MECHANISM OF PRESSURE CHANGE 
patterns and to be detected in the field of pressure 
changes. For the day-to-day pressure variations, except 
for the diurnal change, it would appear that the space 
variation of the transport of air over colder and warmer 
surfaces is the most important of the nonadiabatic proc- 
esses. A satisfactory explanation of the development, 
intensification, and motion of the fields of acceleration 
and pressure change must be based, in part, upon the 
effect of nonadiabatic temperature changes. 
Horizontal Advection. Local changes in temperature 
in one region relative to another may arise through the 
horizontal advection of warm or cold air. The concept 
of thermal advection as a cause of pressure changes has 
been discussed by many meteorologists since Ferrel [8] 
advanced a thermal theory of pressure changes. For 
example, Exner [7], Henry [17], Defant [5], and Mc- 
Donald [21] have illustrated relationships between tem- 
perature advection at the surface and the simultaneous 
pressure change. Austin [1] has shown that regions of 
maximum advection of warm air near the earth’s sur- 
face are accompanied by pressure falls at sea level while 
pressure rises occur in regions of maximum advection 
of cold air. It was further demonstrated that advection 
at high levels did not appear to be associated with sig- 
nificant pressure changes at sea level. 
Even though prominent relationships have been es- 
tablished between the advection of cold and warm air 
in the lower troposphere and the occurrence of pressure 
rises and falls, it is still necessary to explain the mecha- 
nism of the pressure change. Perhaps the process may 
be visualized as the horizontal motion of warm air into 
a region of cold air giving an outflow at high levels and, 
therefore, a pressure fall just as the direct heating in 
Fig. 1 gives rise to a pressure fall. The converse is the 
transport of cold air into a warm region with high-level - 
inflow and a pressure rise at sea level. Such an explana- 
tion replaces the direct heating or cooling of the thermal 
model in Fig. 1 with the localized heating and cooling 
which arises from horizontal transport. The process has 
been described by Douglas [6] as “a convectional over- 
turning between adjacent cold and warm masses.” The 
similarities between the nonadiabatic and advection 
hypotheses may be summarized as follows: 
1. Both types of heating and cooling give rise to sea- 
level pressure changes when the temperature change 
extends upward from the earth’s surface. 
2. High-level temperature changes produced by non- 
adiabatic processes or indicated by geostrophic advec- 
tion are not necessarily associated with prominent sea- 
level pressure changes. 
3. The accelerational fields which arise from nonadia- 
batic processes appear to be similar to those which 
accompany the differential advection of warm or cold 
air. 
4. Both types of heating or cooling give rise to pres- 
sure-change fields only as long as there is a space varia- 
tion in the heating and cooling. 
Two important differences may be stated briefly as 
follows: 
1. The nonadiabatic process gives rise to the pressure 
change and accelerational field from an initially baro- 
635 
tropic state. The advective process requires a particular 
field of motion relative to the temperature field. 
2. The nonadiabatic heat source is stationary whereas 
the advective heating is a moving source. This distinc- 
tion is important in problems concerning the pattern of 
horizontal divergence and vertical motion with moving 
pressure-change systems. 
At this stage it is desirable to consider those aspects 
of atmospheric pressure change which can be attributed 
to the differential advection of temperature at low 
levels. 
The major part of a pressure fall or rise in a concen- 
trated region of large pressure changes at sea level can 
be explained by the advection of warm or cold air in the 
lower atmosphere. Also many features of the propaga- 
tion of pressure-change centers may be attributed to 
this advection process. For example, consider a simple 
model, as in Fig. 2. It will be assumed that the essential 
COLD 
—z 
WARM 
Fic. 2.—A katallobaric system. The dashed lines are 
katallobars and the solid lines are isotherms. 
features of the horizontal advection can be judged from 
the geostrophic flow even though it is recognized that 
the motion cannot be geostrophic. The katallobaric sys- 
tem is associated with a strong advection of warm air 
and consequently it should be expected to move in the 
direction of the flow, that is, in the direction of the 
geostrophic wind. However, the isallobars show that the 
geostrophic field is changing so as to intensify the warm- 
air advection at B and to produce cold-air advection at 
A. This changing field of horizontal motion itself gives 
rise to a motion of the katallobaric center in the direc- 
tion AB. Hence it should follow that the actual motion 
of the pressure-change center is in a direction somewhere 
between the direction AB and the existing geostrophic 
direction at /. Synoptic studies appear to confirm this 
conclusion. On the other hand this differential advec- 
tion, as judged by geostrophic flow, cannot account for 
the development of new isallobaric systems or for the 
intensification or weakening of existing isallobaric 
centers. 
One aspect of the motion which requires further in- 
vestigation in connection with the development of pres- 
sure changes at sea level is the role played by ground 
friction. It seems probable that the retarding effect of 
friction influences the low-level inflow and outflow 
which follow the creation of isallobaric fields. This re- 
tarding force may affect the ultimate pressure change, 
and consequently surface friction requires further con- 
sideration. 
