Impacts of Liquefaction under 20 feet of Dense Fill

[theCorkster]

We're looking at a project where the COE filled in a low lying area adjacent to a river about 50 years ago.

While there are cross sections pre/post construction, there is no available documentation regarding what the depth, make up, or in place consistency of the alluvium was prior to placing 20 feet of fill, nor are there any records regarding the fill placement.

Visual observations in the top 8 to 10 feet of fill in pits suggests the fill is predominantly gravel (80%) with cobbles and boulders to 2+ feet scattered randomly in the fill; the consistency is visually estimated to be medium dense (based on lack of caving during pit excavation). The site has a design acceleration on the order of 0.38g with a magnitude 8+ earthquake.

A project done nearby about 10+ years ago used pits to 10 feet, and based on the observed consistency and lack of mottling in the fill, concluded that settlement associated with liquefaction potential was essentially nil due to 1)alluvium densification beneath the fill, 2) medium dense consistency of the fill, and 3) lack of mottling (in essence, lack of groundwater).

Given the site geology and seismic regime, our approach will be to evaluate both fill and alluvium consistency using Becker Hammer Test (while not perfect, likely the only drilling method to get through cobble/boulders) to assess liquefaction/settlement potential.

To the question for this post, what performance would one expect out of 20 feet of coarse, medium dense, porous fill overlying alluvium if the alluvium liquefied? Does anyone have any experience with similar conditions?

Thanks!

 

[dgillette]

Not sure what precisely you are asking for regarding "performance," but unless the ground is dead flat, lateral spreading is a real possibility, along with settlement, perhaps differential, due to excess PWP.

"0.38g with a magnitude 8+ earthquake" is quite a load. Alaska? New Madrid area?

 

[fattdad]

somewhere in this discussion we need to know about the water table elevation. Otherwise, it's a non-issue.

f-d

¡papá gordo ain’t no madre flaca!

 

[theCorkster]

Thanks for the questions.

The site is northern California.

Groundwater elevation is not known; however, the COE fill was placed to raise the site elevation above the 1964 100-year flood elevation.

It is not likely that the design flood and design MCE would occur at the same time. Given the coarse and porous nature of the fill, liquefaction in the fill is also unlikely. However, the underlying alluvium is of interest as the alluvium is near river levels.

 

[dgillette]

If you're adjacent to a river, you have potential for lateral spreading, with the amount of displacement being a function of alluvium density and site geometry. That's probably a more serious concern for structures on the fill than is settlement due to post-EQ densification. Definitely test your alluvium.

 

[FixedEarth]

A lot depends on depth of the "coarse and porous fill". You are assuming medium dense consistency will save the day, but may be not. For example, if you have GP, GW, GM or GC you may very well liquefy at a depth of 20 ft, even if SPT = 20 & there is no ground water table.

The 0.38g is not the culprit, but the 8.0 earhquake magnitude is high. Check this 8.0 as most 0.38g sites have an earthquake magnitude near 7. Why not quantify and obtain your F.S. against liquefaction by doing a proper analysis?

Then there is the dynamic settlement magnitude. Even if all soil layers do not liquefy, the ones that do will experience seismic settlement. Add this seismic settlement to static settlement and the project will soon be in a difficult soil site.

 

[dgillette]

'...may well liquefy at a depth of 20 ft, even if SPT = 20 & there is no groundwater table."

Are you sure? If the fill and alluvium aren't saturated, how can they generate high excess PWP? GP, SP, GW, or SW, or typical GM or SM above the water table would have air voids due to gravity drainage, even if they started out saturated and the water table dropped below them.

Also, you say PHA=0.38 isn't the culprit. That's a heck of a shake, and it could be a big problem, even if produced by a much smaller earthquake magnitude. Sheffield Dam failed with PHA variously estimated as 0.15 to 0.25, and that was in a much smaller earthquake (6.3). A PHA of 0.38 can give a CSR well over 0.3, (even including the magnitude scaling factor for smaller earthquakes), and there are plenty of dots below 0.3 on Seed, Lee, and Idriss's chart (and its descendants).

 

[FixedEarth]

dg;

Unless you supply me a site specific field data (SPT's), lab data(few gradation and some Atterberg limits), and historical ground water elevation, this site liquefies. Attached is my interpretation.

 

[dgillette]

Your example shows water table at 10 feet. In your earlier post you say there is no water table. Huge difference.