What Color is the Orion Nebula?

On the Cloudy Nights discussion board, the question of the nominally “true” color of Messier 42 as well as many other nebulae and stars has been tossed up for discussion. One frequent contributor to the board insisted that he could see pink in the Orion Nebula and, moreover, that anyone who could not suffered from defective vision. Another stalwart relied on standards that he derived from measurements made during daylight photography of other objects and from those defended his interpretive photographic processing. These are examples of confirmation bias: the operator is first pleased by an image and then finds objective reasons supporting the choice. 

At the same time, chatrooms across the internet, apparently starting at Reddit and soon including Gizmodo and Ars Technica, were abuzz about Samsung’s claim that its artificial intelligence learning algorithm enhanced a photograph of the Moon taken with a Galaxy S21 Ultra. In fact, the software recognized the object as the Moon and added details from other images. This was not an isolated example of artificial enhancement. 

• “Sorry, Internet, Some of Your Favorite Space Pictures are Fake” by Ethan Siegel on Medium.com here https://medium.com/starts-with-a-bang/sorry-internet-some-of-your-favorite-space-pictures-are-fakes-d24baf2232a1
• “The Night Sky is Always Getting Faked,” by Emma Roth on The Verge here: https://www.theverge.com/2023/3/16/23640343/samsung-astrophotography-night-sky-moon-fake
• “This Milky Way Photo on Nat Geo is Raising Eyebrows,” by Michael Zhang on PetaPixel here: https://petapixel.com/2019/05/07/this-milky-way-photo-on-nat-geo-is-raising-eyebrows/

When photography became available for astronomy, subjective experience was supposed to have been removed. Perhaps the best case was the Orion Nebula, first photographed by Henry Draper and Mary Ann Palmer Draper on 30 September 1880. (See “The Drapers’ First Photograph of Messier 42” here: https://aas.org/posts/news/2022/10/month-astronomical-history-september-2022 By that time, the Orion Nebula had been rendered in drawings for almost 200 years. One consequence of all the attention was that the representations tended to converge to a common understanding: they started to look alike. Now they do not.

 

https://phys.org/news/2013-03-astrophoto-beautiful-orion-nebula.html

To me, the most important fact is that astronomers explore and record Messier 42 and the rest of the universe in different wavelengths and we can only reproduce the very narrow visible band: x-rays, ultraviolet, infrared, and radio waves must be transduced into some false color presentation. The 3500 stars otherwise hidden by the dust and gas in the Orion Nebula were only revealed with infrared imaging. (See The Orion Nebula: Where Stars are Born by C. Robert O’Dell, Harvard University Press, 2003.) The nebula itself is a complex space of emissions and reflections. Therefore, many valid transductions are possible.

 

A more informative discussion would start with which species is doing the viewing. 

“In comparison, butterflies and mantis shrimp have a dozen classes
[of sensory cells] to cover a broader range of the spectrum
(from deep ultraviolet to far red light), which gives them hyperspectral vision.” 

Geographers call it "ground truth." When you have a photograph that shows yellow, is that reflection, subtraction, filtering, or emission? You do not know until you go there for yourself. I completed my master's degree in 2010 with two classes in geographic information systems. One of our projects was to present our university stadium. Our school colors were green and white. The concrete was painted green. So were the wooden seats. The field was artificial turf, also green. Surrounding the field were real living green plants, from grass and shrubs to trees. The assignment was to sort out the wavelengths of imaging and deliver an informative presentation.

 

PREVIOUSLY ON NECESSARY FACTS

Why Evidence is not Enough 

Galileo and Saturn: Epistemology not Optics 

Visualizing Complex Data 

Knowledge Maps