AEROLOGY OF EXTRATROPICAL DISTURBANCES 
UPPER LONG WAVES AND CYCLONES 
As has been pointed out, the idealized picture of a 
zonal flow is strongly disturbed in all actual situations. 
Therefore the real pattern of air flow can be regarded 
as the superposition of different wind disturbances 
upon the zonal flow pattern. 
There is still no satisfactory theory for the formation 
of the disturbances of the upper westerlies. However, 
three general sources for large disturbances must be 
considered. The first depends upon the topography of 
the earth’s surface, the second is thermodynamical and 
depends upon the distribution of secondary heat and 
cold sources connected with the geographical distribu- 
tion of continents and oceans, and the third is connected 
with a possible instability of the zonal baroclinic 
current. 
The orographic disturbances are not limited to the 
regions of mountains. Once formed in such a region, a 
disturbance has a tendency to influence other regions 
around the entire hemisphere, as has been pointed out 
by Rossby [53], Yeh [64], and Charney and Eliassen 
[14]. The thermodynamical disturbances depend upon 
the fact that the continents act as secondary cold 
sources in the winter season and as secondary heat 
sources in the summer season. The opposite can be said 
about the oceans. In the cold season the continents are 
regions of surface outflow or divergence and the oceans 
are regions of surface inflow or convergence. Corre- 
spondingly, there must be upper convergence over con- 
tinents and upper divergence over oceans. Because of 
the westerly flow in the middle and upper atmosphere 
there must be a tendency for the formation of upper 
troughs over the eastern part of continents, and ridges 
over the eastern part of oceans. During the summer 
season the opposite must be true.® 
The combined effect of the “orographic” and “‘ther- 
modynamic”’ disturbances presents a very complicated 
problem. The location of the hemispheric disturbances 
dependent on both of these effects is also mfluenced 
by the strength of the west wind and the location of 
the belt of the wind maximum. And because an 
“orographic” or “thermodynamic” disturbance, once 
formed, must influence other regions outside its source 
region, there are a great number of possible combina- 
tions for perturbations around the entire hemisphere. 
Since the influence of these geographical factors can 
never be eliminated, it is difficult to get any idea of 
what kind of atmospheric disturbances would form in an 
atmosphere without any geographical factors influenc- 
ing the atmospheric processes. 
A comparison with the Southern Hemisphere could 
give some clues for the solution of this problem. How- 
ever, the upper-air data from the Southern Hemisphere 
are still so sparse that no satisfactory hemispheric 
upper-level analyses can be made. Although the influ- 
ence of mountain barriers and other orographic effects 
6. The northeastern part of the North American continent 
(the Hudson Bay region) acts as a cold source in summer also; 
in this respect there is a remarkable difference between this 
region and eastern Siberia. 
605 
is not as great in the Southern Hemisphere, the con- 
tinents of Africa, Australia, and especially South 
America, and in addition the whole Antarctic, have a 
considerable disturbing imfluence upon the zonal air 
flow. 
The third type of disturbance, that caused by the 
supposed instability of the zonal current, has been the 
subject of a large number of theoretical studies in the 
last twenty-five years. The instability of “infinitely 
small” disturbances superimposed upon the westerly 
current has been the usual starting point in these in- 
vestigations. A study of these attempts to solve the 
cyclone problem, however, gives the impression that 
there still does not exist any theoretical solution fully 
applicable to the cyclone problem. Therefore the ques- 
tion arises whether it is, in principle, permissible to 
start from infinitely small perturbations in discussing 
the cyclone problem. If such an infinitely small per- 
turbation could cause the development of strong cy- 
clones, it would indicate that the atmosphere is ex- 
tremely unstable. How such an imstability associated 
with storing of useful potential energy could develop in 
an atmosphere where rather strong perturbations of all 
kinds are always present seems difficult to understand. 
If we consider this, it seems more likely that extra- 
tropical cyclones are induced by rather large migrating 
disturbances which were already in existence. These 
large disturbances must always be present. Since cy- 
clones and anticyclones are probably cells for trans- 
forming potential energy into kmetic energy, the pre- 
existence of a large amount of useful potential energy 
is necessary for a strong cyclonic development. How- 
ever, the pre-existing situation must correspond to some 
kind of potential instability that can be released only 
by the influence of finite perturbations.’ 
The available potential energy of the atmosphere is 
concentrated primarily in the solenoid field of the 
polar front. It is also a well-known fact that the polar- 
front region is the principal birthplace of extratropical 
cyclones. The question then arises whether a migrating 
disturbance would induce an irreversible process of the 
type observed in the development of cyclonic disturb- 
ances. The cyclone problem then would become more 
of a problem of the instability of certain large disturb- 
ances always observed on synoptic charts, whereas the 
ultimate cause of these large perturbations could be 
neglected. Similar ideas have been expressed earlier by 
many meteorologists (see, for example, von Ficker [23]). 
The 500-mb level has certain advantages for a study 
of such large disturbances because at that level the 
upper zonal wind is well developed and the polar-front 
zone is still not too diffuse. 
On circumpolar charts for the 500-mb surface, the 
belt of the strongest westerlies has the appearance of a 
“meandering river” situated, on the average, just to 
the south of, or in the zone of, the strongest horizontal 
temperature gradient. This zone has already been de- 
7. In his contribution to the Compendium, J. Bjerknes dis- 
cusses in some detail the interaction between migrating upper 
disturbances and polar-front cyclones. 
