Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

CellBioGuy comments on X risk update, Gliese 710 will pass thru Oort in 1.35 my - Less Wrong Discussion

4 Post author: morganism 08 January 2017 07:58PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (18)

You are viewing a single comment's thread.

Comment author: CellBioGuy 10 January 2017 04:56:35AM *  2 points [-]


I or businessinsider must be missing something with regards to the degree of danger posed by a star going through the oort cloud or with regards to it being a particularly extreme pass, what with businessinsider calling it the strongest perturbation in the future history of the solar system.

The density of stars in our local neighborhood is ~0.004 stars per cubic light year. This close approach is calculated to be ~0.21 light years (13,000 AUs, 440x the distance to Neptune). A sphere that wide has a volume of 0.04 cubic light years. Stars should be at least that close about 0.016% of the time, or for about 713,000 total years out of Earth's history.

Typical relative velocities of nearby stars are in the low tens of km/s. This one is 13.8 kilometers per second. Let's say it spends circa 0.1 LY on a 'close approach' within this volume. Thats a circa 2000 year encounter and you'd expect circa 356 encounters like this one over the solar system's history, one every 13 million years with more continuing into the future. Indeed the first paper mentions other recent approaches that are nearly as close.

Going back to first principles, if you assume a 15 km/s average encounter speed and an approach radius of 0.21 light years, you expect about ~127 encounters at least this extreme over the solar system's history, one every 36 million years or so. More frequently if the average velocity is higher or the spread of velocities is larger. You have to go down to a distance of 0.019 light years (1190 AUs, or 40x the distance to Neptune) before you're pretty sure you aren't going to get an encounter that extreme or more over the lifetime of our solar system so far. Granted one could argue there could be a bit of a selection effect.

This seems to suggest to me that stellar close approaches are likely less disturbing to inner solar systems and/or that impacts are less disturbing to biospheres than some claim. I know that the latter has some support - there are some pretty big impacts in the geological record not associated with mass extinctions, and those that are associated with mass extinctions seem to coincide with long term stressors like flood basalt eruptions.

EDIT i updated the numbers, the timescale comes to circa 10 million years not 3 million years, caught a math burp, but the point still stands. I still can't quite figure out why they're calling it so extreme.

EDIT 2 Actually read the paper. The popular press quotes them as saying this star is the "strongest disrupting encounter in the future and history of the solar system." They ACTUALLY state that the star has the biggest influence on our solar system for at least the last and the next ten million years which is the period of time they can project, not for all time. Events like this HAPPEN, with greater frequency than mass extinctions on Earth.

EDIT 3: Their units for perturbation on the oort cloud are in units of (solar masses / distance squared / velocity ). The perturbation is stronger the slower the star moves and the more massive it is. This relative velocity is on the lower slope of the curve and the mass is higher than many stars (half a solar mass - most stars are dwarfs and I believe the sun is larger than ~80% of stars) so I can see that the SIZE of the perturbation may be a bit larger than distance alone will indicate. I would need a better handle on the distribution of relative velocities and masses to get a good handle on these numbers.

On a related note, I now can't figure out how they're modeling the effects on comets...

Comment author: morganism 14 January 2017 10:45:22PM 0 points [-]

This article points to a paper that shows destabilized comets coming in at 160 kp/s, that is way beyond anything we have seen before,...

"The team’s spectrographic analysis, using Hubble data collected from two observing runs separated by six days, detected carbon gas and silicon in the light of HD 172555 moving across the face of the star at a speed of 160 kilometers per second."

This is a young star, and disk, but still surprising pertubations.

A New Look at ‘Exocomets’


Comment author: CellBioGuy 16 January 2017 12:10:04AM *  2 points [-]

You can't have an oort cloud comet hit the Earth at 160 km/s. The absolute fastest that anything falling from an orbit bound to our sun (oort cloud comets being very lightly bound) in the outer solar system can hit the Earth is about (((1+ square root of 2)* the Earth's orbital velocity)^2 + Earth's escape velocity^2)^0.5, or about 73 kilometers per second. This is if it falls from basically infinity (oort cloud distances) to the Earth's distance from the sun, reaching solar escape velocity at our altitude (square root of two times our orbital velocity) and then hits us head on in our orbit. The true velocity would vary between 73 km/s and 16 km/s with most values somewhere in the middle. Those measured faster velocities came from comets that were falling closer than one AU away from their parent stars.

Granted that's only a factor of 8 in maximum available energy per unit mass (difference between 73 squared and 160+earth's orbital velocity + dealing with earth's escape velocity as above). Still, this has almost certainly happened before over the Earth's history, many times, on ten to low tens of megayear timescales. Slightly less extreme events would be much more common - events with one tenth the calculated perturbation parameter would be ten times as frequent and come every 1-3 million years. EDIT stellar mass lessens this argument a little, this star is relatively large and thus its perturbation parameter is larger than the average stellar pass at this distance

Comment author: morganism 16 January 2017 09:45:23PM 0 points [-]

yes, i have no idea how a 25 my old star and disk could have rocks in-falling at that speed, seems like even a gas giant wouldn't do that, as Jupiter only gives you 30kps,(outside Roche limit).

Still, if Planet 9 is real, and starts slinging stuff around out there, there may be some un-bound bodies in the system soon enough....

Comment author: CellBioGuy 17 January 2017 02:21:52PM *  0 points [-]

You get infall at speeds like that by falling close to the star - it is the strongest gravitational field in the system and the closer an orbit takes you to it the faster you go. Something falling from the Oort cloud to the surface of the sun reaches a speed of 600 km/s at the moment of impact, and at four solar radii away it is moving 300 km/s. Speed goes down with the square root of distance. Sungrazing comets do this all the time in our solar system, but they are for the most part smallish. The data seeing cometary material moving across the face of another star at that speed was interpreted to mean that there were large numbers of large sungrazing comets falling very close to the star at that point in that star system's evolution. Keep in mind that at the distances you are looking at, the entire star system is basically a pixel for most instruments so you are seeing the whole thing superimposed over itself.

When two objects interact gravitationally, the maximum physically possible delta V provided to a small object by the big object by a near miss is equal to twice the orbital velocity of the large object. From far away it basically looks like an elastic collision in basic mechanics, and that maximum speed is provided if it comes in and 'bounces off' exactly along the orbital vector of the larger object with an initial velocity of nearly zero. Planet 9, if it exists, has an orbital velocity on the order of low single digits kilometers per second and if it sent things into the inner solar system that 'elastic collision' would be very much not along the line of its orbit and impart rather less. The biggest thing that could impart velocity changes would be the other star and it would be passing through at 15 kilometers per second, at right angles to the radial direction towards the inner solar system at close approach. Furthermore, near misses would be exceptionally rare. These things are notable for stirring the orbital parameters of very slow moving oort cloud objects by providing a small far-away force for a short time over only a piece of its orbit and allowing some to approach the inner system on orbits with very similar energies rather than adding appreciable energy to the system.

Comment author: morganism 17 January 2017 08:36:21PM *  0 points [-]

I understand this, and as a young system, you would potentially have a lot more rocks affected by the proposed gas giant, but as you also point out, any un-bounded material should have already been ejected from the system. It is difficult, but obviously not impossible to change parabolas into hyper parabolas to enable these kind of speeds, but they obviously got close enough to hit the roche limit, or simply dissolved like the Christmas Comet of 2014.

Planet 9 is also theorized to be near 90 d(edit:30d) to orbital plane also, so tossing things out where we aren't looking for them is another hazard in itself. I think the orbital plane of galaxy is out where Pluto is now, (because of the diffuculty of finding secondary targets for New Horizons was made more difficult by background clutter from MW) and 9 is another 40d around the orbital plane, so with a (edit:15k) orbit, there is not a great chance it is going to be relevant in the double influence scenario.

Comment author: morganism 14 January 2017 09:41:29PM 0 points [-]

I think they use the models of disruption used by the gas giant displacement/movement.

And it appears it is pretty easy to wobble the system, they are saying that the Planet 9 is responsible for the offset of the solar rotation angle, and the disk of the planets (6 degree).

Solar Obliquity Induced by Planet Nine


There are also a LOT of theories that show that just crossing the plane of the galaxy is enough to disturb the Oort, though the last paper i saw on Neptune doing it was a solid not.

Comment author: morganism 14 January 2017 09:48:22PM 0 points [-]

and as an aside, i think the biggest factor of this paper, is that it gives you a hard date for a filter.

Most AP/Cosmologists say we have 4-5 gy ahead , because of solar lifetime, this shortens that horizon considerably. Tough to survive as a planet based society if you have 1-10 my of bombardment by large impactors, 1st one causes nuke winter, 2nd one, heat can't escape because of dust blanket, you end up Venusian...

As the t-shirt says, Dino asks you how your space program is doing....

Comment author: CellBioGuy 16 January 2017 12:30:44AM *  0 points [-]

What do you mean by 'we have' 4-5 billion years? A lot happens in that time. Theres a several percent chance of destructive orbital dynamical chaos in the inner solar system during that timeframe coming from the eccentricity of Mars or Mercury, the Earth will almost certainly pop into Venus mode from the increasing solar luminosity by 1.2 gigayears from now (if some parameters are overly harsh maybe as little as 300 megayears and DEFINITELY by 2 gigayears from now). And amusingly enough if we have a longer period of time before popping into the runaway greenhouse, there is a reasonable chance of a carbon crisis in which atmospheric carbon falls low enough that it becomes the limiting factor in biomass production due to the slowing geosphere and increasing burial of carbonate rock. And the average mammalian genus lasts what, ten, twenty million years?

This being said, I again point out that events like this have almost certainly happened many times over the history of the solar system. Even if events like this are rarer than every few tens of megayears due to the larger than average mass of this star, the sun was only 1% dimmer a hundred million years ago. Runaway greenhouses aren't THAT easy to get going and I believe there is evidence from the KT impact that the atmosphere was only strongly disturbed for a short time, potentially single digit years. And there are large impacts not associated with mass extinctions, leading some paleontologists to suggest that they only are associated with mass extinctions when the biosphere was already strongly stressed by something chronic like a large flood basalt eruption. (I myself am partial to the idea that the KT event in particular represents an ongoing low-level flood basalt eruption that got kicked into high gear for a few thousad years by the more or less worldwide 9-pointer earthquake the impact would've generated, providing a double punch). The biosphere and complex life in general is not threatened by impacts.

Comment author: morganism 16 January 2017 09:21:10PM 0 points [-]

4-5 gy is stellar lifetime that most astrophys guys throw out there when discussion of solar sys/Earth comes up.

I agree with the flood basalt/ volcanics postulation, i was never convinced they were extinction driver by themselves.

I thought the orbital danger was Venus, as it is still so close to us on perihelion that it has gravitational interaction.? Reminds me of the exoplanet system found with 5 planets inside the orbit of Mercury, we are a pretty unusual system...

Comment author: CellBioGuy 17 January 2017 02:14:23PM *  0 points [-]

I recall seeing a paper with orbital dynamical simulations in which they found a 2 percent chance that over the remaining lifetime of the sun, the orbital eccentricity of Mercury would increase such that it interacted with Venus, either hitting Venus or getting boosted onto an Earth-interacting trajectory which could lead to a collision or eject it from the solar system altogether, and which puts Venus on a much more closely Earth-interacting orbit. In one simulation out of the 2,500 they did, Mars's eccentricity was perturbed until it became Earth-crossing.

Remember that there is a bit of a selection effect when it comes to looking at exoplanets - we see the compact large planet systems much more easily than systems like ours. The latest work I've seen has suggested that we are a less common class of star system but that stuff like ours might be something like 10% of star systems. Nobody REALLY has a handle on planet formation yet, and it looks like there may be several very different ways that planets and planet-forming material can migrate around the protoplanetary disc during the planetary accretion stage that people argue bitterly over in the literature.

Comment author: CellBioGuy 10 January 2017 05:03:49AM *  0 points [-]

On another note, you can come up with an estimate of the hard maximum fraction of stars over Earth's history that bear self replicating systems capable of bridging different distance scales with something that can survive!