"Climate models are very good for understanding climate, but they are very poor for predicting climate." - Freeman Dyson
-
-
Replying to @MimeticValue
Climate change because of human action has always been VERY predictable. Have you heard about Svante Arrhénius, a scientist of the end of the 19th century ? He saw the problem with coal. Changes due to gases are EASY to calculate. Found this, in english :https://www.theguardian.com/environment/2005/jun/30/climatechange.climatechangeenvironment2 …
1 reply 0 retweets 0 likes -
Replying to @GirardForum
Yeah, of course peiple can see that there will be environmental problems even 100 years ago, but that's not what Dyson or I mean by prediction. Prediction means that you have confidence in the likelihood of a particular event at a particular time. No model will tell you that.
1 reply 0 retweets 1 like -
Replying to @MimeticValue @GirardForum
It's the same thing as scientists know which zones are more prone to earthquakes. That part is simple. Yet nobody can actually predict actual earthquakes until immediately before they happen. You can only determine general trends, not specific predictions.
1 reply 0 retweets 1 like -
Replying to @MimeticValue @GirardForum
For example, you can say that in 2050, average daily variance in temperature in France would be greater than in 2000 with a 80% confidence, but you can't predict what is the temperature on June 6, 2050 in Paris. Or the date the next big catastrophe will be here.
2 replies 0 retweets 2 likes -
Replying to @MimeticValue
In this example, you are mixing by mistake, like Trump but with much more intelligence, meteorology and climatology, 2 separate (not completely of course) scientific fields : meteorology prediction unit, the day ; climatologic scenarios units, the year, the decade => ≠ scales.
1 reply 0 retweets 0 likes -
Replying to @GirardForum @MimeticValue
I’m not sure what you are mixing but it’s all essentially history and the problem of induction in dynamic systems. Same predictability problems exist across domains. Prediction further out becomes less reliable so years, months worse than days.
2 replies 0 retweets 1 like -
Replying to @Ahimsa_Satya_ @MimeticValue
I agree. But meteorology gives good predictions for some days, about 1 or 2 weeks, even if it deals with dynamic/chaotic stuff requiring huge computers. Climatology studies climate since millions of years (studying very old ice, fossiles) and gives good scenarios for decades.
1 reply 0 retweets 0 likes -
Replying to @GirardForum @MimeticValue
Meteorology gives moderately good results for a few days, the ones on TV actually lie to you about the probabilities. Climate models may be accurate for a short while but not enough to give any real long range predictions. Best approach is precautionary principle imo.
2 replies 0 retweets 1 like
It seems to me that instead of complicated model we don’t seem to understand to make decisions we should just not do stupid things like digging up millions of gallons of dead plant and animals and burning them off into the atmosphere at a high rate. We only have one planet.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.