(SLAR) with no visual 

 confirmatbn. As a result, IIP can 

 only classify icebergs into the 

 broad categories of growler, 

 small, medium, and large, and 

 assume characteristic mass and 

 cross-sectional areas for each 

 category. When visual 

 confirmation is available, it is 

 possible to distinguish between 

 tabular and non-tabular icebergs, 

 resulting in somewhat different 

 mass and cross-sectional areas. 

 Regardless of the size and shape 

 of the iceberg, both the air and 

 water drag coefficients are set to 

 1.5. 



Figure D-1. Area of Study 



Currently, IIP estimates 

 that the model drift error is 1 0nm 

 (~ 18.5km) for the first 24-hour 

 period and an additional 5nm 

 {~9km) for each additional 24 

 hours of drift, up to a maximum 

 error of 30nm (~56km). The 

 accuracy of this error estimate is 

 evaluated in this report. 



In 1983 IIP began using 

 observed-current data derived 

 from the trajectories of freely- 

 drifting satellite-tracked buoys to 

 modify the mean geostrophic 

 field (Summy and Anderson, 



1983) during operational model 

 mns. The modifications are both 

 temporary and localized in that 

 they are only applicable during 

 the period that a buoy is in that 

 specific region, after which the 

 currents revert back to the mean 

 geostrophic currents. It is not the 

 intent of the present report to 

 address this practice directly, but 

 rather to compare the drift-model 

 accuracy using two sets of input 

 data: mean geostrophic data with 

 FNOC wind and on-scene 

 measured data. In doing so, the 

 importance of using on-scene 

 data becomes clear. 



