634 
It is now necessary to investigate the magnitudes of 
the changes produced by heating and cooling. If the 
air column BCNM were heated 5C relative to the en- 
vironment, the magnitude of the sea-level pressure 
change would be somewhat less than the change which 
would arise if the height of the 600-mb surface should 
remain invariant and the air column below 600 mb were 
replaced by a 5C-warmer column, that is, less than 
9 mb. The reasons for the smaller change are the 
raising of the 600-mb surface and the distribution of 
the outflowing air over a finite region. The extent of 
the surroundings also determines the sea-level pressure 
rise in the surroundings. It does appear, therefore, that 
heating and cooling can produce significant pressure 
changes and that nonadiabatic processes may be im- 
portant for the development of pressure fields. 
A brief survey of various types of nonadiabatic proc- 
esses will now be presented. The atmosphere is con- 
tinually gaining and losing heat by radiation. However, 
radiation studies indicate that the net loss of heat by 
the tropospheric air is small and that the space varia- 
tion of this loss is slight. It appears that the cooling is 
greater in regions of high specific humidity and conse- 
quently direct radiative processes in the troposphere 
may produce weak pressure fields. The space variation 
of radiation absorption in the stratosphere, such as in 
the ozone region, should give rise to high-level accelera- 
tional and pressure-change fields (see Haurwitz [16]) 
However, it is not obvious that these fields have a pro- 
nounced effect on the sea-level circulation. A significant 
difference between low- and high-level heating is the 
absence of a fixed boundary at the base of the heated 
layer in the case of the high-level heating. Before it 
may be concluded that such heating can produce sub- 
stantial sea-level changes it is necessary to investigate 
the possible significance of adiabatic processes in the 
troposphere below the level of heating. Adiabatic cool- 
ing can account for a definite pressure fall at high levels 
and an insignificant sea-level change. Hence the answer 
to the question of the effect of radiation absorption in 
the stratosphere upon low-level changes must await the 
results of further theoretical and empirical studies. 
Nonadiabatic temperature changes also arise from 
the flux of heat between the atmosphere and the earth’s 
surface. Because this heating or cooling is not uniform 
over the entire globe it should be expected that this 
thermal process would be accompanied by accelera- 
tional fields and pressure changes. The significance of 
this factor may be illustrated by a few simple examples. 
1. The development of low pressure in a region of 
strong heating is often observed on weather charts. 
These cyclones are usually referred to as thermal lows. 
2. The sea- and land-breeze circulations are good ex- 
amples of the creation of a small-scale pressure-change 
field through surface heating and cooling. 
3. The mean pressure maps of the Northern Hemi- 
sphere demonstrate that the average sea-level pressure 
over the hemisphere is approximately 4 mb lower in 
July than in January. This appreciable pressure change 
between summer and winter is consistent with the heat- 
ing of one hemisphere relative to the other. 
MECHANICS OF PRESSURE SYSTEMS 
4. Perhaps the most striking pressure changes are 
those which occur in middle latitudes between land 
and water areas from summer to winter. In winter, air 
is beg cooled over the land and air from the land is 
being heated over the water. The converse holds during 
the summer season. This distribution of heating and 
cooling requires a pressure rise over the land from 
summer to winter and a pressure fall over the ocean. 
From mean pressure and temperature charts it has 
been possible to determine approximately the pressure 
and temperature differences. The data in Table I clearly 
indicate that substantial pressure changes accompany 
the heating and cooling. 
TaBLE I. SEASONAL VARIATION IN AVERAGE SURFACE-AIR 
TEMPERATURE AND AVERAGE SEA-LEVEL PRESSURE 
Lat. 40°N Lat. 50°N 
Land Ocean Land Ocean 
January 
Pressure (mb)........... 1023 1015 1022 1006 
Temperature (F)....... 36 48 10 37 
July 
Pressure (mb).......... 1011 1020 1011 1016 
Temperature (°F) ...... 82 67 69 55 
5. Wexler [29] has explained the development of a 
polar anticyclone through cooling and its associated 
isallobaric convergence. The analysis of the pressure 
change which accompanies the heating or cooling has 
been extended by Schmidt [27]. Schmidt assumes a cer- 
tain distribution of the heating and computes the pres- 
sure changes from a consideration of radiative density 
variations, isallobaric divergence, and the vertical 
motion of the air. The theoretical formula is tested by 
a study of the seasonal pressure variations over the 
vast continents. The theoretical estimates are in good 
agreement with the observed pressure changes from 
summer to winter. 
6. Estimates [4] have been made of the rate of addi- 
tion of heat to cold polar air as it leaves the mid-Atlan- 
tic coast and moves out over the ocean. This heating 
can be very pronounced and it would appear that this 
process alone could produce substantial pressure 
changes since there is a space variation in the heating. 
Tt should be noted, however, that it is difficult to check 
the importance of the heating directly since a strong 
advection of cold air is usually associated with a pres- 
sure rise (see below). This type of strong heating occurs 
near the Atlantic coast of North America and the Pacific 
coast of Asia. 
The survey suggests that nonadiabatic heating or 
cooling of air, through contact with the earth’s surface, 
may produce significant pressure changes. The horizon- 
tal mixing of different air masses and the cooling or 
heating of air by falling precipitation are other examples 
of nonadiabatic processes which could give rise to pres- 
sure changes. 
The conclusion which is drawn from this review of 
the effect of nonadiabatie processes is that they are of 
a sufficient magnitude to influence the daily pressure 
