SOLAR RADIATION AND WEATHER 



OR 



FORECASTING WEATHER FROM OBSERVATIONS OF THE SUN 



By H. H. CLAYTON 



On my return from Argentina in 1922, where in 1918 I had 

 initiated the making of weekly weather forecasts from solar data 

 combined with the ordinary meteorological observations at the 

 earth's surface, Dr. C. G. Abbot manifested a desire that I should 

 test the possibility of such forecasts for the L^nited States. The 

 cooperation financially and personally of Mr. John A. Roebling has 

 made possible the necessary researches and tests. 



In order to make forecasts for the United States, it was necessary 

 first to determine the meteorological sequences, if any, which follow 

 changes in the amount of solar radiation as observed at the astro- 

 physical observatories of the Smithsonian Institution. 



The first step in the investigation was to divide the observed solar 

 radiation values into grades. This was done for the interval July, 

 T918, to September, 1922, which included all the solar data available 

 at that time since the beginning of continuous observations in Chile. 

 The observations in Chile for the summer of 19 18 were supplemented 

 by observations at Mt. Wilson, and since October, 1920, by observa- 

 tions at Mt. Harqua Hala in Arizona. The grades were taken .010 

 calorie apart, beginning with the lowest measured values, around 

 1.860 calories per square centimeter per minute, and proceeding step 

 by step to the highest values, around 2.030 calories per square centi- 

 meter per minute. The observed frequencies of the different inten- 

 sities are given in table i and are plotted in figure i. The frequency 



Table 1. 



-Frequency of Occurrence of Different Intensities of Solar 

 Radiation, July, 1918 to September, 1922. 



Smithsonian Miscellaneous Collections, Vol. 77, No. 6 



