SUNLIGHT AS A SOURCE OF RADIATION 



111 



3022 A was made in Cleveland in i<j;3() and 1937 by Luekiesh d <il. (1937, 

 1939). The results exhil)ited about the same variation with Z and the 

 same spread as the Washington data of Fig. 3-9 and therefore are not 

 plotted in Fig. 3-10. For wave lengths less than 3022 A, Eh was about 

 one-fourth the Eh for wave lengths less than 3200 A. 



For ultraviolet wave lengths less than 3130 A, Eh was observed during 

 1932 and 1933 by Ives and Gill (1937) in 14 cities scattered over the 

 United States. In Fig. 3-11 their results are given for two groups, where 

 group 1 refers to the most smoky localities and group III refers to the 



200 



50 



t<4 



60.8° 



30 60 90 



AIR MASS Af 



Z, degrees 

 Fig. 3-10. Values of Eh and E,, for wave lengths less than .3200 A. The symbols 

 indicate Ei, for some clear days in Washington during 1942 and 1943. The solid-line 

 curve is for Eh for higher latitudes in 1941. The £„ for clear days in Washington 

 1936-1941 was within the dotted curves. (Coblentz and Stair, 1944; (Coblentz el al., 

 1942.) 



least smoky localities; the data for clear and cloudy skies were plotted 

 separately. It is seen in Fig. 3-11 that the curves for the more smoky 

 localities lie below those for the less smoky both for clear and cloudy skies 

 and also that the curves for cloudy skies lie below those for clear skies. 

 Therefore both smoke and clouds decreased the amount of ultraviolet 

 radiation that reached the surface of the earth. Comparison of Fig. 

 3-10 with the "clear-sky" data of Fig. 3-11 shows good agreement, when 

 it is remembered that Eh for wave lengths less than 3200 A is about 2.5 

 times Eh for wave lengths less than 3130 A. 



The conclusion is therefore that, from Fig. 3-10, a rough estimate may 

 be made, correct perhaps within a factor of 2, of Eh in clear weather for all 

 seasons of the year and all times of the day. If a more e.xact value of 

 Eh is required, provision must be made to measure it. To make the 

 rough estimate, an average curve was drawn through the data of Fig. 



