Care must be taken in evaluating such observed data since often 

 estimates are not too accurate and conditions often do not permit accurate 

 observations „ For example, it is quite difficult to estimate wave heights 

 from a moving point of observation o On the other hand, it is quite 

 possible to make accurate estimates from an anchored lightship o On the 

 basis of this, observations of lightship captains are considered quite 

 accurate o Since not all seagoing personnel are equally capable in this 

 direction, a careful selection is necessary,, Most of the data used here 

 has been collected by a small number of experienced captains. 



bo Comparison of recorded and estimated wave dimensions „ - To be 

 able to evaluate the accuracy of the wave dimensions based upon observa- 

 tion, the results of comparison between measured and observed data are 

 given o First, the data collected by observer I, lightship FEHJIARNBELT 

 during the period, 22-30 September 1936 are compared to data obtained 

 with tested DVL apparatus recorded during the same period „ Data was 

 taken only when conditions of observation were perfects The dimensions 

 required from the observer were the maximum vertical differential (wave 

 height) in meters. While observations were made at certain specific 

 times (8 a.mo noon, and 4- pom.) the recording instrument operated contimi- 

 ously over the 8-day period o The ten highest waves were determined from 

 all waves recorded during a period starting one-half hour before the time 

 of observation and ending one-half hour after the time of observations 



Figure 1 shows the observed value of the greatest difference in 

 height and the measured wave height which was surpassed 10 times during 

 one houTo This corresponds to the value given for the evaluation of 

 measurements which were carried out to establish a scale for the relative 

 state of the sea (DVL report covering periods measurements in the LUEBECKSR 

 BQCHT)o In most cases, the agreement between measured and observed values 

 is good; e„go, for two- thirds of all cases differences are less than 20 cm. 

 In individual cases, the estimated wave heights are lower than those 

 measured In the extreme case, a wave measured to be 92 cm„ in height was 

 estimated to be 4.0 cm„ It must be borne in mind that because of the 

 irregularity of storm waves, the value for the hei^t surpassed 10 times 

 per hour is not altogether identical bo the maximum height observed during 

 the time of observation » 



In figure 2 a comparison is shown of wave lengths measured and 

 observed concurrently » No observed values are available for certain 

 periods of time, probably because unfavorable observation conditions made 

 accurate estimates impossibles Wherever possible wave lengths were 

 estimated; also a count was taken of the number of crests moving past 

 the point of observation per minute. From this frequency coefficient n, 

 the mean period, T = 60/n, in seconds and consequently, the wave length 

 is computed, i.e., the corresponding value for the wave length from the 

 trochoid theory 



L = g T^ = lo56 T^ in meters. 



27r 



