Mostly we're only measuring the bottom 2 or 3 meters of the atmosphere but that's beside the point. Thousands of scientists and statisticians think it's an adequate measure and that's good enough for me. If you think otherwise it's up to you to provide actual scientific evidence.
You can't explain why the assumptions are justified, but now I have a burden of proof?
Nonsense. As an engineer, I have the scientific training to grasp what you can reasonably conclude from a set of data.
Given the assumptions involved, there is very little you can actually conclude form the current "global warming" data.
Your original statement was "It only happens to contradict previous global warming predictions that all the ice would be gone and we'd have massive floods of coastal areas." To me that implies you meant by now. No "past future prediction" that I'm aware of said that.
I heard it, that's good enough for me. If that's not what the scientists meant, they should try a lot harder to control the conclusions being propagated.
Ice on the Earth's surface is a complex thing composed of glaciers, ice sheets and sea ice. It's not particularly surprising that in some places the ice is increasing now. Meanwhile the net amount of ice on the planet is still declining.
The predictions did not say, "some ice will increase". They were wrong, and the popular label of "greenhouse" warming misleads the public. (Greenhouses don't get cold spots amidst their warming)
No one predicted that "global warming" (now "climate change") would result in more ice.
It's been "climate change" since at least the 1950's. Gilbert Plass published a paper in 1956 titled "The Carbon Dioxide Theory of Climate Change".
That's not a prediction that "climate change" will result in more ice.
As I said before the total amount of ice across the whole planet is still declining despite the fact that it is increasing in some areas.
If it's measured like global temperature, there's no credibility here. A quick look also finds that they're talking about AREA, when the important numbers depends on VOLUME. Area can be approximated as volume ... but again we are dealing with margin of error contributed by assumptions exceeding the result of concern.
The proposed mechanism is backfilled speculation about a complex system that is still not fully understood.
So are you saying since we don't fully understand it we should treat it as if we understand nothing?
No, I'm pointing out that zero accountability for false predictions results in no integrity for the process.
Actions based on false understanding are more harmful than actions based on little understanding. We can deal with known unknowns far better than unknown unknowns. False understanding results in far more of the latter, by definition.