depth at which 1 percent of the visible radiation 

 is found. A simple method for estimating this 

 depth, based on measurennents of blue-green 

 downwelling irradiance, is presented below. 



An examination of the transmission charac- 

 teristics of Type I, 11, and III ocean waters 

 (Jerlov, 1951; see also table 9) reveals that 

 it is possible to define the type of ocean water 

 on the basis of a single diffuse attenuation 

 coefficient, providing wavelength specificity 

 may be assigned to the coefficient. Further- 

 more, after the water type has been identified, 

 it becomes possible to determine the per- 

 centage of surface radiation at any depth in 

 the water column provided the attenuation 

 coefficient remains constant with depth (John- 

 son and Kullenberg, 1946). Jerlov (1951) pre- 

 sented such an analysis in graphic and tabular 

 forn-i based on attenuation coefficients ob- 

 tained at 25-m/j intervals. The percentage of 

 total incident solar energy was also given as a 

 function of depth by Jerlov (1951) for the three 

 ocean water types. 



In the present study, the Jerlov data (1951) 

 have been used, but the treatment differs from 

 that of Jerlov in two respects. First, the 

 percentage of visible (rather than total) down- 

 welling energy at various depths has been 

 computed, by use of the incident energy dis- 

 tribution of visible energy given by Moon 

 (1940) for air-mass-2 conditions (table 5). 

 Secondly, a family of curves has been con- 

 structed to yield five subtypes between Jerlov 

 Type I and II, and four subtypes between Type 

 II and III ocean water. As will be seen, these 

 additional subdivisions make the original clas - 

 sification somewhat more useful. 



The calculations carried out to derive this 

 family of curves are similar in principle to 

 those used by Jenkin (1937). The transmission 



values given by Jerlov, 1951, (see table 9) for 

 ocean water Types I, II, and III were inter- 

 polated to yield five intermediate water types 

 between Type I and II, and four types be- 

 tween Type II and III; a linear interpolation 

 was used at each 25-m^ interval between 400 

 and 675 m// . The percentages of radiation 

 remaining at 5, 10, 25, 50, 75, and 100 m. 

 were computed for each of these water types 

 from the interpolated transmission values, on 

 the assumption of homogeneous water and 

 constant energy between 400 and 675 m/i. This 

 percentage was in turn multiplied by the per- 

 centage of energy contained in each 25 mfi be- 

 tween 375 and 700 m^, as derived from the 

 air-mass-2 data of Moon, 1940 (see table 5). 

 The percentage remaining at each depth was 

 then obtained by summing the percentages at 

 each wave-length interval. No corrections 

 were made for losses at the surface. The 

 results of these calculations are given in table 

 10 and presented graphically in figure 3. 



To determine the percentage of downwelling 

 visible irradiance remaining at depth, it is 

 first necessary to be able to specifythe in situ 

 spectral sensitivity of the detector. A k value 

 derived from two or more irradiance meas- 

 urements may then be assigned to an appro- 

 priate peak sensitivity and band width. On the 

 basis of this information, a water type (in the 

 Jerlov sense) may then be selected for which 

 the transmission characteristics are approxi- 

 mately known and the percentage of down- 

 welling visible irradiance obtained as a func- 

 tion of depth. In table 10 the k values at 475 

 m/J (this is approximately the sensitivity of the 

 equipment used) in Types I, II, and III, and the 

 intermediate types, are given. A somewhat 

 more useful graphic presentation of this in- 

 formation is given in figure 3. 



Table 10. — Percentage of Incident visible downwelling irradiance computed at selected depths in 

 ocean water types for a zenith sun possessing the spectral distribution of air mass 2 and the 

 k value at 475 m/j 



15 



