DETERMINATION OF DEPTHS AT WHICH TEMPERATURES WERE MEASURED 



AND WATER SAMPLES COLLECTED 



The length of the wire from the surface to the water 

 bottle gives the exact depth only if conditions are so fa- 

 vorable that the wire remains vertical but if the drift of 

 the vessel is great on account of surface currents or 

 wind, or if considerable subsurface currents occur, the 

 wire cannot be kept in a vertical position. An approxi- 

 mate value of the depth of a water bottle can then be com- 

 puted from the length of the wire to the bottle and the 

 wire angle at the surface, assuming the latter to remain 

 constant. Such computation will generally render erro- 

 neous values for the depth, however, because the wire 

 will, as a rule, not remain straight but will form a curve 

 and the wire angle is, therefore, not constant but varies 

 with depth. In most cases the wire angle decreases with 

 depth and the assumption of a constant wire angle gives, 

 therefore, too small values of the depth. 



The Carnegie could not be manuevered so readily 

 that the wire could be kept approximately vertical under 

 all conditions and great wire angles necessarily oc- 

 curred. The knowledge of the depth at which tempera- 

 tures were measured and from which water samples 

 were brought up would, therefore, in many instances have 

 been inaccurate if this should have been derived from 

 wire lengths and wire angles only. On board the Carne- 

 gie , however, every second water bottle of the deep se- 

 ries was provided with one unprotected and one protected 

 thermometer, and by means of the indications of these 

 thermometers the depths of these water bottles could be 

 found with an accuracy, which, according to the results 

 in a preceding section (p. 11) was about +10 meters at 

 a depth of 1000 meters and +30 meters at a depth of 6000 

 meters. Knowing the depth of several points of the wire 

 with this accuracy, the curvature of the wire could be 

 determined and the depth of the intermediate water bot- 

 tles could be found. When taking the shallow series, 

 down to 300 to 400 meters, two or more of the water bot- 



tles were also provided with unprotected thermometers 

 and the indications of these were used for detecting any 

 conspicuous deviation of the wire from a straight line. 

 It was found, however, that at small depths no great er- 

 rors were introduced by assuming the wire angle to re- 

 main constant. 



The practical method, which was adopted for com- 

 puting the depths on the basis of all available informa- 

 tion, is best explained by means of an example. Table 1 

 contains the data from Carnegie station 71 (latitude 11° 

 57' south, longitude 78° 37' west). At this station, which 

 was occupied on February 6, 1929, two series of water 

 bottles were sent down. The wire angles observed on 

 board are entered in the first column of the table and 

 were 35° for the shallow and 40° for the deep series. 

 The cosines of these angles are entered in the second 

 column of the table. The third column contains the wire 

 lengths to the different water bottles. Four of the seven 

 water bottles of the deep series and two of the water 

 bottles of the shallow series were provided with both 

 protected and unprotected thermometers. From the in- 

 dicat ons of these thermometers, the depths have been 

 computed which are entered in the fourth column of the 

 table. The next column contains the factors by means of 

 which the corresponding wire lengths must be multiplied 

 in order to give these depths. These factors and the 

 cosines of the wire angles have been plotted against wire 

 lengths (fig. 1) and curves have been drawn representing 

 the factors by means of which any wire length has to be 

 multiplied in order to find the depth of that particular 

 point on the wire. From the curves the factors have 

 been read off for the intermediate wire lengths and en- 

 tered in the fifth column of the table. The final depths 

 in column six have been derived by multiplying the wire 

 lengths (column three) with these factors. 



From table 1 it is seen that the ratio between the 



Table 1. Computation of depths at Carnegie station 71 (latitude 11° 57' south, longitude 78° 37' west, 

 February 6, 1929) on the basis of thermometer depths and assuming the wire angle to be constant 



35 



0.819 



40 



0.766 



15 



