Hurricane Pam: the eye of the storm
The Senate committee investigating federal management of Hurricane Katrina is looking closely at the simulation, "Hurricane Pam". Except for the death toll, "Pam's" scenario was very realistic, but what does that mean?
Deaths:
"Pam" :up to 60,000. Katrina: 1200 bodies found, 2700 people still listed as missing
Flooding:
Pam: 10 to 20 feet of water Katina: up to 20 feet of water
Collapsed buildings:
Pam: 234,000; Katrina 250,000
Residents without power:
Pam 786,000; Katrina 881,400
People evacuated or displaced:
Pam more than a million; Katrina more than a million
Does this prove anything? Well, it's quite easy to make realistic simulations of natural phenomena. Depth of flooding, for instance, depends on geography and wave height. Consequential effects can also be fairly easy to work out - eg number of poeple without power or evacuated.
However the problem is to get people to act on your results. This is because simulations are always a 'what if' activity. It's usually possible to do another simulation, based on different assumptions.
If you don't take to the basic premise (eg "there will be a major hurricane") all the maths in the world don't help much.
Credibility is an asset. Presentation also helps. Simulation techniques can directly affect the latter, but not always the former.
ADD YOUR COMMENT:
COMMENTS SO FAR:
There's some guidance on what makes a credible website at Stanford's Persuasive Technology Lab. (I picked this up in Peter Morville's Ambient Findability, page 110.)
... In fact, we shouldn't forget Robert Cialdini's work.
