No, there's no reason to treat error bars differently in climate science compared to any other branch of science.

At least if climate science aspires to be a science. If it doesn't - then by all means pretend that measurement errors don't affect the reliability of the calculations.

We have air samples of ocean pH levels?

The correct answer is no, you cannot infer anything about rapid changes from low resolution proxies. It's quite unphysical (and anti-science) to claim otherwise.

If you know of high resolution proxies, please let us know.

How?

We have no idea whether the rate is unusual. There are no proxies with that resolution available.

(But why let science stand in the way of a good scare story?)

I understand the climate models very well. How do you think the input parameters to the models are derived?

The parametrizations also involve numerical parameters that must be specified as input. Some of these parameters can be measured, at least in principle, while others cannot.

- IPCC AR4 WG1

The values used for parametrization are based on research that begins with measurements. Those measurements have errors - as in any other branch of science - yet those errors are not propagated through the calculations.

Science thus says that climate models cannot do projections more than a few years out, until the combined error exceeds the projection range.

There's no scientific support - whatsoever - for claiming that there's an expectation of weather to keep within 95% of the "confidence intervals". A model is only as good as its inputs - and measurements (en masse) are what those inputs are created from.

This is well known in all other fields of science, where claims of "confidence intervals" based on model runs would rightly get laughed out of all journals. The error bars of your measurements, inherit in all equipment, must be carried forward in all calculations.

For some reason that's not done in climate science. I don't understand why - there's no difference between "climate equipment" and other forms of equipment.

No, it's actually much much worse. Climate scientists create error bars on their projections by running models with different input parameters. They're not using the actual error variables from measurements and propagating the compounded error forwards in the calculations - which is how it's done in every other branch of science.

If you do that, the projections become meaningless just a few years out. The climate system is absolutely nothing like the single variable coin flip.

I'm very worried about the anti-science stance taken in climate discussions just because it doesn't lead to the preconceived result some hope to show (or even effect).

Not according to science.

Through studies of tree rings, sediment and other natural evidence, researchers have documented multiple droughts in California that lasted 10 or 20 years in a row during the past 1,000 years -- compared to the mere three-year duration of the current dry spell. The two most severe megadroughts make the Dust Bowl of the 1930s look tame: a 240-year-long drought that started in 850 and, 50 years after the conclusion of that one, another that stretched at least 180 years.

http://www.mercurynews.com/sci...

Which of course makes me wonder why so many people feel it's important to claim otherwise.

but he only knew German, Dutch, Italian, and Portuguese.

"only"

Also keep in mind that if you can drive a car from the 50ties, you can drive one built in 2015, and inversely. Safely.