>kapitza 19 days ago [-]
>This is slightly misleading. What we've understood for a century is the thermal forcing due to CO2, which is a logarithmic function of the CO2 concentration (ie, forcing == O(log n) in CO2 concentration). What we don't understand (unless we decide GCMs are predictive and can be validated) is the impact of a given thermal forcing on planetary temperature. This is the "climate sensitivity." Unfortunately, this is the number we actually care about. If climate sensitivity is low (negative feedback), Earth is thermally buffered and we have to worry less. If climate sensitivity is high (positive feedback), the opposite. (Note also that Arrhenius, the discoverer of this effect, and pretty much everyone through the '60s, assumed that global warming was a positive side effect of burning fossil fuel, just from the preindustrial assumption that warm winters and high crop yields are good for humanity.)
Climate sensitivity is just shorthand for the question of how harmful global warming is going to be. The concept concedes the correctness of the simple model I outline.
As it should, because basic physics would have to be wrong for the simple model to be wrong.
So, sure, there are questions about how bad it'll be. But those questions are just: will this be bad or really bad. They're not: is global warming occurring or not?
That's why it is so significant that so few scientists doubt it. Not because reality depends on some sort of popular vote of scientists, but because expertise increases the opportunity for doubt.
Also: you are reading about a scientific topic on the blog of someone who got famous for drawing cartoons. Ask yourself, if you are a developer, how seriously you would take a blog post from Scott Adams about whether bubble sort is actually faster than quicksort.
[1] https://news.ycombinator.com/item?id=12961996