Don't Go Being Politically Insane You Climate Change Skeptics

[Tim Sielbeck]

 

[DB]

20 hours ago, sunday said:

It is known that the Earth's atmosphere is not a linear system, so modeling it is very difficult as there is chaotic behavior present. For an example of an extremely simple non-linear system that shows a chaotic behavior, see:

 

Although some systems are chaotic, they can also be predictable in many cases - the behaviour that you observe depends on your starting conditions. Some starting conditions are well-behaved, some are not. When your model is unstable, perturbing the starting conditions by very small amounts can reveal stability, or divergence, which yields a measure of confidence in the prediction.

For weather prediction, rather than climate prediction, a typical multi-day weather prediction will be run several times with perturbed input parameters to give a confidence estimate.

The art of generating the input matrices for weather prediction from sparse and semi-random input point data is an interesting subject on its own.

[sunday]

2 hours ago, DB said:

Although some systems are chaotic, they can also be predictable in many cases - the behaviour that you observe depends on your starting conditions. Some starting conditions are well-behaved, some are not. When your model is unstable, perturbing the starting conditions by very small amounts can reveal stability, or divergence, which yields a measure of confidence in the prediction.

For weather prediction, rather than climate prediction, a typical multi-day weather prediction will be run several times with perturbed input parameters to give a confidence estimate.

The art of generating the input matrices for weather prediction from sparse and semi-random input point data is an interesting subject on its own.

Yes, I think it would not be outrageous to claim that many non-linear systems could be reasonably modeled for a short time after initial conditions, but before the onset of the really chaotic behavior. In the case of the simple system of the video, probably a few seconds, not more than fifteen. But after that, the system could do almost anything.

So yes, weather predictions are feasible, and there are good models of hurricanes. That is weather, however, and in a pretty short term on top of that. Modeling the climate some decades in advance is a whole different kettle of fish.

[rmgill]

4 hours ago, DB said:

Although some systems are chaotic, they can also be predictable in many cases - the behaviour that you observe depends on your starting conditions. Some starting conditions are well-behaved, some are not. When your model is unstable, perturbing the starting conditions by very small amounts can reveal stability, or divergence, which yields a measure of confidence in the prediction.

For weather prediction, rather than climate prediction, a typical multi-day weather prediction will be run several times with perturbed input parameters to give a confidence estimate.

The art of generating the input matrices for weather prediction from sparse and semi-random input point data is an interesting subject on its own.

How long do the weather models run before they go way off due to the input variability? 

[Ivanhoe]

1 hour ago, rmgill said:

How long do the weather models run before they go way off due to the input variability? 

One presidential election cycle.

[DB]

9 hours ago, rmgill said:

How long do the weather models run before they go way off due to the input variability? 

That question isn't really answerable, because potentially chaotic systems that are complex do not behave chaotically consistently... However, unless they've changed their approach since I last discussed this with a friend who worked there, the UK Met Office runs something close to the following. They may have changed the durations of the core run as computing power increases, or they may have increased the density and complexity of the model since he worked there.

Input data at time zero uses the weather station data (and balloon sounding data and satellite data, etc.), plus the output of the previous model with some clever levelling done to make the input model match the real data to create a three dimensional input data matrix. The model then runs forward for three days. The output from the three day run is used as input to the next three days, and eventually they provide a 9-day forecast that they have reasonably high confidence in.

It's worth differentiating between chaotic behaviour and wildly divergent behaviour - a system can be chaotic and yet similarly predictable (to some extent).

This is well described here, I think. Don't be seduced by the use of the "butterfly effect" in the title - they explain how the popular perception of that is incomplete.

 

[rmgill]

11 hours ago, DB said:

That question isn't really answerable, because potentially chaotic systems that are complex do not behave chaotically consistently...

I have heard distinctly different from someone who used to forecast weather for the USAF. He put the number at about 2 weeks iirc. It DOES relate to the input variability and the lack of enough data points to even start the simulations. 

Suffice to say the models are good for a finite time span and have to be re-freshed, restarted with new baselines/current data and then work well for another 2 weeks. 

I can see that number moving, but it is a rather short period overall. To try to extrapolate that to geological time scales and then look at how they've had to massage the input data AND still can't get results that match up to observed data sets. 

The climate scientists should show us how they can get rich playing ETF's on the day trading markets with their models. When they're all broke they can gloat at how it was just not quite the right data and that their models DO work if you just jigger the graphs. 

 

[DB]

I'm not sure that you understood me, but what your USAF forecaster said is pretty much exactly what I described for weather predictions, except for his determination that they could be good for 2 weeks whereas I said the UK Met Office did 9 days - and that was several years ago.

So, are you replying in a contradictory sense just so you can push the tin-foil claim that climate scientists are all in it for the money, or was that simply a bonus to go alongside your lack of reading comprehension skills?

[rmgill]

Maybe I'm confused that you assert two things that contradict. Should I have gone with your second statement of 9 days vs the "we can't know"? Are you channeling Stuart for comedic effect? 


As to if people are in things for the money, yeah. That's often a thing. We see those who stand out and then there's the push to drive them out of the field because they speak heterodoxy. We've seen this over an over. It doesn't take a rocket surgeon to figure that out. Why is Judith Curry not at Georgia Tech any more? She was controversial on it because she questioned some of the "science". 

We've gone over numerous bits in here before. Money is a STRONG incentive. Being able to get a job is an even stronger incentive. With the push by big media/tech to push one Narrative on both covid AND climate science, you'd have to be Mr Macgoo to not see some of the problems here. 

[lucklucky]

What quality and intensity past climate have in today's climate?

 

If climate was stopped by one hour if it restarted again will continue as it would be without that stoppage?

[DB]

Ryan, the only thing going on here is your attempt to manipulate any form of discussion into some sort of fratboy debating exercise to massage your own prostate.

You've demonstrated time and again that you only ever engage in these types of topic to bait and switch, so have at it.

The idea that most academics working on climatology are rolling around in masses of cash from some the evil climate change cabal suggests that you don't need a tinfoil hat, you should instead have worn a crash helmet as you were clearly dropped on your head when you were younger.

[Ssnake]

Nevertheless, career advancement, academic acknowledgement and potential for limited fame are strong motivators to support a political "consensus" agenda. Even if you're fundamentally honest but work with manipulated historical temperature records, you're still contributing to junk science. "Cash" in my understanding is the shorthand word to round up all these additional non-monetary motivations.

[rmgill]

2 hours ago, DB said:

Ryan, the only thing going on here is your attempt to manipulate any form of discussion into some sort of fratboy debating exercise to massage your own prostate.

Right, I cite real examples of what a real climate scientist from a major university in the US and it's massaging my prostate.

How erudite of you. 
 

2 hours ago, DB said:

You've demonstrated time and again that you only ever engage in these types of topic to bait and switch, so have at it.

You come up with assertions of things that are beyond the pale and then you get bent out of shape when I make points that rebut or show that your "beyond the pale".

Try to be intellectually honest yourself here. You're not. 

2 hours ago, DB said:

The idea that most academics working on climatology are rolling around in masses of cash from some the evil climate change cabal suggests that you don't need a tinfoil hat, you should instead have worn a crash helmet as you were clearly dropped on your head when you were younger.

Straw man fallacy. Noone asserted they were rolling around in masses of cash, but if their research funding is predicated upon studying certain things AND with having certain results, then yeah, you're going to be baking certain assumptions into the cake. This isn't beyond the pale by any means. 

More so, you can't sit there and honestly ignore the Mann emails from the Climate Unit in the UK. Those DEMONSTRATED deliberate behavior on the part of Mann and the others at that research facility. The simple act of hiding data was a problem that they did so and asserted that open examination of their research was not to be allowed was a larger problem. There were other issues showed by those emails. All of it goes to credibility and honesty. 

By all means keep asserting that there's nothing wrong there and that there's no credibility issues. 

More so, again, when theire 20 years of predictions don't match up, again....sure. We should definitely just ignore that. 

How utterly silly. 

Edited January 7, 2022 by rmgill

[Mikel2]

25 minutes ago, rmgill said:

More so, you can't sit there and honestly ignore the Mann emails from the Climate Unit in the UK. Those DEMONSTRATED deliberate behavior on the part of Mann and the others at that research facility.

 

Got a link to those in particular? Thanks.

[rmgill]

Here ya go.

https://www.lavoisier.com.au/articles/greenhouse-science/climate-change/climategate-emails.pdf

The issue was a bunch of leaked emails from the Climate Research Unit at the University of East Anglia.

When it comes to quality and integrity of work, if material comes to light that cast doubt on both of someone's work, do you think in DB's world that that means that person and possibly their entire department is out of a job? What about in any other person's world? Do you think that perhaps CASH as a reason to hide dishonest or fraudulent scientific findings and that that is a powerful motivator to hide and continue to hide the fraudulent basis for the scientific work? 

[rmgill]

Judith Curry's perspective on it. 



https://judithcurry.com/2019/11/12/legacy-of-climategate-10-years-later/

By Way of Dr Curry, she points out Ross McKitrick's observations below:

Exoneration?

The mainstream media and the Climategater scientists themselves claim complete exoneration by the various ‘inquiries’. Were they exonerated?

There was no exoneration by any objective analysis of the various inquiries. Ross McKitrick lays all this out in his article Understanding the Climategate Inquiries 

 

“The evidence points to some clear conclusions.

1. The scientists involved in the email exchanges manipulated evidence in IPCC and WMO reports with the effect of misleading readers, including policymakers. The divergence problem was concealed by deleting data to “hide the decline.” The panels that examined the issue in detail, namely Muir Russell’s panel, concurred that the graph was “misleading.” The ridiculous attempt by the Penn State Inquiry to defend an instance of deleting data and splicing in other data to conceal a divergence problem only discredits their claims to have investigated the issue.
2. Phil Jones admitted deleting emails, and it appears to have been directed towards preventing disclosure of information subject to Freedom of Information laws, and he asked his colleagues to do the same. The inquiries largely fumbled this question, or averted their eyes.
3. The scientists privately expressed greater doubts or uncertainties about the science in their own professional writings and in their interactions with one another than they allowed to be stated in reports of the IPCC or WMO that were intended for policymakers. Rather than criticise the scientists for this, the inquiries (particularly the House of Commons and Oxburgh inquiries) took the astonishing view that as long as scientists expressed doubts and uncertainties in their academic papers and among themselves, it was acceptable for them to conceal those uncertainties in documents prepared for policy makers.
4. The scientists took steps individually or in collusion to block access to data or methodologies in order to prevent external examination of their work. This point was accepted by the Commons Inquiry and Muir Russell, and the authors were admonished and encouraged to improve their conduct in the future.
5. The inquiries were largely unable to deal with the issue of the issue of blocking publication of papers, or intimidating journals. But academics reading the emails could see quite clearly the tribalism at work, and in comparison to other fields, climatology comes off looking juvenile, corrupt and in the grip of a handful of self-appointed gatekeepers and bullies.
   
    

*******

If anyone here was involved in ballistics engineering research and deleted data, evidence and other information casting doubt on say and then as a result of that massaged data got contracts for years for employment do you think there would be a larger question of long term employment if all of this came out and if people held you to account for lying about your engineering research? 

 

Edited January 7, 2022 by rmgill

[rmgill]

And on the subject of modeling, the thing about some of the climate models from folks like Mann was that if you fed them random noise data they would STILL show a hockey stick progression of trend out of that data set of random noise. 


https://climateaudit.files.wordpress.com/2005/09/mcintyre.ee.2005.pdf

 

Once again, MBH98 contained a misrepresentation, this time about their PC method. After the University of Virginia FTP site was made publicly available following MM03, by examining PC series archived there and, by examining source code for PC calculations,1 we were able to determine that MBH98 had not carried out a “conventional” PC calculation, but had modified the PC algorithm, by, among other things, subtracting the 1902-1980 mean, rather than the 1400-1980 column mean, prior to PC calculations, so that the columns were no longer centered on a zero mean in the 1400–1980 step. By this procedure, series are more decentered, and their variance more inflated, the larger is the difference between the series mean and the mean of the 20th century subset. The effect of this transformation would have been mitigated if they had carried out a singular value decomposition on the covariance matrix, but they carried it out on the de-centered data matrix. We have shown elsewhere that this method re-allocates variance so that the PC algorithm then strongly over-weights hockey stick-shaped proxies and that it is so efficient in mining a hockey stick shape that it nearly always produces a hockey-stick shaped PC1 even from persistent red noise [McIntyre and McKitrick, 2005; discussed in Muller, 2004].

But, yes, by all means, TRUST THE SCIENTISTS AND THEIR MODELS. 

[sunday]

Those three posts are a pretty good recap of the state of the issue, Ryan.

 

[RETAC21]

 

[Ssnake]

Another option is to sift through historical records and to see how often a certain extreme was reached or exceeded in a given period of time. But that's usually not being done because it doesn't deliver a clear trend, or contradicts the presumption that extreme weather events are happening more often. Specifically, the Ahr flood is a good example where two similarly high flood levels of the same rivers and creeks were reported in the last two hundred years, 1806 and 1910. But this time it's climate change. You don't need an academic degree to see that such a statement is hogwash. Same with the Heat Dome.

[NickM]

On 1/7/2022 at 8:36 AM, rmgill said:

And on the subject of modeling, the thing about some of the climate models from folks like Mann was that if you fed them random noise data they would STILL show a hockey stick progression of trend out of that data set of random noise. 


https://climateaudit.files.wordpress.com/2005/09/mcintyre.ee.2005.pdf

 

Once again, MBH98 contained a misrepresentation, this time about their PC method. After the University of Virginia FTP site was made publicly available following MM03, by examining PC series archived there and, by examining source code for PC calculations,1 we were able to determine that MBH98 had not carried out a “conventional” PC calculation, but had modified the PC algorithm, by, among other things, subtracting the 1902-1980 mean, rather than the 1400-1980 column mean, prior to PC calculations, so that the columns were no longer centered on a zero mean in the 1400–1980 step. By this procedure, series are more decentered, and their variance more inflated, the larger is the difference between the series mean and the mean of the 20th century subset. The effect of this transformation would have been mitigated if they had carried out a singular value decomposition on the covariance matrix, but they carried it out on the de-centered data matrix. We have shown elsewhere that this method re-allocates variance so that the PC algorithm then strongly over-weights hockey stick-shaped proxies and that it is so efficient in mining a hockey stick shape that it nearly always produces a hockey-stick shaped PC1 even from persistent red noise [McIntyre and McKitrick, 2005; discussed in Muller, 2004].

But, yes, by all means, TRUST THE SCIENTISTS AND THEIR MODELS. 

I am remembering something==I might have 'misremembered it' but, didn't NASA/NOAA have a recent 'chart' showing some dates in the early/mid 1930s as being 'unusually cold', while historical texts talk about the Depression Era Dust Bowl having the States Suffering from an unprecedented heat wave and drought?

[Ssnake]

That's what Tony Heller has been reporting for a while.

[rmgill]

14 hours ago, Ssnake said:

Another option is to sift through historical records and to see how often a certain extreme was reached or exceeded in a given period of time. But that's usually not being done because it doesn't deliver a clear trend, or contradicts the presumption that extreme weather events are happening more often. Specifically, the Ahr flood is a good example where two similarly high flood levels of the same rivers and creeks were reported in the last two hundred years, 1806 and 1910. But this time it's climate change. You don't need an academic degree to see that such a statement is hogwash. Same with the Heat Dome.

Yep. The challenge with using proxies is they don't necessarily work for comparing to modern temperature data. 

Using one instrument to measure something that's a gross value and another instrument to measure a fine value isn't very precise. 

You could say, use a hand marked ruler to measure something. And then go and make a pattern using those measurements extrapolated to precise values to .0001 and then use a micromenter as the instrument. How good is your comparison going to be? 

[Ssnake]

No matter how many decimals you add, the precision of the hand ruler won't be increased. But I guess that was a rhetorical question, since this is 7th grade school knowledge.

[Ivanhoe]

2 hours ago, Ssnake said:

No matter how many decimals you add, the precision of the hand ruler won't be increased. But I guess that was a rhetorical question, since this is 7th grade school knowledge.

Your assertion is 93.785% correct.

A friend on the east coast posted the local weather forecast a few days ago; something along the lines of "3.51 inches of snow." Very precise, not very accurate. 

Large swaths of American school children no longer learn this sort of thing.