What color do you see?
27 July 2007
One of the essential ingredients of any tour of Mauna Kea observatories includes talking about adaptive optics, including use of the laser guide star. The wavelength of light from the laser is tuned to some ridiculously precise number of decimal places when represented in nanometers instead of wavenumbers (not sure exactly how many, but probably more than five according to Rob, Kevin, and Allen). Somewhere back in the mists of time when I started giving summit tours, I heard the laser light described as ‘yellow-green,’ and have been repeating that nearly every time I talk about adaptive optics.
As usually happens, my authority was undermined when someone then pointed to a digital picture of the laser in operation (all you wags out there are ordered to say ‘propagating,’ not ‘firing,’ ) and asked why, if it was yellow-green, it appeared as orange in the picture. I sure as heck didn’t know, and admitted as much. I then queried some of our laser folks, who also professed ignorance. The quick answer was that it was some artifact of the way the eye perceives color, but that was knocked out by color appearing in digital images. I started offering that disparity as a bit of ‘salt-and-pepper,’ during my tour, and received some thoughtful looks and puzzled discussion in return. I was even graced with an answer from one of my least favorite kinds of visitors, the sixty-ish male techno-weenie blowhard who has an answer for everything usually involving a number with five significant figures: a long and complex explanation involving something like ‘secondary excitation,’ and other jargon and harking back to his glory days working for some defense contractor.
Then, the other day, I gave a tour to a couple of knowledgeable amateur astronomers and an artist.
In astronomy, adaptive optics ‘de-twinkles,’ the light from a target object by bouncing it off of a mirror that changes shape hundreds of times per second and sending the light into into a separate detector used to take a picture or spectrum of the target. The deformable mirror is (very quickly) shaped to reduce the depradations caused by the passage of the light through the atmosphere. The necessary shape of the mirror is calculated by taking the light from an adjascent bright star and measuring how it is deformed from an ideal point. Since the light from the star and from the target pass through more or less the same piece of atmosphere, the corrections will be the same for the target as for the star.
The bad news: only about 5-10% of targets have bright enough stars adjascent enough to them to use the adaptive optics that way. In order to spread the technique around the sky, a laser that emits light at the wavelength necessary to excite atoms in the sodium layer of the atmosphere is used to, well, excite atoms in the sodium layer of the atmosphere. This layer exists at about 90km above our heads, and seems to be deposited by meteors vaporizing as they encounter the atmosphere that already exists there (not much, but still). Thus, anywhere the telescope points, the laser creates a guide star to run the adaptive optics system with.
When I posed the conundrum of the mismatched colors to the third culture tour group, the artist suggested that something involving the color complement was at work, since that functions differently for light than for pigment. None of us could figure out how a laser could produce any color but the one it was supposed to, but it seemed to me that if the dark orange appearance of the laser was the complement of yellow-green light, then perhaps we could identify the response in our eye causing the confusion.
I did some digging, and found some fascinating articles on color vision, learned what metameric light sources were, and in general was impressed by how complicated some simple-seeming things can be. A suspicion was growing, however.
It turns out that we perceive color via three kinds of molecules in our eye, each kind is responsive to a different range of wavelengths of light. The combination of signals from these pigments feeds into our brain where, like much else involved with our vision, a bunch of complicated stuff happens that we don’t necessarily understand yet. To further muddy the waters, how a color comes about in the first place depends on the agent of its coming about. Suffice it to say that the mechanism for generating colors caused by pigments (mixing paint, for the classic example) is different than that caused by light (e.g., patterns from stained glass windows).
When I looked more closely at which colors go with which wavelengths, it turns out that 589 nanometers (the designated wavelength of the Gemini laser) is right in the range of orange light. Thus, all the fuss was caused by a misapprehension about the color in the first place. Ah, well. One of the most famous negative results in science it isn’t, but I learned quite a bit along the way. The example also made me grateful to Carol for another illustration of the value of different ways of looking at the problem. And reminded me that sometimes mundane things are mysterious and sometimes mysterious things are mundane. Chalk it up to another summit-induced bout of hypoxia.