Had an interesting morning here in San Francisco. After staying up late following the recent earthquake in Japan, I was awoken early by a text message from a good friend in the neighborhood concerned about calls and texts she was receiving from friends about a tsunami. She lives on the top of a hill located well outside our 20 foot run up zone and I told her I'd be by for breakfast since her home was safer than mine which is only a block outside the run up zone.
A few minutes later, I received a call from another friend in Iowa advising me to start filling my bathtub and pots and pans in case we lost our water supply. After chuckling a bit, I reminded her what I did for a living.
These contacts from concerned friends reminded me once again that we really don't do a good job of public warning. Amanda Ripley points out in her excellent book, The Unthinkable, that our warnings contain fact and actions but don't always stress "why". In this case, the local authorities provided a timely warning with good information and appropriate actions. What they didn't do was couch the warning in a way that allayed concern.
Our tsunami warning include a potential for a 2-3 foot wave. The problem is the average citizen has no idea what a 2-3 foot wave would do. They tend to think in terms of a tsunami that would wash over the entire peninsula, something that is not even geologically possible given the shape of our shoreline. A wave of the size predicted would probably not even have gotten over our seawall. The EOC staff took a reasonable precaution by closing the Great Highway adjacent to the seawall and elected not to evacuate the immediate area which demonstrates they understood the risks. Unfortunately, this information wasn't really shared well with the public and the media.
This should not be taken as criticism of my local colleagues who did everything appropriately according to our accepted practices. Instead, I'm suggesting that we need to rethink the way we issue warnings and the mechanisms we use to distribute those warnings. We need to share our reasoning with the public rather than just telling them what to do. Our studies show that people don't really trust our messages and will seek verification from other sources before acting. Understanding why they should take the actions we recommend could help reduce this lag.
So the next time you need to issue a warning, give some thought to explaining what the potential threat really means. Your warnings will be a lot more effective.