The same water masses exist along this section as along the previous one, but 

 to a greater extent. Intermediate Water at the surface is associated with the center 

 of high salinity (31.50 °/oo) and a tendency for cyclonic circulation. At this distance 

 from the coast there has occurred less mixing with Alaskan Coastal Water so that 

 the central part of this Intermediate Water region still contains a high percentage of 

 the Modified Shelf Water. 



A water mass having the same characteristics as, but not geographically con- 

 nected with, the Deep Shelf Water of the Bering Sea appears along the bottom at 

 the seaward end of the section. The nomenclature is retained because we believe that 

 the two masses have the same general origin (see discussion in Conclusions). 



In the region of the drifting ice along the periphery of the ice pack is cold 

 low-salinity Ice Melt. This water has the same salinity characteristics as the previously 

 defined Modified Ice Melt (fig. 2) and thus is differentiated from it only by temperature, 

 indicating that it warms rapidly as soon as it is isolated from the melting ice pack. 

 On the other hand, the Ice Melt water has the same temperature characteristic as the 

 Deep Shelf Water and is differentiated from that mass only by salinity. 



In summary, the three water masses, Alaskan Coastal, Intermediate, and Modified 

 Shelf Waters, are continuous from the Bering Sea through the Chukchi Sea. Deep Shelf 

 Water occurs in both seas but has no continuity through the Strait at this season. The 

 Siberian Coastal Water, Modified Ice Melt, and Ice Melt are found in the Chukchi but 

 not in the eastern Bering Sea. 



surface currents 



Movements of the water masses are generally determined by dynamic compu- 

 tations, by direct current observations, or qualitatively by the temperature and salinity 

 distributions. 



The dynamic topography of the surface, computed over a selected reference 

 level, indicates the speed and direction of the current at the surface relative to that 

 of the reference level. 18 The relative current was computed over a reference level of 

 130 feet to determine if this method would agree with the few observed current measure- 

 ments and thus give a valid picture of the surface-current pattern throughout the area. 

 In most instances, except in Bering Strait where the relative current was large, the 

 computed current was only a small fraction of that observed. It was evident that the 

 movement at the reference level was usually much greater than the relative current 

 computed between the 130-foot level and the surface. 



Another means of determining the speed of the current is by the geographic 

 displacement of water of a given type during a limited time. This method was used 

 to determine the speed of the coastal current from Kotzebue Sound around Pt. Hope. 

 Low-salinity water of less than 29 °/oo was found at station 67 on 10 August 1949, 

 0405Z, but did not appear northward along the coast. Similar low-salinity water was 

 found at station 149 on 22 August 1949, 1735Z. These measurements imply a progress 

 of about 200 miles (a minimum value because of the mixing) along the coast in 12Vi 

 days, an average speed of between 0.5 and 0.9 knot depending on the course traveled. 



18 J. W. Sandstrom and B. Helland-Hansen "Uber die Berechung von Meeresstromungen" Nor- 

 wegian Fishery and Marine Investigations, vol. 2, no. 4, 1903. 



17 



