Sh2-68 (Flaming Skull Nebula) bombed by Starlink

[gorann]

Before my main target came up I had the crazy idea of shooting something near the horizon that I could only reach this time of the year. So I aimed at Sh2-68. I am at 60°N and this nebula is at 0°, so on the celestial equator. Astro darkness still starts late here so I could only start at 21.00 and then it fell below the forest at about an hour later and seeing was bad since I had to get through a lot of atmosphere (I had a record high RMS of 2" in PHD2). I was using my dual-RASA8 rig so it still gave me 2 hours of data. What I had not considered was Elon's satellites over the equator. Here is first what it looked like after some initial processing. I then tried some different image integration options in PI, but nothing really helped. To allow the algorithms to cancel out the trails I would probably have needed many more than my 28 subs (5 min each). So lesson learned: never aim close to the equator where Elon rules and the atmosphere is thick.

So far the image integration in PI has always been very effective for me up here in removing satellite trails, but this was something else. What do people do when imaging near the equator?

Cheers, Göran

First impression:

And here after a major crop, rotation and a lot of clone stamping in PS:

Edited Wednesday at 18:44 by gorann

[symmetal]

That's bad Göran 😬

It's not so much the celestial equator but towards the horizon in general, where you're getting more satellites in your FOV.

I use Astro Pixel Processor for stacking and must admit I haven't had a real problem with satellites, even when I stack just10 or so as i check the results as they come in. Though I haven't aimed close to the horizon for a while. 

Alan

 

[gorann]

2 hours ago, symmetal said:

That's bad Göran 😬

It's not so much the celestial equator but towards the horizon in general, where you're getting more satellites in your FOV.

I use Astro Pixel Processor for stacking and must admit I haven't had a real problem with satellites, even when I stack just10 or so as i check the results as they come in. Though I haven't aimed close to the horizon for a while.

 

Good point, it may mainly be that I was shooting through a lot of near earth space and therefor a lot of satellites. However, as I understand it stationary satellites need to be at the equator or else they cannot be stationary.

[symmetal]

They aren't stationary as they are in a low earth orbit around 340 miles up (550 km). They each complete an earth orbit in 90 to 100 mins depending on their altitude so are moving around 17,000 mph. (27,500 km/h) Their orbit inclination is 53 degrees  from the equator, which is why there are none operational over the poles. Also that's why there are so many, so that any point on Earth (almost) has a satellite close by.

If they were geostationary the signal latency would be excessive at 250mS there and back, compared to around 3mS in low Earth orbit.

Also, if geostationary, they wouldn't create trails across your images. 🙂

On the website I posted the image above from, you can view their position in real time.

Alan 

Edited Wednesday at 23:51 by symmetal

[Roy Foreman]

Around 50% of my subs contain at least one satellite trail where ever I aim in the sky, apart from the polar region.  Sometimes I have to resort to tedious manual removal of them.  It is only a matter of time before satellite pollution will surpass light pollution as the main obstacle to DSO imaging.  What a horrible thought !

[Laurin Dave]

Shorter exposures higher gain and more subs will help..   or a slower scope 😉

Dave

[lukebl]

7 hours ago, symmetal said:

Also, if geostationary, they wouldn't create trails across your images. 🙂

Not so. Geostationary satellites do create trails if you’re using longish exposures on a tracked equatorial mount. When imaging, you are following the stars, so the satellite gets left behind, leaving a trail.

They would only remain stationary if you were using a fixed mount, when it would be the stars creating trails.

[ollypenrice]

9 hours ago, symmetal said:

 

Also, if geostationary, they wouldn't create trails across your images. 🙂

 

Alan 

I'd have said they would. We are tracking the stars from a moving earth. If we have any horizon in our images from a tracking mount, it trails during the integration. A geostationary satelite is, in this respect, a terrestrial object, I'd have said. Geostationaries produce very bright trails in my experience.

Olly

[gorann]

2 minutes ago, ollypenrice said:

I'd have said they would. We are tracking the stars from a moving earth. If we have any horizon in our images from a tracking mount, it trails during the integration. A geostationary satelite is, in this respect, a terrestrial object, I'd have said. Geostationaries produce very bright trails in my experience.

Olly

That is also what I thought. They are stationary in relation to the earth but then obviously not to the sky since our planet rotates last time I checked.

[Ouroboros]

To be stationary wrt the stars a satellite would have to be in sidereal orbit - if such a term exists. Wouldn’t that require a satellite to be at the same distance from the earth as satellites in geostationary orbits, but going in the opposite direction? Sounds to me like a recipe for satellite billiards!   

[ollypenrice]

2 hours ago, Ouroboros said:

To be stationary wrt the stars a satellite would have to be in sidereal orbit - if such a term exists. Wouldn’t that require a satellite to be at the same distance from the earth as satellites in geostationary orbits, but going in the opposite direction? Sounds to me like a recipe for satellite billiards!   

Yes, to be fixed against the stars a satellite would need to orbit every 23hrs, 56 minutes and 4 seconds and in the earth's equatorial plane. A Musk exterminator!

lly

Edited Thursday at 11:24 by ollypenrice
Clarification

[ONIKKINEN]

Most starlinks orbit the Earth at an inclination of 53 degrees, which places them at the horizon for those of us further north than that. I avoid imaging this far down the horizon, but have spent a few hours on the Horsehead and flame region which was similarly photobombed by trails that were not removed by stacking.

I just used the good old lasso tool and content aware fill combo for them which works ok, does take some time though.

[symmetal]

OK, geostationaries wouldn't create long trails 'randomly' crossing the image, which was the reason for the topic. 🙂

I've taken images around Dec 0 and not noticed short horizontal trails, though it seems at the UK latitude geostationary satellites would be around Dec -7 degrees due to parallax. At the north pole they would be at Dec  -8 deg 37 mins so would be below the horizon. Also, left or right of the meridian changes the satellite declination slightly too if you're not at the poles or equator by some complicated formulas shown here.

Alan

 

[Ouroboros]

5 hours ago, ollypenrice said:

Yes, to be fixed against the stars a satellite would need to orbit every 23hrs, 56 minutes and 4 seconds and in the earth's equatorial plane. A Musk exterminator!

lly

Chuckle. I could comment but it would be outside the rules of this site.  

Don’t we need to allow an additional ~4 minutes to the earth’s rotation period because the sidereal position advances by four minutes every day due to the earth orbiting the sun. 

Edited Thursday at 16:39 by Ouroboros

[ollypenrice]

1 hour ago, Ouroboros said:

Chuckle. I could comment but it would be outside the rules of this site.  

Don’t we need to allow an additional ~4 minutes to the earth’s rotation period because the sidereal position advances by four minutes every day due to the earth orbiting the sun. 

Correcting errors does not lie outside the rules of the forum. Only doing so rudely contravenes the rules. Please go ahead!

1 hour ago, symmetal said:

OK, geostationaries wouldn't create long trails 'randomly' crossing the image, which was the reason for the topic. 🙂

I've taken images around Dec 0 and not noticed short horizontal trails, though it seems at the UK latitude geostationary satellites would be around Dec -7 degrees due to parallax. At the north pole they would be at Dec  -8 deg 37 mins so would be below the horizon. Also, left or right of the meridian changes the satellite declination slightly too if you're not at the poles or equator by some complicated formulas shown here.

Alan

 

If we image a target all night, that target will pass from one side of the observer's horizon to the other, or a substantial part thereof.  Say you have a nearby church steeple on your southern meridian and you are imaging for four hours, two each side of that point. If imaging low down that steeple will at some point appear in your image as you track towards it and disappear as you track past it. It will pass from one side of your image to the other, as will geostationaries. Or so it seems to me.

Olly

Edited Thursday at 17:46 by ollypenrice

[symmetal]

2 minutes ago, ollypenrice said:

If we image a target all night, that target will pass from one side of the observer's horizon to the other, or a substantial part thereof.  Say you have a nearby church steeple on your southern meridian and you are imaging for four hours, two each side of that point. If imaging low down that steeple will at some point appear in your image as you track towards it and disappear as you track past from it. It will pass from one side of your image to the other, as will geostationaries. Or so it seems to me.

But a single sub wouldn't be 4 hours so you'd end up with a hundred subs or more each with a short trail which would be removed fairly easily in stacking. 🤔

Anyone like to post a sub showing geostationary satellites being a nuisance? 🙂

Alan

[Recretos]

12 hours ago, Roy Foreman said:

Around 50% of my subs contain at least one satellite trail where ever I aim in the sky, apart from the polar region.  Sometimes I have to resort to tedious manual removal of them.  It is only a matter of time before satellite pollution will surpass light pollution as the main obstacle to DSO imaging.  What a horrible thought !

Doesnt the Pixel Rejection in stacking help to remove those trails?

[ollypenrice]

8 minutes ago, symmetal said:

But a single sub wouldn't be 4 hours so you'd end up with a hundred subs or more each with a short trail which would be removed fairly easily in stacking. 🤔

Anyone like to post a sub showing geostationary satellites being a nuisance? 🙂

Alan

The sky rotates against the earth at 15 degrees per hour or 0.25 degrees per minute. A 10 minute sub will track past 2.5 degrees of terrestrial horizon and this easily exceeds the full FOV of many DS imaging systems. A geostationary easily produces an edge to edge trail. I'm not saying they won't be removed by outlier rejection. I remember my Witch Head being plagued by them but the improvements in the sigma routine between one version of AstroArt and the later one made them disappear.

Olly

[symmetal]

1 hour ago, ollypenrice said:

The sky rotates against the earth at 15 degrees per hour or 0.25 degrees per minute. A 10 minute sub will track past 2.5 degrees of terrestrial horizon and this easily exceeds the full FOV of many DS imaging systems. A geostationary easily produces an edge to edge trail. I'm not saying they won't be removed by outlier rejection. I remember my Witch Head being plagued by them but the improvements in the sigma routine between one version of AstroArt and the later one made them disappear.

Olly

OK Olly. I think I'll give you that one.   

[Ouroboros]

2 hours ago, ollypenrice said:

Correcting errors does not lie outside the rules of the forum. Only doing so rudely contravenes the rules. Please go ahead!

Oh, yes. I endeavour to correct errors courteously. Not that I do that often.

No, it was my potential response about Musky that might well have contravened the rules.  😁

[Roy Foreman]

9 hours ago, Recretos said:

Doesnt the Pixel Rejection in stacking help to remove those trails?

Yes it does, mostly, as long as I have a decent number of subs to stack. I was just pointing out the sheer number of subs affected,  and it's getting worse.

[ollypenrice]

10 hours ago, Ouroboros said:

Oh, yes. I endeavour to correct errors courteously. Not that I do that often.

No, it was my potential response about Musky that might well have contravened the rules.  😁

Ah, then I'm sure I would face the same difficulty! 

Olly