K + A £ (l - K ) sin* + = — cose 

 1 o f o ' a 



1 + (2A - 1) siiKj 



(3) 



where S = undrained shear strength 



u 



a = vertical effective stress during K 

 consolidation (equal to field 

 overburden pressure) 



A = a pore pressure parameter 



c, <j>, and K are as defined previously 



To use this equation it is necessary to determine a distribution 

 for the pore pressure parameter, A f (defined by Skempton, 1954). For 

 the NCEL tests, A f is equal to the change in pore water pressure divided 

 by the change in the difference between a and a_ during shear. The 

 parameter A f was obtained from each of the sets of triaxial test re- 

 sults, considering only the portions of the stress paths above the K 

 line. In Figure 9 A is plotted versus the vertical effective stress 



following K consolidation. A line has been constructed between the 



o 

 data points for use in interpolation. In the clearly normally consoli- 

 dated region, A f is independent of a vo (Ladd, 1964, Figure 111-18) . In 

 the pseudo-overconsolidated region a linear interpolation between the 

 data points has been constructed. 



An estimated undrained shear strength profile can be constructed 

 using the data which have been developed to this point. The strength, 

 S u , as a function of overburden pressure, a vo , is obtained by inserting 

 the previously obtained values of c, cj>, K Q , and A f into Equation 3. 

 The overburden pressure as a function of sub-bottom depth is given by 

 Figure 6. It is possible, therefore, to obtain shear strength as a 

 function of sub-bottom depth. This strength profile is given by 

 Figure 10. 



This profile appears to be the best achievable with the present 

 information. Most of the assumptions made in its derivation are rela- 

 tively insignificant. Two important assumptions are that (1) the basic 

 soil type remains constant with depth and (2) consolidation in a tri- 

 axial cell simulates field consolidation. It is planned to check the 

 validity of these assumptions by obtaining long piston cores and deter- 

 mining the variation of water content and index properties with depth. 

 If the assumptions of this report are correct, the index properties 

 will be nearly constant and the water content will vary in a consistent 

 manner predicable by consolidation testing. 



For short-term use the strength profile of Figure 10 is recommended 

 for such endeavors as direct embedment anchor analysis. It should be 

 noted that several characteristics of the profile are not conducive to 



20 



