3 ms·

You can never predict something as complex as the earth climate. It's not a question of computing power: it's impossible to predict a complex system at all. W

You can never predict something as complex as the earth climate. It's not a question of computing power: it's impossible to predict a complex system at all. We can't even predict a simple double pendulum, now imagine a climate simulation: it's basicaly billions of double pendulums, all dependent one on another. Even the smallest error in initial conditions or the models and even over a short period it will totaly diverge from what will actually happen on Earth. This will be nothing other than a money sink. Invenst into clean tech, not these useless models that are incapable of predicting anything at all. Eg how can you take in account a solar cycle which could change the energy that the Earth recieves more than any human made intervention/pollution? You cannot. All of these models will fail totaly. But politicians and the media love them for fear mongering, which is how the money gets pumped into this.

JoeAltmaier 4y ago
It's a matter of degree, precision and resolution. Of course, models can be made, and of course they can be useful. The modern attitude of "It won't be perfect, so I can claim it's no good" is today's religion. Deny science, because it can get the details wrong.
- oifjsidjf 4y ago
  You don't get it: it's NOT a matter od "degree" and "resolution". It's that the model will DIVERGE COMPLETELY in a way that you cannot predict because you can never fully take in account everything. So a resolution increase of eg 100x will NOT help as the model will STILL DIVERGE. Even if you take in account 100x more variables in a double pendulum or decrease the timestep by 1000x it will STILL DIVERGE sooner or later.
  - JoeAltmaier 4y ago
    There it is again. Not perfect; no matter how far ahead the model predicts its said to be not enough. This absolutism is so, so easy to spout. And shows no understanding of the value of models and the many useful results.

JoeAltmaier 4y ago

It's a matter of degree, precision and resolution. Of course, models can be made, and of course they can be useful. The modern attitude of "It won't be perfect, so I can claim it's no good" is today's religion. Deny science, because it can get the details wrong.

oifjsidjf 4y ago
You don't get it: it's NOT a matter od "degree" and "resolution". It's that the model will DIVERGE COMPLETELY in a way that you cannot predict because you can never fully take in account everything. So a resolution increase of eg 100x will NOT help as the model will STILL DIVERGE. Even if you take in account 100x more variables in a double pendulum or decrease the timestep by 1000x it will STILL DIVERGE sooner or later.
- JoeAltmaier 4y ago
  There it is again. Not perfect; no matter how far ahead the model predicts its said to be not enough. This absolutism is so, so easy to spout. And shows no understanding of the value of models and the many useful results.