In a report issued just this week (20 Sept 2011), the National Oceanic and Atmospheric Administration (NOAA) said that many who died as a result of the EF-5 tornado that struck Joplin, MO, on May 22, 2011, might have lived if they had paid attention to warnings issued in advance of the storm. The Joplin storm resulted in the greatest loss of life in one location due to a tornado in almost 60 years, according to the National Weather Service. The number of dead currently exceeds 160.
However, the size and fury of the storm was such that even those who heeded warnings meant that your chance of survival was slim, regardless of where you were at the time it struck. Buildings that would be considered safe shelters for most storms simply weren’t strong enough to withstand this one. The storm system was so powerful that St. John’s Hospital, a pretty large and substantial building, had to be condemned. The building was moved on its foundation! (Imagine a force of that magnitude being exerted on your office building, warehouse, or home.)
And, lest we forget, Japan dealt with an even larger catastrophe less than three months earlier. There, two natural forces – an offshore earthquake, spawning a massive tsunami – combined to cause a meltdown at a nuclear power plant. As long as it will take Joplin to rebuild, efforts there will be dwarfed by those around the northeast coast of Honshu. In either case, much will never be recovered.
We all know how risk assessment and risk management are done. We identify risks, determine their likelihood and level of impact, prioritize them, and do what can reasonably be done to eliminate, minimize, or mitigate them. Sometimes, we do nothing and simply accept the risk. Generally, this system works well. But what about our ability to assess the likelihood that a high-impact (i.e., catastrophic) event will occur?
It seems we have to reevaluate the way we determine the likelihood or probability of certain events, given recent history. In the case of tornadoes and thunderstorms, 2011 appeared to be a particularly brutal year in the USA. With major earthquakes in Japan and New Zealand also taking place, some reinsurers have taken a beating. They must be taking a hard look at their risk models, don’t you think? Or, do you think they’ll chalk these disasters up to “a number of one-time events”?
Are natural disasters trending upward in frequency and severity? We need a better answer to this key question than “We’re not sure.” Improving our ability to assign more accurate probabilities to critical events – hence, improving our planning for and response to those events – is nothing less than a life-and-death issue.
- “Aerial Photographs of Joplin (MO) Before and After the Tornado”, New York Times, 25 May 2011 – http://www.nytimes.com/interactive/2011/05/25/us/joplin-aerial.html
- “Satellite Photos of Japan, Before and After the Quake and Tsunami”, New York Times, 15 March 2011 – http://www.nytimes.com/interactive/2011/03/13/world/asia/satellite-photos-japan-before-and-after-tsunami.html
- Risk Assessment Matrix, Texas State University-San Marcos Student Organizations Council – http://www.studentorgs.txstate.edu/forms/risk_assessment_matrix.pdf
- Berkowitz, Ben, and Mortimer, Sarah, “Analysis: Reinsurers at Risk if Hurricane Season Big”, Reuters, 1 July 2011 – http://www.reuters.com/article/2011/07/01/us-insurance-florida-idUSTRE7602GD20110701