KIC 8462852 Observations Options

[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.

[dudley]

If I recall correctly, the Kepler Space telescope found no periodicities in the dips in light from this star. We're assuming, it seems, that the plane in which any planets, or their shattered debris would travel, aligns with the star from our point of view.

Even if two planets had collided and destroyed themselves, we might have expected other planets, or even just one, to be found, mightn't we? If there was a complete absence of planets around this star, these collisional scenarios would seem to be unworkable.

[HSchirmer]

If I recall correctly, the Kepler Space telescope found no periodicities in the dips in light from this star. We're assuming, it seems, that the plane in which any planets, or their shattered debris would travel, aligns with the star from our point of view.

The did see something with a 750 day period.
For reference, mars' orbit is 687 days, so assuming the "something big" is in orbit,
that would be, eh, roughly around the inner edge of our asteroid belt.

Even if two planets had collided and destroyed themselves, we might have expected other planets, or even just one, to be found, mightn't we? If there was a complete absence of planets around this star, these collisional scenarios would seem to be unworkable.

Brilliant question!
Complex answer.

Rephrased-
"If Kepler was looking at our own solar system from Tabby's planet, how many of our planets would it see transit?"

If all the planets around a sun have the same angle, and the same 'phase' (argument of perihelion?) then yes.
In practice, no.

Here are our planet's orbital tilts relative to earth.

Mercury 7.01°
Venus 3.39°
Earth 0
Mars 1.85°
Jupiter 1.31°
Saturn 2.49°
Uranus 0.77°
Neptune 1.77°
https://en.wikipedia.org/wiki/Orbital_inclination

It's a difference of a few degrees, but at several million KM away from the sun, a small angle leads to
a big distance above or below the orbital plane.

And, each tilt is going to be oriented in a different direction-
imagine looking down at the solar system as if it were a compass.
So, Mercury's tilt is above the plane towards North and below towards South,
but Venus' might be above at East and below at West.

Another way to think about it- why are solar eclipses special?
Even though we have a (relatively) large moon, that is close to the same plane as the sun,
it's actually quite rare for the orbit of the moon to line up with the orbit of the sun.

[JRehling]

The did see something with a 750 day period.

No. The first dip, which by itself resembled one planetary transit, was seen at mission day ~792 and then a very strange series of several dips took place roughly from days 1510 to 1570. Nothing in the later series of dips is obviously a repeat of the first observed dimming event. The paper has a section discussing the possibility of the second event being related to the first, with a ~750 day period; in the discussion, that possibility is called "problematic." Follow-up observations showing a lack of IR excess further discredit that possibility.

Without a third observation, the inference that two light curve dips indicate one specific "something" is doubtful even in a less strange case. An observer elsewhere could see Venus and Earth both transit the Sun, but wouldn't know whether those two transits were caused by one planet or two planets until a third transit was seen.

Whatever was seen between days ~1510-1570 was seen only once, and what was seen at day ~792 was not obviously seen even one more time, much less two more times.

[HSchirmer]

No. The first dip, which by itself resembled one planetary transit, was seen at mission day ~792 and then a very strange series of several dips took place roughly from days 1510 to 1570. Nothing in the later series of dips is obviously a repeat of the first observed dimming event. The paper has a section discussing the possibility of the second event being related to the first, with a ~750 day period; in the discussion, that possibility is called "problematic."
...

Yep, I should have said "they're looking" at a 750 day period, but more data is needed.

The dip at D1500 is then interpreted as the same material seen one orbit later,
with the 750 day period implying an orbit at 1.6 AU.
...
A more robust prediction is that future dimming events should occur roughly every 750 days,
with one in 2015 April and another in 2017 May.
..,.
Thus, while this scenario is attractive because it is predictive, the periodicity argument may be inconsistent, and the probability of witnessing such an event may be very low
...

Likewise, if the two deep dipping events at D800 and D1500 are from the same orbiting body (or bodies), a
period of 700 – 800 days remains a possibility.
...

http://arxiv.org/pdf/1509.03622.pdf

So, suppose two super-Earth sized rocky planets were ejected to the outer part of that star system, and then happened to collide very recently. Could that be a reasonable scenario?

Eh, I think that would be overkill.
The paper mentions that you don't need a really huge impactor to create a suitable dust cloud-

A broad range of scenarios for the dipping behavior that involve occultation by circumstellar dust clumps was considered. Among these, we find that the break-up of one or more massive exocomets (or planetesimals on comet-like orbits) provides the most compelling explanation consistent with the data in hand.
The required mass of the original body may have been in excess of 3×10^21 grams (only 0.3% the mass of Ceres, and perhaps 100 km in diameter).

It's not that hard to conceptualize an area where several 100 km objects orbit, we call it the asteroid belt. The problem seems to be that there is no known mechanism for comets or asteroids to break up in a coordinated manner to generate the 100 year dimming, or the series of Kepler dips. The papers about KIC 8462852 follow current ideas about comets and assume that extra-solar comet clouds have stable circular orbits which require passing stars to trigger comet showers.

That's why I though the comet-ejected-by-gas-giant versus comet-scattered-by-ice-giant issue was interesting.
It's a tangent about the circularization of hyperbolic comet orbits to create the Oort cloud. The rule stated above

Inner planets are not ejected into orbits with periapses much farther out than the body that ejected them. They are either sent into elliptical orbits that return to the inner system or they are ejected from the system entirely.

doesn't seem to apply to Oort cloud comets. Comets that were ejected from the solar system are though to somehow have circular orbits which require another perturbation to return them to the inner solar system.

Computer simulations of our solar system suggest that 1) planets like to form near the "snow line" where ices first condense, 2) planets can shift because of orbital resonances, and 3) when planets shift they toss lots of small stuff like comets around. Gas giants like Jupiter and Saturn have enough gravity to scatter comets so far out (50k - 100k au) that the comets stay out there in the Oort cloud, and the Oort cloud is estimated to have about 5 earth masses worth of comets, divided into trillions of bodies, with several billion 20km or larger. In contrast, when ice giants like Uranus and Neptune scatter comets, they don't go as far and are assumed to return along the elliptical orbits they started on.

Point is: current theories about Sol like systems indicate that when planets shift with gas giant(s) to eject comets you end up with 5 earth masses of comets into trillions of distant, stable, circular orbits. Then you need a passing star to disturb those circular orbits and create killer comet showers. However, it seems that ice giants only scatter comets, so you should end up with 5 earth masses of comets into trillions of nearby, unstable, elliptical orbits. Then you don't need any second event to get killer comet showers.
So, could KIC 8462852 be the model for a "Kessler catastrophe" solar system?