RADIATIVE TEMPERATURE CHANGES IN THE OZONE LAYER 
minimum follow the solstices. Dobson, Brewer, and 
Cwilong [11] and Goody [19] explain this behavior as 
a balance between two conflicting factors. They point 
out that the temperature of the lower stratosphere is 
primarily controlled by radiative processes, and is 
heated at least in part by ozone absorption of infrared 
radiation from the troposphere. The flux of this radia- 
tion reaches a maximum (or minimum) after the sol- 
stices. On the other hand, the amount of ozone reaches 
a maximum (or minimum) much earlier, just after the 
equinoxes. Therefore the temperature extremes occur 
at intermediate times. 
The temperature variations in the upper, warm, 
region of the ozone layer are not known. Direct meas- 
urements on a routine basis are not as yet obtainable 
there.* Indirect evidence, however, indicates that the 
seasonal variations, at least, may be very large. Whipple, 
Jacchia, and Kopal [60] have shown from their meteor 
studies that the density of the atmosphere at 70 km is 
approximately 2-38 times greater in summer than in 
winter in middle latitudes. This astonishing fact can 
be explained only in terms of a summertime expansion 
of the ozone layer above 30 km, corresponding to a 
mean temperature some 50C higher than in winter. 
Furthermore, Gowan’s calculations of radiative equilib- 
rium temperature at 50°N [26], discussed more fully be- 
low, indicate a summer-winter temperature difference 
of the same magnitude. 
Ozone Absorption. The explanation for the warming 
of the upper part of the ozone layer lies in the absorb- 
ing qualities of ozone. Laboratory measurements of 
ozone absorption reveal intense absorption bands in 
the ultraviolet and less intense bands in the visible. 
The Hartley bands of ozone, the most intense of all, 
lie between 2000 and 3200 A, with a strong maximum 
near 2500 A. The Huggins bands occur in the 3200- 
3600 A region. In the visible are the Chappuis bands 
at 4800-7800 A. The most detailed and homogeneous 
set of laboratory-derived absorption coefficients stem 
from the work of Ny Tsi-zé and Choong Shin-piaw 
[45, 46] and of Vassy [56]. Many other investigators 
[16, 34, 36, 41] have obtained results in agreement with 
theirs. 
Photochemistry of Ozone. The existence of ozone in 
the upper atmosphere may be explained on the basis 
of photochemical principles. The photochemistry of 
atmospheric oxygen has been discussed by many in- 
vestigators, for example, Chapman [7], Bamford [1], 
and Wulf [61]. 
The primary reaction leading to the formation of 
ozone is the dissociation of the oxygen molecule by 
solar energy at wave lengths less than 2400 A:5 
O2 + hy (\ < 2400 A) > O + O, (1) 
4. Recent work at the Evans Signal Laboratory [3] gives 
hope that observations from improved radiosondes may soon 
become available up to 40-50 km. 
5. Not all absorption below this wave length produces 
direct dissociation, but it all at least excites the molecule to 
the point where it is easily dissociated. 
293 
where » is the frequency of the incident light and h is 
Planck’s constant. Ozone is then formed by the collision 
of an oxygen atom, an oxygen molecule, and any third 
body: 
O36 © 4 WP Oy Se ve (2) 
Ozone in turn can be dissociated by absorption of solar 
energy or by collision with an oxygen atom: 
O; + hy (\ < 11,000 A) > O, + O, (3) 
From these four reactions, the rates of change of the 
amounts of ozone and atomic oxygen per unit volume 
are 
dm/dt = 2Q2 + Qs — kynimantm — kignins, (5) 
dn3/dt = = Q3 + kyonynoMm = kignins. (6) 
The symbols 1, m2, and n3 represent the numbers of 
molecules per unit volume of atomic oxygen, molecular 
oxygen, and ozone, respectively; n, represents the total 
number of molecules per unit volume in the air. The 
numbers of quanta absorbed from the solar beam per 
unit volume and unit time are given by Q» (for O2) and 
Q; (for O;). The rate of production or destruction of 
molecules from collisions is proportional to the num- 
bers of collidmg molecules in the unit volume, the 
photochemical reaction factors 1, and kj; being the 
constants of proportionality for the collisions repre- 
sented by (2) and (4), respectively. 
Under equilibrium conditions, dni/dt = dn3/dt = 0. 
From (5) and (6), then, 
= kas p Q> 
This equation was derived and numerically inte- 
grated by Wulf and Deming [61, 62, 63]. More recently 
Diitsch [13], Nicolet [44], and Craig [10] have repeated 
the calculations, making use of more recent and detailed 
information about the parameters entering into the 
calculation. All these computations give the equilibrium 
amounts of ozone to be expected at various levels in 
the atmosphere. Despite some uncertainties in the com- 
putations, the results are generally compatible with 
observation. 
Among the most interesting information derived from 
computations of equilibrum ozone amounts is that 
concerning the degree to which actual ozone amounts 
may be expected to correspond to equilibrium amounts. 
At and above the level of maximum ozone, any devia- 
tions from equilibrium conditions could last only a few 
hours. However, below the level of maximum ozone den- 
sity this time interval rapidly increases until, in the 
lower part of the layer, it is extremely large. Thus, 
below perhaps 25 km, the ozone is never necessarily in 
equilibrium with the sun. It isin just this region that the 
large variations in ozone amount are observed to occur. 
Possible Solar Effects on the Ozone Layer. One inter- 
esting aspect of the question of radiative changes in 
the ozone layer is the suggestion, made most recently 
