A PROCEDURE OF SHORT-RANGE WEATHER FORECASTING 
By ROBERT C. BUNDGAARD 
Air Weather Service, Washington, D. C. 
INTRODUCTION 
The practical procedure of short-range weather fore- 
casting is simply the extrapolation, in a partly geo- 
metrical way, of sequential weather situations. On the 
other hand, the ideal mathematical prognosis is based 
completely upon physical considerations and uses only 
a single set of initial data. Since practical weather fore- 
casting nevertheless aims at accomplishing the same 
effect as integrating the equations of atmospheric dy- 
namics and thermodynamics, it too must take into 
account the physical relationships existing among the 
various meteorological variables, such as wind, pres- 
sure, and temperature. The extrapolation process there- 
fore must also take physical considerations into ac- 
count. 
In this article we shall first consider certain general 
principles of practical weather forecasting. In order 
to make 24-hr to 36-hr forecasts, the complex analytic- 
prognostic process (leading from the observational data 
to the prognostic maps) must be centralized so as to 
permit the beneficial use of an operational procedure 
known as the synoptic cycle. By systematic use of the 
synoptic cycle the analysis as well as the construction 
of the prognostic maps is facilitated, since the prog- 
nostic maps form the first approximation to the analysis 
of the next set of weather maps. On the other hand, the 
prognostic-forecasting process (leading from the prog- 
nostic maps to the explicit forecasts) is decentralized 
so that the predictions of the different weather ele- 
ments can be formulated locally. Such predictions can 
then be synthesized into explicit weather forecasts, 
such as general forecasts giving summary predictions 
for a large region, area forecasts giving more detailed 
local information, and special forecasts for various fields 
of human endeavor. 
Although the involved procedure of weather fore- 
casting should be undertaken systematically, it must 
necessarily vary with the prognostic region and weather 
situation under consideration. Still we have, partly for 
expository reasons, introduced a certain order of seven 
prognostic operations which can be recommended, in 
many cases at least, for use in temperate latitudes. This 
order serves to remind the forecaster of the many 
different rules, of a theoretical or practical character, 
which he should use. We shall see how the prognosis 
of the surface map can be based on geometrical extra- 
polation together with certain physical considerations. 
The prognosis of the upper-air maps will then be pre- 
sented (geometrical extrapolation and physical con- 
siderations). Finally, we shall recapitulate the more 
important considerations to be borne in mind by the 
. forecaster in predicting the various weather constitu- 
ents: temperature; general hydrometeors; fog, strati- 
form clouds, and drizzle; cumuliform clouds, showers, 
thunderstorms, and hail; wind, turbulence, and tor- 
nadoes; and aircraft icing. 
This article is essentially an epitome based primarily 
upon certain chapters of Dynamic Meteorology and 
Weather Forecasting, soon to appear as a publication 
of the Carnegie Institution. Most sincere thanks go 
to Professor C. L. Godske, who, as the principal author 
of the book, has so very generously and warmly ap- 
proved the use of the book in this article. Some of the 
subject matter presented in this article has arisen 
through an intimate collaboration among C. L. Godske, 
T. Bergeron, and R. C. Bundgaard. Some new material 
has been introduced by the author, partly for the 
purpose of taking into account the scope and objec- 
tives of this Compendium. The author also wishes to 
acknowledge his deep indebtedness to Professors T. 
Bergeron, J. Bjerknes, and C. L. Godske for the great, 
inspiration he has received through being associated 
with them. 
THE GENERAL PRINCIPLES OF PRACTICAL 
WEATHER FORECASTING 
The Synoptic Cycle. From just @ single initial state 
of the atmosphere, its subsequent development can be 
mathematico-physically predicted by means of a certain 
system of differential equations and boundary condi- 
tions. Practical weather forecasting, on the other hand, 
is based upon prognostic maps, first sketched by geo- 
metrically extrapolating (in a purely formal way) a 
sequence of completely analyzed weather maps, and 
then improved and amplified by considerations based 
on experience and/or general physical principles. The 
forecaster should, therefore, first survey carefully the 
analyzed map sequences, to amend all recognized errors 
and to add the finishing touch to the maps if time 
limitations had forced him to leave them in an unsatis- 
factory condition. Through the complete, detailed, and 
correct analysis of the maps, dynamical and thermo- 
dynamical considerations can indirectly be taken into 
account in the preparation of the prognostic maps pre- 
liminarily based only upon pure extrapolation. Owing 
to the importance of map analysis as the foundation for 
the prognosis, we have introduced a synoptic cycle 
involving the three basic operations of ‘“‘epignosis,” diag- 
nosis, and prognosis. 
Let us first consider the synoptic cycle as it presents 
itself in connection with the upper-air maps. The ob- 
servations at time ¢ = t are plotted on the prognostic 
upper-air maps referring to the time ft and prepared 
at to) — 24"; we shall refer to these as maps I. These 
maps represent a first approximation to the analysis, 
which, like that of the surface map, has the form of 
766 
