Welcome, Guest. Please Login.
Gravity Simulator
11/18/17 at 06:12:23
News: Registration for new users has been disabled to discourage spam. If you would like to join the forum please send me an email with your desired screen name to tony at gravitysimulator dot com.
Home Help Search Login


Pages: 1
Send Topic Print
Alpha Centauri should harbor Earth-like planets (Read 3570 times)
APODman
Uploader





Posts: 35
Gender: male
Alpha Centauri should harbor Earth-like planets
03/25/08 at 05:59:04
 
Many astronomical works in the last times are using astronomical integrators ( especially Mercury of John Chambers ) had used to determine the possibility of the formation or existence of exoplanets in another solar systems.
 
One of the more recently researchs says that a planetary system can be formed in Alpha Centauri system. A cool detail is that this planets are like terrestrial planets and can be situated in habitable zone around Alfa Centauri B:
 
Quote:
UCSC graduate student Javiera Guedes used computer simulations of planet formation to show that terrestrial planets are likely to have formed around the star Alpha Centauri B and to be orbiting in the "habitable zone" where liquid water can exist on the planet's surface. The researchers then showed that such planets could be observed using a dedicated telescope.
...
To study planet formation around Alpha Centauri B, the team ran repeated computer simulations, evolving the system for the equivalent of 200 million years each time. Because of variations in the initial conditions, each simulation led to the formation of a different planetary system. In every case, however, a system of multiple planets evolved with at least one planet about the size of Earth. In many cases, the simulated planets had orbits lying within the habitable zone of the star.

 
font: http://www.spaceref.com/news/viewpr.html?pid=24935
 
In your site Javiera offear the paper of the research. He made a simples work: create between 400 to 900 planetoids with the mass of the moon in a circunstelar disk of 3AU ( another details how inclination, excentricity, node, perifocus, mean anomaly are also exposed in the paper ).
 
They run the simulation until 200 million years to test the stability of the system but normally in only 70 million years the planets are formed.
 
The data bellow shows the results to each number of particles ( "RUN" column ). The planet formed are showed in "M" columns:
 

 
42% of the planets are formed inside the habitable zone ( the green mark is the habitable zone limits ):
 

 
 
I try replicate the results with Gravit Simulator. I construct the Alpha Centauri A/B system  ( the data of the system here: http://homepage.mac.com/andjames/PageAlphaCen002.htm )  than create 500 planetoids with the same mass of Javiera paper around Alpha Cen B ( the file of simulation is annex ).
 
 
But I see after some time of simulation that the particles are ejected of the disk and none collision occur.
 
Can the Gravity Simulator reproduce the results of Javiera ?  Can Gravity Simulator be used for professional research ?
 
 
More info here:
- http://www.spaceref.com/news/viewpr.html?pid=24935
 
 
[ ]īs
Back to top
 

The diversity of the phenomena of nature is so great,and the treasures hidden in the heavens so rich,precisely in order that the human mind shall never be lacking in fresh nourishment.
Kepler
View Profile   IP Logged
Tony
YaBB Administrator
*****




Posts: 1049
Gender: male
Re: Alpha Centauri should harbor Earth-like planet
Reply #1 - 03/25/08 at 11:31:49
 
Most professional astronomers will write their own code, or use code produced by other professionals.  This gives them the freedom to modify the code as needed.  Gravity Simulator was designed to give amateurs access n-body software that could do many things so modifying the code would not be required, and people with no programming experience can run detailed simulations too.  But there's no reason a professional couldn't use it.  Several have, although I'm not aware of any papers published where Gravity Simulator was used.  But that's not to say it couldn't happen.
 
Regarding your simulation:
To increase the odds of a collision, make the diameters larger.  1000 is rather small for a moon-mass object.  The real moon is about 3500 km in diameter.
 
The biggest problem is that you need to take a slow time step.  In your attached sim, you're using 32768.  The moons are moving at about 12 km / second relative to each other.  So with each time step they are moving about:  12 km / second * 32768 seconds/timestep =~400000 km / timestep.  This is a problem when the moonlets are only 3500 km wide.  The odds are that two moonlets that are destined to hit will simply jump over each other instead.  
 
There's three solutions to this problem:
 
1.  Decrease the timestep to about 128 seconds.  Now your moonlets are moving 1536 km/timestep, making it very likely that two moons on a collision course will actually collide.  But the downside is that you will only be simulating about 1 year per hour.  So to go 200 million years would take 200 million hours, or 22 years.
 
2. Write modified code that automatically lowers the time step when two objects approach each other.  That's something I've thought about doing before, but just never got around to it.
 
3.  (Best method) Set up an analgous system and extrapolate your conclusions.  For example, instead of making each moonlet 3500 km in diameter, make them 350000, or 1000 times larger than real life.  This increases their cross sectional area by 1000^2 or 1 million times.  So they'll be colliding 1 million times as often.  And now you can pump the time step back up to about 8K seconds since the objects have a more difficult time of simply skipping over each other.  After everything is finished colliding, you'll have to draw your own conclusions as to what this exaggerated model tells you about the real world.  You can always play around with different variations, perhaps making the Moons half as big (175000 km) and comparing the final systems.  Do you have as many planets?  Where are they?  How long did it take?  Perhaps you can spot a trend that you can extrapolate all the way down to realistic-sized (3500 km) moons.
Back to top
 
 
Email View Profile WWW   IP Logged
frankuitaalst
Ultimate Member
*****


Great site

Posts: 1507
Gender: male
Re: Alpha Centauri should harbor Earth-like planet
Reply #2 - 03/27/08 at 12:21:03
 
Concerning the timestep question : I have a question about the integration ( Euler method ) .  
Orbits are calculated according to the Euler method which calculates the next position according to :  
Vnew= Vold+ a* Timestep  ( a= acceleration ) . ( I assume a = piecewise constant )  
How however does GravSim calculates the new position ?  
Is this done  : Xnew= Xold + Vold*Timestep + a* Timestepē/2 or Xnew=Xold+Vold* Timestep or something in between ?
Back to top
 
 
Email View Profile   IP Logged
Tony
YaBB Administrator
*****




Posts: 1049
Gender: male
Re: Alpha Centauri should harbor Earth-like planet
Reply #3 - 03/27/08 at 15:20:13
 
do for all object pairs:
a=GM/d2
v=v+a
 
do for all objects:
x=x+v
Back to top
 
 
Email View Profile WWW   IP Logged
Pages: 1
Send Topic Print