is a 0.88d cycle is likely a starspot, which assume the star revolves around its axis a fast 0.88d.
Infrared data doesn't support collisions or even comets.

My speculation which doesn't involve alien constructs or other woo hypotheses:

The star has several exoplanets which orbit it at various distances with periods of many years, so each exoplanet may have only been observed once in the data set.
Some of the exoplanets are ringed systems with very opaque rings tilted out of the ecliptic with no noticeable ring gaps. The exoplanet at 793d with the 22% obscuration has a HUGE dense ring system that is inclined such that it can eclipse 22% of the star's illumination as see from earth.
Some of the exoplanets have very large exomoons orbiting them. (this accounts for the secondary obscurations observed in close time proximity to the major obscurations, for example in the data between 1500-1600 days).
Maybe the large exoplanet at 793d and 1520d is the same exoplanet, just the ring angle changed, and at 1519d an exomoon was observed with it.
JRehling
QUOTE (nprev @ Oct 15 2015, 09:44 PM) *
What spectral class is this star? Higher rotation speeds seem to be associated with brighter, more massive, and generally more youthful stars.

If it's both young and luminous, this fast rotation may mean that considerable amounts of material are being shed from the star. Perhaps what we're seeing are massive CMEs interacting with circumstellar gas and dust...?


It's an F3, larger and brighter than the Sun, similar to Procyon. It doesn't show the IR brightness that would imply a large dust disc, which seems to eliminate one of the possible explanations.

I've wondered about a highly elliptical orbit for the "junk" that's causing the dimming. If it's on a long period orbit, then it might remain cool because it spends very little time near the star. That is essentially the spirit of the authors' explanation, a massive swarm of long-period comets. I wonder if the debris from a collision, now in a long-period orbit, might not be able to produce the same result.
JRehling
QUOTE (silylene @ Oct 16 2015, 07:43 AM) *
My speculation which doesn't involve alien constructs or other woo hypotheses:
[...]
Some of the exoplanets are ringed systems with very opaque rings tilted out of the ecliptic with no noticeable ring gaps.


The third event, at 1541 days, is very intriguingly symmetrical: It has three peaks, with the two outer ones being approximately equal in depth and distance from the larger, central one. That makes me wonder about a ringed planet.

But the other three events don't show that pattern. Event #1 is single-peaked and quite smooth, while Event #4 is double-peaked. Event #2 is pretty messy: It has three peaks, but no symmetry.

So I wonder if Event #3 is due to a ringed planet with the planet and rings participating, and Event #4 due to a ringed planet with only the apsa of the rings participating, and the planet "missing" the star during the event.

Keep in mind that for something opaque to cause a dip of 22%, for a star with 1.58 times the radius of the Sun, the cross sectional area of the transiting object has to be at least 55% the cross section of the Sun, or about 55 Jupiters' worth of cross section. This is a big object or set of objects! A planet + ring system would have to have at least 4 times the diameter of Saturn's rings to accomplish this, like rings extending from Jupiter's cloud tops out to Europa or from Saturn's cloud tops out to Rhea.

And that scenario with rings that we see face-on from Earth would be provided only if the rings were perpendicular to the planet's orbit, which is unlikely to occur by chance; any deviation from that geometry would require a still larger ring system to provide the observed darkening.

That might just work if we only had one such event to explain, but four of them means that there would be at least three different giant planets with three gigantic ring systems. I can't see how that would happen.
HSchirmer
QUOTE (JRehling @ Oct 16 2015, 05:27 PM) *
...
Keep in mind that for something opaque to cause a dip of 22%, for a star with 1.58 times the radius of the Sun, the cross sectional area of the transiting object has to be at least 55% the cross section of the Sun, or about 55 Jupiters' worth of cross section. This is a big object or set of objects! A planet + ring system would have to have at least 4 times the diameter of Saturn's rings to accomplish this, like rings extending from Jupiter's cloud tops out to Europa or from Saturn's cloud tops out to Rhea.
...
That might just work if we only had one such event to explain, but four of them means that there would be at least three different giant planets with three gigantic ring systems. I can't see how that would happen.


Eh, guess we might be seeing the system during a late heavy bombardment / nice-model period of orbital chaos?
Hmm, two hot jupiters swapping places, each dragging along debris fields of trojans. In the nice model, is there a scenario where Jupiter and Saturn don't scatter, but merge, or end up a close binary; basically something huge and gaseous and opaque, that might be {relatively} cool, at least in contrast to simply smashing two earth sized planets together?

Thinking about outside effects, what happens if a brown dwarf binary, e.g. Luhman 16, drifts into the system?
scalbers
Here's a web page that helps explain Epsilon Aurigae that might have some parallels as suggested earlier:

http://www.daviddarling.info/encyclopedia/...on_Aurigae.html
ngunn
Accepting the idea of eclipsing objects of some kind, why do they have to be in orbit around the star? Couldn't they be anywhere in the line of sight? So far there is no evidence of periodicity that would arise from regular orbits, only a sequence of apparently one-off events. If the objects are in a much closer but very faint system that just happens to be aligned with the star then they wouldn't have to be so big to produce the observed degree of obscuration - and of course they'd only pass once.
HSchirmer
QUOTE (ngunn @ Oct 17 2015, 08:54 PM) *
Accepting the idea of eclipsing objects of some kind, why do they have to be in orbit around the star?
Couldn't they be anywhere in the line of sight?


I was thinking that same thing, perhaps we're seeing something closer?

This WTF star is estimated to be around 1480 LY from earth, in the Cygnus constellation.
It's just up and right from a star cluster NGC 6866 (arrow below)


First, that's looking towards the galactic center, the densest area for stars, and molecular clouds and other dark stuff, e.g. steppenwolf planets and brown dwarfs. So, perhaps it's a transit by something closer, a brown dwarf with planet(s) or a brown dwarf with comet belts.

Second, turns out that there is a nearby (~14 LY) triple star system V1581 Cygni in the same area as WTF (1450 LY).
So, let's assume the triple system has ejected something, planets or shredded comets. They would be 100 times closer, and thus could be 100 times smaller to cause the occultation. Instead of hypothetical object(s) ~50 Jupiter diameters across, object(s) 1/2 the diameter of Jupiter ( a normal ice-giant planet) could be the explanation.
JRehling
Interesting thoughts, nprev and HSchirmer. The first thing to grapple with is the probability of an object in one system transiting a coincidentally-aligned more distant star. The next thing to grapple with is the probability of four such events happening in a short time. In addition, the shapes of the events are still weird. And, finally, in such a paradigm, it becomes a little odd that no such event completely eclipsed the distant star, reducing its observed brightness to zero.

The last two might be addressed if we imagine that these are comet-like. Then, in fact, the star may be getting completely covered in terms of area, but by a translucent object. And the first objection may be addressed by a "Birthday Theorem" of sorts: Given enough object pairs, some really precise alignments may occur.

It's the second objection that seems thorniest: How can the closer system have so many objects that transit the more distant star? It would require either that the closer system have a dense halo of large objects or that the closer system has a lot of large objects orbiting in the same plane that coincidentally contains the farther star.

I don't know. Something weird is happening. The closer-and-farther system case seems to handle one weird aspect of the data – the size of the objects – but adds another weird thing – the coincidental perfect alignment – and maintains several of the weird things from the single-system hypothesis. I don't know which, if either, of those two explanations is more likely.
ngunn
QUOTE (JRehling @ Oct 19 2015, 09:12 PM) *
It's the second objection that seems thorniest: How can the closer system have so many objects that transit the more distant star? It would require either that the closer system have a dense halo of large objects or that the closer system has a lot of large objects orbiting in the same plane that coincidentally contains the farther star.


I've been thinking about this too. To observe eclipses in a single system you need one coincidence: The plane of the system must be aligned with us. For a nearer system to produce multiple eclipses of a more distant star you need two coincidences. The nearer sustem must be edge on to us as in the ordinary case, but in addition its proper motion relative to the background star must be in the same direction also to carry the whole succession of objects across the star from our viewpoint. If the nearer system is fairly compact, almost like Galilean moons around a brown dwarf for example, that may not be too much to ask. I agree with you it's difficult to call the relative improbability of this versus a single system scenario. It's clearly something improbable or we would be seeing more of them! For the moment, though, I'm noticing a sequence of events that happens only once and a degree of obscuration that suggests to me that the eclipsing objects are much closer than the star.
HSchirmer
QUOTE (JRehling @ Oct 19 2015, 09:12 PM) *
I don't know. Something weird is happening.
The closer-and-farther system case seems to handle one weird aspect of the data – the size of the objects – but adds another weird thing – the coincidental perfect alignment – and maintains several of the weird things from the single-system hypothesis.
I don't know which, if either, of those two explanations is more likely.


Basically, I'm following an old saying "if you see hoof prints, think horses, not unicorns"
Better to go with something known-but-extremely-unlikely rather than something entirely-unprecedented.

Basically, it's a "devil you know" situation- which is more unlikely? What is the least bizarre explanation?

A - A nearby star with multiple opaque objects in a tight orbital plane, that all happen to occult a distant star?
B - A distant star with multiple opaque objects in a tight plane, that all happen to be several Jupiters across?

So, how certain are we about the WTF star? IRC, the classification is F-type, 6k-7k kelvin, located about 1,400 LY away.
Ok, what if it's not a "normal" distant F star, but a variant of procyon, (F star and white dwarf) but with a white dwarf that has a 6k-7k temperature. So, what would a near by white dwarf pulling material from a brown dwarf look like? Could it be something with the temperature and H2 lines that resemble a F class star? The luminous star would be roughly earth sized, and thus the observed occultations could be a result of "standard" rocky planets or dwarf planets.
Hungry4info
Look at the light curve. A dark sphere transiting a luminous sphere doesn't fit the data well.
JRehling
Comet Hale-Bopp's tail had a maximum length of over 60 million km. Comets can easily exceed Jupiter in size, and then some. However, one comet like Hale-Bopp wouldn't obscure a star; the tail is translucent. So this phenomenon couldn't be explained by one Hale-Bopp. But if a larger body, or set of bodies traveling together, were outgassing as much as Hale-Bopp did, we might have an explanation for this which isn't too crazy.
JRehling
Another thought: A comet seen transiting its star would be, most likely, pointing its tail directly away from the star and directly towards us. This would reduce its size, but increase its opacity. We have never observed a comet in this solar system from that geometry.

From Keplerian considerations, the transiting objects, if orbiting KIC 8462852, is more than 3 AU distant. Because this star is much brighter than the Sun, that may correspond to high levels of cometary outgassing.
HSchirmer
QUOTE (JRehling @ Oct 20 2015, 06:46 PM) *
Another thought: A comet seen transiting its star would be, most likely, pointing its tail directly away from the star and directly towards us. This would reduce its size, but increase its opacity. We have never observed a comet in this solar system from that geometry.

From Keplerian considerations, the transiting objects, if orbiting KIC 8462852, is more than 3 AU distant. Because this star is much brighter than the Sun, that may correspond to high levels of cometary outgassing.


Well, comets can have multiple dust tails, and a separate ion tail. Not sure if the ion tail eventually coalesces back to normal material at some great distance.

Shoemaker Levy's beads-on-a-string appearance comes to mind.
Rittmann
I was wondering...

How much obscuring could happen in a system like ours of some big body with plenty of water entered the inner system? Something like KIC 12557548.

http://www.space.com/15849-disintegrating-...er-mission.html

If the orbit change was recent and chaotic, could it leave behind a trail big and wide enough as to obscure the star in such a way? I always wondered how massive would be a tail for something like a Pluto if it fell below the ice line due to some disturbance. And in that system there is a second star some 885AU away, so something like a Sedna (936 AU of aphelion) could be disturbed and end up falling onto the main.

Let's make some comparison to see the numbers. Halley comet has a mean diameter of 11km, with dimensions of 15x8 km. I found some old calculations that suggest a mass loss of 1.8x10_8 metric tons per appearance. We are talking of a comet that approaches the Sun 0.58 AU, and with an expected life of around 40 more appearances, so it's pretty unstable in cosmic terms, but still gives a good idea of the survivavility of such body.

In comets outgassing is caused not only in the surface, but also by the heating of its interior as demonstrated by Rosetta and the jets that appeared in the night side of 67P. Still, we can make - for the sake of making some rough numbers - an approximation to relate mass loss and comet surface. With these numbers, for something the size of - to give some known thing - Ceres, we could relate mass losses by that formula. Of course, we are assuming similar compositions, which we know is not true for Ceres and Halley.

A(sphere) = 4πr_2

Halley (5.5 km_2): 380km_2
Ceres: 2.770.000 km_2

So something the size of Ceres could lose 7300 more mass per visit to the inner solar system. Using the equivalences above, that would be a mass loss of 1.3x10_12 tons per transit.

With a mass so significantly higher than a comet, though, such a body could also be a lot more stable over time. Let's calculate how many orbits would be needed to lose 10% of the mass of such a body if its density was 0.5 g/cm3. Ceres has a mass of 9.393x10_20 kg at 2.16g/cm3 density, so at the density of a comet (more valid for the surface conditions than for the inner rocky mantle) of such a body, we would have an equivalent comet four times lighter, at 2.4x10_20 kg. 10% mass loss would then be 2.4x10_16 metric tons.

So for a mass loss of 10%, we get a value of at approx. 20.000 orbits. At 750 days period that would mean 40.000 years to lose that 10% of mass.

Something like Pluto, 15 times more massive than Ceres, would take probably 10-15 times more time to lose 10% of its mass, giving the possible value for a full disintegration somewhere around 4 million years.

Of course, a period of just 2 years is not plausible for such bodies, which should be born beyond the ice lines and have far longer orbital periods. At 80 years of orbital period we are talking of 160 million years for a full disintegration. That could account for the lifetime of this star, and be a remnant of its chaotic birth.

In any case, such a body would create a massive tail that could obscure the inner system. Not only that, after detaching from the main body, the tiny particles could be subject to the gravity well of passing planets, creating clouds of dust trailing them over the centuries. Planetary gravity wells could very well form trails of their own of huge sizes. I have no idea on how long would they survive, and how opaque they could become, but the peaks may indicate clouds born from different tails created by different orbits of the diving massive comet.

Furthermore, we know very little about the companion star. If the companion is in a circular orbit is one thing, but if it is in an elongated orbit like Sedna, the main star would be bombed by all icy objects from its outer regions constantly. The same could be possible for the companion, so a good indicator would be check the magnitude variations of the companion star.

For a star passing by at that distance, the disturbance would probably be self-explained by the massive infalling material from the outer parts of that system.
ngunn
I'm still not buying the comets idea, sorry.
silylene
QUOTE (JRehling @ Oct 16 2015, 05:27 PM) *
.....
And that scenario with rings that we see face-on from Earth would be provided only if the rings were perpendicular to the planet's orbit, which is unlikely to occur by chance; any deviation from that geometry would require a still larger ring system to provide the observed darkening.

That might just work if we only had one such event to explain, but four of them means that there would be at least three different giant planets with three gigantic ring systems. I can't see how that would happen.


Yes, I am proposing a complicated planetary system with multiple planets with giant ringed systems (at least 3 were observed), plus some planets without ringed systems for the minor occultations, and perhaps even additional exomoons, because if you look closely at the data, there are some shoulder spikes.

And yes, the ring system of the largest one would be 3-5x bigger than Saturn. [I want to remind that exoplanet J1407V has a ringed system 200x larger than Saturn.]

Also ringed object may be more of the rule, rather than the exception. Jupiter, Saturn and Uranus all have rings, although only Saturn has huge opaque rings. Neptune has arcs. At least one and two minor planets have rings (Charliko and 2060 Chiron).

Yes, what I proposed is complicated. I completely agree. And unlikely, I also agree. Perhaps this is why we have only seen *one* of these kinds of systems, it is unlikely?!

But perhaps what I propose remains more likely than other possibilities than odd planet-sized comets which don't absorb in the IR, orbiting rock swarms which are weirdly cold and opaque, or woo proposals which I refuse to consider.

I would like to understand better the long-term stability of large ringed systems in a complex solar system with several large planets and their gravitational interactions. I do suspect that one could have several "super-Saturns" and this would be a stable system.
JRehling
The biggest problem with the rings hypothesis is this:

The rings would be invisible if the rings are in the plane of the planet's orbit. So, the rings must be closer to perpendicular to the plane of the planet's orbit... and, coincidentally, transiting the star near the node when we see them face-on. (Rings perpendicular to the planet's orbit would still appear edge-on to us much of the time. Consider Uranus as an example: The geometry is similar.) And, this is happening with not just one planet, but at least three. So, we're not just getting lucky once or twice, but about eight unrelated times.

If we saw just one of these, I would buy the rings hypothesis in a heartbeat. Three in one system and none anywhere else? No, this seems definitely wrong.
Explorer1
But as long as the parent planet's spin axis is pointed towards us, wouldn't the rings be perpendicular through the entire orbit? Uranus doesn't seem like a good example considering it orbits the sun, and its axis is pointed at us only twice each time.
Here's how I'm imagining it: point a finger at something far away, and then move your arm in a circle in front of you. The finger is the spin axis, the hand is the planet and rings, and the circle is the orbit.
HSchirmer
QUOTE (silylene @ Oct 21 2015, 03:33 PM) *
Yes, I am proposing a complicated planetary system with multiple planets with giant ringed systems (at least 3 were observed), plus some planets without ringed systems for the minor occultations, and perhaps even additional exomoons, because if you look closely at the data, there are some shoulder spikes.
...


Actually, that might fit.


Binary Saturns.

Penn State has some good discussion
but has also posted the light curves for this star.


[img]

Basically -the WTF star (above) looks totally different from the usual transit (below)




1) Usually, only a fraction to about 1% of the star is blocked.
-Here, almost 15 to 20 percent of the starlight is blocked. Whatever is causing the shadows is huge. Something around 55 jupiters across.

2) Usually, the light curve has steep sides as the object starts to move infront of the star, then a a flat "floor" as the object moves across the star, and another steep slope moves away from the star.
- This is different, the light curve has no floor. That suggests we are only seeing a portion of the object pass infront of the star. Whatever it is, it appears that we are only seeing PART of it pass in front of the star, not the entire silouhette.

Best-non-comet-guess? Binary gas giants with rings, edge on?
Yes, unlikely alignment, but this is an unlikely world.
Perhaps this is a pair of binary Saturns, tilted like Uranus, Pluto and Charon, but with a huge ring system.
Think of binary gas giants, each with Saturn's rings, but like Pluto & Charon, a double planet. Oh, and like Uranus and P&C, tilted ~90 degrees.

Kepler's day ~780 event is seeing one gas with maximum ring occultation but missing the other.
The day ~1510 even it seeing the one gas giant edge with maximum ring occultation, then seeing the other.
(yes, epicycles and deferents by another name....)
So, if you have binary gas giants in a loose orbit, you might have a some interesting periodicity.
Mongo
QUOTE (Explorer1 @ Oct 21 2015, 04:13 PM) *
But as long as the parent planet's spin axis is pointed towards us, wouldn't the rings be perpendicular through the entire orbit? Uranus doesn't seem like a good example considering it orbits the sun, and its axis is pointed at us only twice each time.
Here's how I'm imagining it: point a finger at something far away, and then move your arm in a circle in front of you. The finger is the spin axis, the hand is the planet and rings, and the circle is the orbit.


But the planet transits the star, so the axis of the rings must be more-or-less perpendicular to the axis of the planet's orbit -- and also pointed at Earth -- with at least three separate objects. One is easy to believe -- it happens in our own solar system with Uranus -- but three, all with their axis pointing at Earth? Sound unlikely to me.

On the other hand, this phenomenon is quite unlikely already, otherwise we would have seen other examples before this.
HSchirmer
QUOTE (Mongo @ Oct 21 2015, 04:35 PM) *
But the planet transits the star


Which is a bit of a "founder effect" for Kepler, it only picks out transiting stars...

QUOTE (Mongo @ Oct 21 2015, 04:35 PM) *
, so the axis of the rings must be more-or-less perpendicular to the axis of the planet's orbit


Well, we have that with Uranus and Pluto-Charon. Given the putative size of the ring system here, 55 jupiters across,
it's still going to be huge if viewed at 45 degrees or 30 degrees.

Actually, when I read over this, and thought about the denisty of a ring system or an asteroid belt, I was reminded of the proposals for a 'kamikaze' asteroid belt prob- IIRC something that would start from a solar polar orbit and then make a wrong-way run through the asteroid belt to maximize the number of imaging targets.

Basically, if you view a ring or belt edge on, you may get a rather high density of objects in view.

QUOTE (Mongo @ Oct 21 2015, 04:35 PM) *
with at least three separate objects.
One is easy to believe -- it happens in our own solar system with Uranus we would have seen other examples before this.


Well, as I speculated above, what about a binary gas giant, both with rings.
That would naturally create mulitple occulting objects in the same plane.
Explorer1
And now a star getting 40% of its light blocked: http://www.cbc.ca/news/technology/white-dw...eroid-1.3282014
A white dwarf, so less need for dramatic explanations, in a small object, but quite interesting nevertheless! Definitely a tail in this case...
Paper: http://www.nature.com/nature/journal/v526/...ature15527.html
silylene
My concept is that the rings are not edge-on to earth (of course), but tilted as described earlier, as in the Uranus or Pluto-Charon system.

Yes this star would require 3 super-Saturns with the ring planes titled at high angles to the ecliptic. Agreed. Unlikely? Yes. But then we have Uranus and Pluto-Charon. For the sun, 25% of the large planets have a highly inclined orientation. 75% of the large planets have continuous rings. 25% of the large planets have huge opaque rings.


Now we get down to likelihood. A SWAG could be made just using the above numbers, but for a starting point, I would need to know the average number of large planets orbiting a star which is observed to have planets (it's in the Kepler data, I just don't know it).

I bet the result is rare...maybe 0.1% or something like that.

But then of all the thousands of systems Kepler has found, so far we have found just one system like this.
alan
The asymmetry of one of the transits reminds me of the tadpole orbits Jupiter trojans follow relative to Jupiter position. It would require much more mass than in these orbits than in the present solar system, or it most of it to be distributed among smaller objects. I note that the wikipedia article on Jupiter trojans references a model wherein Jupiter might have captured a much larger mass as trojans as it rapidly grew. If these were then broken up by collisions they may block enough light.
JRehling
QUOTE (silylene @ Oct 21 2015, 11:57 AM) *
Yes this star would require 3 super-Saturns with the ring planes titled at high angles to the ecliptic. Agreed. Unlikely? Yes. But then we have Uranus and Pluto-Charon. For the sun, 25% of the large planets have a highly inclined orientation.


That doesn't begin to capture it. The case we've observed would require that the ring systems be highly inclined and relatively face-on as seen from Earth and highly inclined in the same way. It is not at all a given that two highly inclined (with regard to their star) systems would have poles that are mutually relatively aligned and face-on to Earth, much less that three or four would! And all be transiting. This isn't a couple of coincidences or a few coincidences; it's a barrel full of coincidences.

Basically, five or more axes would have to be co-aligned for no damned good reason. Not merely highly inclined WRT their orbits, but highly inclined WRT their orbits and co-aligned WRT one another.

If we see five of these coincidences in one system, we ought to be seeing two or three of them in orders of magnitudes more systems, and we aren't.
AndyG
Huge, dense and opaque ring systems that are tilted during a transit: wouldn't there be a noticeable slight rise in apparent stellar brightness when this planet was near the star, half a year later?

Andy
JRehling
QUOTE (AndyG @ Oct 22 2015, 01:37 AM) *
Huge, dense and opaque ring systems that are tilted during a transit: wouldn't there be a noticeable slight rise in apparent stellar brightness when this planet was near the star, half a year later?


If such a planet were in a circular orbit, yes. The surge due to a planet's reflected light is routinely detected by Kepler for Hot Jupiters. But: How much of the star's light is blocked does not depend upon the star-planet transit, but how much light is reflected by the planet varies with the inverse square of the distance. So, a planet+ring system with a cross section of 50 Jupiters out at a distance of 5 AU would reflect back as much light as a single Jupiter-sized planet with no rings at a distance of 0.8 AU, and much less than such a planet at 0.1 AU.

But, you raise an excellent point. A system packed with giant ring systems could conceivably avoid those surges, but the likelihood of it fitting our observation – already very low – is still another few notches lower given what you've pointed out.

That may say something about the comet-swarm hypothesis as well: If this star has gargantuan numbers of comets blocking its light, the comets performing that should also be reflecting a lot of light. These things should be – at least at times, and from some geometry – contending with the star itself in brightness, and we see nothing like that. I think you've made a great observation that undermines the comet hypothesis, too.

Explanations that would not produce a surge – if the clouds of occluding material were of low albedo, or quite far from the star, such as in another star system which is coincidentally aligned.
HSchirmer
QUOTE (JRehling @ Oct 22 2015, 12:40 AM) *
If we see five of these coincidences in one system, we ought to be seeing two or three of them in orders of magnitudes more systems, and we aren't.



Something just occurred to me

"If we see" isn't necessarily the question..
When might we next see something?
That's a better question.

Looks like the effects have a 2 year period, occultations occurred in 2009, 2011, 2013


After two small dips in 2009, which had attracted the notice of the Planet Hunters, there was another major dip of about 15 percent in 2011, and it lasted nearly a week. Finally, there was a whole series of dips in 2013, one of them managing to dim the star’s light by 22 percent.


So, it is 2015 now, is that putative 2015 occulation coming up, or has it already occurred?
Guess there will be alot of glass pointed that way this year...

Hmm, hubble has taken some planetary nebula photos at 1,400 LY


So, what size objects hubble can resolve, (not worded quite right) well, what size objects can hubble detect...
Hungry4info
The relevant question is angular resolution, but that's not the right avenue of investigation. The discovery paper shows some good imagery of the star with a ground-based telescope, showing the star to have a companion. Transmission spectroscopy would make more sense here.
Explorer1
I was browsing the extreme exoplanets list on Wikipedia and stumbled upon this; it has to be a measurement mistake, right? https://en.wikipedia.org/wiki/K2-22b.
It would be denser then the element mercury and have around 70 G surface gravity, if true! Like that old Hal Clement novel....
NASA page here: http://exoplanetarchive.ipac.caltech.edu/c...ONFIRMED_PLANET
Hungry4info
No, it isn't right at all. The original paper used those mass and radius values as upper limits, because not only is the planet too small to detect in transit, but it's also too low in mass to detect with RV. The planet is one of those tiny (sub-)Mercury planets that are evaporating large amounts of dust into space and producing an anomalous transit light curve. This feature requires a rather low surface gravity.

The K2-ESPRINT Project. I. Discovery of the Disintegrating Rocky Planet K2-22b with a Cometary Head and Leading Tail
http://arxiv.org/abs/1504.04379v2

The mass and radius values come from Table 4.

Update. I have corrected the Wikipedia article.
Explorer1
Yes, just as I suspected. I did a quick search t to find out if there were any journal articles, but found nothing. Looks like yet again truth is stranger than (science) fiction.
JRehling
There are probably some planets made largely/entirely of gaseous rock and/or metal, which is pretty strange to consider. But I'd be very surprised if we find any planets much denser than iron (that is, iron's various denser states, under pressure). Iron is far more common than any elements heavier than iron, and I can't think of many processes that would discard iron and keep the heavier stuff. The center of the Earth's core is modeled to have a density of about 14 g/cm^3. I think the absolute maximum for any extrasolar planets would be around that level or less.
HSchirmer
QUOTE (JRehling @ Nov 2 2015, 05:01 PM) *
...I can't think of many processes that would discard iron and keep the heavier stuff.
... I think the absolute maximum for any extrasolar planets would be around that level or less.


Well, what about a white dwarf that hits something and goes splat.

So, is there an inverse of the Chandrashekar limit, that requires small pieces of electron degenerate matter revert to "ordinary" matter below a specific mass? I recall theories about chunks of material made with charmed or strange quark properties remaining stable once formed
JRehling
QUOTE (HSchirmer @ Nov 2 2015, 08:44 PM) *
Well, what about a white dwarf that hits something and goes splat.


The escape velocity of a white dwarf is about 6,000 km/s, white dwarfs have a mass of over 100 Jupiters, and collisions are inelastic. That is never going to produce planets.
HSchirmer
QUOTE (JRehling @ Nov 3 2015, 04:20 PM) *
The escape velocity of a white dwarf is about 6,000 km/s, white dwarfs have a mass of over 100 Jupiters, and collisions are inelastic. That is never going to produce planets.


Agreed. You can't hit a white dwarf and break it up, unless you have a really big hammer.
Consider a white dwarf in a loose binary system where the other star goes supernova.
That shock wave should disrupt the dwarf, possibly leaving planetary mass chunks, or blasting it into smaller chunks.
Since we see planets around pulsars, even after a supernova, the disrupted material should reform planetary mass objects.

I'm not sure what the minimum gravitational force is to keep degenerate matter "packed down", thinking about it, it would seem
that the degenerate matter would "re inflate" violently when a white dwarf is disrupted.
JRehling
QUOTE (HSchirmer @ Nov 3 2015, 09:59 AM) *
Consider a white dwarf in a loose binary system where the other star goes supernova.
That shock wave should disrupt the dwarf, possibly leaving planetary mass chunks, or blasting it into smaller chunks.


It's probably not the strongest use of this board to offer qualitative assumptions about never-observed phenomena. The outcomes of huge, cosmic collisions require pretty detailed research to determine, and math-free estimates of how they would proceed doesn't seem likely to be productive. A little math suggests to me that a supernova would not break a white dwarf apart, but that's not really on-topic for this board even if the issue were investigated thoroughly, much less when assumptions are made without math.
Hungry4info
Besides, we know of numerous pulsar+WD binaries.
nprev
JRehling is correct. Please stay on topic.
ZLD
JPL posted a short article yesterday further suggesting the comet swarm hypothesis for the strange occurrence with KIC 8462852. The referenced paper is here for those with access. The reasoning is a continued lack of infrared evidence that would point toward other possibilities, mainly the breakup of a large rocky body.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2017 Invision Power Services, Inc.