General - Forums - AstroBin

Topic starter

Jul 10, 2023 · Edited: Jul 10, 2023

The test lunation of the Astrobin Community was launched in the June 2023 lunation.

https://www.astrobin.com/forum/c/astrophotography/other/the-astrobin-community-abc-all-sky-survey-announcement/

Over 100 people expressed a willingness to take part. Understandably those numbers have not yet materialised. Nevertheless in the first lunation approximately 5% of the survey has been taken. The observing and linear processing pipelines appear robust. While further work remains on the stitching of field centres, good results have been achieved using PI's tools over a areas exceeding 12 or more fields [1% of the sky].

As a example here is a 1% patch of the sky during the past lunation in the field of Lupus, covering approximately 1% of the sky. This is the simply the linear data output from PI's WBPP, stitched using PI's mosaic routines and stretched using STF.

A little more non-linear processing gives the following [cropped to reduce file size, but retain survey resolution of 10arcsec

Team members are actively working on on this aspect of the survey.

However, we do need more volunteers. To observe, QC and assist with the stitching.

It would be great to have as many people on board as possible. Even as little as one field will help. We want to make this truly a community survey.

Thank you.

Brian Boyle

Topic starter

Jul 11, 2023 · Edited: Jul 11, 2023

Group now made public…. please consider joining. Everyone welcome.

CS Brian

Daniele Borsari

Jul 26, 2023

So, I just received my first dedicated astro camera, a ZWO 533MC Pro, and due to the small sensor I should make a 4 panel mosaic to fit the required 9 x 6 degrees field. I'm in the situation of trying and experimenting new gear so maybe in the next lunation I could try to image a field for the mosaic.

Michael Ring

Aug 19, 2023

I am so stupid…. We have several clear nights in a row and I forgot my Samyang at my parent´s home…. So unless others start work in the northern hemisphere there will not be much change for this month.

Michael

James Tickner

Aug 20, 2023 · Edited: Aug 22, 2023

@Brian Boyle @Michael Ring I've updated the booking field form to include a column showing the progress with processing. I've now downloaded all of the images files from Brian's Dropbox account that I can find. Any fields that are marked 'Complete' in column A but show 'Not started' in column L represent files that I can't find (or have managed to miss!).

Would you both be able to double check and upload any outstanding files? Once these are in place I'll download them and add them to the processed archive.

Thanks!

Update: I've added explicit options for 'missing files' and 'corrupt files'

Brian Boyle

Topic starter

Sep 11, 2023

Thanks to everyone who has so far joined this forum.

We are entering the last dark of moon period before the September equinox. After this our Northern Hemisphere friends will have the lion's share of the darkness and so we in south are hoping they take up the challenge.

To encourage those of you who have yet to sign up to take images, here is a brief progress report.

Note that all this has been achieved by five people in three months.

1) Over 15% of the sky has been imaged to the proposed depth.
2) The vast majority of the images are below Dec -10, with the area below -60 deg close to having a complete set of initial observations.
3) QC and mosaicing are actively being worked on, and we would hope to have our first survey products in the next few weeks.
4) Our approach to date has been to acquire as much data as possible, to have a large number of fields with with to assess QC as a function of geography, equipment, sky bringthenss, zd etc etc.

Going at this rate, our best estimate would be to complete the Southern sky in 18months. Northern Sky completion time is closer to 18 years.

Although @Michael Ring is doing great work, and @Todd Stephens has been very generous with the provision of his NCP data, the fact remains that the ABC survey will not happen in the North if we don't have more volunteers.

Initially over 100 people responded to the poll saying they were interested in taking part, so this is a bit of a call to encourage those who did and are still following along on the journey to join in.

The Southern Hemisphere will complete. Don't let the Northern Hemisphere fall behind.

If you have access to a dark site [Bortle 4 or less] please consider contributing. Even just one field is one field closer to completion. If you worry that your data wont be good enough, please don't. Some of the data already taken may not meet the QC threshold, but all data is helpful in scoping the observational requirements of the survey. @James Tickner is working hard on the mosaicing and QC and we should have some fabulous extremely-wide-field images and a robust QC pipeline soon.

The survey is designed such that all focal lengths from 100-200mm, APS-C and Full frame sensors in Bortle 2-4 skies can all make a significant contribution.

The principle of inclusivity and community on which the survey is based is still very much alive. Please join us.

Many thanks and clear skies

Brian

Helpful Engaging Supportive

Todd Stephens

Sep 11, 2023

I have quite a few more panels that I'd be happy to contribute for the Northern hemisphere. If you folks are ready for some more data, I can start putting it together for upload.

Well Written

Daniele Borsari

Sep 11, 2023

I have a question: is a 5x10 degrees panel a "problem" for the mosaic? This would speed things up since I would have to do only two sub-panels.

Well Written

Brian Boyle

Topic starter

Sep 11, 2023

Daniele Borsari:
I have a question: is a 5x10 degrees panel a "problem" for the mosaic? This would speed things up since I would have to do only two sub-panels.

Hi Daniele,

What set up are you using? The 9x6 size was driven by the need to have a good overlap region between field centres spaced at 7.5 x 5 deg. So a 5 x 10 [or 10 x 5 for PA = 0] wouldn't allow for any overlap - unless of course the adjacent field in the short direction was > 5 deg wide. But the overlap would still be reduced.

@James Tickner may also wish to reply as he is actively experimenting with the mosaic tehcniques.

Brian

Helpful

Brian Boyle

Topic starter

Sep 11, 2023

Todd Stephens:
I have quite a few more panels that I'd be happy to contribute for the Northern hemisphere. If you folks are ready for some more data, I can start putting it together for upload.

Hi Todd, Many thanks - I will let @James Tickner comment on when he is ready, and we have had a few good nights here in the south!

CS Brian

James Tickner

Sep 12, 2023

Brian Boyle:
Daniele Borsari:
I have a question: is a 5x10 degrees panel a "problem" for the mosaic? This would speed things up since I would have to do only two sub-panels.

Hi Daniele,

What set up are you using? The 9x6 size was driven by the need to have a good overlap region between field centres spaced at 7.5 x 5 deg. So a 5 x 10 [or 10 x 5 for PA = 0] wouldn't allow for any overlap - unless of course the adjacent field in the short direction was > 5 deg wide. But the overlap would still be reduced.

@James Tickner may also wish to reply as he is actively experimenting with the mosaic tehcniques.

Brian

Hi Daniele

All data is good data in my book!

The 6 x 9 fields were chosen to be a good match with some common setups (135 mm/APS-C and 200 mm/FF) but of course they won't match every lens/camera combination. From a mosaicing perspective, the software that we're building can pretty much cope with any arrangement and orientation of fields. For example, near the poles the arrangement of fields is pretty complex.

However, what the field spreadsheet does give us is a 'logistics management' tool - in other words, a way of checking that we've covered the whole sky with reasonable overlaps between neighbouring regions.

What I'd suggest is that you let us know your setup (either focal length + sensor size, or sky coverage for a single sub-panel as a * b degrees). I can then figure out a special arrangement for you for maximum efficiency. For example, maybe 1 x 2 or 2 x 2 neigbouring 6 x 9 degree fields might nicely correspond to a whole number of panels for your setup. You then then send through your panels 'unstitched' and our software should be able to figure things out.

Hope that all makes sense!

James

Helpful Engaging Supportive

James Tickner

Sep 12, 2023

Brian Boyle:
Todd Stephens:
I have quite a few more panels that I'd be happy to contribute for the Northern hemisphere. If you folks are ready for some more data, I can start putting it together for upload.

Hi Todd, Many thanks - I will let @James Tickner comment on when he is ready, and we have had a few good nights here in the south!

CS Brian

Hi Todd

Love to receive any data you're happy to share (dark- and flat-corrected but otherwise 'raw' 32-bit float RGB preferred if possible). If you can let me know the approximate RA/DEC centres (or equivalently the approximate corresponding field numbers from our list) that will help speed up the plate solving!

I've shared a Google Drive link with you via PM.

James

Helpful

Michael Ring

Sep 12, 2023

@Todd Stephens : In case you do not have all data in stacked form and you do not want to go through the stacking alone I can send you a SSD and do all or parts of the stacking for you, willing to do anything to get the north away from an ETA of 18 years....

Michael

Daniele Borsari

Sep 12, 2023

James Tickner:
What I'd suggest is that you let us know your setup (either focal length + sensor size, or sky coverage for a single sub-panel as a * b degrees)

I have a ZWO ASI533MC Pro and Samyang 135mm (129mm focal length from plate solving). Field is 5*5 degrees, pixel scale 6"/px.

Todd Stephens

Sep 12, 2023

James Tickner:
Hi Todd

Love to receive any data you're happy to share (dark- and flat-corrected but otherwise 'raw' 32-bit float RGB preferred if possible). If you can let me know the approximate RA/DEC centres (or equivalently the approximate corresponding field numbers from our list) that will help speed up the plate solving!

I've shared a Google Drive link with you via PM.

James

Thanks James, link received. The original stacks were done in Astro Pixel Processor, but at a minimum I can bring them into PI and solve them if that would help. Of course, I've gotten such better results lately with WBPP that I'm tempted to just re-stack the data... give me a few days to get organized and I'll start uploading files.

Well Written Respectful

Todd Stephens

Sep 12, 2023

Michael Ring:
@Todd Stephens : In case you do not have all data in stacked form and you do not want to go through the stacking alone I can send you a SSD and do all or parts of the stacking for you, willing to do anything to get the north away from an ETA of 18 years....

Michael

Thanks Michael, I'll keep that in mind. Let me see how much work it's going to be first to re-stack the data. It's not very well organized at the moment so I should probably go through the effort so that the data is indexed properly.

Well Written

Daniele Borsari

Sep 30, 2023

Do you have any updates on some possible special fields for my setup?

Daniele

James Tickner

Sep 30, 2023

Daniele Borsari:
Do you have any updates on some possible special fields for my setup?

Daniele

My apologies - I missed your earlier reply. I’m just heading out at the moment but will work out a suitable field group for you and send through later today.

Well Written Respectful

James Tickner

Oct 2, 2023

Daniele Borsari:
Do you have any updates on some possible special fields for my setup?

Daniele

Hi Daniele

Our nominal field size is 9 (RA) x 6 (DEC) degrees and we aim for a 20% overlap between neighbouring fields.

So working in the RA direction, two of your 5 x 5 fields with a 1 degree overlap between then would be a perfect match for our 9-degree width. For our 6-degree height (DEC direction), there are a few options depending on how many fields you are happy to contribute. This is summarised in the list below:

- One 9 x 6 field = four 5 x 5 fields with 3 degrees 'wasted' (33% waste)
- Two 9 x 6 fields (same RA) = six 5 x 5 fields with 2 degrees 'wasted' (15% waste)
- Three 9 x 6 fields (same RA) = eight 5 x 5 fields with 1 degree wasted (6% waste)
- Four 9 x 6 fields (same RA) = ten 5 x 5 fields with no waste

In the equatorial region (field strips at DEC = -10, -5, 0, 5 and 10 degrees) we have 48 fields in each strip, so the RA boundaries of the fields line up exactly. This would make the equatorial region perfect for you to build a 'stack' of fields with the same RA value and adjacent DEC values.

For example, a 'four-stack' covering the survey fields 489, 537, 585 and 633 (corresponding to centres at RA 00h00m and DEC = -5, 0, 5, and 10 deg respectively would involve you collecting 10 fields centred at RA = 00h08m / RA = 23h52m and DEC = - 5.5 deg, -1.5 deg, 2.5 deg, 6.5 deg and 10.5 deg.

Let me know if this makes sense and is something you'd be happy to contribute. We can easily designate similar stacks for you at different RA values as well.

James

Helpful

Daniele Borsari

Oct 2, 2023

Thank you for your work!
Since I live at 45° latitude something a bit farther from the equator and closer to 45° DEC would be better as it gets higher in the sky.

Daniele

James Tickner

Feb 10, 2024

Maybe my favourite single field so far of the ones I've collected. This ran late at night and I didn't check what was going to be in view, so it was a nice surprise in the morning!

Well Written Engaging

James Tickner

Feb 22, 2024

As of last night, we've reached the 1/3 mark in the survey! Baring a couple of holes which should be filled in the next lunation, we're also complete from 35 degrees south.

Unsurprisingly, Brian heads the leaderboard with 208 of the 373 fields imaged so far. A combination of dedication, a monster 200 mm F/2.0 lens, and those beautiful dark New Zealand skies. 😊

There are currently 708 fields left to image; 254 of these are on or south of the equator, and so within Brian's and my reach. A further 81 fields are between 5 and 10 degrees and so potentially accessible from my site in South Australia. So that leaves 373 fields north of +10 degrees for our northern hemisphere colleagues.

I think we should be able to knock this project off this year!

Well Written Helpful Engaging Supportive

Michael Ring

Feb 22, 2024

Congrats to you guys in the South… Please send some of those beautyfull clear skies in our direction, we need them… I missed 3 clear nights for personal reasoms this lunation, unfortunately they were the only 3 night available to me at all….

Brian Boyle

Topic starter

Feb 22, 2024

It is pleasing to have got this far, but we still have the issue of creating the mosaic and re-observing those fields that don’t meet our QC.

Fortunately software like BXT will help in the image quality, but it also poses photometric issues.

I suspect we will also have to be prepared to re-run a revised processing pipeline, in the light of what we learning in the post-processing.

I am certainly preparing myself to re-process most of my fields, to fix an overly fierce cosmetic correction.

But it certainly good to have this much data to play with. @James Tickner time to issue a revised progress map? If we can also produce a wide-field mosaic of suitable quality, it might attract more N. Hemisphere imagers (although dark skies are a very rare resource up there).

CS Brian

Alan Brunelle

Mar 22, 2024

Hi Brian, James, and others,

In the dark dank cloudy depths of the Pacific Northwest of North America, I have been getting my rig up and running and ready to travel if need be. I am hoping that I can begin to contribute during the next lunar opening from either home, or if need be traveling to my US state's (Oregon) new, large dark sky sanctuary. Now that I feel that I can do this, I wanted to introduce my setup, get any feedback and advice and ask a few questions. I have been practicing with the signup spreadsheet and loading panel coordinates into my rig computer (NINA-driven).

A photo of my rig can be seen here. It is a Rokinon 135mm, f/2.0. So far I can get it to work pretty well wide open. Maybe a little tweak with some spacers, if need be. A QHY268c camera. This is a OSC camera with a C-sized sensor. Pixel pitch @ 3.76 uM. The calculations I used tells me I am capturing a field of 9.95 X 6.66 degrees with a resolution of 5.74"/pixel. Let me know if this sounds wrong to you. From what I have read here, I am assuming that I will not be doing any mosaicing with my data? My experience doing my own imaging with this setup is that because of the undersampling nature of the rig, I have benefitted from dithering and doing a 2X drizzle with my data. Many stars (the faintest ones) appear blocky. However, I did read here that you suggest doing 1X drizzle. Is the 2X drizzle unnecessary for your needs?

I have been working with the spreadsheet today, as I mentioned. I should get pretty efficient with it eventually! As far as sign up, sometimes my weather openings are unexpected and panic sets in. So I may not want to spend the time to do a sign up prior to imaging. Considering the lower contribution from my northern compatriots, I feel the risk of my duplicating someone else's work is fairly low. I will notify ASAP after I go for a panel. Also, if I can capture only isolated panels, without adjacent ones, will you accept them? Again my weather can close fast here and be closed for weeks to months on end. That hopefully will be less of a problem when I get to Bortle 1 and I can knock off a number each night. If this launches on my end, I will follow up with how to get the data to you guys...

Best and wish me CS!
Alan

Helpful Respectful Engaging Supportive