SHORT-RANGE WEATHER FORECASTING 
at a critical zone in middle latitudes, has been suggested 
by the Chicago group, and the “‘confluence” theory by 
Namias and Clapp [88], as possible explanations of the 
mechanism of this current. Which are the causes and 
which are the effects have not been definitely estab- 
lished. Riehl [45] has suggested that the ascent of tropo- 
spheric air below and slightly to the north of the jet 
might affect the distribution of precipitation. Starrett 
[52], mvestigating this relationship, found a significant 
concentration of precipitation activity under the jet 
stream, subject to the normal variations of local dynam- 
ical fields of convergence and divergence associated 
with the short baroclinic waves within the westerlies. 
Tt seems probable that the jet stream and George’s 
“Ssotherm ribbon” [26] are all manifestations of the 
same phenomenon, namely, the frontal zone, and that 
they are usually located along the same latitude at 
any one time and that this zone is favorable for cyclo- 
genesis. 
Pilot-Balloon and Other Upper-Air Charts. The prin- 
cipal use of the 6-hr pilot-balloon charts is obviously 
the forecasting of upper-air winds and, to a much lesser 
extent, surface winds. The interim charts between the 
regular constant-pressure charts are useful in checking 
the movement of troughs and ridges and the latest 
trends. 
Tsentropic analysis was introduced around 1937 by 
Rossby and collaborators [48]. Isentropic charts were 
prepared generally by forecast centers for a number of 
years in accordance with procedures outlined by Namias 
(see Petterssen [42, Chap. VIII]), but were eventually 
abandoned when forecasters failed to find in them the 
hoped-for precise tool for precipitation forecasting. The 
poor results have been blamed on the nonadiabatic proc- 
esses common in the atmosphere, which tend to destroy 
the conservatism of isentropic surfaces, principally (1) 
radiational cooling and heating, (2) evaporation and 
condensation, and (3) convection: Also one thin, and 
often wavy, isentropic surface, arbitrarily selected, may 
be unrepresentative of the principal upslope area. In 
the Middle West, determination of the motion of air 
particles, some distance away from but in line of flow 
toward a station, relative to the contour lines of the 
isentropic surface over the station, was frequently very 
difficult. Contours might move in the same direction 
and at the same rate as the air particles and the ex- 
pected upslope movement would not occur, with resul- 
tant failure in the precipitation forecast. There is some 
question whether isentropic analysis has been given a 
fair trial by forecasters, since (1) time was rarely avail- 
able for careful construction and analysis, and (2) many 
forecasters never fully understood the meaning and 
Significance of all the information on the isentropic 
chart. 
Clouds. With the considerable increase in the amount 
of upper-air information now available to the fore- 
caster, the importance of cloud data has decreased 
somewhat. However, as Brooks [10] states: “Clouds 
are indications of humidity, lapse-rate, (and temper- 
ature), direction and velocity (including shear and 
turbulence) in the free air. As such, they are, of course, 
753 
valuable aids in air-mass analysis and forecasting.” 
Thus, as indirect aerology, clouds provide the local 
forecaster particularly, and the district forecaster and 
analyst as well, with valuable information regarding 
the stability and moisture trends of the lower atmos- 
phere. A close watch on the clouds as shown on the 
hourly sequences and interim charts will provide the 
forecaster with valuable clues on the rate of moistening 
of a previously dry trough. 
PREPARATION OF PROGNOSTIC CHARTS 
It is desirable to prepare, in a formal fashion, prog- 
nostic charts for 24-hr mtervals. These charts should 
include isobars, frontal positions, high and low centers, 
and precipitation areas. The positions of fronts and 
pressure systems can be indicated informally on the 
regular six-hourly synoptic chart for six or twelve 
hours or for any other time interval. Upper-air prog- 
nostic charts should include contours and ridge and 
trough lines. 
Preparation of the 700-mb Prognostic Chart. Because 
of their complexity and the time involved, the prepa- 
ration of upper-air prognostic charts is logically a fune- 
tion of weather centrals and not of forecast centers. 
Techniques have been suggested by Starr [51] but a 
description of those in use by American meteorologists 
has not, as yet, been published for general distribution. 
However, the principal procedure appears to be extra- 
polation with a check for consistency by drawing the 
implied mean virtual temperature between the 1000- 
and 700-mb surfaces. The forecasting of mean virtual 
temperature and mean density patterns should provide 
a better indication of future’ height changes than pure 
extrapolation of changes at one level. Other techniques 
which might be used include: 
1. Extrapolation of trough and ridge lines. The pat- 
tern at 700-mb is more conservative than at the sur- 
face and conservatism increases further with height. 
2. Extrapolation of isallobaric centers, applying cor- 
rections for indicated intensification or diminution. 
3. Application of the formula for the movement of 
long waves in the westerlies, developed by Rossby [47]. 
This formula is c = U — 6L?/4q?, in which c is the 
speed of the wave, U is the zonal wind speed (best 
results are apparently obtained at approximately 600 
mb), L is the wave length, and @ is the rate of change 
of the Coriolis parameter northward with latitude. 
Recent studies by Cressman [17, 18] indicate that good 
to excellent results may be obtained if adequate daily 
700-mb charts are available on a three-quarters or full 
hemispheric scale. Use of this formula is not practicable 
for an area as small as North America. 
4. Isotherm-isobar relationship. Another principle 
based on the kinematics of wave motion provides some 
indication of the rate of movement of minor waves in 
the westerlies. Assuming that there is no convergence 
or divergence and that temperature is a conservative 
element, Rossby [46] derived the equation U/(U — c) 
= Ar/Ap in which U and c have the same significance 
as before, Ar is the amplitude of the isotherms, and 
A,» is the amplitude of the isobars. It can readily be 
