Figure 2-8 is a comparison of the limit of the slick on December 21 as 

 observed on overflights and the "worst case" estimate of the limit predicted 

 from the winds as of 2400, December 20 (figure VII-11) In figure VII-13, the 

 observed limit of the slick on December 23 is compared with an estimate of 

 the limit as of 0700, December 23 (figure VII-12) . Again, the observed limit 

 of the slick was obtained from overflights of the area. In these two fig- 

 ures, the predicted direction of movement of the oil is in agreement with the 

 observed movement, and the predicted areal coverage is about twice the ob- 

 served. The entire prediction was accomplished in approximately 3 hours 

 after the time of request from MSO, Boston. It indicates that vectoral 

 addition of the forces that move the oil is an excellent method for a quick 

 answer to where the oil will go and when it will get there. Had the winds 

 been onshore instead of of offshore, this method would have enabled cleanup 

 equipment to be placed at strategic areas before the oil came ashore. In 

 addition, it seems likely that had actual on-scene winds been used in the 

 forecast rather than long-term predictions from NWS, the predicted movement 

 and dispersion would have been even more precise. 



Figure VII-14 is a progressive hourly wind vector diagram derived from 

 3.5% of the on-scene wind speed data for the period 1600, December 15, to 

 0700, December 23, collected by the USCGC Vigilant. These wind data to- 

 gether with tidal data can be used to compare actual short-term movement of 

 the oil with the predictive technique of adding winds and tides vectorially. 

 In addition, the validity of using 3.5% of the wind speed data in a downwind 

 direction can be examined. 



For short-term drift. Figure 2-9 gives a comparison of observed slick 

 and predicted movement caused by tides and winds. The vector shows the pre- 

 dicted movement of the oil for the period 2400, December 15, 1976, to 0900, 

 December 17, 1976. The outline of the slick was taken from slick map IV-1 in 

 Appendix IV, which shows its location near noon on December 17. There is 

 excellent agreement between the actual direction of the oil movement and the 

 predicted movement as determined from tides and wind. In fact, it appears 

 that 3.5% of the wind speed data adequately describes this transport vector. 

 The greatest error occurs in predicting the tidal component for each hour of 

 movement . 



From December 15 to December 19, the total extent of the spill was not 

 clearly defined by overflights. However, several observations of the move- 

 ment of the oil during and after this period verify the techniques used by 

 the R&D Center. These observations are shown in Table 2-6. The observed 

 movement of the oil spill is documented in the maps contained in Appendix IV. 

 Table 2-6 indicates that the oil was moving westward, bearing 240°T, on 

 December 16. On December 18, pancakes were found 27 miles east of the ship 

 (090°T) . The maximum eastward movement of the oil caused by the wind (Figure 

 VII-14) was 31 miles, bearing 130°T. Assuming this is the maximum eastward 

 extent of the oil, the computed wind factor for moving it would be 3.05%. 

 Beginning with December 20, the daily wind factor for the observed movement 

 and the direction of movement of the oil are given in Table 2-6. A summary 

 of these values compared with predicted values is shown in Table 2-7. 



44 



