Earth shadow at sat altitude issues

[axd1967]

Expected Behaviour

See #1246 and #2360

Actual Behaviour

Umbra should start where satellites become reported as "not sunlit".

Also, "Show umbra at distance" (which makes one wonder whether distance to the observer is intended) should probably read "Show umbra at altitude".

Also, it is odd that the umbra sticks above the horizon when the Sun is visible (or e.g. rising/setting). The umbra should be visible above the horizon during the night only, and should touch the horizon at sunrise/sunset (notwithstanding refraction issues).

It could be (but I haven't checked it) that the umbra should always be near/touch the ASP at any moment.

For geo sats the feature seems to work:

The CU locus seems to converge to the Earth at Moon umbra, which is as expected.

Steps to reproduce

System

• Stellarium version: <Name of downloaded installable file?> 9275de9
  Operating System: Kubuntu 20.04
  KDE Plasma Version: 5.18.8
  KDE Frameworks Version: 5.68.0
  Qt Version: 5.12.8
  Kernel Version: 5.4.0-120-generic
  OS Type: 64-bit
  Processors: 8 × Intel® Core™ i7-10510U CPU @ 1.80GHz
  Memory: 15.3 GiB of RAM

Logfile

If possible, attach the logfile log.txt from your user data directory. Look into the Guide for its location.

[alex-w]

Please share information about used projection

[axd1967]

Projection is stereographic, but that shouldn't matter? It looks like the drawing logic does take projection into account: e.g. fisheye gives this:

The image shows the ISS about to enter the Earth shadow (red, marked "3") in a few seconds. That is where the Earth shadow should start. According to the current implementation, the ISS should enter the Earth shadow at "2", which clearly conflicts with the calculations of the satellite tool.

[alex-w]

Please remember about the trigonometry and projections (or sines/cosines in triangles)...

[Atque]

@alex-w Wouldn't the shadow markings fit anyway?

[alex-w]

@alex-w Wouldn't the shadow markings fit anyway?

See the placement of satellite at points A and B on the drawing:

We render the umbra's circle at altitude H and we do not compute projection the placement of satellite to axis of shadow cone.

[gzotti]

I started investigating this yesterday. The circle should be drawn not at distance "earth_radius+H" but at "(earth_radius+H)*cos(earth_radius+H, antisolar_direction)" where "earth_radius+H" is of course the (geocentric) satellite position.

[Atque]

I don't see the point in showing the shadow, if it does not reflect the border between visible and invisible.

[gzotti]

In the drawing, when the satellite is at B, the shadow circle likewise must go through B. This can be achieved with the above change.

[github-actions]

Hello @axd1967!

The bug or issue has been fixed! You may test it via building Stellarium from source code or wait the weekly development snapshot...

[github-actions]

Hello @axd1967!

Please check the fresh version (development snapshot) of Stellarium:
https://github.com/Stellarium/stellarium-data/releases/tag/weekly-snapshot

[github-actions]

Hello @axd1967!

Please check the latest stable version of Stellarium:
https://github.com/Stellarium/stellarium/releases/latest