Pete - It is clear from the caption that this is a composite but I don't think this kind of thing should be encouraged. To me it looks a bit silly. As observers and imagers we have always been keen to point out fake images on social media. I think this comes dangerously close and doesn't set a good example.  

While I understand your comment Nick, it's out of place and unfair. First, if I create an image like this, I always make it clear it's a composite. I recall a few years back doing this, posting such an image and having similar comments to yours levied at me. A colleague posted a similar image, without explanation or clarification it was a composite, and he got nothing but praise. However, the most unfair part about your comment was the fact that I put this up as a personal image on my Facebook page. I originally created it as a reminder of the event - I'd not seen a lunar occultation of Mars before. It was seen and requested as a BAA PotW. I didn't actually submit it, but I was happy to share it as requested. That's worked out well, hasn't it?!  

As PotW editor I carefully consider whether candidates are worthy of the choice, and this decision includes whether the image is, or is not misleading.  In this case Pete L has demonstrated a not insignificant amount of skill in producing what is clearly an excellent representation of the event. The reason I posted the support frame as a supplementary was to make it clear that this is a "compiled" image to better display the event. And I would also ask the question whether any of the images we post these days, which are often subject to star removal and replacement and selective sharpening of key objects using masking and other techniques, can truly be claimed to be "genuine and un-doctored".  I am at peace with my choice, but this is an interesting discussion!

Indeed an interesting discussion, guys,

The degree to which you manage the rendition of these sort of transient events is a thorny subject. With comets it is acceptable to stack on the comet then go back and stack on the stars; then combine the sharp stars with the sharp comet. To make a decent image of this kind of occultation event one could follow the same route and I feel would be acceptable in the same way, as the data is all taken at about the same time.

I think Pete has said that the Mars image was taken shortly before the event but I wonder when the Moon image was taken. Comparing the shadows at the edge of the Moon to other images taken of the event it looks like it is a lunar image taken at a different time. If so I think that should have been made a little clearer in the description, as it might affect one's opinion of the image.

Having said that, Pete says the purpose of the image 'was to reproduce something which was similar to how it looked to me while observing', which does sort of flag that it was more impressionistic than faithful.

Martin

The image was not intended or submitted as a scientific rendition of the event. In fact, technically it wasn’t submitted at all! The component images were taken within a short period of each other, the scientific data (if there is any science to be gleaned from an occultation event such as this), is contained within the 4,624 frames of the 10Gb occultation ingress video – a single frame of which was used as a reference to construct this occultation composite is linked above under the ‘Files associated with this observation’ heading as ‘wk4_23_lawrence_marsoccult_ref_081222.png’.

In terms of generating composites such as this, to describe them as ‘a bit silly’ is a bit rich. I’m sure comet images which show point stars and spectacularly detailed comet structures are not turned away as being a ‘bit silly’. Martin has already mentioned that it is deemed acceptable to stack on the comet, then go back to stack on the stars. This somewhat glosses over a common technique where the stars are digitally removed from each comet subframe using a process such as Starnet++ or Star Xterminator, leaving just the comet’s data which can then be stacked and re-composed into a sharp star frame, built without the comet. Both components – star field and comet will have had data manipulated to get to this point. If you remove a star from an image, you need to fill the gap with something. The results are spectacular of course – and I admire those who produce such images. But you cannot dismiss one technique which technically doesn’t actually remove or artificially add data to form the base images for a composite, while accepting another which does. That *would* be silly.

I could go on and point out that many composite or multi frame images rely on the creator’s experience and judgement to show what they do. Consider a total solar eclipse where multiple exposures are combined to create a high dynamic range result which shows lunar surface data together with coronal structure heading out many solar radii from the Sun’s centre. The resulting images are often spectacular but are false in that no observer has ever seen such a scene nor been able to photograph one in a single shot. They take great skill to produce but inevitably, what you are presented with is an interpretation of how the imager believes the components should go together.

I’m not dismissing any of these techniques and welcome them because I feel they expand the way we see and interpret the night sky and all its wonders. If done with honest disclosure, they also hopefully serve as an inspiration for others to try. I do not condone creating false images for submission and never have, especially where the sole intention is simply to garner likes. Some will know very well, that I have been very vocal in challenging such images when I see them.  

Since I started this discussion I guess I had better respond.

I actually said "To me it looks a bit silly" and, to me, it does. Others, clearly, have different opinions. I really don't like these type of composites. I know that you didn't submit this directly and I know that you were very clear on how it was done but I do feel that there is a danger here. Given your prominence in the astronomy world, I think this example weakens our argument when we try to call out others who do the same kind of thing but who are not honest about what they have done.

At present online there are many images purporting to show comet C/2022 E3 in the sky front of some kind of horizon or other. They were clearly taken with different instruments and have been composited but that is rarely stated by the people who post them. The posters claim that they represent their view from wherever. To me they look silly. Even some of our best comet imagers are compositing images taken with very different focal lengths. They are clear that they have done it but I don't like that either. Some images of comets are over-processed and show considerable artefacts. That is a rather different situation and, again, in most, cases there is no intention to mislead but, with the vast collection of image manipulation techniques now available it is difficult to know what is real and what is not. The night sky, and comets in particular, have plenty of real wonders without inventing new ones in a computer.

To me, extracting two stacks of a comet from the same input data, one tracked on the stars and one tracked on the comet and then combining them is OK as long as each stack is manipulated in the same way and combined without manually "patching" the image. The same applies to the highly processed HDR images of the solar corona. That may be a bit illogical but I guess it is something to do with the fact that the images were all taken at the same time, at least that is how I would justify my view! 

There is clearly a continuum between "good" and "bad" image manipulation and different people will have different thresholds. If the approach used is clearly stated, as you have done here, then that is fine. Personally, I think the results look silly, others don't. My main point in starting this discussion is that we need to be very careful about this given that there are people out there who could exploit it when we call them out.

While I understand and respect your opinion, I think you’re using selective sight to a degree. In terms of composites, modern imaging kit means that genie has left the bottle. My prominence (which I think you may have over-stated a tad) will have little effect either way. Besides, I’ll be the one who decides whether or not I put my own images out, submitted or not. You can’t stop people composing images and there will be plenty of fakes that arise as people want to ‘join the club’ and be loved for their results. In my opinion, the best way forward for this is to educate that you need to include details when you do it. I have written a few articles on creating such composites and always try to stress that you need to explain what you’ve done when you create such an image.

I guess the parallel is trying to get people to put at least the date, preferably the date and time and even more details on an image. Look at many of the social media images of the Sun, Moon, planets, comets, etc. and you’ll see none of this. The word hasn’t got out properly or has been swamped by the desire to be shown – ‘get the picture done and put it out as quickly as possible’ now rules it would seem.

I mentioned that I thought you were using selective sight. Creating comet images is pushing the limit a bit. One process requires you to do a comet align, create a set of results where stars have been removed, stack the comet only result, subtract the starless comet images from their comet + star versions to create the comet-less star field and stack the star field. Then re-combine the comet and stars. That’s a lot of manipulation. You say you can forgive it because “the fact that the images were all taken at the same time”. That’s shaky ground my friend. Does it mean a result comprising several hundred hour’s exposure time isn’t valid because of the time taken to gather all the components? Does it mean the Hubble Deep Fields aren’t okay? Some compile deep sky results by combining versions taken days, weeks, months and years apart.

And those comet images – they aren’t instant. They may span 10, 20, 30, 40+ minutes, all condensed into an image which purports to show the comet as it would appear in an instant. All of the components of my composite were taken close in time over the same session. I’ve also supplied a single frame from the full occultation video as an example of the positional information. That and the description of the process used is a lot more than many will provide with their results. Anyway, I’ll leave it there as I guess the week is nearly up. Thank you for the discussion.  

Indeed a good discussion and a lot that we can agree on. Probably best continued over a beer the next time we meet up!