Comet NEOWISE, and "What is Real?" Part II
This is an unexpected continuation of my opinion piece written last year discussing how astrophotos are made and the (hopefully) practical way to discern “real” images from artistic compositions. To read part 1, please click here. This article will focus less on the overall concept of “What is Real” and more so on artistic integrity.
You would be hard-pressed to find an astronomer anywhere in the world right now not currently aware of the naked-eye Comet we have quickly making its way past the Sun. At the time I write this, Comet C/2020 F3 NEOWISE (named after the satellite which discovered it in March 2020) is being subject to near-continuous capture from astrophotographers across the Northern hemisphere since it is the first bright comet to grace our skies in over 20 years. What prompted me to write this post is one particular image taken of this comet and the extreme and, in my eyes, inappropriate response it received.
Bowen Cameron, an astrophotographer from the Netherlands, has taken to astrophotographry something like a fish to water. Despite procuring his own dedicated astronomy camera only a few months ago he has produced some impressive images, and more than one of these impressive images was taken with a DSLR. In his own words, he pushes his processing to its limits, and it shows in the final result. The colors are striking, the contrast is strong, yet the backgrounds and star sizes are kept relatively tame as far as I can tell - and he does it all much quicker than I can, considering my processing seems to take months per image.
Four recent images by Bowen, the Crescent (Upper Left), Heart & Soul (Upper Right), North American & Pelican (Lower Left), and Orion’s Belt region (Lower Right)
Yesterday he posted another image - one of NEOWISE, which displayed some interesting details in the comet’s out-gassing tail which were not found in any other images so far. I was impressed when I saw it; I’ve only been taking night sky photos for a few years at this point, but have spent enough time in the processing software suites used for capture, calibration and editing to think I’m starting to have some semblance of competency. So seeing Bowen’s image certainly sparked a little jealousy, and a little uncertainty over my own skills in photo editing.
The image that launched a thousand tendies as the Reeeeeee’s shook the Earth
Accusations & A retraction
Bowen had posted this image to both Instagram and Reddit, and on encouragement from some of us (‘us’ being some fellow astrophotographers who all share a group chat on Instagram) he even submitted it for an APOD, where it was shared on the Sky page on Facebook (which shares notable images submitted for APOD). A day later came the response. His posts across all 3 platforms were receiving a heaping of both praise and accusation, the latter of which proved to be justified in motive, but was periodically vitriolic in execution.
The overall accusation was that the photo was fake, and the more specific reasons were that 1. the outgassing tails of comets do not have the physical characteristics to show twists and bends (though the ion tail can, but that is a different topic) and 2. that Bowen had outright stolen an image from Damien Peach (another excellent astrophotographer with ample experience shooting comets, not to mention his work on planets) and overlaid the detail from Damien’s image of Comet C/2016 R2 PANSTARRS onto his own. This second accusation was primarily sourced by an article written by David Rankin titled “How to Spot Bullshit Comet Photos” and a major factor in my decision to write what you are currently reading, though only after examining the evidence myself to be certain.
Stealing an image is particularly insulting to astrophotographers given the time, cost, and effort required to take such images (of course I’m sure any kind of photographer would be insulted by this, I simply don’t branch out of my little niche corner of shooting the night sky). Second to an accusation of image theft is the allegation that an image being otherwise fake, which is particularly aggravating since many such assertions are hurled by those who don’t know a single thing about astrophotography in the first place.
Over on Instagram, Bowen was seeing these responses and having a reasonable amount of self-doubt, so he began double-checking his work out of fear that he made a mistake while also posting possible explanations, such as the Comet’s relatively high altitude in his northern latitude. He also posted his stacked (but unedited) image for the public to download and invited anyone interested to process it themselves and see if they could replicate his work, all the while being inundated with demands in his chat from people wanting to know his processing steps. Eventually, through examining his image with Range Selection masks on call with Mustafa Aydin and William Burns, Bowen discovered that he had been inadvertently sharpening and increasing contrast on noise patterns - instead of on real detail.
Upon finding this, Bowen edited the caption on his Instagram photo to reflect this discovery, cautioning viewers to interpret the photo as artwork, and also posted a few stories basically saying the same thing.
An image shared by Bowen displaying his suspicion that noise patterns corresponded to the tail features
What Bowen Did:
- Received accusations from others (a Null Hypothesis conflicting with his Hypothesis)
- Asked for help from others & Reexamined his image (Researched / Constructed an Experiment)
- Found that he had sharpened noise artifacts (Analyzed Data / Formed Conclusions & Rejected his Hypothesis)
- Immediately notified people on his social media of his results (Communicated Results)
If that sounds a little like the scientific method, that’s not a coincidence. Bowen acted ethically during this entire process, at first claiming the details in the image were real, because, well, there was little evidence to the contrary aside from some rather angry people in the comments (and we are all on the Internet, the place is infested with them). Then, when encountering people who accused him of fabrication, he acted logically and with an open mind, found them to be correct, and admitted his mistake. Personally, I think if more people were willing to change their minds when faced with new evidence, I think the world would be a better place.
What Bowen did not do:
- Purposefully or maliciously mislead anyone
- Double down in the face of contradicting evidence that the image was real
- Post hostile comments in response to those accusing him of image theft
What Bowen Did Not Do, Part II:
- Steal Damien Peach’s Comet photo
I want to go over this final accusation in detail. A couple distant years of high school debate and a generous helping of online discussions have taught me to be cautious in my assertions. I like to have evidence on hand or a nearby link I can post in support of any argument I make because if the the world’s Internet users are skilled at one thing, it is in the angry application of Cunningham's Law, which states "the best way to get the right answer on the internet is not to ask a question; it's to post the wrong answer." In this case, as best I can tell, David Rankin posted the wrong answer. I do want to be clear on this, Bowen’s photo would quality as a “Bullshit Comet Photo” per the title of David’s article, but this is a point on which Bowen himself agrees.
What Bowen’s image is not though, is stolen. Upon first reading this article I tried to keep an open mind. Astrophotographers have stolen images before. You see it occasionally on Instagram or other places, though it is often quickly and harshly called out - and rightfully so. Astrophotographers have also been falsely accused of theft - sometimes there are only so many ways you can shoot a nebula before it starts to look like someone else’s photo.
So I opened Bowen’s NEOWISE image as well as Damien’s PanSTARRS image and tried to compare the two. I did make a few edits to each image regarding increasing contrast and darkening the background to make lining up the two images easier, and here is what I found:
The supporting evidence for the accusation seems to be the criss-cross pattern near the center of the GIF, but unless Damien’s image was significantly warped or distorted it doesn’t match up with any other detail in Bowen’s image. Again, I’m not claiming the detail is real - but neither is Bowen at this point. If that’s still not clear, just scroll on up and check out that section titled “What Bowen Did” one more time!
Now, David was quite insistent on this point, but refused to contact Bowen directly for an explanation per the Facebook comments below. I will note that David partly walked back this accusation in an update to his blog post, saying, “While I can't be 100% certain [Bowen] stole the data from the Peach image, I doubt that such similar features would coincidentally just show up in his image when it's painfully clear he added features to the image.”
Still, while David refused to ask the artist directly for an explanation as others did, Bowen was busy answering questions from others along these lines, posting his own data, and asking other for comparative processing work to see if he had made a mistake.
One other nuance to David’s accusation was that he apparently found stars in the compared images that lined up between the two. I asked him if he could provide screenshots, but he told me the theft was self-evident and apparently left the conversation.
I attempted to find any position where multiple stars were matched but was unsuccessful, and even PixInsight seemed unable to find any alignment, despite setting the alignment tolerances to their most maximum value. David, if you’re reading this, I am happy to consider new evidence, but could not find any in this regard on my own. Extraordinary claims require extraordinary evidence, and I’m still waiting on that last bit!
Update:
[David did end up adding an update to his Blog post regarding my article here, including an overlay showing the comparison of the single star he found, so if you have not seen the updated article please check it out for full context. To add my own thoughts after seeing his new comparison, I was somewhat expecting an alignment of more than one star - most alignment programs may require dozens or even hundreds for the best alignment.
However, as I said above after my own comparison image, this does not preclude significant distortions or warping edits (which would naturally not involve a star-based alignment), so as with all else in this article (and his), consider the evidence on your own. Unfortunately both of our sides require speculation, but I feel as though this lacks the burden of proof required for an accusation of theft]
Data Integrity versus Art
In part 1 of this topic, I finished my thoughts by comparing the possible difference between an image which is artistic and an image which shows reality, using an image of an enlarged Moon placed onto a widefield seascape as an example. The truth can be nuanced - the vast majority of astrophotos are “enhanced” in multiple ways for multiple reasons (again, see Part 1, etc, etc), but that does not necessarily mean they don’t represent reality. A false color image is still generated from real photons, and even if the resulting color pallet appears unnatural, the purpose is not to show “real” color as our pathetic human eyeballs experience it - but to highlight the contrast and composition of the nebula in question.
To digress, I finished off the previous article with this statement referencing the photo below:
“Astrophotographers work through many obstacles, including light pollution, camera noise, and atmospheric blurring, among others, to produce the best representation of their night sky object of choice.
Despite our best efforts, each photo we take, whether of the night sky or anything else, is simply the best guess made by a camera which has been made to emulate our natural human vision. In my opinion, as long as you are not trying to claim that a 1500mm Lunar image pasted onto a 20mm landscape is “real,” keep on taking photos of the night sky.”
I tried reverse-searching this for credit, by the way, but couldn’t find the original. Feel free to comment at the end of this article if you know the artist and I’ll add their name here.
Bowen worked through a lot of obstacles on his comet photo, including significant amounts of camera noise, and did as he always did, pushing his processing to what he thought was the limit to overcome the same obstacles astrophotographers have been dealing with for years. When he found out that his image had essentially lost its objective integrity (as with the image above) - he immediately labeled it as such.
One aspect of my previous “What is Real?” piece not sufficiently covered was the dance between science and art. I spent all my time explaining how astrophotography works and forgot to give any real input on the difference or the relationship between the two. Astrophotography meanders far too close to both science and art to be entirely on one side or another - it is irrevocably tied to both capturing accurate and clean data, and making that data look good. Many of the edits made to astrophotos are done to reduce obstacles in the way of the actual detail (light pollution gradients, camera noise, etc), but to stop there is to ignore an often vast and overwhelming section of processing which relies entirely on aesthetic choice. Using a Curve to alter a nebula’s contrast or saturation level, starless processing, or the so-called “tone mapping,” all of these edits rely on the eye of the artist with little to no regard for scientific integrity - yet a thousand new images of this caliber will be posted to social media every day without the barest complaint. If you’ve ever layered in HDR exposures to recover detail from Orion’s core, you’re an artist first and foremost, yet along comes a comet with enhanced noise patterns and we have a mob of people claiming it will be the end of reputable astrophotography?
“Witchhunts” and encouragement
Finding a hobby with a welcoming community is one of the most valuable experiences you can have. Every hobby has those excited to share their craft with newcomers as well as the experienced “old timers” who won’t give you the time of day, but astrophotography tends to be on the more welcoming side.
I, and thousands of others, have written what amount to short essays in response to simple online questions like “I want to take photos of planets, what telescope do I buy?” only to then scroll down, find another unanswered question - and start typing yet again. We are all united by our love of the night sky, the thrill in capturing these impossible dim and distant constructs hanging against the stars, and nearly all of us will jump at the chance to help another astrophotographer improve their capture or processing.
Unfortunately, despite Bowen realizing his error and publicly stating his image should not be confused with reality, one astrophotographer in particular still posted a video 12 hours later which tried both calling Bowen out for posting a fake image while at the same making a request to everyone who watches the video to kindly not send any more hate Bowen’s way.
“There are several space artists who I actually like, but the difference is they’re not trying to pass off their images as astrophotography. And what this does, is make the rest of us who are doing the right thing and trying to keep that integrity in our processing - it makes all of us look bad” - Dylan O’Donnell, on Bowen’s image of NEOWISE
“I don’t like calling out people like this” - man who made a 9-minute video calling a person out
I’d be surprised if Dylan ever reads this article. Of the literal ones of people who visit my website on an average day, a grand total of 22 have apparently used IP addresses from Australia. Still, I want to say I’m disappointed. The video itself is 9 minutes long and Dylan discusses the “Comet Controversy” for perhaps half of it, sandwiched between an update on his attempt to shoot the Centaurus A galaxy. I generally respected Dylan, he seems to have a good sense of humor for sure, but I feel his reputation may rightly suffer for this.
Ask yourself, who benefits from this video? It has no constructive criticism, so Bowen is given nothing useful to improve his craft, and despite the pinned comment from Dylan discouraging anyone else from doing what he proceeds to do for about 4 minutes, I can’t help but feel that the request comes off as hollow - for the sole reason that the video is even up.
Now, teaching others how to take good photos is not Dylan’s responsibility - he makes videos and post astrophotophotos because he enjoys it, like all of us do! So what benefit is gained by a 4-minute segment on calling out an astrophotographer for an error for which they already apologized half a day earlier, all the while claiming to dislike calling out said astrophotographer? He could have filled the video with scientific information - one of the reason the details in the tail were immediately seen as fake was that only a Comet’s ion tail exhibits these types of swirls (something I learned recently as well - most astrophotographers aren’t actual scientists, after all). That would have been informative and would have taught others how to spot fake comet photos in the future, but Dylan seemed content on claiming that he could immediately spot a fake and left it at that. What can I conclude from this aside the possibility that Dylan is capitalizing on the publicity at Bowen’s expense?
Leading by example is one of the best ways to influence others, and Dylan - I don’t feel you’re setting a good example here. I hope your heart was in the right place, at least.
Final Thoughts
Dylan is fortunately the minority in this case - most people, upon realizing Bowen’s attempt was genuine, as was his mistake, simply said “ah, well then” and moved on. One comment on Facebook was even relieved he wasn’t some “upstart sad sack trying to pull a fast one” (as they put it). So I’d like to think that the majority of astrophotographers will still be willing to be patient, thoughtful, and encouraging in their responses when something like this happens again (this hobby is particularly good at making us patient, if anything).
Bowen learned a few lessons today. He probably learned some new processing techniques for noise reduction, learned that calibration frames will be more helpful in future projects, and likely learned a little about the physics of comet tails as well - he didn’t learn anything from the hostile accusations, though. Unfortunately they still seem to be coming in.
I think I’ve said more than what I need to on this, so I’ll let a fellow astrophotographer end this piece for me:
”Over the last 24 hours it seems everyone under the sun is queueing up to take a swing based on nothing more than a glance - “that’s not real” and “it’s been blended with another image.” Whatever happened to constructive criticism? Would we all be happy to take the same accusations from others simply because we had processed something in a way deemed incorrect? If that’s the case, let’s just get the accepted versions of each target documented and draw a line under the whole hobby.
I didn’t get involved with astro to see people take a load of abuse for trying their best, surely no one else did either, so let’s remember that and have a bit more respect and care when it comes to approaching images that push our own limits of what we thought possible.” - Ross Clark
Comet NEOWISE, captured by Bowen, processed by Me