Question 2 of 6: Shear-strength testing for an earth dam on normally consolidated clay

Question 2: Shear-strength testing for an earth dam on normally consolidated clay (10 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

The foundation is a saturated, normally consolidated clay of great depth, and it is loaded by a relatively rapidly placed earth dam. Two stability conditions must be examined. The end-of-construction (short-term, undrained) condition is critical for a normally consolidated clay: the low-permeability clay cannot drain during construction, large positive excess pore pressures build up under the embankment load, and — because a NC clay has essentially no effective cohesion ($c' \approx 0$) — the soil is at its weakest just after the fill is placed and before any consolidation gain occurs. The long-term (drained) condition, after the excess pore pressures have dissipated and the clay has gained strength, must also be checked, together with any intermediate rapid-drawdown or staged-loading cases.

To capture both, the recommended laboratory programme is consolidated–undrained (CIU) triaxial compression tests with pore-water-pressure measurement, performed on specimens reconsolidated to their estimated in-situ effective stresses (and, ideally, over a range of confining pressures spanning the field stress increase). This single test type yields everything the design needs: (1) the effective-stress strength parameters $c' \approx 0$ and $\phi'$, from the effective-stress Mohr circles, for the long-term drained (effective-stress) stability analysis; and (2) the undrained strength $s_u$ and the Skempton pore-pressure parameter $A$, from the total-stress response, for the end-of-construction analysis and for predicting the excess pore pressures generated by the fill. A consolidated–drained (CD) test would in principle give $c', \phi'$ directly but is impractically slow in a low-permeability clay (days to weeks per stage); an unconsolidated–undrained (UU) or unconfined test alone gives only $s_u$ at the sampling condition and no effective parameters, so it cannot support the long-term analysis. An in-situ field vane shear programme is a valuable complement, giving a continuous $s_u$ profile with depth (corrected for plasticity) that is used to calibrate the laboratory undrained strengths.

The samples must be high-quality undisturbed samples — thin-walled (Shelby) tubes, or preferably fixed-piston samplers, pushed (not driven) and recovered at representative depths within the influence zone of the dam, then sealed and stored at in-situ moisture content. Undisturbed sampling is both feasible and essential in a soft-to-firm NC clay: the measured strength, compressibility and pore-pressure response all depend on preserving the natural fabric, water content and stress history, which disturbance would destroy (sample disturbance typically reduces the measured $s_u$ and rounds the consolidation curve, giving unconservative settlement but conservative strength). Because the deposit is deep and normally consolidated, the effective-stress parameters govern long-term safety while realistic prediction of construction pore pressures — hence the CIU-with-pore-pressure choice — governs the short-term case.