Weather and Climate change

[Dill]

(08-07-2019, 07:23 PM)JustWinBaby Wrote: I asked you to go google ECS and then look at where IPCC gets there values from.  The IPCC takes those values primarily from the climate models.  That's inherently problematic.  The long and short of it is the science DOES NOT reject the default assumption of low climate sensitivity, which as I already pointed out is most likely between 1 and 1.5 based on the actual temperature data.  But then the climate alarmism falls apart, and with it a giant deflation of a $2T+ industry (among other things).

Those models aren't validated.  That's the hugely critical difference between data mining and science.  If their models had actual power, if they could be validated, they would do so with out-of-sample testing.  At a high-level, their model development has a computer chug thru different linear combinations for parameter estimates until it fits the data within the different constraints.  There's not a unique solution, which is why the IPCC uses something like 70 different models.  And there could be many more, so the range of ECS estimates is driven by assumptions and preferences of the modeler (or more deliberately, putting one's "thumb on the scale).  Many, many ways to bias a model, unintentionally or not, which is why validation is critical.

In other fields, assumptions and forecasts that come from an unvalidated model are dismissed out of hand.  The model is almost certainly wrong and certain to fail, so such research is a total non-starter.  I'm sure you won't believe me, but check out pages 2&3 of this UChicago link - this is what I've always said when referring to the fundamental building block of this as junk science:
http://jtac.uchicago.edu/conferences/05/resources/V&V_macal_pres.pdf

I'm sure you can find a lot of justification and explaining away the need to validate climate models.  But you need to pay very close attention to what is being said.  That a model matches the observed data is meaningless - it was forced to fit the data.  Plugging in new, observed values and getting an accurate forecast is a different model - the actual model had a wrong forecast/simulation for those variables because only CO2 is considered external forcing.  Fitting the model to include the new data is a different model (i.e. "re-tuned").

So in summary climate alarmism is not based on science, but on data mining (likely to produce a pre-determined result).

How are you defining "climate alarmism" here? Are you referring to pronouncements of some environmentalist groups, or to the IPCC? Are you following a journalist agreement among some publications to use the term?

Without reference to specific studies and arguments, it is hard to follow a claim that the IPCC "takes its values from climate models," and that that is  "inherently problematic" in some way not already acknowledged and addressed by the IPCC.  If you have seen and can specify some exact point where your claim is substantiated, I don't understand why you can't refer me directly to that. Perhaps you do not know how to cite or incorporate textual evidence into an argument? (My complaint here is the same as Breech's; you make general claims which look like they came from specific support, but at the moment any scientist would cite that support, you gesture vaguely towards Google while saying we obviously don't understand how science works. That looks like you are working with someone else's summary.)

You speak of an "unvalidated model" as if there were only one in question here, as if validation of open-ended systems is the same as validation of closed, and as if the global warming thesis depended only on models.

One can grant all the criticism of data mining in your links to non-climate science discussions without assuming that, in itself, invalidates climate modeling--an ongoing project of many models and many forms/sources of data.  That climate scientists themselves, working with physicists and mathematicians, criticize such models to constantly improve them, is hardly in itself sufficient evidence of "junk science."

So you are still a LONG WAYS from establishing that an undefined "climate alarmism" is, somewhere out there, based primarily upon data-mining, which is bad because discussions of V & V unrelated to climate science say it is bad.  I am not disputing it is bad, just saying you have really made no specific connection.