768 
method (see [1]). Fixing his attention upon the atmos- 
pheric fields within his area, the area forecaster predicts 
their form at any instant during his forecast period and 
thereby issues a more detailed and accurate area fore- 
cast, which describes in a semi-Hulerian way the local 
peculiarities of the weather and the time of arrival of 
the weather changes—the what and when of the weather. 
The preparation of the general forecast is primarily 
the employment of the more theoretical aspects of 
meteorology for sketching in broad terms the probable 
future meteorological development and a rather ideal- 
ized weather state associated with the development. 
The area forecast, on the other hand, applies the general 
forecast together with supplementary information so 
as to satisfy the requirements of its areas as to detailed 
predictions. The general forecast is a synthesis and a 
generalization, whereas the area forecast is a division 
and a specialization; they complement each other. 
The area forecast bulletin should be perspicuously 
expressed, especially so as take into account the extreme 
local variability in time and space of the weather 
phenomena. Although brevity is essential, the forecast 
should not be resolved into such simple but ambiguous 
terms as “clear,” “rain,” or “snow.” For example, a 
light cirrus overcast might be accepted as ‘‘clear” by 
the average person but would likely be regarded as 
“cloudy” by, say, a photographer. The ordinary area 
forecasts are regularly issued several times each day. 
Transmitted by radio, they are announced preferably 
by the forecaster himself (otherwise read verbatim by 
the radio announcer). Under the personal direction of 
the area forecaster, televised discussions of the map 
and other visual weather aids make possible a clearer 
presentation of the forecast than the verbal radio emis- 
sion, and a more rapid dissemination than through the 
newspapers. 
Further, the area forecast is particularized for each 
of several forecast districts into which the forecast area 
is subdivided. For delimiting the forecast districts on 
the basis only of weather similarity, it would be tempt- 
ing to apply a rational method such as Schaffer’s 
criterion [70], namely, the mean of the probable number 
of coincidences of rain days and of dry days for any two 
district stations. But at the present time the subdivision 
into districts is determined mostly by political reasons, 
so that the area forecaster must be at liberty to vary 
their extent whenever necessary. Although the district 
forecast bulletins are generally issued together in a 
prescribed order, it is sometimes advantageous to col- 
lect the districts into moving ‘‘zones” having practically 
the same weather (e.g., a zone parallel to the general 
direction of an approaching front) and to give the same 
forecast for the whole zone. 
The optimum extent of the forecast area and districts 
depends on its latitude and topography as well as the 
needs and activities of its nhabitants. To apply success- 
fully the supplementary data furnished by the special 
area network of unrepresentative observations and by 
the local method, the area forecaster must be familiar 
with the degree of unrepresentativeness of the different 
meteorological stations within his districts and must 
WEATHER FORECASTING 
possess an intimate knowledge of the local topographic 
factors affecting the weather in the vicinity of these 
stations. Then too, the forecasting work should not 
be unnecessarily complicated by having a forecast area 
so large as to be simultaneously influenced by different 
weather systems. 
The Local Forecast. Although the area forecast is 
still partly a general forecast, it tends considerably 
toward specialization in the operational interests of 
such area-wide activities as communication, transporta- 
tion, aviation, hydrology, and fire control, and especially 
of such district-wide activities as, say, fishing and 
planting. But if the area forecast were to be sufficiently 
detailed to meet the manifold needs of the populace of 
the area, 1t would be, especially in the United States, 
so voluminous and involved that the individual layman 
(e.g., editor, shipper, farmer) could not easily extract 
from it for his particular needs. To localize, detail, and 
amplify further the area forecast for a particular fore- 
cast locality, and especially to specialize it for the benefit 
of various local endeavors, is instead the task of a 
locally residmg forecaster, who, so to speak, acts as 
the local agent or representative of the weather service, 
supplying the local weather consumers with what they 
really need. 
Every six hours the local forecast office receives from 
its forecast center the area forecast—but not the area 
map analysis or prognosis—which is then further de- 
tailed by the local forecaster as to arrival time of local 
weather phenomena (e.g., showers, etc.) and is also 
amplified and specialized for the needs of the various 
and particular endeavors of his locality. The Eulerian 
considerations are thus, to a lesser extent, applied also 
by the local forecaster. Using mainly the district forecast 
of the area forecast center as guidance material, the 
local forecaster formulates his many local and special 
forecasts through an intimate knowledge of the weather 
peculiarities and activities of his locality. To some ex- 
tent he also utilizes various supplementary, detailed, 
and reliable local observations that he himself can make 
up to the very moment of issuing the local forecasts 
when the area forecast may be up to three hours old. 
As a matter of fact, the local forecast office generally 
takes part in an observing program contributing to the 
area and/or international network, so that at least 
a part of these observations can be applied toward his 
local forecast problems. Although he does not prepare 
maps, he may often apply specialized techniques of 
analyzing and prognosticating locally the various 
weather elements, such as his locally and empirically 
developed diagrams for predicting minimum and maxi- 
mum temperatures. Finally, the local forecaster must 
have, in addition to his intimate knowledge of the 
local topography as well as his in-station study of local 
forecasting problems [1], a thorough acquaintance with 
the various weather needs of his community and must 
be ready to supply special forecasts regularly and when- 
ever called upon. 
But the local forecaster himself cannot provide all 
the information wanted for special purposes. Therefore, 
in addition to the area forecast centers, special fore- 
