techniques has heen extensively described (e.g., Dix, 1955: Clav and Rona, 

 1965). The basis of this method is the familiar T 2 versus X 2 eouation of 

 reflection seismology: 



9 



T2 = T o 2 + fz~ " 2T ° I sin * (26) 



where T Q = vertical reflection time 



T = reflection time at distance X from the source 



V = interval velocity in layer 



= slope of lower interface with respect to upper 

 interface 



If D is the arrival time of the direct wave to the sonobuov and V 

 is the horizontal water velocity 



T 2 t 2 4. ° 2v H 2 2T D V H . 

 T = T o + — o — — g-Ln ^ 



H 



(27) 



If we assume that the interfaces of the lavers are parallel (0 = o), 

 the above equation reduces to 



2 

 T 2 = T 2 + n 2 £%--) (28) 



which is an equation of a straight line in D 2 and T . The interval 

 velocity is determined, therefore, by the inverse of the slope. 



More sophisticated analyses utilize least-snuare curve-fitting 

 approaches and computer techniques. For example, in the manner of Clay 

 and Rona (1965) , the odd powers of X disappear for horizontal layers so 

 that the T 2 versus X 2 equation becomes 



T 2 = T 2 + L__ + KX 4 (29) 



V 2y2 

 v an A 



where V an = inverse of slope of T 2 ,X 2 line at the 

 origin 



K = constant 



from which the following equation is derived 



18 



