The Naked Scientists

The Naked Scientists Forum

Author Topic: So how will the climate be? 2100?  (Read 1906 times)

Offline yor_on

  • Naked Science Forum GOD!
  • *******
  • Posts: 11993
  • Thanked: 4 times
  • (Ah, yes:) *a table is always good to hide under*
    • View Profile
So how will the climate be? 2100?
« on: 07/08/2011 18:22:36 »
The idea behind this site is statistics. Actually, to me, it resembles the way they found the dropped atom bombs outside Spain. I think they named it the Monte Carlo method. It was quite simple, using their best predictions and computer methods they still couldn't find the last bomb, so what they did was making it into a educated guessing contest, where people bet on where they thought the bomb might be. And then they took the result of the competition and started a search where the most people had guessed it might be, and found it, in the approximate area. It might seem slightly confusing, but it has be shown that when you take a large number of opinions on a matter, educated or not, you will find that the answer that crystallises out more often than not, have a statistical significance, coincident with the correct answer, all as I understands it.

So what this site does is to work with all of those slight variations of climate models we already have, spreading them out on home computers, all over the world, to then collect the results and look at how they 'clump together' into a pattern. Those 'patterns' will, according to this model, show a statistical significance pointing to what will become the most probable temperature increase. And it works, the theory as such has been proven several times, on several different occasions. So, this guy sat and tried to create a program such as those climate models could be run on our normal pc:s. and while he sat there our computers became more and more powerful, and in the end those two meet, and this is the result.

It's a brilliant idea and you joining it will be one of the hundreds of thousand creating a virtual 'super computer' doing what no super computer today can do, count on those probabilities and where the significance becomes largest. Do you get it, by joining you will do something NASA, ECHELON, NORAD, NSA, ad infinitum, can't. Your PC will be a part of one of the coolest virtual super computers on Earth. And all for a really good purpose. And we need this answer.

Climateprediction.net


 

Offline CliffordK

  • Neilep Level Member
  • ******
  • Posts: 6321
  • Thanked: 3 times
  • Site Moderator
    • View Profile
So how will the climate be? 2100?
« Reply #1 on: 07/08/2011 21:41:56 »
This is a little different idea...
I think this is the "virtual supercomputer" idea.

Have a million people run a small application on their computer during off hours.
And, get the equivalent of a computer a million times more powerful than a PC.

I have enough problems with my computer heading off into la la land.  I don't need to add more apps to help it do that.
 

Offline graham.d

  • Neilep Level Member
  • ******
  • Posts: 2208
    • View Profile
So how will the climate be? 2100?
« Reply #2 on: 08/08/2011 09:24:32 »
Monte Carlo analysis is used in lots of applications and we use it a lot in IC design. It does not usually involve people's guesses or differing models (though maybe it could in this case). All that it does is randomly vary each parameter within its predicted distribution and then run a simulation. With enough simulations it is then possible to see the resulting outcomes with statistical spreads. In this case there area lot of variables and the simulations are long, but because each simulation is essentially independent from the rest it is an ideal candidate for this sort of distributed processing.
 

Offline yor_on

  • Naked Science Forum GOD!
  • *******
  • Posts: 11993
  • Thanked: 4 times
  • (Ah, yes:) *a table is always good to hide under*
    • View Profile
So how will the climate be? 2100?
« Reply #3 on: 10/08/2011 12:27:18 »
Yeah, I think it will do the job myself.

As a added bonus you get your own little earth, with its very own little climate system, spinning, developing on your computer. Why Clifford reacted I don't know? As I see it all of the climate models existing today are based on the observations we've made so far. They're not 'truths' in them selves but still based on what we know. I don't know how much it will 'steal' from your computer in form of memory and processes but is created as being a background process as I understands it, meaning that it will not stop you from doing what you want to do as any input from you will have a higher importance to the OS. Think of it as a screen saver, coming into its forte when you're not doing anything special.

We need some answers as I think, and this is a brilliant way of making them possible. And yes Graham, it's just a weak memory I have, naming it the 'Monte Carlo' method. Got it from a very good book about submarines and the cold war that I read several years ago. Extremely interesting book containing a surplus of facts around the decisions, etc, defining Countries political outlines. As always my memory is like a sieve :)
 

The Naked Scientists Forum

So how will the climate be? 2100?
« Reply #3 on: 10/08/2011 12:27:18 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums