Confused about Stonyhurst Grid and jHelioViewer

[Arbu]

TL;DR

jhelioviewer, stonyhurst

So I understand that the Earth is north of the Sun's equatorial plane in around September to March. So a geostationary satellite observing the Sun would seem more of the Sun's northern Hemisphere than its southern. But, using jHelioViewer to show the Sun with its equator marked on as a Stonyhurst grid shows more of the southern Hemisphere than northern. Can anyone shed any light on this?

 

[Tom.G]

Perhaps the 23.5 degree tilt of the Earth's axis?

 

[Arbu]

Perhaps the 23.5 degree tilt of the Earth's axis?

No. The satellite that takes the images is in geostationary orbit and looks straight at the Sun, the tilt of the Earth's axis is irrelevant.

 

[Bandersnatch]

Great. I thought I'd look into this, and now I'm confused too.

Let me rephrase the issue.
The software shows the solar equator as seen from below between early December and early June, and from above the other half of the year. The maximum displacement angle corresponds to the ~7 deg axial tilt of the Sun w/r to the ecliptic. That this is correct is acknowledged e.g. here: https://solar-center.stanford.edu/solar-images/latlong.html
This implies that the axis of rotation is tilted roughly in the direction to the vernal equinox, since about two weeks earlier in March the solar equator is maximally displaced upwards.

Now, sources (e.g. https://link.springer.com/article/10.1007/s10569-007-9072-y) list the direction of the solar north pole in celestial coordinates as ~286 deg RA, ~64 deg dec. I.e. roughly in the opposite direction than the axial tilt of the Earth (w/r to the ecliptic). Which should make the equator visible from below from September to March (opposite to what @Arbu wrote in the OP, I think).

So there's like ~60-90 (?) degrees/quarter of a year shift between what my reading of the coordinates suggests and what happens in reality. Where's the mistake?
I don't know about the OP, but myself, I suspect I might be misreading the coordinates. But I don't see in what way.

 

[berkeman]

Thread is closed for Moderation...

 

[berkeman]

An attempt to use an AI chatbot conversation as a technical reference has been deleted (that is against the rules in PF technical forums), and the thread is reopened provisionally.

 

[Arbu]

Great. I thought I'd look into this, and now I'm confused too.

Let me rephrase the issue.
The software shows the solar equator as seen from below between early December and early June, and from above the other half of the year. The maximum displacement angle corresponds to the ~7 deg axial tilt of the Sun w/r to the ecliptic. That this is correct is acknowledged e.g. here: https://solar-center.stanford.edu/solar-images/latlong.html
This implies that the axis of rotation is tilted roughly in the direction to the vernal equinox, since about two weeks earlier in March the solar equator is maximally displaced upwards.

Now, sources (e.g. https://link.springer.com/article/10.1007/s10569-007-9072-y) list the direction of the solar north pole in celestial coordinates as ~286 deg RA, ~64 deg dec. I.e. roughly in the opposite direction than the axial tilt of the Earth (w/r to the ecliptic). Which should make the equator visible from below from September to March (opposite to what @Arbu wrote in the OP, I think).

So there's like ~60-90 (?) degrees/quarter of a year shift between what my reading of the coordinates suggests and what happens in reality. Where's the mistake?
I don't know about the OP, but myself, I suspect I might be misreading the coordinates. But I don't see in what way.

I now understand the issue. But in the interests of scientific rigour, it has been decreed that you, and anyone else who comes across this thread, are to remain in the dark. Sorry. I'm expecting the thread to be closed permanently now.

 

[Bandersnatch]

Come now. Can't you just explain what you came to understand?

 

[Arbu]

No, because it came from ChatGPT, and I've been given a warning for quoting that. So I would just be repeating the harm that I did in quoting ChatGPT if I explain what I understand from it.

 

[berkeman]

No, because it came from ChatGPT, and I've been given a warning for quoting that. So I would just be repeating the harm that I did in quoting ChatGPT if I explain what I understand from it.

The problem that we have to deal with here at PF is that AI chatbots often give incorrect answers, so that is why we currently do not allow them to be used as references in the technical forums. Hopefully you can see why we have made that decision at the current time in the development of LLMs.

Have you been able to corroborate the information you got from the LLM with any reputable sources, or are you just believing what the LLM said?

 

[Arbu]

Most of the answers that I find online confuse the spring and autumn equinoxes with the dates I am looking for. There are one or two answers in the quora article that I referenced in the ChatGPT chat that I now understand to be correct. This is of course not a scientific publication either. It seems incredibly hard to get a reliable answer to this seemingly simple question. But I am happy with the answer that ChatGPT finally gave me.

 

[pbuk]

A search for "stonyhurst grid" easily yielded this reputable source:

Earth crosses the Sun's equator about December 7 and June 7 each year.

 

[Bandersnatch]

...which was already a part of the discussion.

 

[Arbu]

good-oh. There are lots of things you could search for. I searched for "when is the earth north of the sun's equator" and got loads of stuff about the equinoxes, so tried ChatGPT, which gave a more sensible seeming answer, but unfortunately that was wrong too. All it required was someone to point out the error in the first post here, but no-one did. I finally got the right answer out of ChatGPT.

 

[WWGD]

It would be good to know wether more "intelligent/advanced" versions of ChatGpt are effectively more accurate, more likely to provide an accurate reply.

 

[Tom.G]

It would be good to know wether more i)"intelligent/advanced" versions of ChatGpt are effectively more accurate, more likely to provide an accurate reply.

So ask it!

 

[pbuk]

...which was already a part of the discussion.

Oh yes, so it was. So despite having the answer posted here, with a reference to an authoratitive source, the OP insists that ChatGPT is superior because it guessed right the second time. Wow.

 

[WWGD]

good-oh. There are lots of things you could search for. I searched for "when is the earth north of the sun's equator" and got loads of stuff about the equinoxes, so tried ChatGPT, which gave a more sensible seeming answer, but unfortunately that was wrong too. All it required was someone to point out the error in the first post here, but no-one did. I finally got the right answer out of ChatGPT.

I

So ask it!

https://www.physicsforums.com/threads/chatgpt-examples-good-and-bad.1048757/

Think that's close to it.