on the Death Star

HIP 85605
Behold your (possible) doom! (it’s the slightly dimmer one just right of center)

So this story is going around: http://www.iflscience.com/space/more-dozen-stars-will-be-making-close-encounters-earth http://www.forbes.com/sites/brucedorminey/2014/12/29/outer-solar-system-likely-to-collide-with-orange-dwarf-star/ http://www.nbcnews.com/science/space/death-star-coming-us-study-says-its-possible-dont-panic-n277741

A star is going to collide with the solar system! The Death Star is coming! We’re all doomed! Dooooooooomed!


Well, not really. This is an ordinary case of a press release taking things a little too far because it sounds more dramatic.

This story actually goes back a long way. Back in the 1980s, some paleontologists realized that mass extinctions on the Earth seemed to happen at intervals of 26 million years. Astronomers hypothesized that the Sun has a companion (at the time, it was considered unusual for the Sun to NOT have a companion; now we know that 58% of all sunlike stars are single) that would disturb the Oort cloud of comets once per orbit and sling a whole bunch into the inner solar system, where they would bombard the Earth.

The downside to this picture is that the companion star, which they named Nemesis, should have been visible in infrared surveys. The fact that we’ve found nothing in the last 30 years of increasingly powerful surveys suggests that the largest remaining undiscovered object in the solar system should be a gas or ice giant. Nobody puts much stock in the Nemesis hypothesis now.

There are alternative theories to explain mass extinctions astronomically. Other stars – not orbiting the Sun at all – frequently come close enough to disturb the Oort cloud anyway. That’s the scenario Baylor-Jones was looking at.

He’s not the first one to look at it – many previous studies have noted that the star GJ 710 is heading almost directly for us and should (by most accounts) pass within half a light year… 1.3 million years from now.  (The current closest star is Proxima Centauri, at 4.2 light years). The novelty in this study is that he’s found a different star – HIP 85605 – is possibly going to come even closer, within 1/3 of a light year.  That’s still 20,000 AU away, 1500 times more distant than Neptune, but certainly within what we think the Oort cloud is.  Getting that result required careful analysis of the radial (towards and away from us) and proper (sideways) motion, and Baylor-Jones included the effects of galactic orbital motion and thousands of randomly selected trials.

I’m not sure I believe Baylor-Jones’ main result though. There are some issues with Baylor-Jones’ data on that particular star, most of which he points out in the paper itself (and even the abstract summary, although all the press results ignore that). Due to some details of its construction, Hipparcos had problems with a few binary stars that were very near to 20″ apart, and HIP 85605 and its companion HIP 85607 appear to be one of those cases: HIP 85605 is very close with a huge uncertainty on the distance measurement, HIP 85607 has a negative distance that’s really evidence of something having gone wrong.  The supposed sideways (proper) motion of the stars is also very different in the Hipparcos catalog than what other people have found.

Baylor-Jones used those alternative proper motions but still used the Hipparcos parallax, which I don’t feel is justified: Proper motion and parallax are solved at the same time, so if the Hipparcos proper motion is wrong, the Hipparcos parallax should be too. There was a paper that Baylor-Jones mentions that tried to fix the problems with the confused binaries; those authors got the same proper motion other people have seen but put the HIP 85605/HIP 85607 system more like 50-1000 parsecs away. If THEY are correct, Baylor-Jones calculates that HIP 85605 will never come anywhere near the Sun. So, as he notes, it’s possible that GJ 710 is still the closest pass for the next couple million years.

On the other hand, the reason Baylor-Jones published this extraordinary conclusion in the first place is that his numbers were endorsed by one of the pre-emininent experts on Hipparcos data, Floor van Leeuwen. van Leeuwen did some checking and told him that the parallax of HIP 85605 is real, and it’s not actually in the same star system as HIP 85607, and that the proper motion IS incorrect… Given that THAT guy has probably forgotten more than I’ll ever know about astrometry, I don’t know what to think.

van Leeuwen told Baylor-Jones to be careful making conclusions with the data, and he was. The news reports unfortunately make it sound like a certainty. That’s a bit premature. And anyway, this close pass is supposed to be 330,000 years in the future (or 1.3 million for GJ 710). So, we’ve got ample time to build some planetary defense shields or evacuate to some distant habitable planet. If this is the death star, it’s the slow motion death star.


2 Comments Add yours

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s