Is sliding shear failure between deep soil layers a realistic consideration?

[Greenalleycat]

Howdy all,

My boss and I just went down a rabbit hole that ultimately neither of us know the answer to.

Imagine a building over deep, stiff concrete piles that transfer loads to bedrock say 25m below ground level (the numbers don't matter, this is an academic discussion primarily)
Presumably the soil layers between bedrock and ground will vary with depth, some topsoil then maybe various soft and hard sands, clays, or silts - whatever it is

The soil layers will presumably have differing mass, stiffness, and depth per layer, and so under earthquake shaking will respond uniquely to each layer
This is conceptually similar to a multi-storey building with differing properties at every level, right?
Each soil layer will also have a shear capacity that must exceed the shear demands resulting from the earthquake response
In our heads, the most critical shear plans will be between soil layers, so we were wondering

1) Is it possible for a global shear/slip failure to occur between two soils in such a model? What does that look like in practice?

2) If it is possible, imagine the aforementioned stiff concrete piles going through. These piles will be subject to massively different displacement demands across the failure plane
The only way I can really see the pile surviving this is for the soil to deform around the pile...but if you're 20m below ground, will the soils be so dense that the compaction necessary to deform cannot be achieved?
If so, presumably you're pretty much guaranteed a shear failure in the pile..? Anyone have wisdom to share on this

Thanks in advance! Hopefully you guys can help settle our debate

 

[DaveAtkins]

The only time I consider deep seated soil layer issues is with bulkhead design along a shoreline. In this situation, you can get failure of the bulkhead wall itself, OR you can get a deep seated failure of a large mass of soil, which carries the entire bulkhead wall along with the failing soil mass.

I have not done much seismic design in my career, and when I have, I have relied on the recommendations in the geotechnical report. My hunch is that we make simplifying assumptions about the behavior of soil layers during an earthquake, and assign a site class to the entire soil mass. I have never heard of a pile failing deep below the ground surface.

DaveAtkins

 

[MotorCity]

I don't have an answer for you, but I would argue that the same phenomena could happen within a soil layer, not just across different soil layers.

This is probably even more of a concern if the water table is located in the middle of a uniform layer of soil.

You may also want to read up on liquefaction. That is exactly what happens during a seismic event.
The soil laterally spreads, causing a horizontal shearing action within the soil.

Don't mean to muddy the waters....pun intended.

 

[Greenalleycat]

Don't worry, we are very well versed with liquefaction in my city...spent a week of my final highschool year digging houses out of liquefaction sand!
That is the ultimate origin of my question, except that I was trying to generalise it to situations outside of liquefaction

@DaveAtkins, I've never been able to verify this, but during the recovery works afrer the 2010/11 earthquakes here I spoke to a guy who was a demolition operator on one of the largest buildings in town
He reckoned that the top 5m of pile came out attached to the pile cap but the lower bits stayed in because they'd broken (sheared) off at that depth
Never been able to verify whether that was true, but it was a plausible story anyway, given the absolute shitshow of ground we decided to build a city on

 

[geotechguy1]

The piles that sheared off probably sheared off or deformed significantly around the boundary of liquefiable soil to non-liquefiable soil.

 

[Greenalleycat]

Yep I imagine that's right. What are your thoughts on whether that is likely to occur outside of liquefiable layers?

 

[geotechguy1]

Are you talking about a completely flat site with a building on it?

 

[Greenalleycat]

Yep I am