New Data on the Composition and Temperature of the Upper Layers of the Atmosphere

One of the most direct and effective methods for studying the composition and state of the upper layers of the atmosphere is undoubtedly the spectroscopy of auroras. However, the low intensity of the glow requires the use of very high-aperture apparatus; to achieve this, its resolving power has to be sacrificed. As a result, deciphering the auroral spectrograms obtained up to now (in fairly large numbers) has presented great difficulties, and many lines have been identified unreliably. As an example, let us recall the well-known discussion over many years concerning atomic nitrogen.[^1] Therefore, Wegard’s recent preliminary report[^3] is of great interest: using a new, specially constructed spectrograph, he succeeded in obtaining an auroral spectrum (Oslo, night of 23–24 February 1950) in which the wavelengths could be measured with an error of less than 1 Å (see figure). In all, in the region from 3900 to 6300 Å, he measured 114 lines, 54 of which were observed for the first time. The complete decoding of the spectrum, a reproduction of which may be seen in the figure, has not yet been given. Of importance is the establishment of the undoubted presence of atomic nitrogen—altogether 9 lines of N I were identified (7 of them indisputable) and 21 lines of N II (18 of them indisputable)—of atomic oxygen—12 lines of O I and 25 lines of O II—and also of the ions N$_2^+$ and O$_2^+$, whose presence had not previously been detected.[^1] In addition, lines of molecular and atomic hydrogen were identified. The H$_\beta$ line is strongly broadened and shifted toward the short-wavelength end of the spectrum, which is the result of the Doppler effect and testifies to the rapid motion of hydrogen atoms in the direction of the Earth’s surface. This last result is a strong argument in favor of the hypothesis put forward by Wegard as early as 1933[^1] that hydrogen enters the upper layer of the atmosphere not from the troposphere, but intrudes into the Earth’s atmosphere as part of corpuscular streams ejected by the Sun; at the same time it strengthens the hypothesis concerning the formation of so-called noctilucent clouds from droplets of water arising from the chemical reaction of such “cosmic” hydrogen with the oxygen of the air.

Finally, in the spectrum obtained by Wegard, the R-branch of the negative nitrogen band with $\lambda = 3914$ Å proved to be distinctly resolved, which made it possible to carry out an accurate determination of the temperature. The results were as follows:

by the position of the maximum intensity: $T = 219.9^\circ$ abs.;
by the distribution of intensity: $T = 217.9^\circ$ abs.

Thus, the temperature of the corresponding layers of the atmosphere proved to be equal to $-54^\circ$ C.

This value is difficult to compare with other data (especially since in the region of aurorae the temperature, as is known,^1 undergoes considerable changes with altitude), because the author indicates neither the altitude of the aurora he studied nor its type.

Aurora  Argon

Thus, to a certain extent, the data on the composition of the atmosphere are also devalued. It may be hoped that in a fuller communication this gap will be filled.

G. Rosenberg

CITED LITERATURE

1. See, for example, I. A. Khvostikov, The Luminescence of the Night Sky, Publ. House of the Academy of Sciences of the USSR (1948).
2. L. Vegard, Nature 165, 1012 (1950).