Jump to content


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recording Earthshine on a Crescent Moon (Examples)

  • Please log in to reply
2 replies to this topic

#1 james7ca



  • *****
  • topic starter
  • Posts: 9,413
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 25 January 2021 - 05:26 AM

There was a thread a few weeks ago ( __HERE__ ) that touched on whether you could record earthshine using the same exposure as used for the sunlit side of a crescent moon. I've always believed that would be difficult given the limited dynamic range of today's CMOS cameras but I decided yesterday to give it a try on some data that I recorded earlier this month using a Stellarvue SV80ST2 scope and a cooled ZWO ASI183MM Pro camera.


I have already posted an image showing a "normal" processing for the sunlit side of the January 17th crescent moon ( __HERE__ ) and so I went back to that same data and tried a high dynamic range processing using PixInsight's ArcsinhStretch and HDRMultiscaleTransform processes. The end result was close to what I would call acceptable, with detail on both the sunlit and earth-lit sides of the moon. It's not perfect by any means, but it's close enough (I think) to show that there is potential in this technique.


In the case of this image that was taken on January 17, the moon was getting near to a rather "late" early crescent and that makes the recording of the earthshine even more difficult. Also, this is certainly not the first time that techniques like this have been posted to CN. Here is a thread from about one year ago where several others posted images using similar techniques (processing a single exposure to show earthshine):




One area where I could improve on my result is by taking more frames, since I had only 133 subs to stack to produce detail on the dark side of the moon. Plus, I think the base exposure could be changed to record more shadow detail and I'm not even sure whether a red filter was the right choice for this type of image (since the shadow side of the moon is probably more blue than red). Lastly, this certainly isn't the limit of what could be done within PixInsight, as tweaks to the processing could produce better results. Notably, I spent very little time with masking to produce this image (in fact, I used NO masking with ArcsinhStretch).


One other possibility would be to blend the normally processed sunlit side of this same sequence with the enhanced processing for the dark side.


Just to be clear, this image was produced using one set of 133 subs that were all exposed for for the same 15ms (that's what I selected to produce a properly exposed image of the sunlit side of the moon, when I originally took this sequence I wasn't trying to record the earthshine). Typically, when I go out to record earthshine I either expose for the dark side ( __HERE__ ) and/or use multiple sequences with the thought of doing an HDR composite with varying exposure times ( __HERE__, not a very good attempt). The image below is NOT any of the latter, just one, short sequence using a fixed 15ms exposure.


Processing was done in PixInsight and Photoshop CC2021. Image capture was done in Sequence Generator Pro (an app that I almost never use for planetary capture, but I took these prior to a session to image the Orion Nebula and Bode's Galaxies).


C&C welcomed.

Attached Thumbnails

  • Waxing Crescent Moon with Earthshine (small).jpg

Edited by james7ca, 25 January 2021 - 07:38 AM.

  • Kenny V., John_Moore, PirateMike and 5 others like this

#2 aeroman4907


    Surveyor 1

  • -----
  • Posts: 1,528
  • Joined: 23 Nov 2017
  • Loc: Castle Rock, Colorado

Posted 25 January 2021 - 06:28 PM

Thanks for posting this James and putting the effort in to produce this image.  Normally I would have expected another response to your post by now, but it is a Monday.


I think your results overall are quite good.  That is pretty impressive for a single exposure value for just 133 subs.


While 15ms is still relatively short, have you thought about incorporating some sort of calibration frame in the AS!3 stack to help with NR in the Earthshine?  I didn't see it as a technique you may have used in your processing.


The sunlit portion came out pretty well overall.  My only real thought for improvement is that the processing makes it looks a bit more like a pancake with flat lighting along the majority of the disk until you hit the limb and it then has more of a rounded appearance.  It has been awhile since I have used PI for DSO processing.  Could what I be seeing be related to the use of the HDRMT?


A very fine result nonetheless.  Just a few thoughts on what might really make a very good image under challenging conditions stand out even more!



#3 james7ca



  • *****
  • topic starter
  • Posts: 9,413
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 26 January 2021 - 01:19 AM

Steve, I agree that there is an unnatural look to the sunlit side of the moon. Most of that came from the HDRMultiscaleTransform (HDRMT) but that part of the processing was necessary since after the ArcsinhStretch that side of the moon was essentially pure white. I think a better approach would be to use some masking during the stretch and then apply the HDRMT in steps between progressive adjustments in the histogram stretch.


As for the dark side, I doubt that calibration would help to remove any of the noise. It would be better, I think, to increase the exposure slightly and collect more frames.


I have some past captures on the crescent moon where I used SharpCap to collect thousands of frames and I need to go back and try a similar set of steps on one of those sessions.

Edited by james7ca, 26 January 2021 - 01:23 AM.

  • aeroman4907 likes this

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Recent Topics

Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics