[irthomasthomas]

The IPCC has been making dire predictions now for 3 decades. Can you point to any that came true?

Off the top of my head I know that

1. Glacier park was to vanish by 2020. Its still there, barely changed, some glaciers shrank, others grew.

2. Sea level didn't rise as expected.

3 Polar bears are strong, and are not on the endangered list.

4. Temperatures did not rise as much as expected. We are still way below the Roman and Bronze age warm periods.

5. Damage from weather has increased in America, but that is a result of a policy to rebuild, instead of abandoning areas prone to bad weather. Increasing the probability of being damaged by bad weather. Actual weather severity has not changed much.

6. In school, I was made to feel terrified of desertification. The opposite happened, the deserts retreated. The globe is greening.

As I said, that is off the top of my head, but it can all be backed up by data. Or prove me wrong. Can you show me a few ipcc predictions which actually came true? Thanks.

[jaggs]

The IPCC does not make predictions. Never has, never will. The whole point, which real climate scientists repeat as nauseum, is we can never know absolutes because of the huge number of variables and timeframes. So what they do instead is provide scenarios. And each scenario has a series of weightings or probabilities. The summaries spell these out in rough terms, the full reports go into intricate detail and sources. If you spend the time to read the summaries you will find a good number of scenarios which are currently playing out as we speak.

[nostrebored]

Thousands of researchers making predictions with ever shifting probabilities is what you call unfalsifiable.

The predictions we’re talking about play out over decades. The sources cite other predictions. It’s a feedback loop of assumptions and peer review from people who you know already share the same assumptions.

Climate science is much more akin to sociology than a hard science.

[adrianN]

Except climate science has tons and tons of hard data that the models need to agree with to be publishable, whereas sociology has questionnaires administered to a group of 20 sociology students.

[nostrebored]

Making a model that agrees with historical data is literally the base case for modeling. Like it is a feature of all forecasting solutions