E-5 

 the net water density) . Pressure depth had the largest 

 uncertainty in measurement primarily because our pressure 

 gauge had a resolution of 7 meters and a repeatability of 

 3% of the total depth range. As a consequence, along some 

 of the deepest tracks, such as in Oceanographer Canyon, the 

 pressure depths differed from the sonic depth by as much as 

 20 meters. 



Following the computation of the horizontal separation 

 of the ship from the sled, a horizontal range circle was 

 plotted at each 5 minute advance of the ship track. Since 

 camera runs were made at speeds of 0.8 to 1.2 knots, there 

 is a horizontal range circle for approximately each 150 

 meter segment of the track. The size of the horizontal 

 range circle is dependent on the speed of the ship and the 

 camera sled. Photographs were automatically taken every 

 15 seconds. Thus, one picture is taken for each 8 meter 

 segment of the track, and there is a new range circle cal- 

 culation for about every 20 frames. The determination of 

 the point that the sled track intersects each circle was 

 achieved using two approaches. The first involves compu- 

 tation of the physical forces acting on the towed cable with 

 knowledge of the details of the surface track, the length of 

 cable and the depth of the sled. The second was the dead- 

 reckoned sled track computed entirely from the on-board 

 dead-reckoning sensors. A computer program was devised 

 which fits the dead-reckoned sled track to the every sixth 

 range circle and compares the deviation from all the in- 



