756 
in a field where the contour pattern is expected to move slowly 
with respect to the winds, so that the strong winds will soon 
reach that part of the field where the gradient is much weaker 
than would be necessary to balance the velocity of the wind. 
In these cases the winds will tend to turn to the right because 
the Coriolis force exceeds the pressure-gradient force, and air 
will cross the contours toward higher values with a conse- 
quent decrease in speed. Usually the speed decreases more 
than enough to balance the gradient because the momentum 
of the parcel carries it beyond the point of equilibrium so 
that the parcel is turned to flow toward lower heights with 
an increase in speed. This cyclonic trajectory has the effect 
of forming a new trough or of deepening an already existing 
trough and intensifying the adjoining ridge; it also causes 
stationary or even retrograde conditions to prevail in that 
area of the map. 
The use of the jet stream im the prognosis is very quali- 
tative. Although the jet is at a much higher level, its effects 
are apparent in the stronger flow induced at 700 mb. Thus 
from a consideration of the confluence theory of jet-stream 
formation, it is apparent that any system moving into the 
vicinity of the strong flow will accelerate and usually will not 
deepen. Thus the effects of these storms on the 700-mb con- 
tours are limited to their contributions to the confluence and 
the speed of the jet. Also any storm moving in the vicinity 
of the jet will not cross it but will tend to move with it. 
The 24-hr height changes at 700 mb are used in much the 
same way as sea-level pressure changes. Here again, however, 
the interrelation between the 700-mb and the sea-level 
charts must be studied in order to determine some criterion 
for using the height changes rationally. Any conservative 
feature common to both charts, whether of movement or of 
configuration, is helpful in deciding how much and where to 
apply the changes. 
Along with the dominant role played by the surface anal- 
ysis and the surface prognosis in the making of the 700-mb 
analysis and prognosis, it is important to note that the thick- 
ness chart constitutes an important check on the consistency 
of the prognoses. (The 1000-mb prognosis is quickly made by 
direct extrapolation from the sea-level prognosis.) Thus in 
the absence of strong vertical motions over the prognosticated 
area, the temperature field should change in the direction and 
with the speed of the flow in the ‘‘thickness” layer. The thick- 
-ness lines are usually more conservative than the contours 
or than the sea-level isobars. 
Lennahan concludes that it would certainly be pleas- 
ing to all if some good theoretical or objective method 
could be developed for preparation of the prognoses. 
However, for the present at least, it seems that we must 
be satisfied with our ‘‘cut and try’? methods for short- 
term forecasts. In a review of Scherhag’s new book,” 
E. Hovmoller remarks: “‘A theoretical meteorologist 
would probably, after reading this section [G of first 
part], feel discouraged once more by the contrast be- 
tween the magnificent building of theory itself and the 
modest, almost crippled part of it which is thought to 
be applicable in the weather service.” It is unfortunate 
that this is also equally true of so much of the recently 
developed meteorological theory. 
Although, quantitatively, the 700-mb techniques ap- 
pear to be equal to those available for the surface 
2. Scherhag, R., Newe Methoden der Wetteranalyse und Wet- 
terprognose. Berlin, Springer, 1948. Reviewed in Tellus, Vol. 1, 
No. 4, pp. 70-74 (1949). 
WEATHER FORECASTING 
prognostic charts, and the higher-level chart is some- 
what more conservative, most forecasters believe that 
the 700-mb prognostic charts are not as satisfactory; 
however, this belief is difficult to substantiate. The 
principal weaknesses are the forecasting of the develop- 
ment of closed lows and their future movement, the 
movement and changes in intensity of minor troughs, 
and the emergence of major troughs from mean troughs. - 
The reasons for these inadequacies may be (1) non- 
utilization of all techniques, (2) lack of experience, (3) 
preoccupation with surface prognostic charts and with 
mean troughs and ridges, and (4) lack of satisfactory 
techniques for dealing with closed circulations. There 
is also considerable evidence of confusion in defining 
major, mean, and minor troughs. When does a minor 
trough beceme a major trough? The consideration of 
major troughs and ridges as synonymous with mean 
troughs and ridges by many forecasters and prognosti- 
cators is believed to be erroneous. There appear to be 
certain large-scale controls, not very well understood, 
which during the persistence of any given regime tend 
to result m trough (ridge) formation or intensification 
(weakening) repeatedly at some one geographical loca- 
tion. So long as the regime persists—and it may change 
suddenly—pressure falls (rises) intensify (weaken) as 
they approach the region of the mean trough (ridge). 
The result is a mean trough or ridge in the favorable 
location, but troughs or ridges may move out of the 
mean positions and, apparently depending upon the 
wave length, become migratory major troughs or ridges. 
The crux of the forecasting problem in such cases is 
whether troughs and ridges emerging from the mean 
positions will attain major intensity or remam minor 
waves. 
Recent work by Charney [18, 14] and collaborators 
gives some promise that the new electronic computing 
devices will eventually provide assistance in prognosti- 
cating constant-pressure surfaces. 
Preparation of the Surface Prognostic Chart. In the 
preparation of the surface prognostic chart, the pressure 
field is given primary consideration smce an accurate 
pressure prognosis is the synthesis of all computations 
by, and experience of, the forecaster. The customary 
procedures used in the preparation of this chart include: 
1. Determination of the movement of fronts and 
pressure systems. The six-hourly positions of significant 
fronts and highs and lows are plotted for the past 12 to 
24 hr or more, and changes in direction and rate of 
movement are carefully noted. As a starting point, 
highs and lows might be typed in accordance with, 
and the 24-hr average direction and rate of movement 
ascertained from, some previous statistical study such 
as that of Bowie and Weightman [7] for the United 
States. From the previous history of the system, an 
immediate deduction can be made whether it is opera- 
ting under the average steering indicated by the study. 
A second position may be obtained by modified extra- 
polation, sometimes called the “path” method. The 
six-hourly projected positions should be further modi- 
fied in accordance with changes indicated by the general 
synoptic situation. Another predicted frontal position 
