most of the bergs in the area were driven close to the beach or 

 disintegrated. Subsequently late-season ice did appear and bergs 

 moved far enough south to endanger track C until 13 July. With 

 these facts and circumstances it is of interest to examine the mag- 

 nitude of the barometric pressure gradient in a direction parallel 

 to the coast. 



Daily synoptic charts, prepared by the aerological office of the 

 Naval Air Station at Argentia, for 0630 G. g. t., were available for 

 all but Sundays and holidays. Each chart was scaled as follows; 

 selecting an area of rectangular shape with its length 600 nautical 

 miles and its width 180 nautical miles and stretching along the 

 Labrador-Newfoundland coast with its inshore long side running 

 from 55° N., 57° W., to 46° N., 50° W., the barometric pressure 

 gradients along each pair of parallel sides were averaged for each 

 available chart, to get the gradient components which might be 

 considered to apply to the area, centered at about 51° N., 51° W. 

 Each component of the gradient (normal and parallel to the coast) 

 was averaged for the month and the monthly averages resolved to 

 get a resultant average gradient for each month from November 

 1946, through June 1947, inclusive. The long dimension of the 

 rectangle selected is oriented at about 335V2° and 1551/2° true. As 

 the wind runs across the isobars at an angle of about 30° to the left 

 at the surface in this region, and as the drift of floating sea ice (as 

 a direct effect of the wind) is to the right of the wind by a variable 

 amount approximating 45°, the ice drift has been taken as being 

 directed 15° to the right of the geostrophic wind, the approximate 

 value of which, expressed in knots, has been listed in the following 

 table along with the average gradients of barometric pressure in 

 millibars per nautical mile, and the resultant direction of the wind 

 drift in degrees true estimated as described above. 



From this table it will be noted that the direction, which was con- 

 stant during the months of November, December, and January, 

 changed 180° between January and February and continued to have 

 an onshore component during March, after which it returned to 

 about its previous direction for the rest of the season. To clarify 

 this picture, the drifts are shown in figure 13 in which the lengths 

 of the vectors are shown as proportional to the gradients. The 



60 



