New Orleans Levees:
Hello and welcome to the second episode of Failurology; a podcast about engineering failures. I’m your host, Nicole, and I’m from Calgary, Alberta. I have over a decade of experience in the field of mechanical engineering in building science; directing the construction of plumbing, heating, ventilation, and air conditioning systems in various high-rise, multifamily residential and commercial buildings across western Canada. I love reading about engineering failure case studies. Not just the science behind what went wrong, but what we can learn as an engineering community to prevent future failures from happening. Starting a podcast has been a dream of mine for a long time. And I want to thank each and every one of you for listening. I hope these stories interest you as much as they interest me. If you want to see photos from this week, check out the link in the show notes of this episode.
This week in engineering news; a new paint so reflective, it can cool a surface to almost 2C below the surrounding temperature. As global energy usage continues to increase, passive cooling without using any energy is crucial. Developed by Xiulin Ruan at Purdue University, the paint was able to reflect 95.5%of solar energy, compared to other heat reflective paints which reflect about 80-90%. And the Superwhite paint is thinner and cheaper. Ruan estimates the paint could save about $50/month in cooling costs on a 200 square meter home when compared with existing heat reflective paints. I’ve heard about painting roads and sidewalks with traditional white paint, but this is a whole new level. Check out the link in the show notes for this episode if you want to read more on the superwhite paint.
Now on to this week’s engineering failure; the levees in New Orleans during Hurricane Katrina. American jazz pianist Ellis Marsalis said that “In New Orleans, culture doesn’t come down from on high; it bubbles up from the streets.” And after visiting the city myself in 2017, I have to agree. It’s a mixing pot of cultures that come together near the mouth of Mississippi river, creating a totally different way of life unlike anywhere else on earth. The loss of life experienced during and after Hurricane Katrina was horrendous and for the most part preventable. This is a really sad story, and researching it was quite upsetting. But there’s a lot to be learned here to prevent this type of disaster from happening again.
New Orleans is located 170km upriver from where the Mississippi drains into the Gulf of Mexico. To the north is Lake Pontchartrain, which is not technically a lake, but an estuary open to Gulf. The water is brackish, which means it’s a mix of salt and fresh water. To provide access across the lake, there is a 38km causeway which is the longest continuous bridge over water in the world. According to Google maps, it’s roughly a 30-minute drive across Lake Pontchartrain. The Mississippi River runs west to east through the middle of the region, south of downtown New Orleans. There are several canals from the City that run north to pump rainwater into Lake Pontchartrain, or other nearby bodies of water.
The City of New Orleans, which is built on marshland, sits roughly 2 to 6 metres below sea level, depending on which part of the City you're looking at. It’s also subsiding or sinking at an average rate of 3.8 to 5 millimetres per year. But, it didn’t start out this way. In order to expand and develop the City over the last 300 years, a significant amount of water had to be removed from the original marshland. In addition, as more levees and structures are built upstream and around New Orleans, fresh sediment deposits are not able to keep up with the rate of subsidence. Over time, the removal of groundwater and lack of sediment deposits cause the soil to compress and subside or sink at a rate faster than can naturally be replenished.
Now that you have some understanding of the geography of New Orleans, what about hurricanes? How they are tracked? And how they are classified? Atlantic hurricane categories are measured using the Saffir-Simpson hurricane wind scale. The US National Hurricane Center, located in greater Miami Florida, officially monitors and tracks tropical weather systems in the Atlantic and Eastern Pacific Ocean basins. A category 5 hurricane is classified as a storm having a minimum one minute sustained wind speed greater than 254km per hour. Over 30 tropical cyclones from 1851 to now have reached category 5 levels. The first recorded category 5 hurricane hit Cuba in 1924; before then, the equipment typically blew away before wind speed reached category 5 levels. 2005, the year Katrina hit, saw 14 hurricanes, this is the most hurricanes in a single-season record. And 2005 exceeded the previous record, set in 1969 of 12 hurricanes in a single season. There have been worse hurricanes, in terms of wind speed, before and after Katrina. But the storm, combined with levee failures, makes Katrina one of the most expensive natural disasters to date.
While researching this episode, I read a lot about hurricanes. And I’m still continuing to follow tropical weather systems around the world. Hurricane Eta and then Hurricane Iota have devastated countries along the Caribbean Sea this month. Most notably Nicaragua and Honduras, but also Guatemala, Mexico, Venezuela and Columbia were hit hard. Both hurricanes were a category 5 when they hit the region. If you want to help, the Red Cross is a good place to start.
Each year, 21 names, each starting with a different letter and arranged in alphabetical order, are used to name the storms as they form. If there are more than 21 storms in a year, they use the Greek alphabet to name them. Each list is recycled and reused every 6 years. When a storm makes a significant impact, to the point that reusing it would be insensitive, it’s retired. There were five names retired in 2005, with Katrina being one of them.
Hurricane Katrina formed on Tuesday, August 23, 2005, as a tropical depression over the Bahamas before heading east, gathering strength, and being upgraded to a tropical storm. By the time the storm reached landfall in Florida, it was a category 1 hurricane, with wind speeds ranging from 119 to 153 km per hour. After less than 8 hours over Florida, the storm intensified further as it made its way across the Gulf of Mexico. By the time it reached Louisiana and Mississippi at 6 am on Monday, August 29th it was a category 5 hurricane.
Atmospheric pressure in New Orleans is approximately 101.6 kilopascals or kpa. The hurricane reduced the pressure to 91.6 kpa. For every 3.4 kpa that the atmospheric pressure drops, water levels rise 300 millimetres. That means that sea levels would have already been almost 1 metre higher than normal based on the pressure of the storm alone. In some areas along the Gulf, the Katrina storm surge reached 6m above sea level and along Lake Pontchartrain, to the north, it was about 3.5m above sea level. Heavy rainfall made it worse.
So how does New Orleans protect itself from tropical weather systems? Construction and design of the levee system, under the United States Army Corps of Engineers, goes back as early as the 1920s following the Great Mississippi Flood of 1927 which initiated the Flood Control Act of 1928. Maintenance was overseen by local levee boards depending on the location of the flood wall. This means that there isn’t one group that looks after levees; it is divided amongst many federal, state and municipal groups depending on location, funding and when the wall was constructed. In addition, it was never viewed as one cohesive system, only parts and pieces. Now, while I somewhat understand how this can happen, as the levee system evolves and expands over time, this creates a scenario where no one is really taking responsibility for the integrity of the whole system and looking after the City of New Orleans. When scopes are not clearly defined, or overlapping scopes are assigned to more than one group, it becomes a finger-pointing game if anything goes wrong. Which is pretty much what happened here.
The flood protection system surrounding New Orleans was under and incorrectly designed. I think this is evident based on the amount of destruction. But it wasn’t just one thing that went wrong, there were so many issues and it’s a wonder any of these walls lasted as long as they did. There were problems with the criteria they used to design the floodwalls, the walls were too small and erosion at the base of the wall was not adequately accounted for, the datum they used to set the wall elevations wasn’t right, and on top of that, the pumping stations didn’t withstand the storm and not only couldn’t remove water but in some cases became another path for water to enter the City. It was bad, so bad. And I’m going to get into all of that in more detail, right now.
The initial design parameters for the levee and wall systems were intended to protect the city from a category 3 storm based on standard project hurricane data, rather than a probable maximum hurricane. It doesn’t quite make sense to me, why would they only design for a category 3 storm, when category 5 storms are a high possibility. I realize that always designing for worst-case scenario is expensive and sometimes excessive. But in a city below sea level, I would have expected to see a more robust basis of design. That being said, some of the levees failed even at water levels below what would be expected during a category 3 storm; so it’s tough to say if a category 5 storm design would have made that much of a difference based on all the other design flaws. Also, the design criteria had not updated as new data was made available and some older walls were not updated or improved to match the height or strength of adjacent, newer sections.
Mean sea level, which is assigned an elevation of zero, was supposed to be used to determine the elevation of the flood control systems. But, the designers and installers used land-based elevation, incorrectly assuming it to be the same as mean sea level, resulting in the structures being built roughly 1/3 to 2/3 metres below the intended elevation. This was before we all carried the internet in our pockets, but you’d think someone would have at least double-checked.
There were three types of flood walls that exist around the City. There are pictures on the podcast webpage, link in show notes if you want a visual of each wall type. Firstly, there are I-walls, which are vertical corrugated metal walls, kinda like those used in shipping containers, with Z-shaped steel pilings and compacted soil at the base of the wall to support it and hold it in place. Then there are, T-walls, concrete walls built like an upside-down T, with soil and concrete pilings to support the base and hold the wall in place. T-walls are more expensive to construct than I-walls and therefore, were less common. And lastly, levees are compacted earthen material piled to act as a dam. Which is` a fancy way to say they built a dirt hill at the water’s edge to keep the water out. The taller the levee needs to be, the wider the bases are to accommodate. The hill-style levees are not as common in denser areas where space is limited. In many locations throughout the city, where existing levee heights needed to be raised, they built an I or T wall into the existing levee.
There are over 550 kilometres of floodwalls or levees in and around New Orleans, with 450 kilometres of those falling under federal responsibility. Nearly 170 kilometres of the walls were damaged during Hurricane Katrina– that’s about a third of the walls. Could you imagine building a building with a 30% chance it would fall down? I think not.
There were 3 main design flaws that caused the walls or levees to fail. In addition to overtopping which was the most common cause of failure, water also caused deflection in the wall, creating a water-filled gap that eroded the base, and lastly, soil strength was underestimated and seepage undermined the wall. All of these are in some form or another, resulting from soil erosion.
In several locations, the flood water level was half a metre above the tops of the walls and levees. At the I and T walls, the water washed away compacted soil on the land side of the wall, causing the wall to slide and collapse. At earthen levee locations, the water just overwhelmed the embankment. The compacted clay and grass levees withstood the storm better, while the silt and sand embankments just washed away. There were more than 20 locations of failure along the Mississippi River-Gulf Outlet in Saint Bernard Parish, flooding the entire parish and East Bank of Plaquemines.
The water-filled gap was the second most common cause of failure and significantly impacted the areas adjacent to the 17th St, Industrial, and London Avenue canals. The original engineers did not account for deflection of the floodwall during a storm event. When the flood walls deflect it creates a gap between the compacted material at the base of the wall, and the wall itself. This gap then fills with water and begins to erode the compacted material at the base of the wall, causing the wall to slide. Where the water-filled gap occurred, the safety factor of the wall was reduced by at least 30%. In several locations, the flood water levels were almost 2 metres below the top of the wall at the point of failure.
And the third most common cause of floodwall failure, also impacting areas adjacent to canals, was the engineer's overestimate of soil strength, this led to water seepage underneath the walls in many locations. When evaluating the soil strength at each wall location, engineers took several samples of soil along the length of the intended flood wall and averaged out the data to determine the strength for that section of wall. However, some areas along the wall had weaker soil than others, that was not accounted for. Furthermore, samples were only taken at the wall locations and not in the areas surrounding the walls where weaker soil could impact the integrity of the base of the wall. During the storm, water seeped into the soil below the wall, forming cracks in the weaker marsh layer above the soil on the land side of the wall. The soil erosion ultimately undermined the wall and caused it to slide out. In several locations, the flood water levels were almost 2 metres below the top of the wall when the failures occurred.
In addition to the walls themselves, there were other problems. Because the city is below sea level, all rainfall has to be pumped out through the canals. These pumping stations were not designed for hurricane or even overland flooding, only rain and minimal seepage. Almost 100 pumping stations have been constructed over the last 100 years; some new, some still original. The pumping stations were typically designed for a one in 10-year 24hr storm, or about 225 millimetres of rainfall and most have backup power generators. During and following Hurricane Katrina, the pumping stations became inaccessible, and many operators had been evacuated, leaving the stations inoperable, and many ended up flooding. Older stations were not designed to withstand wind and water forces and received a lot of damage that took time to rebuild. Even if pumping stations were working, there was nowhere for the water to go. The canals used to pump water out were damaged and already full. Due to lack the of check valves or backflow preventers, in some locations water back flowed through the canal outlets to the pumping stations and into the city.
There’s more, there are several road, rail and pipeline crossings throughout the flood mitigation system. Areas where transportation routes have to cross the floodwalls. In these locations, closure systems are provided to maintain a continuous flood wall. But many were either missing or inoperable.
As you can see, many things went wrong, and New Orleans was left to deal with the consequences. Some homes were flooded to the rooftop within minutes. Others were flooding at a rate of 300 millimetres every 10 minutes. Seven oil spills occurred during the storm and the runoff from over 50 US Environmental Protection Agency contaminated cleanup sites polluted the floodwater.
By September 1st, two days after the hurricane hit, 80% of the city was still sitting under 2-3m of water. In some areas, it took weeks to subside. Damage is estimated at 125 billion USD. But that doesn’t even compare to the over 1,000 deaths, some people were never found. Over 500 thousand people left the City, some to never return. Aside from drowning deaths, residents were hurt trying to escape through their roofs, infections occurred from exposure to floodwaters, and then these infections spread through crowded shelters, and access to medical facilities was limited. Electricity was down for weeks and drinking water was compromised.
In 2007 the American Society of Civil Engineers called the flooding “the worst engineering catastrophe in US history”. Now, this podcast is about the engineering shortfalls, and I don’t want to get too political. But I think would be a disservice to the people of New Orleans to talk about Katrina without mentioning the government’s response. Here are some of the highlights or lowlights if you will. Despite being aware of the hurricane's path to New Orleans on August 27th, the evacuation orders weren’t issued until mid-day August 28th, the day before Katrina hit. Meaning they lost an entire day of time to evacuate the City. President Bush, Louisiana Governor Blanco and New Orleans Mayor Nagin were aware before the storm that the levees could overflow. On August 27th, Amtrack moved equipment out of the City, offering spaces on those trains to the City for evacuation. The city declined and the trains left empty. When the evacuation orders finally came on the 28th, there was no plan for homeless, low income or sick people to get out; which represented over 100,000 residents. Most of these people were Black. The evacuation plans relied heavily on the use of cars, not considering that many residents, especially in areas with highest risk of flooding, relied on public transportation. Many white majority communities were able to escape. This is a very clear example of environmental racism.
Various officials from all levels of government could not agree on a specific solution to handle the storm and its aftermath, leading to instability and inaction which left many in New Orleans stranded. They, like the levee engineers, also underestimated the strength of storm and devastation that would hit the City. Hundreds of first responders volunteering to help were re-routed to Atlanta for two days, two days of presentations on sexual harassment and a history lesson on FEMA, the Federal Emergency Management Agency. Later on, then director Michael Brown resigned over his handling of the response. Because less people were able to effectively evacuate than they had expected, the supplies FEMA sent were inadequate. Repair crews restored power to a nearby pipeline before bringing power back to two rural hospitals. You can thank Dick Cheney for that one.
As a side note, in 2010 Ray Nagin was convicted on twenty charges of wire fraud, bribery and money laundering related to bribes from city contractors before and after Katrina. He was released to house arrest in April of this year due to COVID spread in prisons.
There is no excuse for the negligence that occurred resulting in the loss of life and damage caused by Hurricane Katrina. That being said, I do want to talk about risk management. Removing all risk is nearly impossible and very expensive. As a collective of federal, state and city officials, as well as the people of New Orleans, an agreement has to be made to establish the amount of risk that is acceptable. Obviously, the flood mitigation system needs a serious overhaul. A better evacuation plan is necessary to reduce loss of life. And communication between the flood mitigation engineers and the residents of New Orleans needs to be improved. The people of New Orleans should have known what everyone else seemed to have known, that the levees could fail.
The United States Army Corps of Engineers did an analysis following Hurricane Katrina and found that had the levees not failed, and the pumping stations remained operational, almost two-thirds of the deaths would not have occurred. Two thirds. This is especially interesting since they have not really taken full responsibility for the failures they designed.
The American Society of Civil Engineers review panel made the following recommendations in their 2007 report. Public safety, health and welfare has to be top priority. Quantify, regularly evaluate and modify the assessment of risk. Communicate with communities and determine their level of acceptable risk; manage those risks as required. Assign the responsibility of design, construction and maintenance of the flood mitigation system to one party. Independent reviews of design, repair, and maintenance are required to maintain integrity and operation of these life safety systems.
So what's next for the city? What's happened since 2005? Well, even now the City of New Orleans is still recovering. I don’t think you ever just walk away from something like Katrina. Over 55 levee breaches were repaired following the storm. The Army Corps of Engineers continues to work on mitigating risk. Repairs were done quickly to all major breaches. On Sept 23, 2005, high water caused by Hurricane Rita flowed over a temporary closure on the Industrial canal. The structure held but part of eastern New Orleans re-flooded. The temporary repairs were complete by January 2006. After Katrina, the city’s population dropped by about 30%. 15 years later, it’s just about back to pre-Katrina levels. The Army Corps of Engineers brought in a group of Dutch engineers for evaluation, design and construction management of the levee system. I will be covering the Netherlands canals on a future episode; so stay tuned for that.
Almost 15 billion USD has been spent on flood mitigation in New Orleans since Katrina. Some major projects include. The Gulf Intercoastal Waterway West Closure Project which can pump over 63,000 litres per second. That’s over 200 bathtubs per second. The IHNC Lake Borgne Surge Barrier, which the longest storm surge barrier in the United States. And the Seabrook Floodgate at the connection of Lake Ponchartrain to the Industrial Canal. As well as hundreds of levee and pump improvements.
A storm similar to Hurricane Katrina is possible in the next 50 to 500 years. I for one hope that the City is better equipped with the tools required to handle the next one. Check out the podcast page, link in show notes, for photos from this week’s episode. And if you want to chat with me, you can find me @failurology on Twitter. Thanks everyone for listening. And tune in next week to hear about the Hyatt Regency Collapse in Kansas City. Bye everyone, talk soon!