!MeRdFpEonLoCwhoHeT:matrix.org

SunPy

149 Members
Python for Solar Physics - https://sunpy.org - +sunpy:openastronomy.org for other channels29 Servers

Load older messages


SenderMessageTime
6 Jul 2020
@dstansby:matrix.orgdstansby with AIA data I've also seen m.meta['rsun_ref'] = sunpy.sun.constants.radius.to_value(u.m) used to fix some metadata which might help 16:35:52
@dstansby:matrix.orgdstansbye.g. you can see what happens when you don't do that here: https://docs.sunpy.org/en/stable/generated/gallery/map_transformations/reprojection_heliographic_stonyhurst.html#sphx-glr-generated-gallery-map-transformations-reprojection-heliographic-stonyhurst-py16:36:11
@dstansby:matrix.orgdstansbywhich looks similar to what is happening in your figure above16:36:17
@cadair:cadair.comCadairyeah that can be important if you have different assumptions about the radius of the Sun16:36:33
@deadzero:matrix.orgsamaloneythey are different but only at 0.0004 % so I wouldn't think that's the issue but I will test16:40:56
@deadzero:matrix.orgsamaloney
In reply to @dstansby:matrix.org
with AIA data I've also seen m.meta['rsun_ref'] = sunpy.sun.constants.radius.to_value(u.m) used to fix some metadata which might help
Yea that was tiny percent was enough
16:57:13
@dstansby:matrix.orgdstansby maybe it is worth raising a warning in sunpy when Map reads in a rsun_ref that isn't the same as the value in sunpy.sun.constants? 16:58:45
@deadzero:matrix.orgsamaloneyYea I was thinking the same, or at least when it used with reproject it looks like it just masks off extra pixels as being off disk17:02:04
@cadair:cadair.comCadair
In reply to @dstansby:matrix.org
maybe it is worth raising a warning in sunpy when Map reads in a rsun_ref that isn't the same as the value in sunpy.sun.constants?
oh a nice warning that would be raised every time someone opened an AIA image, and you wouldn't want to change it because then your grid lines would be wrong 😉
17:13:26
@cadair:cadair.comCadairI think the reproject idea is better, but I don't know if there is a way we could do that17:13:44
@ayshih:matrix.orgayshihI was just about to write all of the above17:14:11
@dstansby:matrix.orgdstansby where in the transformation pipeline is rsun_ref used? 17:16:07
@cadair:cadair.comCadairwhen calculating the position on the sphere when making the HPC coord 3D17:16:52
@cadair:cadair.comCadairit's only an issue when they mis-match in a transform17:17:18
@ayshih:matrix.orgayshih sunpy.coordinates.Helioprojective.make_3d 17:17:25
@documentation-bot:cadair.comDocumentation Bot sunpy.coordinates.Helioprojective.make_3d 17:17:35
@ayshih:matrix.orgayshihThe issue is that there's nothing wrong with having a non-IAU value for Rsun. It could simply be what the project chose to use for unit conversions (e.g., AIA choosing to round the value). It could also be used to denote an actual physical difference.17:22:13
@ayshih:matrix.orgayshih
In reply to @deadzero:matrix.org
they are different but only at 0.0004 % so I wouldn't think that's the issue but I will test
Can you confirm that you mean 0.04%? If the difference were actually 0.0004%, it'd probably work fine.
17:28:48
@ayshih:matrix.orgayshihs/fine/better/17:29:51
@ayshih:matrix.orgayshih * s/fine/better/17:30:17
@deadzero:matrix.orgsamaloneyyup sorry was 6960.... v 6957.... which 0.04%17:30:29
@sed:maunium.netS. Edbot Can you confirm that you mean 0.04%? If the difference were actually 0.0004%, it'd probably work better. 17:31:30
@sed:maunium.netS. Edbot * Can you confirm that you mean 0.04%? If the difference were actually 0.0004%, it'd probably work better. 17:31:31
@ayshih:matrix.orgayshihRedacted or Malformed Event17:37:44
@ayshih:matrix.orgayshih astrofrog (or Cadair ): See this comment (https://github.com/sunpy/sunpy/issues/3328#issuecomment-528484639). Was anything ever documented in reproject? 17:38:35
@cadair:cadair.comCadairErrrr17:40:29
@cadair:cadair.comCadairDonno17:40:38
@dstansby:matrix.orgdstansbyis rsun_obs a FITS standard keyword? I can only find mentions of it in SDO related documents/websites17:51:13
@dstansby:matrix.orgdstansby I am wondering if sunpy should warn when assuming rsun-obs because it's assuming something about missing metadata 17:52:11
@cadair:cadair.comCadairit's not really standard no18:12:22

There are no newer messages yet.


Back to Room List