March 16, 2011
What does the future look like: 2 – How disasters become Catastrophes
by Shalva Marjanishvili
In my earlier blog I have attempted to question the current approach to risk, reliability and safety, I have briefly discussed the development of the Japanese disaster into a large scale catastrophe. This is the second part of the series of blogs I am writing in attempt to uncover the answers to the following question: how one disaster could possibly become a large scale catastrophe?
To answer this question we need to explore the unusual correlation between the levels of economic development with the levels of devastation caused by a disaster. To this end I would like to discuss the following:
- On August 29, 2005 Hurricane Katrina caused enormous damage to the city of New Orleans, USA. The flooding was accompanied by fire and caused destruction to the level that it became questionable rebuilding the City was feasible. Hurricane Katrina was considered a 1 in a 400 year event and it was not expected that the levees would contain the water. However, the population was unaware of this. The failure was catastrophic and resulted in the death of 1,500 people. Engineers gave the reason that the hurricane was simply too big and was not considered during the original design.
- On January 12, 2010 Haiti was struck by a powerful earthquake, totally leveling the city of Port-au-Prince and affected approximately 3 million people. The effects of this earthquake are still very much apparent today and were aggravated with an outbreak of cholera. Engineers attributed the failures to the lack of quality construction quality and non-existent modern building codes
- On February 27, 2010 Chile was struck by a powerful earthquake killing approximately 500 people and damaging 370,000 homes. A year later Chile has almost recovered. The remarkable performance in Chile was attributed to most structures being built in accordance with modern building codes.
- On March 11, 2011 Japan was struck by a powerful earthquake triggering a series of disasters, which quickly turned into a large scale catastrophe. This catastrophe will continue and the final outcome will not be known for some time.
I would like to point out an unusually strong correlation between the levels of devastation and levels of economic development. We can group these events into three categories based on development level of each country:
- Countries with highly developed economy: US, Japan
- Developing country: Chile
- Poor country: Haiti
It appears that disasters quickly become catastrophes in either poor countries or in highly developed countries. It is expected that poor countries will perform poorly, however it appears as a surprise that highly developed counties also suffered greatly. Developing countries however, recovered fairly quickly. The question is why natural disasters become catastrophes in highly developed countries?
Most of the engineering explanations of failures seem to focus on the magnitude of the event (too big of an earthquake, hurricane, etc) and that the protective construction or the infrastructure was not strong enough. The time has come to re-evaluate this strength based approach. When structures are designed for a given load (say maximum credible load), and when this load is exceeded (maximum is based on probability of occurrence and therefore it can be exceeded) the structures fail unexpectedly. These sudden failures are amplified in highly developed countries leading to a large scale catastrophe primarily due to the interdependency and relative reliability of structures and infrastructure systems in place.
The current approach to safety is based on cost-benefit studies using theories of reliability developed in 1960 and appear to be acceptable for poor or developing countries. However, these theories fail when applied to highly developed countries such as US and Japan.
The conclusion that can be drawn is that the current reliability approach to the reduction of the risk and disaster management appears to be outdated for highly developed counties and needs to be revised. I will discuss possible solutions in my next blog – stay tuned…