search

skyfielders / python-skyfield

Elegant astronomy for Python
MIT License
1.43k stars 213 forks source link

Finding the circumpolar circle #109

Closed tcz closed 7 years ago

tcz commented 8 years ago

This is a question about usage: I'd like to draw the circumpolar circle from a certain point on the surface of Earth. I'm wondering if this is possible to do with skyfield (that is, finding the location of the circle).

ghost commented 8 years ago

The circumpolar circle at latitude L is simply the declination line 90-L (in degrees).

tcz commented 8 years ago

Sorry, I'm not sure I'm following. Don't I need to find the celestial pole first?

davidmikolas commented 8 years ago

@barrycarter Precession & Nutation? I'm not sure if the current axis of the Earth is exactly perpendicular to the "z axis" in Skyfield's coordinate system.

brandon-rhodes commented 8 years ago

@tcz Where are you planning on drawing the circle — on what kind of plot, and against what kind of coordinate system governing the plot? The answer will be different depending on what kind of coordinate needs to be the input to the plot.

tcz commented 8 years ago

I'm drawing a sky map with a stereographic projection, picturing the whole visible celestial hemisphere from a certain surface point on the Earth (and a certain time). I am using right ascension and declination data (and other information) of stellar objects and convert them to apparent position on the sky (altitude, azimuth) using Skyfield.

On Thu, Sep 8, 2016, 20:48 Brandon Rhodes notifications@github.com wrote:

@tcz https://github.com/tcz Where are you planning on drawing the circle — on what kind of plot, and against what kind of coordinate system governing the plot? The answer will be different depending on what kind of coordinate needs to be the input to the plot.

— You are receiving this because you were mentioned.

Reply to this email directly, view it on GitHub https://github.com/skyfielders/python-skyfield/issues/109#issuecomment-245699970, or mute the thread https://github.com/notifications/unsubscribe-auth/AAJxCL1iZkJ6tiuTysMvRlXAKwrLjGvFks5qoFhggaJpZM4J27xC .

brandon-rhodes commented 8 years ago

To draw a circumpolar circle around the J2000 pole, you could create a few hundred Star objects evenly spaced at RAs from 0h through 24h and at the declination where you want the circle. Compute their positions and coordinates on your diagram, and then draw lines between them to create the circle.

For circumpolar circles around poles of other years besides J2000, you will need to apply a rotation because the dynamical coordinate system changes each year. I wonder if Skyfield should in the future include a method that makes it easy to turn RA and dec values for different years into pure ICRS coordinates?

ghost commented 8 years ago

I was actually aware of nutation/precession when I wrote my answer, but was hoping my answer would be accurate enough and no one would catch me out :)

My http://astronomy.stackexchange.com/a/13009/21 may or may not be helpful.

Brandon, re your wondering, yes, yes it should :)

ghost commented 8 years ago

If you need exceptionally high precision https://naif.jpl.nasa.gov/pub/naif/generic_kernels/pck/ may help. Be sure to read the aareadme.txt file for details.

brandon-rhodes commented 7 years ago

@tcz — As far as I know, the approach I suggested (approximating the circle with points distributed in a ring around the celestial pole) should let you draw a circumpolar circle with as high an accuracy as you would like to achieve. So I am going to tentatively close this issue for now as solved. But please re-open if you have further questions or problems with your diagram!