• BananaTrifleViolin@kbin.social
    link
    fedilink
    arrow-up
    22
    arrow-down
    1
    ·
    1 year ago

    This is the problem with believing too much in models. A model can show you anything you want - it’s output is only as good as the parameters and algorithms you set it.

    Modelling the climate in the next 50-100 years is already extremely difficult and fraught with inaccuracies but we have lots of models and data to extrapolate from, so we do have a crude idea where we’re going. But we can’t model next years weather with accuracy, just the base trend. Crucially important warning for climate change but limited otherwise.

    Modelling out to 250 million years is basically a crock of shit. The tectonic movements are predictable and gross predictions that a pangea arrangement might be warmer may have some validity but modelling the climate and evolution and status of mammals is pure conjecture.

    Good thing about modelling that far is you will never have see you model’s accuracy being tested. Publish a paper, play into current fears around climate change with an irrelevant prediction about 250million years away, get an article published in the New York times and egos massaged all round.