region was subnormal. The departures from normal mean tem- 

 perature are not as striking and the subnormality of the heat 

 transport is attributed to the reduced volume transport. Sections 

 U and W were expected to show smaller than normal transports 

 because of the diversion of the Labrador Current by the Atlantic 

 Current salient as shown in figures 7 and 9, but it is evident from 

 table 1 that the amount of Labrador Current reaching section T 

 also was less than normal. The minimum obsei^ed temperatures 

 were somewhat warmer than normal. 



Considering the Bonavista triangle, during the second survey 

 about 80 percent of the water entering the triangle followed the 

 eastern branch of the Labrador Current. Approximately 85 per- 

 cent followed this branch during the post-season cruise. The 

 volume transport past section G was about the same during the 

 second and third surveys, but a considerable part of this water 

 recurved northeastward north of section F during the second sur- 

 vey whereas this diversion occurred farther south in the third 

 survey. The circulation deduced from the values given in Table ] 

 is shown schematically in figure 10 for each of the surveys made 

 in 1951. The computed figures for volume transport have been 

 rounded to the nearest 0.1 million cu. m/sec. and adjusted for the 

 triangles so that the sum of the transport past sections SW and SE 

 equals the transport past section NW\ In the first survey it is 

 considered that the increase in volume at section U over that at 

 section T is the result of a contribution from the west, rather than 

 a closed eddy on the east, because of the lower mean temperature 

 found at section U. The recirculation of water in the closed eddy 

 centered near 43° N., 49° W., in figure 7 seems, from the low mean 

 temperature found at section W, not to have extended as far west- 

 ward as this section. 



In the second survey, the greater flow past section H than that 

 at either sections SE or G indicates additional circulation asso- 

 ciated with the margins of the Labrador Sea, as does also the 

 higher mean temperature. 



In the third survey the addition of water from the westward 

 between sections G and F, resulting in the lower mean temperature 

 at section F, is an indication that this water is not associated with 

 the Grand Banks eddy which by the first of June has undergone 

 appreciable warming. Similarly the addition of water from the 

 Grand Banks between sections T and U is called upon to explain 

 the higher mean temperature at the latter section rather than 



3 Regarding the reliability of the computed volumes of transport, it should be noted that a 

 consistent error of 0.01 ^/oo in salinity at one of a pair of stations will produce an error in 

 the dynamic height of that station of about 7.2 dynamic mm and an error of about 36 X 106 

 cu. m/sec. in the volume of flow between the two stations if 1,000 decibars is the reference 

 ?TnnT ■I^''^^- errors become 10 8 dynamic mm and 0.81 X lOO cu. m/sec. respectively if 

 1,500 decibars is the reference surface. 



28 



