Slashdot Mirror
Simulate "The Day After Tomorrow" On Your PC
kpearson writes "climateprediction.net, a distributed computing project to predict Earth's climate 50 years from now, has a new add-on project to study THC slowdown (how climate might change as CO2 changes in the event of a decrease in the strength of the thermohaline circulation). This kind of rapid, extreme climate change is shown in the movie The Day After Tomorrow, in which New York City is treated to a 10,000-year-long ski season. Anyone can download the project's client software and participate in the simulation. climateprediction.net was previously mentioned in the September 13, 2003 article
Distributed Computing and Climate Change." Clients are available for various varieties of Microsoft Windows, but none are listed for other OSes.
That movie is expected to draw furhter focus on the environment and specifically global warming. This is both good and bad, as too much focus on the environment can draw focus from other points of interesets. Danish Scientist "Bjorn Lomborg" (one of Time Magazines top 100 important persons) has been warning politicians to not forget other points of importance, such as healthcare and clean water. I hope this does not distort the vision of politicians around the globe, lets not forget how er priotitize.
I admit I'm a bit clueless here, but AFAIK climate modelling deals with overall changes at a high level e.g. "It's going to be colder in the north atlantic by about 2 degrees on average in a few decades" and the like.
As opposed to weather prediction which says "It will rain in this spot on this day"
Just because we can't predict the 'noise' in the short term doesn't mean we can't determine overall changes long term.
"There is nothing wrong with the planet. The planet is fine . . . been
here 4 1/2 billion years. We've been here, what, a 100,000 years, maybe
200,000. And we've only been engaged in heavy industry a little over 200
years. 200 years versus 4 1/2 billion. And we have the conceit to think
that somehow we're a threat? The planet isn't going away. We are."
-George Carlin
Is it true that more people vote for the winner of American Idol, than vote for the president? -Ali G.
Hmm...
;-)
Earth may have a self-regulating system we do not yet (fully) understand. When you say its a robust system, you are right. Our climate models may well be worth squat, right again.
Dinosaurs did quite well for quite a long time, until they could...
However, the issue is not whether human interventions can fully derail earth's climate or only tackle the parameters a bit.
The point is, that the changes in the weather we as a civilisation started will most certainly affect our way of life as we know it. Human is considered the most adaptable species, present in all kinds of environments. I cannot argue whether we are going to drive ourselves into extinction or not by what we did to the climate, but surely life (and survival) in either dry deserts, frozen glaciers or stormy swamps has little to do with the current trends in economy. There is a vast difference between a habitable and comfortable climate, you see...
Dont start the IPO on that dino-breeding company just yet
I did a climate modeling project for my dissertation at university. These guys have a good idea - throw massive computing power at it - but it's not that simple. Compared to SETI or protein folding, climate modeling is not as highly parallelisable. Plus what's holding back the state of the art right now is the quality of the algorithms we're using - they loose accuracy pretty fast as the result of feedback between multiple iterations of a process that introduces quantisation noise each time.
IMHO this project will produce the same quantisation noise-ruined results we have now, just more finely ovesampled
Still, they might get some useful insights into how to tackle the problems of parallel dynamic system simulation
foo mane padme hum