taken for approximately 20 min at the outer buoy immediately followed by a 

 similar sample of the inner buoy. This sequence and the time lapse involved 

 is effectively the same as concurrent measurement, since conditions at sea, 

 particularly more extreme conditions, very rarely change that fast. The scat- 

 tergram of HS2 (inner buoy significant wave height) and HS1 (outer buoy) shows 

 some linearity, as indicated in Figure 11 (R = 0.39). TP2 (inner buoy peak 

 period) and TP1 (outer buoy) do not show any statistical relation, though vi- 

 sually the scattergram (Figure 12) indicates TP2 tends to equal TP1 in many 

 cases. Attempts to run scattergrams of HS22 with HSI^ and HS2/gTP2^ with 

 HSI/gTPI^ produced no meaningful information. The significant wave heights 

 were originally computed as representative of energy in the sea state, so 

 the relation of energy levels is inherent in any comparison of HS2 with HS1. 

 Squaring these values apparently only exaggerated the scatter. H/gT , as a 

 measure of wave steepness, showed no statistical trends apparently because 

 the peak periods by themselves didn't either. 



30. Attempts to relate measured wave conditions to concurrently mea- 

 sured wind direction did not yield any dramatic conclusions, but they did re- 

 inforce the intuitive assumption that the most severe wave conditions exist 

 with winds from Chiniak Bay. Figure 13 is a scattergram of HS2 with DIR (di- 

 rection), in which cases with HS1 < 1.0 m and TP1 < 4.0 sec were excluded. 

 The grouping of lower HS2 values in the range 280-360 deg indicates the rela- 

 tive frequency of offshore winds. HS2 values above 1 m are grouped in the 

 range 45-180 deg. Figure 14 is a scattergram of HS1 with DIR, showing simi- 

 lar indications. The ratio TP2/TP1 plotted against DIR in Figure 15 shows 

 the tendency for TP2 to equal TP1 with directional grouping similar to the 

 wave height scattergrams. Womens Bay winds blow toward Puffin Island in the 

 200- to 250-deg range. This range has few points plotted on the three scat- 

 tergrams, indicating severe wave conditions concurrent with winds from Womens 

 Bay are rare. 



31. SPSS procedures attempting to correlate wind velocity at Puffin 

 Island with significant wave heights did not reveal any meaningful conclu- 

 sions. The relationship of wind velocity to significant wave height is known 

 to be highly nonlinear, however, and includes a number of other equally crit- 

 ical independent variables. Wind velocity data are, therefore, treated sep- 

 arately in this report with a view toward application of these nonlinear 

 relationships. 



32 



