IRocbliUG 3fun& 



ON THE CORRECTIONS TO BE APPLIED TO SILVER- 

 DISK PYRHELIOMETRY 



By C. G. abbot 

 Secretary, Smithsonian Institution 



In the Zeitschrift fur Meteorologie for October 1936, Fuessncr has 

 presented evidence which convinced him that the correction factors 

 K and K' given in association with silver-disk pyrheHometers are in- 

 appHcable, and better omitted. If this were true, it would overthrow 

 the results of 25 years of observing, as published in the Annals of 

 the Astrophysical Observatory, and in my recent papers on the 

 dependence of weather on variations of solar radiation/ 

 The following facts contradict Fuessner's conclusion : 

 I. If the corrections had been omitted in reducing pyrheliometry 

 at Montezuma, Chile, with silver-disk pyrheliometers S. I. 30 and 

 S. I. 31, certain notable changes would have occurred in the solar- 

 constant values reported in table 31, volume 5, Annals of the Astro- 

 physical Observatory. To fix ideas I will first refer to these Annals, 

 volume 5, pages 69, 211, and 213. From these references we note 

 that in the months of August and November, 1929 and 1930, the 

 average temperature of the air at 8 o'clock a. m. (which can have 

 differed but little from that at which the pyrheliometers were read) 

 was higher by 15°! F. in 1929 and by i2?o F. in 1930 in November 

 than in August. Transforming these differences to Centigrade and 

 performing the appropriate computations, it appears that the omission 

 of the correction factors would have produced a change in the relative 

 values of the mean solar-constant numbers for these two months of 

 approximately 0.019 calorie in 1929, and 0.016 calorie in 1930. The 

 actual monthly means of the solar-constant values as given on pages 

 211 and 213 for Montezuma observations were 1.931 and 1.936 in 

 the year 1929, and 1.945 and 1.944 in the year 1930. To speak ap- 

 proximately, the published difference of only 0.005 calorie in 1929 

 would have become 0.024 calorie, and the published difference of only 

 o.ooi calorie in 1930 would have become 0.015 calorie if the silver- 

 disk pyrheliometer corrections had been omitted. 



^Smithsonian Miscellaneous Collections, vol. 94, no. 10. 1935: vol. 95, nos. 12, 

 15, 19, 193^1- 



Smithsonian Miscellaneous Collections, Vol. 95, No. 23 



