Friction Angle for Cut Slopes Underwater

[geomane]

I have a question regarding dredging a shipping channel. The existing channel will be deepened and widened by box cut dredging. Based on past dredging events in this channel, after the box cut operations are completed, the slope fails and stabilizes anywhere between 11 degrees to 14 degrees (i.e. 5H:1V and 4H:1V). The soils are very soft to medium stiff lean clay [CL] and fat clay [CH]. To determine the angle the slope will stabilize at, I'm thinking this would be similar to the angle of repose. I am hung up on laboratory testing methods to determine the drained friction angle for the clay soils. CU with pore pressure measurement or CD to determine peak drained friction angle, fully softened strength from direct shear remolded at LL, or residual strength. I don't know if tests to determine the angle of repose of sands also apply to clays.

After the slopes are cut, then fail, I don't think the peak drained friction angle applies. The clays would be somewhat remolded, absorb water when negative pore pressures dissipate, and I think would be close to the fully softened strength. I know the likelihood of reaching the fully softened strength increases with higher PI, higher moisture, higher clay size fraction, lower sand and silt content, presence of fissures or shrinkage cracks, higher activity, etc.; similar to my situation. But since the slope has failed, would we be closer to the residual strength? I want to say yes, but I'm not sure if the failure plane would be oriented parallel to the angle the slope eventually stabilizes at.

Slope stability analyses need to be performed to prove the new cut slopes will stabilize close to the existing slope angles of 5H:1V.

Your thoughts and experiences would be appreciated.

 

[GeoEnvGuy]

My thoughts are you should be using undrained shear strengths for cohesive soils and unless you have complex subsurface conditions the stability charts should give you a reasonable estimate on slope angle. This can be confirmed with stability programs.

https://www.researchgate.net/publication/276078531_Taylor's_Slope_Stability_Charts_Revisited

 

[miningman]

THeres a good reason why dredging tends to be an ongoing process in a given waterway. It's because there is no stable slope.
If this is coastal then daily tides and changes in water velocity continually erode and redeposit material. The issue becomes how wide a shipping channel do you intend to excavate and what rate of redeposition can you live with??? Alternatively
how often do you intend re-dredging ??? Annually or every 50 years???

 

[GeoEnvGuy]

Miningman makes some good points but i just want to add that there is a difference in a larger deep seated slope failures caused by excavation at steep slopes and slow erosion sloughing failures. Typically sloughing doesn't make the news, but larger failures do.

Also as you posted this is a shipping channel previously dredged it will need to be dredged again, just a matter of time.

 

[geomane]

GeoEnvGuy, why would you run an undrained analysis and not a drained analysis? We would have to run both, but I would think the drained analysis would control for cut slopes.

Some of the research I have done so far shows use of undrained shear strengths, but the slopes are temporary. The underwater cut slopes in the San Francisco Bay Mud with work by Duncan (factors of safety and reliability in geotechnical engineering) shows use of undrained shear strengths, but they were cutting slopes and backfilling with sand. Never mentioned drained strengths. Those slopes failed (if I remember correctly) because of creep and dissipation of excess negative pore water pressures due to permeable sands below the bay mud.

 

[GeoEnvGuy]

I think there is a lot of confusion in the industry as to how to apply material strengths in analysis. See the below white paper from the US dam association see strength of embankment materials 2007

https://www.ussdams.org/resource-center/publications/white-papers/

When people say drained strength they infer effective stress and when people say undrained strength they infer total stress, this is not correct.

The most recent version of Canadian guidance I see says to use effective stress with no pore pressure change and undrained strengths. That is for contractive materials and soft clay. Dilative sand materials should be modelled with a residual friction angle and stiff clay with a fully softened strength using effective stress.

If you play around with the stability programs long enough you will see that putting in friction angles or su/sigma' will show you shallow sloughs as the answer. This is because the program interprets T=tan14xsigma'+0 and Su=0.25xsigma' as base shear resistance = 0.25x sigma' which is the same.

Putting in cohesion only will show deep circles that go to the base of the material and come back up as rotational failures.

For cut slopes in clay you need to consider the change in stress state. I am familiar with geostudio a sigma analysis which is a finite element mesh using consolidation parameters before and after cut and long term deformation, if acceptable deformation is predicted. Then a limit equilibrium analysis with those stress states can be completed.

 

[EireChch]

geomane - Give BS 6349-5:2016 Maritime works – Code of practice for dredging and land reclamation a read.

The following table may be of some use.

 

[BigH]

If possible, local experience is best. I was called out (as a neophyte engineer) to a job in New Brunswick when the port owner decided to dredge the firm to stiff clay at, if i remember correctly, 2H:1V. Needless to say, cracks developed about 3 to 4 m behind the crest .. . oops, a pipeline was running parallel to the crest about 2 m behind - so the cracks were on the wrong side . . . Tidal variation were about 8 m or so. Lots of money spent on cranes to hold up the pipeline until remedial measures could be carried out.

 

[Okiryu]

I have also used undrained and drained analysis for dredging projects in soft silts and clays and normally the drained conditions control the design. BTW, EireChch chart is a good reference. Also, the Duncan's reference that you mentioned indicated very steep slopes (I was surprised when I saw it). Anyway, I am curious to see your shear strength parameters from your lab testing program. I think that underground slopes are difficult to analyze and we may need to be a bit more conservative during the analysis.