Apr

22

From Peter Huber:

None of the computer models predicts that CO2 alone can deliver much warming for a long time to come. What the models predict instead is that a small amount of CO2-induced warming will lead to a slight increase in the amount of water vapor in the air, which will then cause much sharper warming because water is a much more powerful greenhouse gas. But more water in the air means more clouds—water aerosols—which, depending on their shape and altitude, may either warm things further (by reflecting more outbound heat back to earth) or cool things down (by reflecting more inbound heat back into outer space). To get things right, the computer must also correctly model the heat-reflecting capabilities of all the other aerosols, ice, tundra, and forests, and the carbon-sinking capabilities of all the life, soil, land, and water on the planet. And correctly model how every predicted change in climate might subtly change all those factors. Any mistake made in what’s predicted for 2020 will amplify the error in what’s predicted for 2030, and so on. Predictive models of all but very simple, stable systems amplify their own mistakes.

Pitt T. Maner III comments:

Anyone who has tried to play the following computer model game has a feel for the complexity and unintended consequences of manipulating single parameters. I always ended up with an ocean -dominated, Neptune-like world populated by intelligent fishes or a burned up planet hotter than Hades.

Are there any new sim games that are useful training tools for market participants? I remember Capitalism was an early business simulation game.

Being able to play with environmental models or business models and observe the effects of changing parameters would seem to be a good educational tool for many these days and an aid toward deeper understanding.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search