RE: engineering questions re. Katrina

From: Kevin Woods <>
Date: Mon Sep 26 2005 - 00:11:42 EDT

Here's a minor point about levees, not especially related to New Orleans,
but perhaps of interest. This was from a conversation I had with a Corps of
Engineers officer at the '93 flood on the Mississippi. I'm a biologist, not
an engineer, but I suppose he knew what he was talking about.

Levees are not designed to hold water permanently, like a dam with a packed
clay core. They are permeable, but they work because as water seeps through
it is deflected downward. This is sufficient for the duration of most flood
events. However, the water will eventually seep through horizontally to the
dry side of the levee. Once the soil is completely saturated, a blow-out can
easily occur. The areas about to fail can be recognized because the soil
becomes like jelly.

I was part of a crew stacking sand bags not only on top of a levee, but also
on the "dry" side trying to stall the seepage.

So there are limitations, in addition to height.

-----Original Message-----
From: [] On
Behalf Of Keith Miller
Sent: Friday, September 23, 2005 10:07 PM
Subject: Re: engineering questions re. Katrina

George Murphy wrote:

> ----- Original Message ----- From: "Bill Yoder"
> <>
>> In my mind the biggest immediate issue in this disaster is that the
>> levees broke. Why did the levees break? They should not have broken.
>> I suspect that the amount of water that flowed over the levees would
>> not have caused anywhere near the amount of flooding that was caused
>> by the water that flowed through the breaches.
>> Apparently the water that overflowed the levees ran down the other
>> side and washed out the foundation of the levees. This soon weakened
>> them and they failed. Is this a reasonable way to design a levee? IMO
>> this is represents a major failure in levee design. I am confident
>> that levees can be designed to preclude this.
> Apropos this could someone in the know clear up a point? On 60
> Minutes ~1.5 weeks ago someone who sounded as if he knew what he was
> talking about argued that it wasn't the _levees_ that had failed but
> the _floodwalls_ - i.e., relatively narrow flat walls rather than the
> broader earthen levees. I gathered - but was not sure - that these
> floodwalls, not having been tall &/or strong enough, broke down, &
> that the resulting flooding then caused failures in the levees. Is
> this right, was he wrong, or was I just misunderstanding things?
The NYT article referenced by Rich Blinne does a pretty good job of
explaining things but omits one thing: Bill or Rich, please correct me
if I'm wrong but my understanding is any levee is not built to
withstand water pouring over the top of it: it will erode, and quickly.
  The aim is to dam up water, and it is very difficult to build an
earthen dam that can stand being overtopped. The flood walls as
described in the article (concrete "teeth" anchored in the top of a
levee) are even more susceptible if topped b/c they lack wide bases.
The levee failures we saw in the Mississippi and Missouri river systems
in 1993 all started by water topping the levee: the resulting streams
rapidly eroded gaps in the levees.

The talk on levees, however, obscures what Glenn described so well: the
land is below sea level and will keep on sinking, and there is a
continued demand for money if the only "solution" offered is to keep
building the walls higher. Houses on stilts and the river freed to
flood and deposit more mud, as it used to, is the only long-term
solution I can envision working. That and eliminate canals conducting
water above and thru populated land below sea level!

Holland is an important parallel. Why doesn't it look like a disaster
waiting to happen? Or does it, if you live in Europe? A couple
details: 1) is Amsterdam below sea level? My understanding is most of
the lowlying land is farmland, not heavily populated. 2) what's the
strongest storm one can expect in the Northeast Atlantic? Hurricanes
very very rarely make it to England, and England is itself a buffer for
France, Belgium and Holland from nasty high-wind storms. But
hurricanes are a fact of life in the Gulf. Amsterdam is very much
older than New Orleans--a longer history to count storms over.


Ruth Douglas Miller
Associate Professor
Dept of Electrical and Computer Engineering
261 Rathbone Hall
Kansas State University
Manhattan, KS 66506-5204
ph 785-532-4596
fx 785-532-1188
Received on Mon Sep 26 00:14:45 2005

This archive was generated by hypermail 2.1.8 : Mon Sep 26 2005 - 00:14:45 EDT