NO. 10 SOLAR RADIATION AND WEATHER STUDIES ABBOT 



5 



I and 2. Their sources of error, corrections to their direct readings, 

 and other details regarding them are published in volumes 2, 4, and 5 

 of the Annals of the Smithsonian Astrophysical Observatory and in 

 papers nos. 3182 and 3288 of the Smithsonian Miscellaneous Collec- 

 tions. Intercomparisons of silver-disk pyrheliometers made at inter- 

 vals over a period of about 20 years indicate that the scale of ob- 

 serving has not changed appreciably. These intercomparisons are 



Diagram of water-flow pyrheliometer. 



Solar rays are mainly absorbed on the cone in A, but some are scattered about 

 the walls of AA. Their heat is given up to water which flows in a spiral channel 

 about the cone and tube AA. The rise of temperature of the water due to solar 

 heating is measured by the electrical thermometer D1D2. Test quantities of 

 electrical heat introduced at G or H may be measured as a check. 



published extensively in the Annals, volume 4, pages 94-97, and vol- 

 ume 5, pages 139-145. Table i gives one typical example. 



Table i. — Long-continued Series of Intcrconiparison of PyrJxcliometcrs 



S.I. I ivith A.P.O. S, 



Year igii 



Ratio 1.0357 



1911 

 1.0246 



; 1913 191S 



8 1.0324 1.0343 



» It is believed that owing to maladjustment S.I. i 

 occasion. 



1916 " 

 1.0119 



1917 

 1.0360 



1917 

 1-0330 



1920 

 1-0352 



was not properly exposed on this 



The distribution of energy in the solar spectrum before it enters 

 the atmosphere approximates roughly that of the perfect radiator at 

 6,000° K. Hence, nearly all of its energy is contained between wave 

 lengths 0.3 and 3.0 microns. Rays beyond 0.3 micron in the ultraviolet 

 are almost wholly cut ofT by ozone in the higher atmosphere, and those 

 beyond 3.0 microns in the infrared by water vapor in the lower at- 

 mosphere. Between these limits not only these and other atmospheric 

 vapors, but also dust and even the permanent gaseous molecules of 

 the air, absorb or scatter the sun's rays both selectively and generally, 

 so that the solar beam is both changed in spectral distribution and 

 generally weakened during its passage through the atmosphere. In 

 order to evaluate these losses, energy spectral measurements are re- 



