current-induced dip of the taut wire buoy, would cause a consistent 

 amplitude deviation at a particular station. From perusal of the 

 records, it seems more reasonable to assume that the first and third 

 explanations both apply. The internal wave surface or zone might 

 therefore be analogous to the sea-swell regime of the air-sea interface. 



B. SIGNATURE VELOCITY DISTRIBUTION 



Computed signature speeds and directions are summarized in 

 Appendix II for the fall data and in Appendix III for the spring data. 

 Cumulative frequencies of occurrence of speed within 45° sectors of 

 direction (set) have been calculated from these data and the results 

 are graphically displayed in Figures 13 and 14 for the fall of 1959 

 and the spring of I960, respectively. 



The assessment of internal wave or temperature signature velocity 

 distribution must be tempered by an estimate of the motion of the 

 medium itself. The only reasonably reliable current data collected are 

 given in Appendix I, and cover the period 12-25 April I960. These data 

 represent about five-minute averages observed every hour with a 

 meter suspended at 100 feet from one of the guide cables. The meter 

 support (bracket and cable) was not completely free from lateral 

 movement. In tank tests, the meter was found incapable of orienting 

 to within 30° at constant tow speeds less than about 0.15 knots. Both 

 of these factors contribute to uncertainty of the data. In addition, 

 the effect of the platform on magnetic reference of the meter was not 

 determined. 



The following averages for current speeds greater than 0.01 knot 

 have been computed from the data in Appendix I: NE = 0.23 knot; 

 SE = 0.27 knot; SW = 0.31 knot; and NW = 0.38 knot. Of the 268 

 observations used, 7.5% were in the NE quadrant, 26.0% in the SE, 

 27.5% in the SW, and 39.0% in the NW. Less than 8% of all hourly 

 observations were at speeds below 0.1 knot. These ranges of speed 

 correspond reasonably well to the vector spread of velocities 

 summarized in Figures 13, 14, and 15. 



The statistical treatment of the current data allows for the 

 approximate nature of the basic data in that rough estimates are 

 made of average speeds within broad bands of direction. There are 

 several breaks in the hourly observations for which no adjustments 

 were made. It is felt that the single record period of two weeks at 

 least gives some notion as to probable speed ranges and can serve 



