Right ascension: can someone explain where I'm going wrong here please?

[BrendanC]

Hi all,

So, I was just thinking about right ascension. I thought I understood it, but now I'm not so sure.

My understanding was that celestial objects have their right ascensions set according to where they were at the exact vernal equinox relative to the Greenwich meridian, currently using the year 2000 as the base year.

In theory then, I thought that meant an object with an RA very close to 00:00:00 would appear directly north at that time and place.

One of the objects in my target list is close enough to test this: NGC 7822, with an RA of 00:01:09. So, given that the vernal equinox in 2000 was at 7:35 on 20th March, I thought it would be a simple case of firing up Stellarium, setting my location as Greenwich, the time and date for the equinox, locating NGC 7822, and it being almost exactly North. 

But it isn't. Here's where it is according to Stellarium:

I'm clearly misunderstanding something quite basic here!

I mean, I don't actually really need to understand this because, well, we just use our fancy-pants software to find things for us nowadays, don't we? But I'd still like to know where my reasoning is falling down here. Just for peace of mind, you know, to be reassured that the heavens do actually behave like clockwork, as they should.

Can someone help please?

Thanks.

[andrew s]

What you are missing is local siderial time LST. A stars RA is, for a given epoch fixed, and your LST will tell you which RA is now on your meridian.  As stars moves due to precession and proper motion you need to use Jnow but your program will take that into account for you.

Regards Andrew 

[BrendanC]

Sorry, don't understand!

NGC 7822's JNow and J2000 RA values are 00:01:08.59 and 00:02:12.65 respectively - not enough of a difference to account for how far out of North it appears on that screenshot.

I'm just trying to determine whether I'm even right that, if I go back to the date of the vernal equinox in 2000 (ie the J2000 epoch), objects with an RA close to 00:00:00 would appear North. Am I right about that? If so, then why doesn't Stellarium show that? And if I'm wrong, which Stellarium is indicating, then where am I going wrong? An actual screenshot would really help show me!

[dph1nm]

J2000 coords are relative to the position of the equinox at 12:00 1st Jan 2000 i.e. the position at which the equator and the ecliptic cross at that time. This moves continuously due to precession.

NigelM

[dph1nm]

The other way to think of it is that by definition the Sun is at 0h on March 21st - it obviously crosses the meridian at midday, so will anything else with 0 hrs RA. So NGC7822 will be due N at midday on the 21st.

[BrendanC]

14 minutes ago, dph1nm said:

The other way to think of it is that by definition the Sun is at 0h on March 21st - it obviously crosses the meridian at midday, so will anything else with 0 hrs RA. So NGC7822 will be due N at midday on the 21st.

Yes, this works, I just tried it on Stellarium and that's correct.

So, there's clearly a misunderstanding on my part about what RA is regarding the vernal equinox. If the definition I thought was right, then it would be due North at the precise date/time of vernal equinox in the year 2000. But it isn't, it's at midday on the day of the vernal equinox ie when it crosses the meridian.

I guess I'll have to do some more thinking...! It'll probably make sense at some point.

Thanks for the attempts to explain it to my feeble brain...

Edited January 29, 2021 by BrendanC

[kens]

From Wikipaedia: https://en.wikipedia.org/wiki/Right_ascension

Quote

Right ascension (abbreviated RA; symbol α) is the angular distance of a particular point measured eastward along the celestial equator from the Sun at the March equinox to the (hour circle of the) point in question above the earth

So it is measured from the position of the Sun at that date/time

[BrendanC]

Correct. Believe me, Wikipedia was my first port of call when I realised I didn't quite get this.

So, in that case, why does an object with 00:00:00 not appear directly north at that date/time?

Given that the sun is at RA 00:00:00, and is overhead at midday on the equinox date, then this would appear to be the definition of RA. But it isn't. It should be the date and time of the equinox.

There's clearly something missing from my understanding here.

I just need someone to explain why I'm wrong, and to give me the correct answer! Why doesn't NGC 7822 appear due north at the exact date and time of the vernal equinox, as it should according to its RA and the definition of what RA is?

I should add that I've been involved in astronomy and astrophotography for over two years now, have done an intro course at Oxford University, and have used RA and Dec coordinates quite a lot, for example when calculating the day that a given object will be directly overhead me at midnight based on its RA. It's just this oddity I've realised, this gap in my understanding. 

Edited January 29, 2021 by BrendanC

[malc-c]

This may help...http://www.pas.rochester.edu/~blackman/ast104/coordinates.html

[BrendanC]

Thanks, but which bit?

I've read just about every web page I can find. They all give exactly the same definition of RA which is what my understanding is (or rather, was).

Here's my dilemma, in a nutshell...

EXHIBIT A - Location of NGC 7822 at the exact time of the equinox last year. Right definition, wrong location, if - as I think it should be - it should be due north. I could be wrong. I know I'm wrong. I want someone to tell me why I'm wrong.

EXHIBIT B - Location of NGC 7822 at the actual exact time of the equinox according to J2000. Absolutely right definition, wrong location again.

EXHIBIT C - Location of NGC 7822 at midday on equinox last year. Wrong definition, right location. How can this be? I want someone to tell me!

It seems to me that the only way to square this circle is to adopt the definition that the RA is the location of an object when the sun reaches the meridian on the day of the equinox. That's the only way I can see that the RA actually reflects where NGC 7822 is on the day of the equinox. That is, noon, not 9:37am or 7:35am.

I've read pages and pages and pages and watched videos and videos and videos, but none really discuss the origins of how RA is derived from first principles. And certainly none of them help me understand where I'm going wrong here.

I just know that the answer is probably really simple and if someone could just explain this to me, I'll suddenly see the light and slap myself on the forehead.

Edited January 30, 2021 by BrendanC

[Merlin66]

I think your over thinking the whole issue of what RA as a coordinate system is all about.

You're familiar with Long and Lat position to define a location on the Earth's surface.

In the sky we have a slight problem.....

The stars and galaxies don't move much over the years relative to each other...but...the coordinate grid we "overlay" on the sky, unlike the fixed Long and Lat on Earth does move due to precession.

To "agree" on a location of a star/ galaxy using a coordinate system we must therefore all agree on a "base" for the timing and hence positioning of the RA grid on the sky.

It's usual to use either J2000 or Jnow to "reset" the coordinate system. Depending on your planetarium program and drive interface. It doesn't matter which one you use, as long as you're consistent.

A star's RA is the sidereal time when it crosses the meridian.

 

 

[BrendanC]

4 minutes ago, Merlin66 said:

A star's RA is the sidereal time when it crosses the meridian.

So, that's the same as this, right? 'The RA is the location of an object when the sun reaches the meridian on the day of the equinox.'

I know I'm wrapping myself up in ribbons, but the full story is this: the other night I gave an online talk to a group of people, just giving them a tour of the night sky via Stellarium, sharing my photos, that sort of thing, with Q&A afterwards. It went really well, even fielding questions like "Is the universe expanding and if so, what's it expanding into?' (short pause, then wax lyrical about redshift, Hubble, dark energy) and they might want me to do another one. So, I was just thinking about 'stuff', as you do, mainly arcmin, degrees, RA, Dec, and so on. Then I got to thinking about how I'd explain RA to people. It went, in my head, something like this: "The RA is the position of an object relative to the sun at the exact time and date of the vernal equinox. So, if an object has zero RA, that means it's exactly... er... where?" I thought it made sense for it to be due North. But it isn't. And I didn't know why.

It's one of those things that you sort of take for granted, using RA and Dec day in, day out, doing clever calculations and feeling very smug. Then you suddenly stop, and think about it, and realise you don't quite understand something very fundamental, which you really should. It's like realising your cat is really a dog.

However, I think I do now. As you say, it's the sidereal time when it crosses the meridian. Which, I think, is what I'm saying too: it's the location of an object when the sun reaches the meridian on the day of the equinox. 

Does this work? Or should I just go back to banging the rocks together?

 

[Merlin66]

Forget the Sun as a reference point.

It's just another astronomical body which has it's RA/ Dec defined by the Epoch used for the coordinate grid system.

[malc-c]

When anyone asks me what the numbers mean, and how things are found in the night sky I just tell them it's like longitude and latitude on the surface, but "transposed" onto the "dome" of the night sky.  Otherwise it can get complicated.  It may not be  totally correct, but they can grasp the principle.

(not that I get that hung up on it - I just right click and tell the scope to slew to target  ) 

[kens]

3 hours ago, BrendanC said:

I just need someone to explain why I'm wrong, and to give me the correct answer! Why doesn't NGC 7822 appear due north at the exact date and time of the vernal equinox, as it should according to its RA and the definition of what RA is?

If you look at the Stellarium screen shot from the time of the equinox in 2000 you can see that NGC7822 is close to the same hour angle as the Sun and which is by definition RA 0h0m in J2000. It is the position of the sun at that exact date and time that defines RA 0h0m for J2000. And that is low in the SouthEast. At that date/time the meridian is at an RA of about 19h18m. So why would you expect an object near RA 0h0m to appear near the meridian?

If you advance the Stellarium clock to midday of the same day the sun appears overhead as does NGC7822. The Sun is no longer exactly at 0h0m but still quite close at 0h0m40s or so. The RA of NGC7822 does not change.

It seems you are confusing two different approaches to RA and appear to have redefined RA:

1. At the exact time of the equinox the sun is at RA 0h0m. No meridian or midday involved in this definition.
2. When the sun is at the meridian its RA is the same as the sidereal time. Applies on any day at all. Not related to the equinox. Also, sidereal time is not the same as terrestrial time.

Does that help?

[BrendanC]

I think this helps. Thank you. I'm going to sleep on it and see if it still makes sense tomorrow.

[BrendanC]

10 hours ago, kens said:

It seems you are confusing two different approaches to RA and appear to have redefined RA

I know! That's what I've been saying all along! That's what I need help with.

So, having slept on it, I think I get it now. The vernal equinox happens at a certain date/time each year, generally 20/21st March. That's the precise date/time at which the sun crosses the celestial equator. Right Ascension for an object is taken (in practice, was taken in the year 2000, is taken according to JNow) as its distance from the sun at that precise point, expressed as hours/mins/secs. The direction it's in is immaterial, and this is where my 'hunch' at the start of all this thread was wrong. An object will only be due north/south as it crosses the meridian, which is nothing to do with the point at which it crosses the celestial equator.

Makes sense, because having been to Avebury for the summer solstice for over 20 years, I know that the solstice can happen any time, whether a convenient PM or a less convenient AM (there's that pesky meridian popping up again...)

Thanks!

These things matter. It's like when I realised the moon's phases couldn't be described by a perfect sine wave, and someone pointed out that's because it's an elliptical orbit. It's just noticing small things that don't quite add up, and needing to know them because they're pretty fundamental.