NO. 



SOLAR RADIATION AND WEATHER — CLAYTON 



6i 



made three to five days in advance. A forecast of + .005 calorie or 

 more was considered a high value, + .004 to — .004 a normal value, 

 and —.005 calorie or below was considered a low value. 



The mean results for six months (May to October, 1924) are 

 given in table 20 : 



Table 20. — Mean Observed Values of Solar Radiation Following 

 Forecasts Five Days Ahead. 



Days before Forecasted Days after 



Forecasts Cases —2 — i 012 



Above normal 39 +.0008 +.0013 +.0024 +.0018 +.0004 



Normal 97 +.0003 +.0001 +.0004 — .0006 — .0003 



Below Normal 21 — .0008 — .0011 — .0038 — .0015 — .0016 



+.002 



+.001 



.000 



-.00 J 



''Zdays 



Fig. 44. — Mean observed solar radiation following forecasts 5 days ahead. 



These results are plotted in figure 44. They prove conclusively 

 that the solar radiation can be predicted with some success for five 

 days ahead from observations of visual phenomena on the sun. The 

 mean of the observed departures rises to a sharp maximum on the 

 day for which the forecast was made (zero-day). The mean 

 observed solar radiation following normal forecasts is seen to average 

 normal, and that following forecasts of low values is seen to fall to 

 a sharp minimum on the day for which the forecast was made. 



