760 
persistency of wet and dry weather has been given by 
Newnham [39] for Kew Observatory in Hngland and 
Aberdeen in Scotland, as shown in Table V. 
The forecaster should prepare a skeleton prognostic 
chart for each 24-hr period after the first 24-hr period, 
TABLE V. PROBABILITY OF A Rain Day 
Number of preceding successive fine days 
Station 
1 2 3) 4 5 6 7 8 9 
Aberdeen.......... 0. 50/0. 41/0. 37/0. 39/0. 37|0.36)0.37) — | — 
Kew..............|0.45}0.34)0.36)0.32\0. 260. 27|0.27|0.22) — 
Number of preceding successive rain days 
Station 
1 2 3 4 5 6 7 8 9 
Aberdeen.......... 0.67\0. 700. 76\0. 70\0. 66)0. 67\0. 72|0. 75)|0. 78 
Kew............../0.56/0.61/0.61/0. 64/0. 64/0. 67/0. 73)0. 69/0. 70 
if one is not available from the weather central. The 
actual timing of frontal passages, wave development, 
precipitation surges and similar phenomena becomes 
mcreasingly difficult with time and, consequently, the 
forecast should be couched in more general terms. 
Usefulness of Various Charts and Techniques. An 
analysis by Elhott, Olson, and Strauss [21] of question- 
Tasie VI. USEFULNESS OF CHARTS AND TECHNIQUES 
Often Occa- 
useful sionally 
Charts 
OO-mibychianty.: setae tees aerate cea cacpeier your 95.3 | 3.0 
SOOzmonchar ies Nets sec tees tete so ote aes 57.5 | 26.2 
S50=mbicharty Sau are ontae here ce eeeee 37* — 
S002mbY chartietays svoctpaskuactessicrig tide pe onion 22* = 
ZOQsmio ech ants cow tenses Garechow eae evseer ce seekeeatere 16* — 
@rossisections ee Fee. A een Aen cate 22.8 | 21.0 
AMONGST ClOBWHIS 5 noc oo ou sgnsoconansccenacval) 1O).4 | @.6 
Mean virtual temperature charts...........| 5.0] 4.2 
Tsenitropicrchantee. remnants Di 4.9 
AGH OTE CHOPTENIME), 5 ooccoscnooecacveseescoce 92.0 | 4.0 
Ma Paciiapramsy sees eels «roche webs eS or eee |e 
IRON? CUMMINS, . 255 oonccccscoccocsuococos 8.6 | 13.7 
Otherichantsrondiderdms eee eee asa oul i 
Techniques 
Petterssen’s kinematic technique............. 10.3 | 26.4 
Steering of surface system along upper-air flow| 76.2 | 19.1 
Extrapolation based upon history............ 81.0 | 15.1 
Frontal movement from gradient winds...... 61.6 | 31.3 
Upper-level isobar-isotherm relationship for 
TANOWVEWNEINI OH WAVES. osaccondcanecccsas 50000 37.6 | 36.1 
Upper-level isobar-isotherm relationship for 
deepening or filling........................ 53.5 | 29.4 
Average cyclone and anticyclone tracks and 
MOVIE DUS se: cin cries Sen ee eee 32.3 | 42.0 
Analog irestate Sivas. atlas sem rre eee En eee 4.7 | 17.5 
Weather itypesi ised Ho! casement 23.6 | 31.1 
Rossby’s constant-vorticity trajectory.......| 2* 7.9 
Othertechniquest2 ee cn ee ee ee 21.5 | — 
* Approximate per cent. 
naires on the usefulness of certain charts and forecasting 
techniques returned by some 1460 forecasters in the 
military, civilian, and private weather services indi- 
cated the percentages of forecasters finding the several 
charts and techniques “‘often useful” or ‘occasionally 
useful” as given in Table VI. One may safely say that 
WEATHER FORECASTING 
charts and techniques only ‘occasionally useful’ are 
not regularly used in the preparation of forecasts. The 
question may be asked, Are forecasters using all the 
most valuable available techniques? If not, is it because 
of lack of clerical assistance, lack of time or lack of 
knowledge of how to use them, or do the few simple 
extrapolation and stability estimation techniques pro- 
vide the same accuracy as additional checks using more 
complicated techniques? 
Prognosis of the Weather. The forecasting of the 
various weather elements is the last step in the prep- 
aration of the weather forecast but it is the most 
important and the most difficult. An understanding of 
the current weather, together with the surface and 
upper-air prognostic pressure patterns, forms in a gen- 
eral way the basis of weather forecasting. Tests have 
shown that forecasters demonstrate but little skill in 
forecasting the various weather elements when given a 
“nerfect” prognostic pressure pattern alone.* One must 
again emphasize the individuality of each day’s synoptic 
chart with the infinite variations of lapse rates, tempera- 
ture and moisture, convergence, radiation, and all the 
other factors involved in future weather. 
Cloudiness and Precipitation. Following the develop- 
ment of the concept of air-mass analysis by the Nor- 
wegian meteorologists, J. Bjerknes [4] in 1918 first 
described the arrangement of clouds and precipitation 
around a cyclone. The classical arrangement of hydro- 
meteors around an idealized cyclone was described by 
Bjerknes and Solberg [5] in 1921 and was extended in 
1922 [6] to include all stages of cyclone development. 
Since the ideal cyclone is rarely present on the daily 
weather chart, the marked variations in stability and 
moisture in the air masses normally found around the 
cyclone are of paramount interest to the forecaster. 
These elements frequently possess little relationship to 
the isobaric pattern. In many areas, the forecaster is 
faced with the complicated problem of determiming 
where, as well as when, precipitation will develop in 
various sectors of a cyclone. No very satisfactory tech- 
niques are available for forecasting the transition of a 
dry to a wet trough other than by a rather subjective 
evaluation of moisture and stability factors. 
In summer, particularly, the precipitation pattern is 
even more imperfectly associated with frontal and cy- 
clonic systems. Pure air-mass thunderstorms may be 
forecast with fair success from the thermodynamic 
diagrams, but the general shower and thunderstorm 
activity, often nocturnal in nature, over large areas in 
middle latitudes, is very difficult to handle. Means [83] 
has suggested techniques for forecasting nocturnal 
thundershowers in the Midwest which show a slight 
increase in skill over more subjective methods, but the 
mechanics of the nocturnal thunderstorm in the Great 
Plains and upper Mississippi Valley are still imperfectly 
3. It is not known whether, in this experiment, forecasters 
had access to the preceding weather charts. In any case it is 
not intended to imply that the prognostic pressure chart is not 
useful. Indeed, a good prognostic pressure chart is one of the 
most useful tools available to the forecaster. 
