cnVs. On 20 July (201) the iceberg 

 increased speed to 10-20 crtVs. The 

 TAD failed on 24 July (205). 



One attempt was nriade to relocate and 

 re-sight TAD 4504, but it occurred four 

 hours after its failure on 24 July. Three 

 icebergs were located in tfie vicinity 

 (within 20 km) of 4504's last 

 krxwn positbn, tfie closest of whiich 

 was a large blocky iceberg , which fit the 

 description of the iceberg on wtiich the 

 TAD had been placed. However, the 

 TAD could not be seen on the iceberg. 



TAD 2580 provided a7-day track before 

 it ceased transmitting on 23 June ( 1 74) . 

 Because of the short record, the position 

 data were not filtered. The track pre- 

 sented in Fgure 6 is plotted using po- 

 sitions interpolated at 24 fx)ur intervals 

 (at OOZ for each date). The U and V 

 components represent 3 hourty aver- 

 aged values. 



During most of the 7-day period, 2580 

 nxjved souttTward. In the first half of the 

 period it averaged 1 0-20 cnVs, while in 

 the second half it slowedto 5-8 cnVs. On 

 the day that 2580 failed it reversed 

 direction. 



Unfortunately, the tracks provided by 

 2580 and 4504 are not very useful 

 additions to the data set desired by Ice 

 Patrol. Future deployments should be 

 made south of Hamilton Bank and in 

 water depths greater than 200 m deep. 



Discussion 



Determining tx)w bng TADs remain on 

 icebergs is not always easy. Because 

 TADs nxist survive in the iceberg nrtelt 

 ponds, they are designed not to sink 

 wtien they fall ir<o the sea. Using the 

 qualitative changes in the TAD's trajec- 

 tory, such as an increased responsive- 

 ness to wind forcing, as an indicatorthat 

 tfie TAD is freely drifting is sometimes 

 useful but not conclusive. The internal 



tennperature sensor can sometimes be 

 used to infer that the TAD has entered 

 the sea. However, it is not always 

 possible to distinguish between a TAD 

 in a nnelt pond and one ftoating in cold 

 surface water using temperature . 

 Currently, relocating the iceberg and 

 visually confirming ttiat the TAD is still 

 aboard is the only certain method. To 

 this end, painting some txight stripes on 

 the TAD would help the ice observers 

 see it. In fog, a radar survey of the area 

 is useful, but not conclusive evidence 

 that the TAD is on an iceberg. If tfiere 

 are no radar targets in the area it is quite 

 certain ttiat the TAD is no longer on an 

 iceberg. The existence of a radartaiget 

 in thie area is, however, not conclusive 

 evidence that tfie TAD is on an iceberg. 

 Using tfie SLAR, it is not always pos- 

 sible to distinguish between an iceberg 

 and a ship passing through tfie 

 area, f^oreover, there are navigational 

 errors associated with the aircraft's iner- 

 tial navigatbn system that introduce 

 additonal uncertainty into the target 

 identificatbn process. 



Page 95 



