presence of swell. Nevertheless Figure 34 suggests a linear relationship between the £"s 

 determined by two independent methods. It is concluded that the wind data may be used to 

 determine the sea state, at least qualitatively. 



Figure 35 suggests a linear relation between the characteristic wave heights h, as de- 

 termined by the method of References 5 and 6, and the square root of E, except for very se- 

 vere sea conditions. 



Figure 3 6 would be expected to indicate a linear relationship between the significant 

 wave height and the visual shipboard estimates of the characteristic wave height since the 

 latter is presumably proportional to E in accordance with the indications of Figures 34 and 

 35. Figure 36 does not contradict such a linear relationship. The scatter of values is most 

 likely due to errors in the determination of the significant wave height, inasmuch as Figure 6 

 shows that the visual shipboard estimates are reasonably correct. The computed value of 

 the significant wave height, Item (d), is very much a function of the cutoff length. The 

 UNIMAK stereophotographs did not furnish sufficient data, in the opinion of the Hydrographic 

 Office oceanographers, to permit an evaluation of the significant wave height; and therefore 

 these data were not available for the plots. 



From an overall point of view, consideration of Figures 6, 34, 35, and 36 suggests that 



1. The methods of Reference 8 may be applied to make a rough estimate of wind waves. 



2. Trained observers can, on the average, make reasonably accurate observations of the 

 heights of the larger, well-formed waves that are present in a given sea. 



3. The characteristic wave height reported by trained observers is proportional* to the 

 square root of the statistic E, corresponding to the sea state considered, except for severe 

 sea states. 



4. The so-called "significant" wave height is not particularly significant since it is 

 difficult to compute, although it is statistically well defined. The average height of the pre- 

 dominant wave heights,** as reported by observers, is physically more meaningful and is 

 more easily reproduced on repetitive estimates than is an estimate of the significant wave 

 height. 



*The empirical relationship between the characteristic wave height h reported by the observers and the statistic 

 y/E, for the data plotted in Figures 34 and 35, is approximately as follows: 



h^il.53 \fE when E is derived from wind data, and 

 A^ 1.88 y/E when E is derived from the wave data. 

 **Here designated by the term "significant wave height." 



46 



