By John A. O'Keefe Ph. D., Asst. Chief, Theoretical Division, NASA Goddard Space Flight Center; and Winifred Sawtel Cameron Theoritical Division NASA Goddard Space Flight Center


[35] Summary


The principal results in the field of space science obtained from the MA-7 mission are:

1. The luminous band around the horizon is attributed to airglow; a large part of the light is in the 5,577 -angstrom (Å) line, where maximum intensity is at about 84 kilometers.

2. Space particles, similar in some ways to those reported by Astronaut Glenn, were shown to emanate from the spacecraft. They are probably ice crystals.

3. New photographs showing flattened solar image at sunset were made.




A discussion is presented in this paper of the observations regarding terrestrial space phenomena made by Astronaut M.Scott Carpenter during the MA-7 flight and reported in paper 7. Some of these observations are compared with those made by Astronaut John H. Glenn, Jr., in the first manned Mercury orbital flight and described in reference 1. The principal subjects considered in the field of space science are:

1. The airglow layer at the horizon.

2. The space particles reported by Astronaut Glenn.

3. The flattened solar image at sunset.

An analysis of these and other observations of the astronauts is continuing.


Airglow Layer


Toward the end of the MA-7 flight, between 4 hours and 2 minutes g.e.t. (16 hr and 47 minutes Greenwich mean time) and 4 hours 18 minutes g.e.t., May 24, 1962, Astronaut M. Scott Carpenter made a series of observations of a luminous band visible around the horizon, known as the "airglow" layer. The airglow is a faint general illumination of the sky visible from the ground on a clear, moonless night. The glow is brightest about 10° or 15° above the horizon and becomes fainter toward the zenith. The height of the airglow layer has been investigated by Heppner and Meredith (ref.2) of the Goddard Space Flight Center using an Aerobee sounding rocket. This rocket, which carried a filter that transmitted only the 5,5577-angstrom (Å) line, have indicated that the height of the layer extend from 90 to 118 kilometers above the earth. Their studies were also concerned with the characteristics of other layers of specific wavelengths, such as the sodium layer. The light emitted from the luminous layer is attributed to a forbiden transition, or transition from a metastable state, of the oxygen in the upper atmosphere. A forbiden transition is very difficult to produce in the laboratory because the atoms lose the energy corresponding to the transition through the collision with another atom or with the walls of the container. This effect can be minimized only if the laboratory apparatus is very large and the enclosure is at an extremely high vacuum. Thus a forbidden transition is much more common in space.

The astronaut's observations of this luminous layer permit investigation and identification of three of its physical characteristics. The wavelength of the emitted light is discussed initially, and this is followed by an analysis of the brightness if the airglow layer. Finally, an examination of the height of the luminous band above the earth's surface is presented.




The most significant observation was made with a specially developed filter supplies by the NASA Goddard Space Flight Center. The filter transmits a narrow band of wavelengths, approximately 11 Å wide at the half-power point and centered at the wavelength of the [36] strongest radiation of the night airglow, namely 5,577 Å.

During the flight the astronaut noticed that the filter passed the light of the luminous band with but little attenuation; however, it rejected the light of the moonlit earth. Therefore, the band was identified as the 5577 layer.


Britghness of the Layer


Astronaut Carpenter noted that the airglow layer was relatively bright. An indication of this brightness was derived from a comparison of the brightness of the layer with that of the moonlit horizon.

Astronaut Carpenter also noted that the layer was about as bright as the horizon, which was at that time illuminated by the moon at last quarter. Assuming that the atmospheres at the horizon acts like a perfect diffusing reflector, and noting that the illumination of the moon at last quarter is approximately 2 x 10-2 lux, it is found that the surface brightness is 6 x 10-3 lux per steradian.


Height of the Layer


The astronaut provided evidence on the height of the layer through five separate observations:

1. By making a direct estimate which was from 8° to 10°.

2. By noting that it is approximately twice the height if the twilight layer. Astronaut Carpenter estimated the height of the twilight layer as 5 sun diameters of 2 1/2 °; hence the height of 5577 layer would be 5°.

3. By observing the star Phecda as it passed the middle of the luminous band.

4. By noting the time when Phecda was halfway from the luminous band to the horizon.

5. By noting the fact that when the crossbar of the reticle is scribed on the window set diagonally, the horizontal bar just covers the distance from the band to the horizon.


In method 3, the time of the passage of the star below the brightest part of the luminous layer was used. Through careful timing of the spacecraft tape and conversation with the astronaut, the time has been fixed at approximately 04:05:25 ground elapsed time (g.e.t.) or 16 hours 50 minutes 41 seconds G.m.t. To find the true height at that time, a special set of computations was made at Goddard Space Flight Center, starting from the spacecraft latitude...



diagram illustrating formulas used to determine height of layer

Figure 4-1. Parameters used to calculate height of layer.



...and longitude for each minute of ground elapsed time. By using the standard formulas of spherical astronomy, the angular zenith distance Z, schematically shown in figure 4-1, of Phecda was calculated. The ray from Phecda was considered to be tangent at each moment to an imaginary sphere which is concentric with the earth, and which is situated at a distance h below the observer. The usual formula for the dip of the horizon is h= R(1-sinZ), where R is the radius from the center of the earth to the spacecraft. Since only 3-figure accuracy is needed in h, it is not necessary to enter into refinements in the calculation of R; a mean radius of the earth of 6,371 kilometers plus the spacecraft elevation gives more than sufficient precision. Subtracting h from the spacecraft elevation gives the elevation of the layer. By using the above- mentioned time, the lower boundary of the layer is found to be at 73 kilometers. Other points are less definite; it appears at 04:03:33 g.e.t. or 16 hours 48 minutes 49 seconds (110 kilometers) G.m.t. Phecda had not yet entered the layer, and that at 04:04:52 or 16 hours 50 minutes 8 seconds (84 kilometers) G.m.t. it was approaching the middle of the layer.

These heights are some 10 to 15 kilometers lower than those which result from rocket measurements (ref.2). The discrepancy may be due in part to geometrical effects; for instance, a very thin layer has some intensity at all zenith distances greater than that of the tangent to the layer. Hence the determination of the bottom of the layer is intrinsically uncertain. In a thick layer, these methods are [37] slightly biased toward the lower portions. On the other hand it appears to be physically possible, especially if account is taken of turbulence that the maximum of the oxygen (O) is really lower than 90 kilometers.

The observation of the luminous layer through the filter was made at 04:16:50 g.e.t. Sunrise was witnessed about l minute later while the observation was being conducted. It follows s that the airglow is visible even when the twilight band is very strong. An attempt to observe it in the clay appears to be desirable. In this connection, it should be noted that Astronaut Virgil I. Grissom reported a grayish band at the top of the blue sky layer (see ref. 4). He remembers this layer as narrow and grayish in color, representing an actual increase in intensity. He pointed out the approximate position of the layer on one of the photographs taken by Carpenter at the height of 1.7° above the horizon Astronaut Grissom ma:, have in f let observed the luminous layer during the daytime.

Astronaut Carpenter did not note any vertical or horizontal structures in this layer. He did not attempt n continuous survey around the horizon; however, he did note the layer at several points along the horizon and believes it to be continuous all the way. It does not appear possible that this layer can actually absorb starlight. Any layer at this level capable of absorbing a noticeable fraction of the light (25 percent or more) would also significantly scatter light; it would therefore be a very prominent object on the daylight side. However, it is not definitely visible on the photographs of the day- side. That the decreased visibility of stats passing through the layer was a contrast effect is entirely in agreement with Astronaut Carpenter's impression. This layer is thus assumed to be luminous.

An interesting feature of this observation is the discrepancy between the eye estimates of 8° to 10° for the altitudes above the horizon, on the one hand' and the results of timed observations on the other. The latter indicates altitudes of 2° to 3°, which are clearly correct. For example, Astronaut Carpenter noted that when one arm of his reticle was at an angle of 45° it covered the space between the horizon and the bright band. The crossarm is 1.21 centimeters in length and is 26.2 centimeters from the astronaut's eye. At an angle of 45°, it subtends a vertical angle of about 2.6°.

It thus appears that the well-known illusion which exaggerates angles near the horizon, may also be experienced in orbital flight. It was evidently present during the MA-6 mission, since Astronaut Glenn also reports 7° to 8° as the height of the luminous band.

A summary of the results derived from the five methods of calculating the height of the airglow layer is presented as table 4-I.


Space Particles


Astronaut Carpenter also noticed and photographed white objects resembling snowflakes, or reflecting particles, at sunrise on all three orbits. (See fig. 4-2.) However, he also saw these objects 7 minutes after the first sunrise and again 43 minutes after sunrise and 2, 11, 23 26, 36, and 45 minutes after the second sunrise. It is thus quite clear that they are not related to sunrise, except perhaps in the sense of being most easily visible then.

In the photographs some of the particles were considerably brighter than the moon, which was then very near the first quarter. At this time, the moon is about-10; the particles may have been between -12.6 magnitude (10 times brighter than the moon) and -15 magnitude (100 times brighter than the moon). The second is considered more likely, in view of the appearance of the full moon (-12.6) as shown on photographs taken on the MA-6 mission. At -15 magnitude the particle brightness is consistent with centimeter -size snowflakes. The particles were verbally described by the pilot as having been between 1 millimeter and I centimeter...


[MISSING] Figure 4-2. Space particle photographed by Astronaut Carpenter.




[38] size and having a strong visual resemblance to snowflakes.

Shortly before reentry just at sunrise, Carpenter improvised the decisive experiment of hitting the walls of the spacecraft with his hand. The blows promptly resulted in the liberation of large numbers of particles. It is this clear that at least those particles observed in MA-7 flight emanated from the spacecraft.

The possibility that the particles might be dye marker or shark repellant, both of which are green and both of which are exposed to the vacuum, was considered. Tests were conducted which demonstrated that neither material tended to escape from the package in a vacuum. The possibility that they might be small particles of fiberglass insulator was also considered; in view of the smallness of the fibers, it appears likely that they would have been blown away at once, like the confetti of the balloon experiment. The dynamic pressure of 1 dyne per square centimeter is sufficient to remove at once anything weighing less than about 10 to 100 milligrams per square centimeter, which corresponds to a thickness of the order of 0.3 to 1 millimeter for most ordinary substances.

As mentioned in reference 1, there are two plausible sources within the spacecraft for these particles:

1. Snow formed by condensation of steam from the life support system.

2. Small particles of dust, waste, bits of insulation, and other sweepings.

The latter are very conspicuous in a zero g environment when there is nothing to keep them down, and it is extraordinarily difficult to free the interior of the spacecraft of such material. Undoubtedly, the exterior parts of the spacecraft which are exposed to the environment will contain these particles, and they undoubtedly provide a source for the space particles. In particular, a corkscrew-shaped piece observed by Astronaut Carpenter could possibly have been a bit of metal shaving or perhaps a raveled piece of insulation.

On the other hand, there is considerable evidence which points to snow as the source of the majority of the material. In the first place, water is exhausted from the spacecraft in far larger quantities than any other substance. In the second place, the material looked like snowflakes both to Glenn and Carpenter. In the third place, the frequency with which the particles are reported by Carpenter appears to be correlated with the temperature of the exterior of the spacecraft as recorded by thermocouples in the shingles. The temperature was always lowest at night, falling to temperatures of -35°C just before sunrise, and rising to 10°C just after sunrise.

The condensation probably occurred in the space between the heat shield and the large pressure bulkhead of the spacecraft, rather than outside the spacecraft, because even at the lowest recorded shingle temperature, around -50°C, the vapor pressure over ice amounts to about 0.039 millibar. Although this pressure is very low, it greatly exceeds the ambient pressure at the lowest spacecraft altitudes. Accordingly, it is not possible that snowflakes should form under the circumstances, even though it is true that the spacecraft must be surrounded by an expanding atmosphere of water vapor.

If the water vapor is assumed to expand freely, then the pressure at a distance of 1 meter from a hole 1 centimeter in diameter will be of the order of 1/10,000 of the pressure at the hole. Hence it is fairly clear that the pressure between the heat shield and bulkhead of the spacecraft will be far higher than the outside pressure, in spite of the presence of 18 one-centimeter apertures. Therefore, condensation behind the heat shield is more likely than outside. It is noteworthy that no formation of rime was noticed either on the window or on the balloon string. It is considered most likely that the luminous particles are snowflakes formed in the spacecraft between the cabin bulkhead and the heat shield by the steam exhaust from the life support system. It is suggested that they may have escaped into space through the ports, being driven outward by the expanding vapor. Note that at 2 hours 52 minutes 47 seconds g.e.t., Carpenter noticed a particle moving faster than he. At 2 hours 50 minutes g.e.t., he had planned to observe sunrise and was facing forward. This particle was therefore probably seen at point east to him. Most of the particles were seen behind him and falling back. This supports the idea that the particles probably [39] are pushed outward by the expanding steam from the spacecraft before they begin to stream backward. It is probable that many of the particles lodge on the outside of the spacecraft, since Carpenter is quite sure, from the direction in which the particles streamed across the window, that they came from near the point where he had knocked..


The Flattened Sun


New information regarding the refraction by the earth's atmosphere of celestial objects as seen from space has recently been provided by the Mercury manned orbital flights. Theory predicts that the sun's image near the horizon should he highly flattened. Astronauts Glenn and Carpenter obtained photographs of the setting sun that illustrate this effect rather strikingly . Carpenter recognized the phenomenon visually , but John Glenn did not.

A general procedure for the computation of refraction, in order to construct a theoretical sol solar profile for comparison with the actual photographs, is presented. . The quantities determined are the apparent and true zenith distances as seen from the spacecraft denoted by Zapp, and Ztrue, respectively.

To find these quantities, a ray through the atmosphere to the spacecraft is idealized. The phenomenon takes place effectively only for rays whose perigees are lower than 20 kilometers above the surface of the earth. Figure 4-3 illustrates the geometry employed.

The ray from the sun is traced backward from the spacecraft, C. The first section from...


procedure for the computation of

Figure 4-3. Geometry employed in computation of refraction.



....the spacecraft to the atmosphere, X, is straight. If the ray continued in this direction toward the sun, there would be a point, B, of nearest approach to the center of the earth, O. That distance is denoted by p, and the angle at the center of the earth from the spacecraft to B is denoted as (H) [Note: in the original document, (H) is represented as a circled H, Chris Gamble, html editor]. If B and p are known, the apparent height of any point on the sun as seen from the spacecraft could be calculated.

To make the calculation, the curving optical ray is followed forward until it is refracted co as to be parallel to the surface of the earth. This point is called the perigee of the ray, and is denoted by G. The line OG makes an angle (H) + r with OC, where r is the refraction angle for the sun when an observer at G sees it 90° from the zenith.

If the straight portion of the ray is prolonged it will intersect OG at some point A. Then, the height of D above G is called the refraction height, s. For any given height G, the refraction angle r at the horizon and the refraction height s. which depends on the true height and r. can be calculated . Then the right triangle OBD for the distance p can be solved. The length p is denoted by analogy with the similar dynamic problem, such as the impact parameter.

Given p and the spacecraft height, the apparent angles at the spacecraft can be calculated as a function of (H). The refraction angle 2r = R is added to form the true zenith distances.

The computation of the refraction r = z-z', where z is the true zenith distance and z' the apparent zenith distance, for a fictitious observer stationed at perigee, was based on the rather detailed theory given in reference 5 . The pertinent formulas are:


formula as described above



formula required for determining refraction


T= the absolute temperature divided by 273.0° at height h

P = pressure at height h divided by the ground pressure of 1.013x106 dynes/cm2

B = coefficient involving the index of refraction µ and the polytropic index n.


[40] The temperature, pressure, and density 3 of the atmosphere at altitude h were taken from reference 6 . More recent data on these parameters are available from reference 7.

The parameter s, here called the refractive height, is n refraction correction commonly applied in calculations of times of contact in eclipses. The derivation of s is found on page 515 of reference 4, which gives its relation to the index of refraction as 1 + s/a = µsin z' /sin z. Here is mean radius of tire earth (6,371,020 meters). The index of refraction µ is computed using µ = 1 + 2h.[Greek letter delta] where h. is a constant! and [Greek letter delta] is the density at h divides by the density at the surface (1.17.0x10-3 q/cm3).

Once µ, r, and s have been obtained, then R follows immediately from the simple relation R=2r (a ray is doubly refracted at the space craft ) and p is obtained from the equation p= (a + h + s) cos r. Then (H) is determined from the relation cost (H) = p/H, where H = a+ hc (hc = 257,000 meters as determined by the MA-7 orbit ). Finally, Zapp and Ztrue are related to (H) and R by the equations Zapp = 90° + (H) and Ztrue= 90° + ((H) + R) . Table 4-II summarizes the computed results.

The flattening of the image of the setting sun is best illustrated in the plot of Zapp versus, Ztrue. An image representing the sun to scale may be placed at any Ztrue, and points around the limb extended to the curve may be located on the Zapp axis. This procedure yields the apparent zenith distance of those points. Since the horizontal axis is not affected by refraction, parallels of altitude may be laid off on the unrefracted image of the sun and similarly on the apparent image of the sun. The apparent image may be rectified for easy comparison. The theoretical profiles of four phases of a setting sun are illustrated in figure 4-4, which is a plot of Ztrue vs. Zapp for four true zenith distances of the sun's center. These distances are Ztrue= 105.460°, Ztrue= 106.236°, Ztrue= 106.918° (sun's lower limb on the horizon), and Ztrue= 107.180° (sun's center on horizon). The ratios in percent of the vertical to horizontal diameters are approximately 0.63, 0.46, 0.17, and 0.11, respectively. Considering the spacecraft angular velocity of 4°/ min, it is seen that the entire refraction effect took place in the relatively short interval of about 20 seconds.



chart describing sun refraction relative to observer's position.

Figure 4-4. Stages of the setting sun.



The uncertainty in photography times precludes an exact comparison of theory and observations. However, figure 4-4 (c) most nearly simulates the photographs in figures 4-5 and 4-6 which show the effects of the spacecraft motion and still demonstrate the arresting, effect. Figure 4-5 was photographed by Astronaut Glenn on February 20, 1962. He specifically states that he did not see the sun as a narrow, flat object. He observed it as spreading out about 10° on either side and merging with the twilight band.


[MISSING] Figure 4-5. Flattening of sun photographed by Astronaut Glenn.



Figure 4-6 was photographed by Astronaut Carpenter on the MA-7 flight of May 24, 1962. He stated that the sun definitely appeared somewhat flattened during sunrise and sunset.


41] [MISSING] Figure 4-6. Flattening of sun photographed by Astronaut Carpenter.



Therefore, the flattening effect produced by atmospheric refraction of a celestial body as seen from space has been demonstrated by direct observation . However, it is hoped that future missions will yield photographs with more precise times of observation and perhaps measures of the horizontal and apparent vertical diameters by the astronaut. using a sextant might be feasible. At any rate, the observations by astronauts of future flights will be carefully analyzed and further compared with the theory stated herein to explain refraction phenomena more fully.

Acknowledgments.- Thanks are due to Mr. Lawrence Dunkelman of Goddard Space Flight Center for providing the 5,577 filter; to Professor Joseph W. Chamberlain, University of Chicago, for assistance and advice in the interpretation of the airglow information; to James J. Donegan of Data Operations Division of Goddard Space Flight Center for the provision of the final orbital elements; and to Frederick B. Shaffer of the Theoretical Division of Goddard Space Flight Center for programing and obtaining the orbit on the 7090 computer.

 Table 4-I. Observations of the Height of the 5577 Layer.




1. Eye estimate at angular height

8° to 10° above horizon

Apparently the moon illusion exists even in the absence of a gravitational field; objects look larger near the horizon.

2. Comparison with twilight layer.

5° above horizon

Same as above.

3. Observation of star in the middle of the layer.

101°54' from Zenith

Height about 83 kilometers

4. Observation of star halfway from haze layer to horizon.

Zenith distance is 103°10'

Apparent horizon 1° above geometrical horizon, whose Zenith distance 106°

5. Observation of angular height with reticle.

2.6° above horizon

Confirms methods 3 and 4; i.e. apparent horizon is more than 1° above geometrical horizon.



 [42] Table 4-II. Summary of Refraction Computations

h, meters T P, dynes/cm2 [Greek letter delta], g/cm3 r, minutes µ s, meters p, meters (H), deg Zapp, deg Ztrue,deg



























































































































1. ANON.: Results of the First United States Manned Orbital Space Flight, Feb.20, 1962. NASA Manned Spacecraft Center.

2. HEPPNER, J.P., and MEREDITH, L.H.: Nightglow Emission Altitude From Rocket Measurement. Jour. of Geophysical Research, vol.63, 1958, pp51-65.

3. ANON.: American Institute of Physics Handbook. McGraw-Hill Book Co., Inc., 1957.

4. ANON.: Results of the Second U.S. Manned Suborbital Space Flight, July 21, 1961. NASA Manned Spacecraft Center.

5. GARFINKEL, B.: An Investigation in the Theory of Astronomical Refraction. Astronomical Jour., vol.50, no.8, 1944, p.169.

6. Rocket Panel, Harward College Observatory: Pressures, Densities, and Temperatures in the Upper Atmosphere. Phys. Rev., vol.88, no.5, 1952, p.1027.

7. CIRA: Cospar International Reference Atmosphere Report. North-Holland Pub. Co., Amsterdam, or Interscience Pub. Inc., New York, 1961.

8. CHAUVENET, W.: A Manual of Spherical and Practical Astronomy, vol.1 (5th ed.), Dover Pub. Inc., New York.