Photographs of the Excitement of Geotechnical Engineering (Failures)

July 10, 2009 2:15 PM

Professors Ross W. Boulanger and Dr. James Duncan have put together a Geotechnical Engineering Photo Album, with details of the successes and disasters. The album includes compaction techniques for a highway off-ramp, deep excavation methods, an offshore tank structure, and earthquake hazards of many sorts (mountain landslides, liquefaction damage to ports in Kobe, Japan, surface rupture in Taiwan, and problems with shallow foundations and subsidence in Turkey). (via oi9)
posted by filthy light thief (12 comments total)

Holy cow, this is great stuff. I've always enjoy reading all the details of how they pull off big construction projects, but so much of what I find is very superficial or overly technical. To me the guys working on big construction sites and the engineers that make it happen just don't get enough props in our culture.

Of course one thing I'd like to see is a review of how this accident in China happened.
posted by crapmatic at 2:28 PM on July 10

crapmatic - wow. It looks vaguely similar to this photo from the shallow foundation set, which had this description:

Widespread damage to buildings occurred throughout Adapazari, Turkey, during the 1999 Kocaeli earthquake. A major cause of damage was liquefaction of the recent alluvial deposits that underlaid large portions of the city. The result was excessive settlements and bearing capacity failures for countless buildings, most of which were supported on shallow foundations.

Insufficient foundations plus earthquakes equals buildings toppling over as one structure. The building itself was clearly held together well, but it didn't tie back into anything significant enough.
posted by filthy light thief at 2:43 PM on July 10

That fallen-over building was awesome. It takes a lot of hard work for get failure modes like that.
posted by GuyZero at 2:49 PM on July 10

That kind of thing makes me wonder a lot about long bridges that go through bays, like the Lake Pontchartrain causeway, as they're basically sinking the piers into the same kind of liquified alluvial soil that toppled over those buildings. My recollection of Gulf Coast geology is that it's an alluvial plain and bedrock is many miles down, so I'm not sure how it is those concrete piers are not toppling over. Since there's thousands of them I don't think it's cost-effective for them to overengineer each one. I tried Googling around on this and didn't find anything; sites like this have some good detail but don't explain much about the softness of the soil or how deep the piers go.
posted by crapmatic at 2:57 PM on July 10

crapmatic -- I can possibly help you with that since I drive across the Causeway every day.

The southbound span built in 1956 is definitely sinking a bit unevenly and levelling it is one of the things on the Causeway Commission's long-term agenda. Driving across it is a bit of a humpta - dumpta - dumpta type experience nowadays. However, the amount of sinking is limited because the pilings are driven in far enough, and enough of them are used, so that the friction between the sides of the pilings and the soft earth is enough to keep them from moving with respect to the soil. This is standard engineering practice in the area, and is how all the skyscrapers and tall bridges are supported too. There is as you say no bedrock to support anything. The pilings holding up the tall buildings and river bridges are hundreds of feet long, and splayed outward as they are sunk to increase the amount of soil supporting the structure as a whole. Causeway pilings aren't so dramatic because each piling has a relatively modest load. this site has the dimensions; they are sunk about 70 feet into the lake bed.

What causes the severe deformations seen in the photographs is the liquifaction of the soil, which is not its normal state; normally, settled deposits are pretty stable and sink at a relatively constant rate on the order of inches per decade. What causes the liquefaction is earthquakes, and southern Louisiana is a remarkably stable area in that regard. We do have periods of volcanism at intervals of 20,000 years or so as the river deposits enough crap on the continental shelf to push it down into the mantle and lava flows upward, but as it happens we're just about midway through that cycle right now, so the volcanoes are only visible to geophysicists doing seismic oil exploration.

I can say, knowing what is holding up things like One Shell Square and the Huey Long Bridge, that if we ever did have an earthquake losing the Causeway would be the least of our problems.
posted by localroger at 3:35 PM on July 10 [1 favorite has favorites]

Did I see concrete pilings without rebar reinforcement in the photos crapmatic linked to?

Hope you didn't put a deposit on any units in adjoining buildings...
posted by maxwelton at 3:58 PM on July 10

Wow, it's like looking at the future of California. Maybe I won't work late today...
posted by jewzilla at 4:33 PM on July 10

This is one of those posts that makes me feel like a total weirdo for being so excited by such a mundane-seeming subject.
posted by kiltedtaco at 5:57 PM on July 10

maxwelton: "Did I see concrete pilings without rebar reinforcement in the photos crapmatic linked to?"

The pilings are hollow. It's hard to tell but there appear to be cables coming out of a few of the pile stubs on the building. If there are the piles could have been Post Tensioned.

Also a piling is pretty well under a pure compressive load; I'm not sure but you may not even require reinforcement in that case because there aren't any tension forces acting on the pile.
posted by Mitheral at 6:34 PM on July 10

I've actually worked with Prof Boulanger on a project involving soil slope failure on a marine structure (i'm a structural engineer by training, so my apologies if i completely err in my description of the geotech stuff). I'll have to compliment him on these photos the next time i run into him (it's an incredibly small community). There's nothing a geotech/structural likes better than cool collapse photos (as long as you aren't involved in the project prior to the collapse).

For structures in soft soils (marine structures especially) during seismic events you can get liquefaction, slope failures, and mixtures of those responses. The thing to understand about geotechnical engineering is it's really more of an art than a science. Most designs are created / evaluated based off a limited number of borings (maybe 3 to 10 depending on the site/importance/money to be spent) on a highly variable and inhomogeneous material, so it takes a great deal of interpretation in order to come up with a conclusive opinion (and we like to say you get two geotechs in a room and you're bound to get more than two opinions...don't worry, they think we're overly precise).

Sandy / loose soils with high waterlines gain much of their strength from friction between particles and from boyancy of the water. So during an earthquake the friction breaks down (since it's all vibrating) and the soil turns into mud. If you're dealing with a flat surface this can lead to differential settlements (which wreck the structure). Where it gets interesting is when you have a sloping ground surface (such as with marine structures, where you usually have a steep slope). In this case you can get soils failing along curving planes beneath the structure, which really plays havoc with the structure. This is actually an area of a great deal of ongoing research and difficulty in modeling (why i love my job). So basically, these photos allow engineers to not only say "wow, cool!", but to also gain some knowledge correlating predictions from models with real world responses and (hopefully) calibrate those models to prevent future events (civil engineering is a slowly evolving field as we can't crash test our structures (think about that the next time you debug code)).

As far as the chinese apartment building, my best guess is that it's some form of slope failure / inadequate bearing capacity (think tower of pisa). The photos do show the stumps of piles sticking off the bottom of structure. These are likely precast concrete or steel pipe piles, but it's hard to make them out exactly. They appear really short in the photo, but it's very possible that they were broken while the structure overturned. If this were to happen in the US i could expect to read about it from the ASCE, but since this is China, it'll probably just disappear into history.
posted by NGnerd at 9:16 PM on July 10 [3 favorites has favorites]

NGnerd, thanks for the additional information! Flagged as fantastic, because my knowledge of structural and geo. engineering is pretty limited, and your comment made the pictures more meaningful.
posted by filthy light thief at 9:20 AM on July 11

Awesome. I'd been looking for information on subsidence agriculture.
posted by Eideteker at 11:35 AM on July 12