Will they hold? A discussion of possible levee damage in and around New Orleans following Katrina / Rita

by Russell D. Hoffman

September 29th, 2005

Immediately after Katrina, this author was one of many who felt that New Orleans could have been "dewatered" much faster than the officials were saying it would take.

It's true that more and/or larger temporary pumps could have removed the 27 billion gallons of water which flooded New Orleans after Katrina more quickly.  But swift dewatering would have increased the risk of another levee failure by reducing the "hydrostatic pressure" being exerted on the weakened levees by the water sitting on the low, normally dry, side.  As long as water presses against both sides of the levee, it can actually help support a damaged levee.

Furthermore, the levees were (obviously) holding -- and the tops, at least, were drying.  Then along came Rita.

Around the world some recent causes of levee failures include river floods, tsunamis, and even beavers, who dug into a dike to make dens and caused a 12,000 acre flood in California.

There are nearly 10,000 miles of levees in America.  The U. S. Army Corps of Engineers (USACE) is responsible for maintaining about 3,000 miles of these levees, of which 350 miles are in and around New Orleans.

When a levee fails, IMMEDIATE action is just about the only thing that can stop a small breach from getting larger.  In 1953, a Dutch boatman saw a break forming in a levee, rammed and sank his boat in the forming breach, and saved one of Holland's largest cities (source: NY Times).

But being in the right place at the right time is rare.  When immediate action isn't possible, swift action might prevent the breach in the levee from growing on either side, making subsequent repair easier and cheaper.  And it might keep the flow rate from increasing.  But swift action is less successful at stopping the flow.  After a breach, that doesn't usually happen until the water levels are equalized on both sides of the levee.

Water on the wrong side of a levee can be very damaging.  The cement walls (if there are any) at the top of the levee are supported on dirt and rock normally, and perhaps sand and logs and old buildings and who-knows-what-else.

When there is standing water on both sides of a levee for days and weeks on end, the levee gets muddier.  Large amounts of seepage through the levees (which normally happens all the time, a little at a time) can change the structural integrity of the system.

Sometimes water seepage can create "sand boils" near the "toe" of the levee  (the "toe" is the lowest, furthest-out part of the dry side of the levee).  A sand boil represents the head of a long "snake" or cone of water, which has tunneled all the way through the levee and is coming out the dry side, as a wet spot or even a water spout.  Sand boils are often the first sign of trouble, for one type of levee failure.

In New Orleans, sand boils might have been starting to form during Katrina's storm surge when another part of the levee system failed.  There is no way to be sure, since the "evidence" has been flooded and destroyed.

Levee failure can also occur from long periods under extreme and uneven pressure -- a storm surge may last only a few hours, or it may last much longer.  During Katrina the water reached flood stage many hours before the levees failed, and the pressure against the levees was probably as much as they had ever experienced.

Just before failure from too much pressure for too long, sometimes snakes and other animals suddenly appear, rushing away from the levee.  The ground sometimes swells noticeably before bursting.   In New Orleans, anyone who might have seen these things would probably have been washed away and drowned shortly thereafter.

Sometimes, a levee failure is caused because a "pit" or excessively deep channel in the base of the high (wet) side of the levee allows water to seep horizontally along the water table -- perhaps thousands of feet -- and weaken the toe side of a levee.

During Katrina and again during Rita, many, if not all, of these problems might have occurred.  "Scouring" definitely occurred when the levees were "overtopped" by the storm's waves, including "seiching" (large waves) on Lake Pontchartrain which then ran up and overtopped the canals.  After Katrina and before Rita, the tidal surge and the lake waves were blocked from entering two of three canals by urgently-built blockades which were constructed using bridges which already went over those canals.  The third canal (the Industrial Canal) had no such bridges, and thus had no such blockage built against the storm surge and waves, and some of the temporarily repaired sections in that canal failed during Rita.

During Katrina, ditches more than a yard deep and a yard wide, and thousands of feet long, formed at the base of the top wall on the "dry" side of the levees (see link to USACE photo, below).  After overtopping, the water rushes down the "dry" side of the levee and can do additional damage, including hiding any sand boils that might have been starting to form.  The worst kind of levee damage is hidden levee damage.

Huge volumes of water went over, under, and through the levees, so it will be very difficult to assess their current status.  The only way to safely lower the water level was/is to do it slowly, and watch what moves.  Engineers must constantly monitor the hundreds of miles of dirty, mud-covered levees for damage as they naturally dry out -- slowly.  If anything starts to wash away, it must be fixed immediately or a new catastrophe could occur -- as we saw happen with Rita.

Russell Hoffman
Concerned Citizen
Carlsbad, CA

The author, an independent researcher, is grateful for information provided by the Army Corps of Engineers.  A lot of additional research went into this article and mistakes, errors, or omissions are solely the responsibility of the author.

Photo of damage to the 17th Street Canal in New Orleans, showing the "scouring" effect from "overtopping"  (Source: John Rickey, USACE):
http://www.animatedsoftware.com/environment/katrina/usace%20photos/Steve%20Stone%20photos%2045.JPG

--------------------------------


Additional background information (compiled by Russell Hoffman, who is solely responsible for its contents):

The re-flooding during Rita was apparently due to overtopping and then washing away of the temporary levees that were put in place after Katrina. This appears to be due to an engineering mistake, but in the aftermath, it should be fairly easy to analyze, since these were probably the most photographed and observed temporary levee repairs in history.   Had Rita come into Lake Pontchartrain as a Category 4 or 5 hurricane, this author was ready to RECOMMEND re-flooding the -- now presumed empty -- city intentionally to protect the levee system from stress which would almost inevitably lead to failure anyway, but in the event, Rita did not produce nearly the size of the storm surge that Katrina produced and most of the levees held.

Flood-control professionals have a special name for New Orleans: The Big Pumper.   But those massive pumps -- 174 of them -- are not able to keep up with 200-foot wide levee breaches.  No reasonable system could do so.  For that reason, you could say that New Orleans was a catastrophe waiting to happen.

There is a saying among knowledgeable levee specialists that levees protect property, evacuation orders protect people.  When engineers build a levee system to withstand a "Category 3" storm and then it is subjected to a Category 4 or higher storm, there's a chance it will fail.  Regarding the levees around New Orleans during Katrina, one could say: "they almost held against a force far stronger than the system was designed to protect against."

Historical note: Nearly every levee failure is followed by talk of whether or not the levee should be repaired at all.

A point of order:  No matter how many times you hear it, the New Orleans levee system was not actually built to a "Category Three" storm resistance level.  It was built to resist what was then known as a "Project Flood Storm."  The Sanford System Scale for measuring the strength of hurricanes (where 5 is the worst) had not yet been created, but a Project Flood Storm is generally considered equivalent to a Category 3 hurricane.

Modern technology allows embedded communications cables inside a levee.  The cables also act as sensors.  The information is immediately displayed in a control room, with an indication of the strength of the failure as well as its exact location.  This system has not been retrofitted into New Orleans' century-old system of levees.

It is also true that dewatering New Orleans slowly allows the Corps to be mindful of the human remains, which must be carefully and respectfully handled as the process goes on.  The extra effort might tell someone, somewhere, what happened to someone they love.

Some related topics not discussed in this set of articles include questions about the precise quality of the levees to begin with, whether any levees were blown up by nefarious souls in order to flood some areas while (supposedly) protecting other areas, the reason the temporary repairs after Katrina were inadequate for Rita, and all the political and logistical questions regarding the timing of emergency/repair work after these terrible disasters.

If you want more information about New Orleans' levee system, or have a suggestion for the Army Corps of Engineers, here's a URL you can visit:

www.mvd.usace.army.mil

On the left, is a button labeled GOT A QUESTION, where you can type in a question and the Army Corps of Engineers will read it and email you back a response as quickly as possible.  A USACE representative from the Mississippi Valley Division (who patiently answered about 50 questions I bombarded him with), told me that the idea for dropping huge sand-bags into the New Orleans levee breaches actually came to the Corps through their online system.

--------------------------------

From NY Times article by Bill Broad, Sept. 6th, 2005:

"In humanity's long struggle against the sea, the Dutch experience in 1953 was a grim milestone. The North Sea flood produced the kind of havoc that became all too familiar on the Gulf Coast last week. When a crippled dike threatened to give way and let floodwaters spill into Rotterdam, a boat captain - like the brave little Dutch boy with the quick finger - steered his vessel into the breach, sinking his ship and saving the city."


===========================================================

Copyright (c) 2005 Russell D. Hoffman but may be freely distributed if left unchanged.  The first portion may be distributed separately if proper authorship notes are included.  This article can be found online here:

http://www.animatedsoftware.com/environment/katrina/LeveeDamage20050929.html

email: rhoffman@animatedsoftware.com