90 



80 



w 70 



60 



50 



40 



30 



Full sunlight : 

 5,000 to 13.000ft -c 

 X 33ft-c 



Clear sky 



Sun appeared to ^^ 

 set (upper limb \ 

 on the horizon) 



20-- 



10 



"T" 



The end of 

 civil twilight 

 {0-4ft-c) 



— 0-4ft-c 



33ft-c 



1000 



10-0 



1-0 



Equal to the] 

 brightness 

 of the full 

 I moon at 

 the zenith J-q-i 



0-01 



0-001 



H h 



-4 



-7 -8 



•10 -II 



TRUE ELEVATION OF THE SUN IN RELATION TO THE HORIZON 



I \ \ \ 1 \ \ 



10 20 30 40 50 60 



MINUTES AFTER SUNSET {Summer Solstice, 40° latitude) 



Figure 3-4 

 Light intensity during twilight as a function of the sun's elevation in 

 relation to the true horizon. Clear sky curves were redrawn according 

 to the curves of H. H. Kimball, 1938, Monthly Weather Review 66: 

 279-286. Kimball's curves were drawn by simple estimation through 

 points taken on a number of occasions during 1913 and 1916, and as 

 might be expected there is considerable scatter in the points. Thus the 

 curves are at best only examples of what might occur on any given clear 

 day. The duration of twihght will depend upon latitude and time of year, 

 and the times shown are for 40° latitude on June 21, the summer solstice. 

 The time shown as 50 min would be about 40 min on the spring or fall 

 equinoxes and 45 min on the winter solstice. At 30° latitude on the 

 summer solstice the point indicated as 50 min would be about 41 min, 

 and at 50° latitude the point would be about 69 min. The cloudy day 

 curve is a typical one taken from measurements of A. Takimoto and 

 K. Ikeda, 1960, Bot. Mag. Tokyo 73, 175-181. Their low intensity data 

 are not accurate enough to allow plotting on the logarithmic scale, but 

 near the end of civil twilight light intensities are about the same on clear 

 and cloudy days. 



