99

DESIGNED TO FAIL

Floodgate

2005

On July 31, 2006, The Independent Levee Investigation Team released the results of their investigation of the cause of the August 29, 2005, New Orleans levee failures during Hurricane Katrina. It would be satisfying to blame those failures and floods on crooked politicians or the Army Corps of Engineers. But the reality is that the original mistake that led to all the other failures came long before and was a simple mistake. The levees were designed using a model storm to test their strength and survivability. This mathematical testing of the levee designs was called the Standard Project Hurricane. That was the problem. The testers used only standard hurricanes. The model was simplistic and missed some of the effects of the storm. Worse yet, the math used to determine the power of the standard hurricane excluded the data on the most extreme storms. So it is not hard to understand why levees that were designed using the standard model failed when faced with a much stronger than standard storm. Perhaps the math should have also taken into account that the chances of a 100-year hurricane occurring any year are the same for next year as they are for 100 years from now. Such storms can come anytime.

Had the math worked and the assumptions made proved correct, 2,000 people need not have died and tens of thousands would not have been made homeless. The failures made before, during, and after Katrina affected the entire United States. There were many mistakes made on every level, but the one that started it all was made by a mathematician in some quiet design and testing office years before.

If you find an error please notify us in the comments. Thank you!