\ 



Sfl'iM 3;°w 



— I .J ■ - ,j , , I — ■ , — • 



SCiAl 48°W 4S°W 



48''N tp 



47''N 



TART 



45*" N ->' 



CALIB Data 

 5-1 8 May 2003 

 13 Days 



\ 



49^^ 



■<tr 2cr 



dS'W 



■nr ;cr 



47Va 



Figure 1. CALIB track. 



Comparison Methods 



BAPS is not truly intended to accurately model Ice Island shaped icebergs as they are rare 

 and represent a small fraction of the icebergs seen in the Grand Banks region. Additionally, the 

 above water height of Ice Islands is less than archetypical tabular icebergs which have 

 significantly higher freeboards as well as deeper drafts. Subsequently, the comparisons attempted 

 here were done with limited hope of high levels of correlation even when a very large, tabular 

 iceberg was modeled. Therefore, in order to test the model, many permutations were attempted 

 throughout this analysis. Table 1 details the tests performed. Basically, different iceberg shapes 

 and sizes were modeled using the What-If functionality of BAPS. What-If model runs permit the 

 user to alter virtually all of the model's parameters including iceberg size/shape, environmental 

 forcing data fields and model timeframes. In the case of this experiment, What-Ifs were run using 

 a 72-hour sliding window for each size/shape-forcing combination (Table 1 ). The 72-hour 

 window was used in order to avoid any errors associated with longer model runs but still maintain 

 a time frame allowing for response to changes in local forcing. The 72-hour window provided for 

 13 individual model runs per model permutation. The chosen sizes and shapes were based on the 

 size and shape of the actual iceberg and a hypothesis that a growler might present a good 

 representation (in the model) of real-world Ice Island drift. There were no modifications made to 

 the environmental forcing data except that, for certain permutations, a particular forcing was 

 switched off in order to determine the effect of wind or current alone. 



52 



