© 1975-1979, 2008 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Xenology: An Introduction to the Scientific Study of Extraterrestrial Life, Intelligence, and Civilization, First Edition, Xenology Research Institute, Sacramento, CA, 1979; http://www.xenology.info/Xeno.htm

5.4.2  Sky Colors

What about the color of alien skies? Must they always be blue?²⁰⁵⁹ Of course, ETs will probably have different physiological seeing equipment than ours, but we shall permit ourselves the minor anthropocentric convenience of viewing their world through human eyes.

Light that reaches our eyes from the sky is merely sunlight scattered by the atmosphere. Had the Earth no air, our sky would appear quite black. This explains where the light comes from, but not why it is blue.

In 1899, a famous Englishman by the name of Lord Rayleigh devised an explanation for the color of the sky (Figure 5.9). According to his mathematical theory, scattering from very small particles (such as air molecules) increases as the inverse fourth power of wavelength.¹⁹⁹⁵ This means that blue light, which has a very small wavelength, is highly scattered, while red light, with a relatively long wavelength, is scattered much less -- sixteen times less, in fact.^1990,1991 So the blue light is preferentially removed from sunbeams and spread out uniformly from horizon to horizon. A little red is also present, and some yellow and green too, but blue is clearly predominant.

Figure 5.9 Scattering of Light in Planetary Atmospheres

	SCATTERING OF LIGHT BY AIRBORNE PARTICULATE MATTER1993,1994,1995 The family of curves at left represent the relative amount of light scattered away -- at each particular frequency of light -- by a hypothetical cloud of perfectly spherical, transparent, uniform size droplets suspended in air. However, ideal Mie monodispersions rarely occur in nature. Clouds of particles are of various shapes, textures, sizes, colors and degrees of opacity. Consequently, this graph is an abstracted, idealized version of reality and should be interpreted in that spirit. The beginnings of Rayleigh scattering are seen in the visible portion of the spectrum, in the right-hand section of the graph, where the particle radius r falls below a tenth of a micron or so. The flat parts of the curves stretching horizontally to the left indicate uniform scattering of all frequencies of light. The wavy parts in the middle demonstrate the oscillatory nature of the preferential scattering by color.
EXTINCTION OF LIGHT BY PASSAGE THROUGH A CLOUD OF PARTICLES1995 The three curves at right are simply a slightly different way of looking at the data in the above graph. Here, we compare the relative attenuation of blue, green, and red light as it passes through the same cloud of idealized haze/fog particles we considered before. Note that for small droplets in the air, blue is preferentially scattered giving a blue sky. At large particle sizes, no frequency is preferred and the sky washes out white. For intermediate scatterer radii, red and blue alternate in their supremacy. Note also that green, whenever it predominates, is accompanied by large amounts of the other two colors -- leaving a largely white sky with per haps the faintest of greenish tinges.
Type Radius Concentration (mm) (cm-3) Air molecule 10-4 1019 Aitken nucleus 10-1 - 10-2 104-102 Haze particle 10-2 - 1 103-10 Fog droplet 1 - 10 100-10 Cloud droplet 1 - 10 300-10 Raindrop 102 - 104 10-2 - 10-4	PARTICLES RESPONSIBLE FOR NATURAL ATMOSPHERIC SCATTERING1994 This table lists the approximate size (in microns) and concentration of typical scattering particles normally present in Earth’s air. Aitken nuclei are hygroscopic (water-absorbing) microscopic condensation nuclei, the result of chimney flue gases, tobacco smoke, sulfur dioxide industrial emissions, and s host of natural sources.

We can correct the Rayleigh theory for differences in planetary surface pressure and temperature. It turns out that the amount of light scattered is directly proportional to the atmospheric pressure, and inversely proportional to the temperature.¹⁹⁹⁴ So if we double the pressure we double the amount of light scattered in all colors -- and the sky gets brighter generally. Doubling the temperature has the opposite effect: the intensity of scattering is cut in half. On the surface of a high pressure planet like Venus, the effect would be rather extreme. All colors would be so strongly scattered that the sky be comes a dim, featureless white.²⁰⁵⁹

In a perfectly clear, Earthlike atmosphere, the sky would be a rich blue hue. But we observe it to be a hazy, lighter blue. Why?

The Rayleigh theory applies only to particles which are much smaller than the wavelength of light, say, less than 10-100 Angstroms.¹⁹⁹⁴ If the scatterers in the air are much larger than this (as with dust in the atmosphere), Rayleigh’s formulation breaks down and the vastly more complicated Mie theory must be used¹⁹⁹⁵ -- the details of which are beyond the scope of this book.

Rayleigh’s theory tells us that particles smaller than about 0.1 micron will preferentially scatter blue light. The Mie theory explains the behavior of atmospheres containing particles larger than about 4 microns. Above this critical size all frequencies of light are equally scattered, and the result is a gray or white sky. (Since there is always plenty of particulate matter, water haze and industrial pollutants floating around in the air -- perhaps 100-1000 kg over each square kilometer -- the sky’s sharp natural blueness is washed out unless we move to higher altitudes.)

Between 0.1 and 4 microns, the Mie theory becomes especially complex.¹⁹⁹⁵ The selection by color oscillates, sometimes preferring to scatter more blue and sometimes more red.^1993,1995 This effect is extremely sensitive to particle size. A uniform haze of 0.4 micron particles would scatter more blue (blue sky), but a similar cloud of 0.6 micron particles would produce more red (red sky).^1993,1994

If this is true, why don’t we commonly see such vivid colors in natural Earthly hazes and fogs? The reason is the natural fogs and mists contain a mixture of all sizes of particles, from one to ten microns or larger.¹⁹⁹⁵ As a result, these interesting color effects are added together randomly and average themselves out to a bland whiteness -- which we do observe. If some reasonable mechanism could be proposed to get particles of a single, specific size into the atmosphere (i.e., a “monodispersion”); quite beautiful red and blue sky colors would be possible.

Barring this fascinating alternative, as particles of increasing size are added to a “Rayleigh atmosphere” the sky color will appear to change from dark blue to powdered blue, to whitish blue, and finally to grayish white.

A third factor affects sky color besides Rayleigh and Mie scattering. The color of the particles themselves is very important. A red particle, for instance, absorbs all light but red -- which it reflects. Thus, it appears red in color. A green particle tend: to absorb blue and red but reflect green. (Under red or blue light such a particle would look black, but in green light it looks green.) So an atmosphere heavily laden with, say, green dust particles should also take on a distinct greenish hue.

We are now in a position to understand why the sky of Mars is red.^1989,2035 We add up the contributions from three effects: (1) Rayleigh scattering should give blue sky light, but will only be about 1% of its intensity on Earth because the Martian air pressure is only 0.01 atm;²⁰³⁵ (2) Dust motes an estimated two microns in diameter¹⁹⁸⁹ should produce a bright haze without color by Mie scattering; and (3) Particles in the Martian atmosphere are reddish surface dust, which reflect red light while preferentially absorbing blue and green. Hence, the sky of Mars is unusually bright, and appears a hazy “salmon pink” or “orange cream”¹⁹⁸⁹ (“embarrassed brick”?²⁰³⁵). It is clear that many other sky colors are similarly possible, provided a planet can be found with fine surface dust of the desired color.

There are other ways to get non-blue skies. For instance, we have discussed the process of frequency-selective light absorption by dust particles. Molecules of gas exhibit this property too.^619,620 The sky would no longer be blue under a fluorine atmosphere, to take one example. This gas absorbs blue strongly, and appears pale yellow in color. The sky would take on this color.

Chlorine air should appear green, because it absorbs light preferentially at the blue and orange-red ends of the spectrum. Similarly, an atmosphere of nitrogen dioxide would provide an orange-brown sky. If sulfur vapor is available, the air would alter color dramatically with large temperature changes. Near the boiling point at 720 K the sulfur sky would be dark yellow; as the temperature climbed to 770 K the atmosphere would turn a deep red, returning to straw yellow at about 1120 K.

The problem with using gases such as these is that they absorb light too darned well! At one atm pressure, a few meters of pure chlorine gas would transmit no visible light.²⁰⁵⁹ This is because even though blue and red are removed preferentially, some green is also eliminated. The sulfur vapor fares no better, sadly. At 1 atm pressure, blue light is cut to below human eye visibility in less than half a meter, and the red is gone in fifty meters.

So if the partial pressures of any of the aforementioned gaseous absorbers exceeds perhaps 0.001-0.01 atm, no light of any color will be able to reach the surface of the planet from the outside. Any inhabitants there must find their way around without the assistance of eyeballs.

If we want to use gaseous absorbers, it is better to choose weak absorbers instead of strong ones. For instance, under a deep ozone atmosphere the sky would probably appear reddish, because the gas is known to slightly absorb blue, yellow, and green sunlight rather well. Methane and ammonia, weak absorbers as they are, would provide a lovely blue-green sky (because absorption is mainly in the red) assuming the atmosphere was thick enough.²⁰⁵⁹

If the temperature at the surface is sufficiently high, another factor must be taken into account: blackbody radiation. Just as a stove’s heating element glows red when it is hot, so will the surface of a fried world like Venus. On Venus, red light emitted by the hot rocks could be orders of magnitude brighter than terrestrial moonlight -- about like Earth on a dark, rainy day. In the blue the intensity would be about 100,000 times less than in the red, roughly 10% as bright as moonlight. Since red clearly predominates, reflections off the cloud base will give the appearance of a red sky, assuming fair or good visibility.

Still another trick to get colorful skies is to arrange for permanent colored cloud covers. Arthur Clarke suggests in Imperial Earth that the skies of Titan may be white with beautiful orange and red streaks and whorls, because of the presence of hydrocarbons and other organics in the atmosphere.¹⁹⁴⁷ This is similar to what is believed to impart coloration both to the orange bands and the Red Spot of Jupiter. Unfortunately, 20th-century humans are unlikely to find photochemical smog a very attractive method of obtaining unusual sky colors.

More aesthetically appealing are the possibilities of continuous luminescence, phosphorescence, and fluorescence as an adjunct to sky color phenomena.¹⁹⁹¹ But perhaps the most intriguing of all is the striking sunset effect called the “green flash,” which occurs just after the sun has dropped below the horizon.¹⁹⁹² The red and yellow light is not refracted enough to reach the observer at this point, and the blue has all been scattered away. This leaves only green, which is experienced as a brilliant flash during optimal viewing conditions.²⁰⁵⁹

But flashes on other planets could appear vastly different. Even on Earth, blue and violet flashes have been seen at higher altitudes.¹⁹⁹² On low-pressure worlds, where blue is scattered less (as on Mars), blue flashes may be the rule. the planet's rotation is slow enough, the “flash” could become a “glow,” lasting for seconds or even minutes.

It might be supposed that by changing stars one might be able to affect the color of the sky. After all, sky light is just scattered sunlight, and a class K sun puts out a lot more red than a class F star. However, as we see in Table 5.11 on the next page, the consequences of illuminating an Earthlike atmosphere under the light of different stars are not great. Blue will predominate in the Rayleigh sky color, even if light from the coolest, reddest class M sun is used. On the other hand, we note that a terrestrial planet circling an F5 star will have skies of much deeper blue than a world associated with, say, a K2 sun. Stellar class is at best a very fine adjustment to sky color, in capable of countermanding the dictates of the atmosphere.

What about the appearance of the primary itself, as viewed from the planet’s surface? If the planet is in orbit around an orange or red star, the sun would seem bigger and redder than Sol does in our sky. Colors at the surface, illuminated by sunlight, would appear slightly different -- the blues darker and the reds brighter. Shadows would have blurrier outlines than those on Earth. But an F5 star might cast sharper shadows, with a slight bluish tinge.⁸⁷⁷

As far as color is concerned, if the observers are beneath an atmosphere which either scatters the blue (blue sky) or absorbs blue preferentially (red sky), then light from the star will lose blueness and appear redder.¹⁹⁹¹ This effect is most striking at sunset on Earth, when the blue in Sol's rays is so completely attenuated that fiery red alone remains.¹⁹⁹⁰ Were the surface pressure perhaps five or ten times greater, Sol would appear similarly reddish at high noon and deep crimson at sunset (but much dimmer). Wispy puffs of clouds would catch the ruddy solar rays throughout the day, streaking and mottling the luminous azure sky with magnificent ever-changing patterns of coralline and cerise.

If the observers are at the bottom of an atmosphere which absorbs the red (blue sky) or scatters the red preferentially (red sky), the sun will appear bluer than normal.¹⁹⁹³ This effect has been seen, albeit rather infrequently, on Earth from time to time. Owing to the presence of particles at high altitudes following the great volcanic eruption at Krakatoa in August, 1883, the Moon took on a distinct blue-green color. This phenomenon of “blue moon” was observed in Great Britain on September 26, 1950, due to widespread fires covering a quarter-million acres of forestland in northern British Columbia, and on other occasions elsewhere.³⁶⁰ Blue suns and green suns are also possible in the same manner,¹⁹⁹³ and have been observed infrequently.²⁰⁷⁷

Last updated on 4 November 2016