Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Verifying extrasolar planets?

  1. Jun 16, 2010 #1

    I'm a programmer with a strong grasp of maths and wish to verify the existence of extrasolar planets. I assume they just use basic Newtonian physics.

    Where can I download a dataset which has confirmed extrasolar planets so I can replicate the method they use to find them? What method do they use so I can recreate it?

    NASA just released loads of Kepler data at http://archive.stsci.edu/pub/kepler/lightcurves/tarfiles/

    I want to eventually assist in analysing the data so thanks for any suggestions

    Bonus points if you give me the dataset for the coolest extrasolar star system- Gilese 581 ;)
  2. jcsd
  3. Jun 16, 2010 #2
    Here's a link for what I asked http://oklo.org/downloadable-console/

    And for anyone else interested here's a person on irc teaching me (thanks)

  4. Jun 16, 2010 #3


    User Avatar

    Staff: Mentor

    I stopped reading shortly after "Does the word "photometry" mean anything to you?"

    It appears not...

    Photometry is using a telescope and ccd camera to measure the brightness of an object. So then the most common method is to take continuous measurements of the brightness of a star and detect periodic changes in that brightness. If the star is not a known variable and the brightness occasionally drops, there is a good chance that the drop is due to the existence of a planet.

    So you don't need any data to confirm the existence of such a planet, just a foreknowledge of if such a planet has already been found around a star you are looking at. Calculating properties is a little more difficult (but not too difficult) and is where Newton comes in. If you know the mass of the star and the period of the transits, you can calculate the orbital distance. If you know the brightness of the star and measure the brightness change, you can calculate the size of the planet.
  5. Jun 17, 2010 #4
    yep thanks that works too. But afaik you have to be coplanar with the star-planets orbit which is rare. the method I was asking about was the one using the wobble in a stars movement (away and towards the viewer).

    [PLAIN]http://img8.imageshack.us/img8/9383/starsystem.png [Broken]

    My first discovered planet this morning. very happy about that
    Last edited by a moderator: May 4, 2017
  6. Jun 20, 2010 #5
    Basically, you can detect and determine characteristics of an existing planet. providing that your instruments are sensitive enough and the planet not too small. This workd basically for any orbit except one that is normal to the line-of-sight path from us. It takes allowance of the fact that the star and planet are a two-body system, and they might be described to orbit each other - - - like a spinning dumbbel, in which one mass is usually considerably larger than the other. In other words, the system turns about a point usually somewhere inside the star, but not at its center of mass. This means that the star itself is pivoting around that point, and as long as that motion is not normal to our sight line - - - it means that the star is moving toward us for a while, then away for a while, and so on.

    This all results in a continual shift in frequencies of the spectral lines of that star (doppler shift). This shift, along with the considerations Russ gave, allow the planet's parameters to be calculated, if it isn't too small to make out.

    If the orbit is normal, to us, calculation can still possibly be made, by measuring the star's lateral shift, but this is often not so easy.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook