1294 
or so at a level in the atmosphere not more than a few 
thousand feet above the earth’s surface and often much 
less. If the lapse rate of specific humidity becomes suffi- 
ciently negative, superrefraction is replaced by subre- 
fraction. The degree of superrefraction is vitally af- 
fected by the interval of height over which steep gra- 
dients are maintained and by the closeness to the 
earth’s surface at which they occur. 
The Meteorological Problem 
From what has been said it is clear that we need to be 
able 
1. To recognize those situations which involve a 
temperature inversion exceeding about 5F per 100 ft 
within a few thousand feet of the earth’s surface. 
2. To recognize those situations which involve a 
lapse rate of humidity exceeding about 144 g kg per 
100 ft within a few thousand feet of the earth’s surface. 
3. To state, within say 10 ft, over what interval of 
height the gradients exceed the values mentioned. 
4. To state to an accuracy of say 10 per cent or 10 
ft, whichever is the cruder, the height above the earth’s 
surface at which the steep gradients occur. 
The above requirements are the least that would make 
a definite contribution to quantitative radiometeorol- 
ogy, and are insufficient to provide all the information 
desirable. The complete profiles of temperature and 
humidity, with an accuracy indicated by the require- 
ments listed above, are highly desirable, together with 
their statistical variations. 
There is, of course, no particular difficulty in measur- 
ing the profiles of temperature and humidity as re- 
quired above at a particular location by means of 
meteorological instruments attached to towers, bal- 
loons, kites, aircraft, etc. Important as such experi- 
ments are, however, they do not by themselves consti- 
tute an adequate understanding of the problem for 
radiometeorological purposes. The real problem is to 
be able to specify the profiles of temperature and 
humidity with sufficient accuracy purely from the 
weather data ordinarily available in synoptic meteor- 
ology. From this source, it would usually be possible 
to decide with reasonable accuracy what are the values 
of parameters such as temperature excess and humid- 
ity deficit, illustrated in Figs. 1 and 3. In addition 
wind speed and direction near the surface and at an 
appreciable height would normally be available, as 
well as information about the type of terrain. A good 
deal of additional information of a general character 
would also be available, and it is reasonable to suppose 
that contained in all these data are the values of the 
parameters which control those features of the profiles 
of temperature and humidity near the earth’s surface 
that are vital in radiometeorology. The problem is 
therefore to obtain such a fundamental understanding 
of how the profiles of temperature and humidity near 
the base of the troposphere are controlled by such 
parameters as temperature excess, humidity deficit, 
wind speed, etc., that knowledge of the values of these 
parameters is adequate to provide answers to at least 
RADIOMETEOROLOGY 
the four requirements stated at the beginning of this 
section. 
This can be achieved, if at all, by developing a 
theory of the profiles of temperature and humidity 
near the earth’s surface capable of standing the test 
of experimental verification. Such a theory should, 
ideally, specify the required profiles as a function of 
certain parameters; if these parameters could be de- 
duced from ordinary synoptic weather data, and if the 
theory were to give the profiles with sufficient accuracy, 
the problem would be solved. 
There are, of course, certain aspects of the problem 
which, even at the outset, are not in doubt. The most 
important physical process directly mvolved in con- 
trolling profiles of temperature and humidity is eddy 
diffusion. High eddy diffusion means almost uniform 
potential temperature and specific humidity, and con- 
sequently orthodox propagation. Low eddy diffusion 
permits the existence of substantial atmospheric gra- 
dients and so makes superrefraction possible. Eddy 
diffusion is low in a temperature inversion. Thus a tem- 
perature inversion is Important in connection with 
superrefraction for two reasons. First, the gradient of 
temperature itself causes some downward refraction. 
But, more important, the associated stability of the 
atmosphere permits the existence of a steep gradient of 
humidity, and it is this that is frequently the imme- 
diate cause of unorthodox propagation. 
The important ways in which marked inversions of 
temperature and associated hydrolapses can occur near, 
or relatively near, the surface of the earth arise from 
the processes of subsidence, advection, and nocturnal 
radiation. Over land, nocturnal radiation inversions 
are the principal feature associated with unorthodox 
propagation. Over sea, particularly in coastal waters, 
advection of warm dry air from land to sea is of great 
importance, often complicated by a coastal circulation 
arising from the land-sea difference of temperature. 
Subsidence in many cases is a prerequisite for surface 
inversions for which radiation or advection is the more 
immediate cause. Subsidence keeps the sky clear and 
permits the land to heat up to a high temperature 
during the afternoon and cool rapidly at night. This 
leads to a marked radiation inversion inland at night 
and often makes available, during the afternoon, a 
large excess of land temperature over sea temperature 
for production of advection inversions over the 
sea. Moreover subsidence brings warm dry air down rela- 
tively close to the earth’s surface, frequently forming a 
subsidence inversion. A well-developed subsidence in- 
version is itself a potent cause of superrefraction propa- 
gation between points on the earth’s surface provided 
it exists at a level less than say 5000 ft above the sur- 
face, but many subsidence inversions are too high to be 
of much direct practical importance in ordinary radio 
communication. 
The problem therefore reduces to this: we need to 
understand the phenomena of subsidence, advection, 
and nocturnal radiation sufficiently well to form a 
workable theory of the associated profiles of tempera- 
ture and humidity. 
