dy as TOD #2610 on 24 May and was carried 

 southward and then southeastward. The sensor 

 data indicated that its drogue became discon- 

 nected on 24 May 1983. TOD #2632 remained in 

 the eddy averaging 27 cm/sec until exiting the 

 eddy on 20 July. It followed the meanders of the 

 North Atlantic Current northeastward at 65 

 cm/sec until 6 August when it turned and 

 drifted northwestward. TOD #2632 made two 

 counter-clockwise circles about two separate 

 eddies and then began its eastward journey 

 across the Atlantic Ocean in the North Atlantic 

 Drift on 12 September. Between 6 August and 

 12 September, TOD #2632 averaged 55 cm/sec. 

 The Canadian METOC SST charts support the 

 existence of all the eddies shown by TOD 

 #2632's tracks. 



In summary, all of the TODs drifting 

 southward along the northern extremity of the 

 Grand Banks were guided bathymetrically in 

 their journey south. The past years TODs in this 

 region appear to have been guided 

 bathymetrically (Shuhy, 1981; Summy, 1982). All 

 four of the TODs drifting through the Flemish 

 Pass this year, entered circulation dominated 

 by the North Atlantic Current at nearly the 

 same position, independent of the time of year. 

 With the exception of TOD #2633, the TOD 

 velocity distributions were not significantly dif- 

 ferent (Figure B-4). Of the five TODs which 

 crossed 42°W between 46°N and 47°N, only 

 one (TOD #2613) did not circle back west of 

 42°W and then travel northward. This feature 

 has been observed in the previous years' drift 

 tracks. All of the TODs that have travelled 

 northward, including drifts of previous years, 

 have been caught in what appears to be a sta- 

 tionary meander of the North Atlantic Drift 

 centered roughly at 50°N, 43°W. This feature 

 was clearly portrayed by the Canadian METOC 

 SST charts of the area for most of the 1983 IIP 

 season. 



Modification of the IIP Historical Current Field 



The validity of certain regions of the IIP 

 historical current field had come under suspi- 

 cion because of the differences between 

 computer-predicted iceberg positions and 

 those of actual resights. The most prevalent 

 area in which this occurred was north of the 

 Flemish Cap. 



The criteria selected for making changes to 

 the historical geostrophic current field were: 



a. A minimum of five TOD tracks passed 

 through the same 1° latitude by 1° 

 longitude rectangle. 



b. The speed and direction of the TODs 

 through each selected rectangle must 

 have been reasonably similar. 



The time of the year of the TODs' passage 

 through the rectangle was not a factor since 

 the current field used by IIP is considered in- 

 dependent of time. This selection method 

 should eliminate those differences caused 

 solely by eddies or other short term 

 oceanographic features. 



Three 1° latitude by 1° longitude rectangles 

 met the criteria described above for making 

 changes to the current field. The three rec- 

 tangles selected were between 48°N and 49°N 

 and 46°W and 49°W (Figure B-5). TODs from 

 four different years passed through this area 

 and are included in Figure B-5. Some signifi- 

 cant changes were made to the historical cur- 

 rent field, changing directions by 180° and 

 velocities by an order of magnitude. The 

 changes to this region of the current field were 

 made prior to the 1983 ice season. Future 

 changes of the current field using the above 

 criteria will be done as necessary. 



Summary 



The real-time current information provided by 

 the TODs allowed the International Ice Patrol to 

 improve the accuracy of the iceberg informa- 

 tion disseminated to the maritime community. 

 The ability to obtain real-time current informa- 

 tion is a valuable tool and not available by any 

 other more cost-effective means. The ability to 

 modify with some confidence the IIP historical 

 current base solely on TOD drifts was an unan- 

 ticipated use of this data. As seen in this year's 

 drifts and those of previous years (Shuhy, 1981; 

 Summy, 1982) topographic steering appears to 

 be an important process in guiding drifting ob- 

 jects in the area of the Grand Banks. 



For the upcoming 1984 Ice Patrol season, a 

 subroutine to handle icebergs grounding on the 

 coast has been added. In the past, the ICEPLOT 

 program treated land as an area of cm/sec 

 geostrophic current. If the wind driven current 

 was from the "wrong" direction, icebergs 

 would "drift" across land. Icebergs would have 

 to be moved by the operator back to the posi- 

 tion they entered land. For the 1984 Ice Patrol 

 season, icebergs will automatically be stopped 

 at the coastline and will be flagged as 

 grounded. During the 1984 season, another 

 subroutine will be added to ground icebergs 

 based upon their size and the bathymetry of the 

 area. These two additional routines should 

 enhance the accuracy of the iceberg informa- 

 tion disseminated by the International Ice 

 Patrol. 



63 



