SUNLIGHT AS A SOURCE OF RADIATION 



109 



let radiation falling on the earth's surface far less than it reduces the 

 effectiveness of the direct rays of the sun. The reason is that the light 

 scattered out of the direct solar 



mm Oi 



rays by haze is not entirely lost, 

 since haze exerts little absorption, 

 but reappears as sk}^ light, which 

 for thin haze is mainly directed 

 downward. 



A fraction of the sky light is 

 scattered outward to space and is 

 lost to the earth; the fraction in- 

 creases with increasing haze and 

 clouds. As has been mentioned, an 

 exact calculation of all this is com- 

 plicated and would require a com- 

 plete theory of sky brightness and 

 polarization for ultraviolet wave 

 lengths in terms of the ultraviolet 

 optical constants of the atmosphere 

 in all stages of haziness. Such a 

 theory has not been formulated, and 

 such constants have not been determined; therefore only a survey of the 

 observational material is presented in the following section. 



o 

 o 



V 



^ 



I mm O3 ^^ 



+ "FAIRLY CLEAR" ^ 



J I I I 1 L 



30 60 



Z, deqrees 

 Fig. 3-8. Calculated solar ultraviolet 

 energy E for wave length.s loss than 

 3200 A for several cases. 



OBSERVED ULTRAVIOLET INTENSITY AT THE EARTH'S SURFACE 



Many measurements of ultraviolet radiation from the sun and sky were 

 carried out at Washington, D.C., by Coblentz and Stair (1943). The 

 first series of measurements were made with photocells arranged to record 

 En, the radiation received on a plane normal to the rays of the sun, from 

 the sun and a circular region of sky around the sun as a center 22° in 

 diameter. Data were taken during the clearest days over the years 

 1936 to 1941; for illustration the values of En for 1937 are shown in Fig. 

 3-9. Two types of photocells were used, one sensitive to wave lengths 

 less than 3200 A and the other sensitive to wave lengths less than 3132 A; 

 the spectral-sensitivity curves of the photocells were not reported. In 

 Fig. 3-9 the abscissas are the air mass M and the zenith angle of the sun 

 Z, and the two scales of ordinates are the En in microwatts per square 

 centimeter for the respective types of photocells. It is seen that they 

 were approximately proportional and that the ultraviolet intensities of 

 wave lengths below 3200 A were roughly 2.5 times the intensities below 

 3132 A. The data of the other years from 193() 1941 were similar to 

 those of Fig. 3-9 for 1937. For any zenith angle the variation of the 

 ultraviolet intensitv of a factor of about 2, shown in Fig. 3-7, was proba- 



