348 



The Representation of Oceanic Movements and Kinematics 



The individual values obtained from current measurements as discussed above will 

 contain errors due to the simultaneous movement of the vessel, and correction to the 

 true current can only be made if the movement of the ship is known with some 

 accuracy. Since for current measurements in the open ocean only one current meter 

 records on board ship, the correction method of determining the true current cannot 

 usually be used. If the average true current changes only slowly, the smoothing method 

 of ehminating short period movements of the vessel must be appHed. How strongly 

 the observations have to be smoothed has been shown by Thorade (1934) with 

 observations made by the research vessel "Poseidon" in the Kattegat (August, 1931). 

 The Rauschelbach current meter was used here to give continuous records of the 

 current every 10 sec over a long period. Plotting all these current vectors starting from 

 a single zero point of an appropriate co-ordinate system gives a current diagram of 

 the type shown in Fig. 142. The individual current vectors are strongly scattering and 



Fig. 142. Recordings of the Rauschelbach current meter at the anchor station of the 

 "Poseidon" in the southern Kattegat during I h for each 10 sec. (10 August 1931; 18.30- 

 19.30 h). The current arrows must be drawn from the point O towards the crosses. The 

 indicated arrow refers to the start of the observations. The dotted line shows the movement 

 of the arrowhead during the following 3 min. The dashed-dotted line indicates the position 

 of the arrowhead after smoothing, the point O shows the mean position during the h h. 



their end-points form a point cloud covering a relatively large area. It can hardly be 

 assumed by the values given in the diagram that the true current has altered significantly 

 within the half-hour observational time. The dashed line joins the end-points of the 

 vectors for the first three minutes. Even for this short interval the vectors cover almost 

 the entire area of the point cloud. This shows that single current measurements made 

 from an anchored vessel differing widely in the observation time are more or less worth- 

 less. It is rather different, however, if for short observation intervals mean values are 

 taken for more or less long intervals in time. Fig. 143 shows that for the same values 

 as in Fig. 142 the individual means for each minute are rather scattered, but the means 

 for intervals of 5 min, on the other hand, show only small variations during the half 



