Spotting a star's wobble

Bullgrit · Feb 3, 2012

So, I understand that astronomers are detecting some extra-solar planets by measuring how much the star wobbles from the orbit. But isn't the amount of wobble equal to or [much] less than the size of the planet causing the wobble? If our instruments can detect the wobble, how is it they can't see the planet? Isn't the wobble much smaller than the size of the planet?

Bullgrit

Umbran · Feb 3, 2012

They are tricky in detecting the wobble.

Generally, they do *not* take two pictures, and note that the star is in a different position in each one. You are correct, if we could do that, we could just see the planet. But, the resolution on our telescopes isn't good enough to measure that.

Instead, they look at light coming from the star, and measure very small doppler shifts in the color of the light. When the star is coming towards us, the light shifts slightly blue, and when it moves away, it shifts slightly red. If you watch over time, and see a regular pattern of red to blue to red to red to blue, you know the period of the wobble.

Ed_Laprade · Feb 3, 2012

And I read somewhere or other that they can't release any findings until they've observed the star for several years. Seven, I think? So all these 'new' reports were actually started quite some tme ago.

Umbran · Feb 3, 2012

Can't? Well, it isn't physically impossible for them to come to a conclusion more quickly - how long it takes depends on the star and planets invovled. Nor is it like there's some law against it, and they're going to get arrested if they release findings sooner than that.

That being said, this method does take a while. Good measurements have to factor out the Earth's motion around the Sun, and that's easiest to do if you take measurements over more than one of our years.

And if you want the best data, you also want to take data over the course of multiple wobble-cycles of the target star. That means multiple orbits of the suspected planet. Depending on the planet's orbit, that could take several of our years. If you were trying to detect Jupiter using this method, for example, you're talking about taking measurements over 12 years or more to get just one wobble-cycle.

For interested folks, the Wikipedia page on the subject looks pretty good, at least to my cursory inspection:

Doppler spectroscopy - Wikipedia, the free encyclopedia

Dannyalcatraz · Feb 3, 2012

I thought this was going to be a thread about drunk celebrities...

(

)

Fast Learner · Feb 4, 2012

Dannyalcatraz said:
I thought this was going to be a thread about drunk celebrities...

That's what I was going to pretend, too, and make a remark about paparazzi doing it all the time.

Dioltach · Feb 4, 2012

I just want to add that I love the sound of the word "wobble". One of the best things about getting older is that I can use the word so often in conversation.

Dannyalcatraz · Feb 4, 2012

Well, here's another way to do that!

Virus B- Jah Wobble / Bill Laswell - YouTube

Ed_Laprade · Feb 4, 2012

Umbran said:
Can't? Well, it isn't physically impossible for them to come to a conclusion more quickly - how long it takes depends on the star and planets invovled. Nor is it like there's some law against it, and they're going to get arrested if they release findings sooner than that.
Doppler spectroscopy - Wikipedia, the free encyclopedia

I don't recall if the article mentioned who made the rule, don't think so, but the 'enforcement' would have to be self-policing. I gathered that it was ment to keep anyone from jumping the gun by making an announcement using insufficient data.