Repository navigation
Ecliptic coordinates missing #1430
Description
The astropy.coordinates package does not yet support transformation to or from ecliptic coordinates. While the ecliptic is a bit of a slippery concept at very high precision, it's nonetheless frequently seen and often useful, so I think it makes sense for astropy to support it.
I may be able to assist with an implementation myself (though am ludicrously overcommitted at the moment, so no promises) but in any case I wanted to ask raise this issue so others working on coordinates can comment.
Activity
embray commentedon Sep 10, 2013
Last I saw this discussed I know there were plans to add it, but there were some other intermediate steps that needed to be completed first. I don't really know too much about it though.
astrofrog commentedon Sep 12, 2013
I know @eteq is working on refactoring some aspects of
astropy.coordinatesbut I'm not sure what the status is and whether he is planning to add support for ecliptic or alt/az conversions yet, so any help will likely be welcome.eteq commentedon Sep 12, 2013
Ecliptic is indeed first on the list of coordinate systems to add after the IAU2000 stack to get from ICRS to Alt/Az. The main catch I see in that is exactly as you said, @mperrin - deciding on which definition we want for what ecliptic actually means and fitting it into the IAU2000 scheme.
The implementation catch is that it would be unwise to do this before coordinates have been made array-compatible (currently in-progress on a branch of mine), as it has required some changes to how transformations are defined (that's mainly what's taking it a while to get done). So perhaps we should revisit this subject as soon as that is done?
mhvk commentedon Sep 13, 2013
@eteq -- off-topic, but as you mention Alt/Az: it would be lovely to have that working, and if you want to go the full monkey in term of precision, I'm happy to help with getting Earth-orientation parameters out of IERS (#1145); of course also having sidereal time should help (#1418) -- I think we have enough to make this good at the pulsar timing level; with a bit more effort, perhaps even at that required for VLBI.
mtbannister commentedon Sep 20, 2013
I'd love to see ecliptic coordinates supported; the solar system community aren't going to be keen on astropy.coordinates until that exists. Having a transform to the invariant plane would be a handy addition as well.
eteq commentedon Sep 25, 2013
@mhvk - I like that as a goal (pulsar timing at least, with VLBI even better). It's possible some subtle implementation issues will come up to make this difficult, but until we know that, we'll say that's the target!
@mtbannister -good to know this would indeed. Once arrayification is in and the baseline ICRS <-> Alt/Az is going, I may well ping you for feedback/implementation assistance.
mtbannister commentedon Sep 25, 2013
@eteq I'd be happy to help. Once the coords exist, it'll get immediate implementation over in astroquery (cf. astropy/astroquery#222).
astrofrog commentedon Jan 14, 2015
This is not going to happen for 1.0 so removing the milestone.
@mtbannister - I'm not sure if you are still interested in helping with this, but if so, do get in touch with @eteq as he will be able to provide information on what needs doing.
mtbannister commentedon Jan 14, 2015
Cheers @astrofrog. @eteq, @ijiraq has also mentioned he would be interested in helping with this issue.
astrofrog commentedon Jan 15, 2015
@mtbannister @ijiraq - thanks!
eteq commentedon Jan 22, 2015
@mtbannister, @ijiraq - if you want some ideas on how to most easily implement this with the frames added in astropy v1.0, I suggest you check out http://www.iausofa.org/2013_1202_C/sofa/sofa_pn.pdf, particularly the diagram on page 6. (note that v1.0 uses only the "right side" of that transformation, through CIRS). Depending on exactly how you're meaning to define ecliptic, a few options present themselves:
astropy._erfa.epv00to get the sun->earth or bary->earth vectorastropy._erfa.epv00to get the sun->earth vector and/or the bary->sun vector