by Roger N. Clark
The true color of the Pleiades nebulosity, in M45, is shown to be deep blue, and best described as a deep-high altitude daytime sky blue. Digital cameras can record excellent color for the Pleiades nebulosity.
The Night Photography Series:
The Color of the Pleiades Nebulosity
Terrestrial Blue Sky
References and Further Reading
As we saw in Parts 2a - 2f of this series, the night sky is filled with objects displaying many beautiful natural visual colors. In this article I show the natural true color of the Pleiades nebulosity, M45. The true color presented here is a direct calculation using spectra of the nebulae and the spectral response of the human eye (which was presented in Figure 8 of Part 2c). I will compare the color response of digital cameras to the calculated colors.
The human visual color ranges from about 4000 Angstroms in the blue (400 nm, 0.4 microns) to about 7200 Angstroms in the deep red (720 nm, 0.72 microns). See "What Wavelength Goes With a Color?" for more information.
An image of the Pleiades is shown in Figure 1. The colors seem quite saturated. Are they realistic? As of 2017, such an image would probably be considered overly saturated. On the internet one can see many images of the Pleiades with many different colors. The images in Figure 2 illustrate some common differences (excluding processing we see on the internet that can range from white to green, yellow to red). Probably the most common color we see for the nebulosity is bluish-white, sort of like cirrus clouds in our blue daytime skies.
Published studies show that the nebula for the Pleiades reflection is due to dust in at least two layers in front of the Pleiades star cluster (Odell, 1965; Gibson and Nordsieck, 2003; Witt, 1985 and references therein). The dust appears to be silicate and carbon in the 0.01 to 0.1 micron size range. Carbon Monoxide is also detected. The scattered light is distinctly blue from photometric measurements. The Carbon Monoxide molecules will scatter similar to the nitrogen and oxygen molecules in our atmosphere which make the daytime sky blue. This scattering process by tiny particles is called Rayleigh scattering. The 0.01 micron particles will scatter similar to Rayleigh scattering, but the 0.1 micron grains, while also scattering blue light more efficiently, would be less blue than Rayleigh scattered light. While the intrinsic scattering properties of the dust is to scatter blue light more efficiently, the Pleiades stars illuminating the dust are also blue, creating an even bluer result. But how blue? There are also small changes in color of the reflection nebulae with angular distance from the stars. This has been attributed to the changing scattering angles and not due to changes in composition ( e..g. Witt, 1985; Gibson and Nordsieck, 2003).
Using spectra of the nebulosity, we can determine the true colors in the scene. The first challenge is to find published, calibrated spectra of the nebula. The only relevant data I have found is the spectrophotometry from Odell (1965). Odell's data include 6 positions south of the star Merope and are relative to the spectrum of Merope. Merope is a B6 star, so using spectra from part 2b, Color of Stars, the spectrum of star HD30584, a B6 star which has the same B-V color index as Merope, was used as a substitute. The spectra of such stars are quite uniform. The spectrophotometry of Odell times the B6 spectrum provides an excellent relative spectrum of the nebula. Because we perceive the natural world with daylight white balance, as discussed in the first parts of this series, the nebula spectra divided by the solar spectrum shows the relative intensities for daylight white balance.
The relative spectra of the Pleiades for the 6 Odell (1965) positions are shown in Figure 3. Note the comparison of the Rayleigh curves in Figure 3. Rayleigh scattering is the reason our daytime terrestrial sky is blue. As you can see, for positions 1 - 3, the Pleiades nebulae have blue response greater than Rayleigh scattering, thus bluer that Rayleigh scattered terrestrial atmosphere.
For comparison a spectrum of terrestrial blue sky is compared to the Pleiades nebula, position 1 in Figure 4. Note the Colorado blue sky spectrum is not quite a Rayleigh spectrum, but slightly bluer. The Pleiades nebula is slightly bluer than the Colorado blue sky spectrum. I measured the Colorado spectrum using my Oceanoptics flame spectrometer on a very clear day in January 2018.
Now that we have compared spectra, the spectral responses were converted to color (using the same methods as for the Trapezion true color study, Part 2f) and compared to camera response. It should be noted that colors and RGB values will vary depending on the color space used. Formally, the blue color from Rayleigh scattering and the colors in the Pleiades and terrestrial blue sky are not well described by sRGB color space. Adobe RGB is better but still not perfectly described. The camera values I report here are Adobe RGB values. Conversion by photoshop from Adobe RGB to sRGB significantly alters the values, especially reducing the red value in blue sky data.
The colors of the 6 Pleiades spectra are compared to the colors in the image in Figure 1 at he same locations as the Odell measurements, and the results are shown in Figure 6. The colors are very close, within a few percent.
The results of colors calculated from the spectra are reasonably close to the colors in Figure 1. But different camera models show the blue sky slightly differently. This is illustrated in Figure 7. The point of Figure 7 is to show that the small color differences between cameras are to be expected. The differences between the color calculated from spectra and those measured from stretched images like the colors seen in Figure 1 are similar in magnitude and of little consequence compared to the larger differences in color we see in online images of the Pleiades, and as illustrated in Figure 2, panels a and b.
So what is the perceived color of the blue sky? Photographing the blue sky is one thing, but are the colors in an image what we actually see? Further confusion is brought by color space, like sRGB or Adobe RGB, and even in the same color space, different cameras show slightly different colors (e.g. Figure 7).
First consider exposure. Figure 8 shows zenith blue sky images from a Canon 7D Mark II at different exposure levels. The saturation changed with exposure. The metered exposure on the sky was the same as a landscape with trees and grass. I made the images near noon and quickly downloaded them onto 3 computers, a laptop and two desktops with high gamut IPS calibrated monitors. The perceived color, when linearly brightened was approximately the -1/3 stop exposure, with G/B ratio near 0.62. After spending 1/2 hour outside in the sunlight (walking my dogs), then the perceived color appeared bluer, more like the -2/3 stop when brightened.
The blue sky is bluer at the zenith than at lower altitudes (Figure 9). The typical landscape photo includes sky near the horizon, and usually not near or at the zenith. To compare blue sky color in images I compared the images on my computer screens (as discussed above) to my view of the sky within minutes of obtaining the images. To compare blue sky with the Pleiades, the results here indicate that a high-altitude site on a very clear day should be imaged at the zenith, using metered level -1/3 stop (with cameras similar to the Canon 7D2) then increase brightness in post processing by multiplication (e.g. levels tool move the right slider to the left). Use sRGB color space. Increasing the exposure level in a raw converter or camera exposure will desaturate the color and would not be the perceived blue.
Figure 10 shows another blue sky image and a link to the raw data is given in the caption. This blue-sky day is good, but still not as blue as the Pleiades spectrum. Note the color difference with the images in Figure 9. On the day that the images in Figure 9 were made, the sky was not as blue and that is also reflected in the color of the tree.
Spectral analysis shows that the colors of the Pleiades nebulosity near the star Merope are slightly bluer than terrestrial daytime sky dominated by Rayleigh scattering. High altitude (greater than 7000 feet) very clear day overhead terrestrial daytime sky can produce similar blue colors to that in the Pleiades nebulosity. The color of the Pleiades nebulosity to the southwest from Merope shows colors that are less blue, but still blue. The colors in Figure 1 are close to true colors. The blue in the Pleiades nebulosity is bluer than the typical color seen in online images circa 2017.
The ratio of green divided by blue intensity gives an indication of the blueness, assuming red / green is even lower. The Pleiades position 1 computed color shows green / blue ~ 0.43. Colorado very clear blue sky at the zenith shows green / blue in the 0.49 to 0.55 range at 6000 feet. At 7600 feet that ratio drops to the 0.47 to 0.49 range, so the Pleiades nebula would be close to this color but a little bluer. The red / green value should be less than 0.65. Positions 5 and 6 show green / blue ~ 0.6, so still blue.
References and Further Reading
Gibson, S. J. and Nordsieck, K. H., 2003, The Pleiades Reflection Nebula. II. Simple Model Constraints on Dust Properties and Scattering Geometry, Astrophys. J., v589, 362-377.
Odell, C. R. 1965, Photoelectric Spectrophotometry of Gaseous Nebulae II. The Reflection Nebula Around Merope, Astrophys. J., v142, p.604-608.
Witt, A. N., 1985, Colors of Reflection Nebulae. I. Phase Function Effects in the Merope Nebula, Astrophys. J., v294, 216-224.
Clarkvision.com Astrophoto Gallery.
Clarkvision.com Nightscapes Gallery.
The Night Photography Series:
First Published January 19, 2018
Last updated March 8, 2018