Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.physics > #710549 > unrolled thread

Analysis of climate models

Back to article view | Back to sci.physics

Contents

  Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-10-31 10:22 -0700
    UsedCar salesman cite facts. Jeff-Relf.Me  @. - 2018-10-31 12:37 -0700
    Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-10-31 14:49 -0500
      Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-01 13:00 -0700
    Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-10-31 18:32 -0400
      Re: Analysis of climate models Libor Striz <poutnik4REMOVEnntp@CAPITALSgmail.com> - 2018-11-01 06:32 +0100
        Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-01 02:06 -0400
          Re: Analysis of climate models Libor Striz <poutnik4REMOVEnntp@CAPITALSgmail.com> - 2018-11-01 07:33 +0100
            Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-01 10:08 -0400
              Re: Analysis of climate models Libor Striz <poutnik4REMOVEnntp@CAPITALSgmail.com> - 2018-11-01 15:33 +0100
              Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-01 13:22 -0700
                Re: Analysis of climate models Wally W. <ww84wa@aim.com> - 2018-11-14 00:28 -0500
                  Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-15 16:47 -0800
            Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-01 10:16 -0500
      Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-01 13:16 -0700
      Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-01 16:02 -0700
        Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-02 20:19 -0500
        Re: Analysis of climate models James McGinn <jimmcginn9@gmail.com> - 2018-11-04 12:56 -0800
          Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-05 08:39 -0800
            Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-05 12:59 -0600
            Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-05 14:45 -0500
              Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-13 13:07 -0800
            Re: Analysis of climate models Claudius Denk <claudiusdenk@gmail.com> - 2018-11-05 12:34 -0800
              Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-13 13:09 -0800
                Re: Analysis of climate models Claudius Denk <claudiusdenk@gmail.com> - 2018-11-16 09:50 -0800
          Re: Analysis of climate models pnalsing@gmail.com - 2018-11-05 19:17 -0800
            Re: Analysis of climate models Claudius Denk <claudiusdenk@gmail.com> - 2018-11-06 07:16 -0800
    Re: Archimedes "Village Idiot" Plutonium flunked the Math Test of a lifetime-generation test moroney@world.std.spaamtrap.com (Michael Moroney) - 2018-11-01 21:23 +0000
    Re: Analysis of climate models Boris Mohar <borism_void_@sympatico.ca> - 2018-11-02 05:21 -0400
      Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-05 08:04 -0800
    Re: Analysis of climate models Wally W. <ww84wa@aim.com> - 2018-11-02 08:31 -0400
      Re: Analysis of climate models Dan Riley <dan131riley@gmail.com> - 2018-11-02 17:15 -0400
        Re: Analysis of climate models Claudius Denk <claudiusdenk@gmail.com> - 2018-11-05 12:40 -0800
          Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-13 13:13 -0800
            Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-18 17:22 -0600
              Re: Analysis of climate models James McGinn <jimmcginn9@gmail.com> - 2018-11-18 17:48 -0800
              Re: Analysis of climate models Libor Striz <poutnik4REMOVEnntp@CAPITALSgmail.com> - 2018-11-19 05:05 +0100
                Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-18 22:46 -0600
                  Re: Analysis of climate models Libor Striz <poutnik4REMOVEnntp@CAPITALSgmail.com> - 2018-11-19 06:59 +0100
                  Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-20 07:42 -0800
                    Re: Analysis of climate models James McGinn <jimmcginn9@gmail.com> - 2018-11-20 08:58 -0800
                      Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-20 19:17 -0500
                    Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-20 12:22 -0500
                      Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-20 21:02 -0600
              Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-19 11:01 -0500
                Re: Analysis of climate models Claudius Denk <claudiusdenk@gmail.com> - 2018-11-19 12:58 -0800
                  Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-20 08:10 -0800
                    Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-20 12:04 -0500
                      Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-20 22:02 -0600
        Re: Analysis of climate models Wally W. <ww84wa@aim.com> - 2018-11-19 07:00 -0500
          Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-20 08:00 -0800
            Re: Analysis of climate models Wally W. <ww84wa@aim.com> - 2018-11-20 21:28 -0500
              Re: Analysis of climate models James McGinn <jimmcginn9@gmail.com> - 2018-11-21 10:38 -0800
              Re: Analysis of climate models Wally W. <ww84wa@aim.com> - 2018-11-25 00:53 -0500
      Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-05 08:21 -0800
    Re: Analysis of climate models Dan Riley <dan131riley@gmail.com> - 2018-11-02 17:08 -0400
      Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-05 08:23 -0800
        Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-05 13:00 -0600
          Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-05 13:04 -0600
            Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-05 14:51 -0500
              Re: Analysis of climate models Sergio <invalid@invalid.com> - 2018-11-05 13:59 -0600
    Re: Analysis of climate models Wally W. <ww84wa@aim.com> - 2018-11-19 07:27 -0500
      Re: Analysis of climate models Edward Prochak <edprochak@gmail.com> - 2018-11-20 08:06 -0800
        Re: Analysis of climate models benj <benj.fake@gmail.com> - 2018-11-20 11:59 -0500

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

#713513

On Tuesday, November 20, 2018 at 7:42:49 AM UTC-8, Edward Prochak wrote:

> Thanks for all the contributions, folks. I will try to keep up,
> but I have a lot of other things happening.

This is the busy season for Walmart greeters.

[toc] | [prev] | [next] | [standalone]

#713633

On 11/20/2018 11:58 AM, James McGinn wrote:
> On Tuesday, November 20, 2018 at 7:42:49 AM UTC-8, Edward Prochak wrote:
> 
>> Thanks for all the contributions, folks. I will try to keep up,
>> but I have a lot of other things happening.
> 
> This is the busy season for Walmart greeters.
> 
:-)

[toc] | [prev] | [next] | [standalone]

#713521

On 11/20/2018 10:42 AM, Edward Prochak wrote:
> On Sunday, November 18, 2018 at 11:46:45 PM UTC-5, Sergio wrote:
>> On 11/18/2018 10:05 PM, Libor Striz wrote:
> []
>>>
>>> Inaccurate Initial and boundary values have so big impact, that it
>>>   does not matter much if the model is accurate or not.
>>>   Completeness and accuracy of data have is much more
>>>   important.
>>>
>>> Even if models were perfect, the dynamic instability nature of the
>>>   modelled system progressively amplifies data input errors to
>>>   chaotic results.
>>>
>>> It has nice analogy in putting a small ball on top of a big ball
>>>   and predicting the direction and distance where it will stop.
>>>   
>>>
>>> You can use as much accurate model as you want, all stands and
>>>   falls on accuracy of the initial ball position.
>>>
>>>
>>
>> agree,
>>
>> Balancing a pencil gets into Plancks constant, HUP, others. No matter
>> how clever you are, and no matter how much money you spend
>> on the newest, cutting-edge pencil-balancing equipment, you can never
>> get a pencil to balance for more than about four seconds.
>>
>>   problem
>> https://www.physics.harvard.edu/uploads/files/undergrad/probweek/prob3.pdf
>>
>>
>> solution
>>
>> https://www.physics.harvard.edu/uploads/files/undergrad/probweek/sol3.pdf
> 
> This is an interesting problem, but I would not call this the
> balancing a pencil problem.  This is more clearly the problem
> of statically balancing a pencil.  IOW, once the pencil is stood up,
> you are not allowed to change its motion.
> 
> A closer analog to climate is dynamically balancing a pencil
> on a movable platform (say your finger). Such a system includes
> feedback cycles and perturbations. So you can achieve a stable
> state given even irregular perturbations from light gusts of
> wind. But if a sudden strong gust pushes the pencil past a
> certain point, then it will fall.
> 
> So the climate question in analogy is:
> Has there been a global change that will push global climate
> to a new region?
> 
> Note that I think several of us agree that in the feedback
> processes, the pencil has basically one, moving the base.
> But climate has CO2 levels, CH4 levels, H2O levels, moderating
> sources like polar ice and open oceans. This is not a simple
> problem of one kind of perturbation and one feedback loop.
> 
> So thanks Sergio for the pencil balance example. You were
> making the different point that even with the best data,
> we cannot know 100% what will happen.
> 
> I would say this: that we can be pretty certain global climate
> is changing. But is the current change something outside the
> normal patterns? I don't know yet. I don't think anyone in the
> discussion so far has the knowledge to answer the question
> one way or the other.

Climate is changing? Duh! Now THAT is where nobody disagrees. I'd call 
that "consensus".

As for degree, well that thing in front of you is called a computer. 
There is lots of science data available as to whether there is 
significant movement of climate. Journalist "science" (which is 
non-science) is no place to look. Just check the science. The problem, 
of course, is that this issue has TONS of money involved in MANY ways so 
that an enormous amount of lies (propaganda) has been generated. Your 
task should you choose to accept it is to use your best forensic methods 
to try to distinguish true science from political propaganda. Not such 
an easy job.

https://www.youtube.com/watch?v=oN_oynx1D8w

So next time there is a hurricane and breathless news tells you how much 
it "proves" climate change is dooming us all, now you know better, right?

> Thanks for all the contributions, folks. I will try to keep up,
> but I have a lot of other things happening.

Best is to take the Odd Boinker approach and refuse to examine any 
information that disagrees with your current preconceived ideas. This is 
a lot more comfortable than actually taking time to study the facts that 
may cause you to change your ideas.

> Happy Thanksgiving to all!

And yes, Happy Thanksgiving to you too Ed.
We all want the best for you too.

>     Ed
> 

[toc] | [prev] | [next] | [standalone]

#713681

On 11/20/2018 11:22 AM, benj wrote:
> On 11/20/2018 10:42 AM, Edward Prochak wrote:
>> On Sunday, November 18, 2018 at 11:46:45 PM UTC-5, Sergio wrote:
>>> On 11/18/2018 10:05 PM, Libor Striz wrote:
>> []
>>>>
>>>> Inaccurate Initial and boundary values have so big impact, that it
>>>>   does not matter much if the model is accurate or not.
>>>>   Completeness and accuracy of data have is much more
>>>>   important.
>>>>
>>>> Even if models were perfect, the dynamic instability nature of the
>>>>   modelled system progressively amplifies data input errors to
>>>>   chaotic results.
>>>>
>>>> It has nice analogy in putting a small ball on top of a big ball
>>>>   and predicting the direction and distance where it will stop.
>>>>  
>>>> You can use as much accurate model as you want, all stands and
>>>>   falls on accuracy of the initial ball position.
>>>>
>>>>
>>>
>>> agree,
>>>
>>> Balancing a pencil gets into Plancks constant, HUP, others. No matter
>>> how clever you are, and no matter how much money you spend
>>> on the newest, cutting-edge pencil-balancing equipment, you can never
>>> get a pencil to balance for more than about four seconds.
>>>
>>>   problem
>>> https://www.physics.harvard.edu/uploads/files/undergrad/probweek/prob3.pdf
>>>
>>>
>>>
>>> solution
>>>
>>> https://www.physics.harvard.edu/uploads/files/undergrad/probweek/sol3.pdf
>>>
>>
>> This is an interesting problem, but I would not call this the
>> balancing a pencil problem.  This is more clearly the problem
>> of statically balancing a pencil.  IOW, once the pencil is stood up,
>> you are not allowed to change its motion.
>>
>> A closer analog to climate is dynamically balancing a pencil
>> on a movable platform (say your finger). Such a system includes
>> feedback cycles and perturbations. So you can achieve a stable
>> state given even irregular perturbations from light gusts of
>> wind. But if a sudden strong gust pushes the pencil past a
>> certain point, then it will fall.
>>
>> So the climate question in analogy is:
>> Has there been a global change that will push global climate
>> to a new region?
>>
>> Note that I think several of us agree that in the feedback
>> processes, the pencil has basically one, moving the base.
>> But climate has CO2 levels, CH4 levels, H2O levels, moderating
>> sources like polar ice and open oceans. This is not a simple
>> problem of one kind of perturbation and one feedback loop.
>>
>> So thanks Sergio for the pencil balance example. You were
>> making the different point that even with the best data,
>> we cannot know 100% what will happen.
>>
>> I would say this: that we can be pretty certain global climate
>> is changing. But is the current change something outside the
>> normal patterns? I don't know yet. I don't think anyone in the
>> discussion so far has the knowledge to answer the question
>> one way or the other.
> 
> Climate is changing? Duh! Now THAT is where nobody disagrees. I'd call
> that "consensus".
> 
> As for degree, well that thing in front of you is called a computer.
> There is lots of science data available as to whether there is
> significant movement of climate. Journalist "science" (which is
> non-science) is no place to look. Just check the science. The problem,
> of course, is that this issue has TONS of money involved in MANY ways so
> that an enormous amount of lies (propaganda) has been generated. Your
> task should you choose to accept it is to use your best forensic methods
> to try to distinguish true science from political propaganda. Not such
> an easy job.
> 
> https://www.youtube.com/watch?v=oN_oynx1D8w
> 
> So next time there is a hurricane and breathless news tells you how much
> it "proves" climate change is dooming us all, now you know better, right?
> 
>> Thanks for all the contributions, folks. I will try to keep up,
>> but I have a lot of other things happening.
> 
> Best is to take the Odd Boinker approach and refuse to examine any
> information that disagrees with your current preconceived ideas. This is
> a lot more comfortable than actually taking time to study the facts that
> may cause you to change your ideas.
> 
>> Happy Thanksgiving to all!
> 
> And yes, Happy Thanksgiving to you too Ed.
> We all want the best for you too.
> 
>>     Ed
>>
> 

yes  Happy Thanksgiveing to all,

remember I found that link on how anyone could publish articles on
climate change for a science mag in here somewhere, like how to pick a
grabber headlines, etc, etc....

[toc] | [prev] | [next] | [standalone]

#713333

On 11/18/2018 6:22 PM, Sergio wrote:
> On 11/13/2018 3:13 PM, Edward Prochak wrote:
>> On Monday, November 5, 2018 at 3:40:19 PM UTC-5, Claudius Denk (aka james McGinn) wrote:
>>> On Friday, November 2, 2018 at 2:15:51 PM UTC-7, Dan Riley wrote:
>>>> Wally W. <ww84wa@aim.com> writes:
>>>>> A modeling "experiment" observes a computer program. Said computer
>>>>> program embodies an approximation of an incomplete understanding of a
>>>>> limited number of the relevant interacting natural processes, to a
>>>>> finite number of decimal places that can't track even one chaotic
>>>>> process very far before the program's output will diverge wildly from
>>>>> how Nature actually behaves.
>>>>
>>>> This is basically the "we can't model the weather for more than a few
>>>> days, how can we expect to model the climate" argument.  One answer is
>>>> that modeling the weather is an initial value problem, while modeling
>>>> the climate is a boundary value problem.  Well posed boundary value
>>>> models don't diverge wildly.
>>>>
>>>> -dan
>>>
>>> This is an absurdly stupid comment.  A "boundary value problem" is just a more vaguely defined, "initial value problem." Climate scientists are experts in this kind of evasive, meaningless rhetoric.
>>
>> Now James, be polite.
>> Please.
>>
>>    ed
>>
> 
> prediction using models;
> 
> it is obvious that weather or climate is not an initial value problem,
> nor is it a boundary value problem.  That implies that the Model is
> accurate.
> 
> however, both weather and climate involve turbulent processes, and we
> can only grossly model them with resulting poor accuracy.
> 
> 
Accuracy doesn't matter, Sergio. Facts don't matter. Today only 
"journalist science" matters.

[toc] | [prev] | [next] | [standalone]

#713399

> Accuracy doesn't matter, Sergio. Facts don't matter. Today only 
> "journalist science" matters.

It's the church lady effect.  IOW, science killed religion and so now many have turned science into a religion.  

https://groups.google.com/forum/#!searchin/sci.physics/church$20lady$20church$20ladies%7Csort:date

[toc] | [prev] | [next] | [standalone]

#713499

On Monday, November 19, 2018 at 3:58:41 PM UTC-5, Claudius Denk wrote:
> > Accuracy doesn't matter, Sergio. Facts don't matter. Today only 
> > "journalist science" matters.
> 
> It's the church lady effect.  IOW, science killed religion
> and so now many have turned science into a religion.  
> 
(link snipped)

Hi James,
Sorry but this is not a thread for self promotion. you and Benj
should start your own thread for the religion debate.

ed

[toc] | [prev] | [next] | [standalone]

#713515

On 11/20/2018 11:10 AM, Edward Prochak wrote:
> On Monday, November 19, 2018 at 3:58:41 PM UTC-5, Claudius Denk wrote:
>>> Accuracy doesn't matter, Sergio. Facts don't matter. Today only
>>> "journalist science" matters.
>>
>> It's the church lady effect.  IOW, science killed religion
>> and so now many have turned science into a religion.
>>
> (link snipped)
> 
> Hi James,
> Sorry but this is not a thread for self promotion. you and Benj
> should start your own thread for the religion debate.
> 
> ed
> 
Sorry ed, but if science in now a religion (clearly is the trend) then 
this is THE place for such a discussion. How you get "self-promotion" 
out of such a discussion is as ill thought out as your opinion that Denk 
and I are some how in disagreement and having a "debate". YOU are the 
one we disagree with.

[toc] | [prev] | [next] | [standalone]

#713690

On 11/20/2018 11:04 AM, benj wrote:
> On 11/20/2018 11:10 AM, Edward Prochak wrote:
>> On Monday, November 19, 2018 at 3:58:41 PM UTC-5, Claudius Denk wrote:
>>>> Accuracy doesn't matter, Sergio. Facts don't matter. Today only
>>>> "journalist science" matters.
>>>
>>> It's the church lady effect.  IOW, science killed religion
>>> and so now many have turned science into a religion.
>>>
>> (link snipped)
>>
>> Hi James,
>> Sorry but this is not a thread for self promotion. you and Benj
>> should start your own thread for the religion debate.
>>
>> ed
>>
> Sorry ed, but if science in now a religion (clearly is the trend) then
> this is THE place for such a discussion. How you get "self-promotion"
> out of such a discussion is as ill thought out as your opinion that Denk
> and I are some how in disagreement and having a "debate". YOU are the
> one we disagree with.

this thread is not about the thousands of church ladies in black pointy
shoes and pointy hats marching in rows 20 across to obey McFly's every
command, nor the one where James McGinn sells his dehydrated water
vapor, either.
(note for those who were asleep:=> James McGinn, McFly, and Cladius Dink
are all the same person)

[toc] | [prev] | [next] | [standalone]

#713319

On Fri, 02 Nov 2018 17:15:48 -0400, Dan Riley wrote:

>Wally W. <ww84wa@aim.com> writes:
>> A modeling "experiment" observes a computer program. Said computer
>> program embodies an approximation of an incomplete understanding of a
>> limited number of the relevant interacting natural processes, to a
>> finite number of decimal places that can't track even one chaotic
>> process very far before the program's output will diverge wildly from
>> how Nature actually behaves.
>
>This is basically the "we can't model the weather for more than a few
>days, how can we expect to model the climate" argument.  One answer is
>that modeling the weather is an initial value problem, while modeling
>the climate is a boundary value problem.  Well posed boundary value
>models don't diverge wildly.
>
>-dan

What is the "boundary" in the "boundary value problem?"

If it is the edge of weather behaviour for the current climate in one
location, then "extreme weather" (so-called for the old climate, but
it is "normal" weather for the emerging climate) is moving the
boundary ... AKA, "climate change."

Mash a bunch of adjacent, interacting, two-dimensional "boundary value
problems" together, and what do you have that is worthy of use to
justify civilization-killing policies?

What is a "well posed boundary value model" in the context of climate
models, where the boundaries are moving?

There are large uncertainties in how hard "extreme weather" will push
on the boundaries because the "initial value" problems for predicting
"extreme weather" at all the boundaries are unstable.

Don't forget to include the transient effects due to shading from
clouds, which were neglected when the IPCC started whingeing in
"assessments reports."

How have you refuted the "we can't model the weather for more than a
few days, how can we expect to model the climate" argument?

[toc] | [prev] | [next] | [standalone]

#713496

On Monday, November 19, 2018 at 7:00:04 AM UTC-5, Wally W. wrote:
> On Fri, 02 Nov 2018 17:15:48 -0400, Dan Riley wrote:
> 
> >Wally W. <ww84wa@aim.com> writes:
> >> A modeling "experiment" observes a computer program. Said computer
> >> program embodies an approximation of an incomplete understanding of a
> >> limited number of the relevant interacting natural processes, to a
> >> finite number of decimal places that can't track even one chaotic
> >> process very far before the program's output will diverge wildly from
> >> how Nature actually behaves.
> >
> >This is basically the "we can't model the weather for more than a few
> >days, how can we expect to model the climate" argument.  One answer is
> >that modeling the weather is an initial value problem, while modeling
> >the climate is a boundary value problem.  Well posed boundary value
> >models don't diverge wildly.
> >
> >-dan
> 
> What is the "boundary" in the "boundary value problem?"
> 
> If it is the edge of weather behaviour for the current climate in one
> location, then "extreme weather" (so-called for the old climate, but
> it is "normal" weather for the emerging climate) is moving the
> boundary ... AKA, "climate change."
> 
> Mash a bunch of adjacent, interacting, two-dimensional "boundary value
> problems" together, and what do you have that is worthy of use to
> justify civilization-killing policies?

just pointing out this ^^^ is argument by fear mongering,
AKA the F in FUD.

> 
> What is a "well posed boundary value model" in the context of climate
> models, where the boundaries are moving?

I have never built a climate model, but I would expect the boundaries
are not between locations such as Europe and Asia, but between the
atmosphere and ground, Atmosphere and oceans, Surfaces and solar,
surfaces and night sky. Maybe others.

So in my mind it is not so much moving boundaries, as conditions
continually changing at the boundaries.

> 
> There are large uncertainties in how hard "extreme weather" will push
> on the boundaries because the "initial value" problems for predicting
> "extreme weather" at all the boundaries are unstable.
> 
> Don't forget to include the transient effects due to shading from
> clouds, which were neglected when the IPCC started whingeing in
> "assessments reports."

Are you sure that was neglected in all the climate models used?

> 
> How have you refuted the "we can't model the weather for more than a
> few days, how can we expect to model the climate" argument?

I think Dan's point was simply that these are two very different problems.

The fact is weather prediction has actually gotten very good over
the last couple decades, partly with improved models, but also
because of hardware advances (multi-Gigabyte memories and
multi=terabyte drives) and system advances (grid computing
and multi-core processors).

Applying similar advances to the climate models will reduce a
good portion of the uncertainty, IMHO.

Ed

[toc] | [prev] | [next] | [standalone]

#713669

On Tue, 20 Nov 2018 08:00:29 -0800 (PST), Edward Prochak wrote:

>On Monday, November 19, 2018 at 7:00:04 AM UTC-5, Wally W. wrote:
>> On Fri, 02 Nov 2018 17:15:48 -0400, Dan Riley wrote:
>> 
>> >Wally W. <ww84wa@aim.com> writes:
>> >> A modeling "experiment" observes a computer program. Said computer
>> >> program embodies an approximation of an incomplete understanding of a
>> >> limited number of the relevant interacting natural processes, to a
>> >> finite number of decimal places that can't track even one chaotic
>> >> process very far before the program's output will diverge wildly from
>> >> how Nature actually behaves.
>> >
>> >This is basically the "we can't model the weather for more than a few
>> >days, how can we expect to model the climate" argument.  One answer is
>> >that modeling the weather is an initial value problem, while modeling
>> >the climate is a boundary value problem.  Well posed boundary value
>> >models don't diverge wildly.
>> >
>> >-dan
>> 
>> What is the "boundary" in the "boundary value problem?"
>> 
>> If it is the edge of weather behaviour for the current climate in one
>> location, then "extreme weather" (so-called for the old climate, but
>> it is "normal" weather for the emerging climate) is moving the
>> boundary ... AKA, "climate change."
>> 
>> Mash a bunch of adjacent, interacting, two-dimensional "boundary value
>> problems" together, and what do you have that is worthy of use to
>> justify civilization-killing policies?
>
>just pointing out this ^^^ is argument by fear mongering,
>AKA the F in FUD.

Actually, it was emphasizing a question.

Let's develop it further but leave out the bit you commented on
without addressing the question: 

Mash a bunch of adjacent, interacting, multi-factored,
two-dimensional, incomplete "boundary value problems" together, and
what do you have that is acurate enough over time scales of longer
than decades to justify any policy that would impose new rules on a
society to forestall (instead of possibly spending less later to adapt
to) consequences that *might* arise decades hence?

How well has back-casting validated said models?


>> 
>> What is a "well posed boundary value model" in the context of climate
>> models, where the boundaries are moving?
>
>I have never built a climate model, 

Neither have I.

Has Dan?

>but I would expect the boundaries
>are not between locations such as Europe and Asia, but between the
>atmosphere and ground, Atmosphere and oceans, Surfaces and solar,
>surfaces and night sky. Maybe others.

The oceans aren't stable boundaries. Witness the frequent whingeing in
the alt.global-warming group about how ocean temperatures last month
were the 4th highest since records have been taken (by force,
tortured, and defiled by the Ministry of Truth).

>So in my mind it is not so much moving boundaries, as conditions
>continually changing at the boundaries.

My understanding of a boundary value problem is that the conditions do
not change at the boundary:

https://en.wikipedia.org/wiki/Boundary_value_problem

An example from wiki: "if one end of an iron rod is held at absolute
zero, then the value of the problem would be known at that point in
space."

The temperature isn't changing at the end "held at absolute zero."

>
>> 
>> There are large uncertainties in how hard "extreme weather" will push
>> on the boundaries because the "initial value" problems for predicting
>> "extreme weather" at all the boundaries are unstable.
>> 
>> Don't forget to include the transient effects due to shading from
>> clouds, which were neglected when the IPCC started whingeing in
>> "assessments reports."
>
>Are you sure that was neglected in all the climate models used?

Why do you ask about "***all*** the climate models used?"

Neglecting the effects of clouds is a major shortcoming in **any**
climate model. It should make a thinking person have severe
reservations about its predictions a century hence.

 
>> How have you refuted the "we can't model the weather for more than a
>> few days, how can we expect to model the climate" argument?
>
>I think Dan's point was simply that these are two very different problems.

They are not unrelated problems. And the "boundary value problem"
isn't a refuge for defenders of the models. 

It is internally inconsistent to assume stable boundaries in an effort
to predict changes in those boundaries.


>The fact is weather prediction has actually gotten very good over
>the last couple decades, partly with improved models, but also
>because of hardware advances (multi-Gigabyte memories and
>multi=terabyte drives) and system advances (grid computing
>and multi-core processors).
>
>Applying similar advances to the climate models will reduce a
>good portion of the uncertainty, IMHO.

Chaos theory seems to fly in the face of this opinion, however humble
it may be.

See Sergio's quote and the post from Libor Striz before it:
On Sun, 18 Nov 2018 22:46:41 -0600, Sergio wrote:
:>> Even if models were perfect, the dynamic instability nature of the
:>>  modelled system progressively amplifies data input errors to
:>>  chaotic results.

:>No matter how clever you are, and no matter how much money you spend
:>on the newest, cutting-edge pencil-balancing equipment, you can never
:>get a pencil to balance for more than about four seconds.

Forecasting harware advances, etc., may be comparable to arguing how
many angels can fit on the head of a pin.

One might expect the time frame over which climate models can make
accurate predictions is eventually going to maximize at some duration
longer than four seconds, but **much** shorter than the tens of
decades that AGW alarmists keep whingeing about ... between their
warnings about the [next] tipping point -- before which we **must**
take "urgent action" [of an unspecified variety].

[toc] | [prev] | [next] | [standalone]

#713777

On Tuesday, November 20, 2018 at 6:28:01 PM UTC-8, Wally W. wrote:
> >The fact is weather prediction has actually gotten very good over
> >the last couple decades, partly with improved models, but also
> >because of hardware advances (multi-Gigabyte memories and
> >multi=terabyte drives) and system advances (grid computing
> >and multi-core processors).
> >
> >Applying similar advances to the climate models will reduce a
> >good portion of the uncertainty, IMHO.
> 
> Chaos theory seems to fly in the face of this opinion, however humble
> it may be.

Dopes don't understand chaos theory.  Uncertaintly is amplified exponentially with each iteration. If we double processing power we don't get twice the certainty.  These idiots barely understand statistics and so they think computers are a crystal ball.

It's incredibly easy to fool the church-lady public. Ed's comment here perfectly exemplifies the dimwittedness of the public.

One of the tricks the global warming frauds employ is to not calculate and not discuss the incredibly low certainty of their model runs. Instead they will talk about "CO2 sensivity."  The public is way too dumb to know that CO2 sensitivity has nothing to do with certainty.  Moreover, discussing CO2 sensitivity is meaningless if you haven't even established a causal connection to CO2 and warming.  The public is so ignorant of statistical principles and so enamoured with computation that it takes very little effort to make these models valuable when I fact (and by the IPCC's own statements) these models are worthless for prediction.

But there are a lot of Ed Prochaks out there, eager to believe anything stated by a "scientist".

James McGinn / Solving Tornadoes



> 
> See Sergio's quote and the post from Libor Striz before it:
> On Sun, 18 Nov 2018 22:46:41 -0600, Sergio wrote:
> :>> Even if models were perfect, the dynamic instability nature of the
> :>>  modelled system progressively amplifies data input errors to
> :>>  chaotic results.
> 
> :>No matter how clever you are, and no matter how much money you spend
> :>on the newest, cutting-edge pencil-balancing equipment, you can never
> :>get a pencil to balance for more than about four seconds.
> 
> Forecasting harware advances, etc., may be comparable to arguing how
> many angels can fit on the head of a pin.
> 
> One might expect the time frame over which climate models can make
> accurate predictions is eventually going to maximize at some duration
> longer than four seconds, but **much** shorter than the tens of
> decades that AGW alarmists keep whingeing about ... between their
> warnings about the [next] tipping point -- before which we **must**
> take "urgent action" [of an unspecified variety].

[toc] | [prev] | [next] | [standalone]

#714594

On Tue, 20 Nov 2018 21:28:01 -0500, Wally W. wrote:

>On Tue, 20 Nov 2018 08:00:29 -0800 (PST), Edward Prochak wrote:
>
>>On Monday, November 19, 2018 at 7:00:04 AM UTC-5, Wally W. wrote:
>>> On Fri, 02 Nov 2018 17:15:48 -0400, Dan Riley wrote:
>>> >This is basically the "we can't model the weather for more than a few
>>> >days, how can we expect to model the climate" argument.  One answer is
>>> >that modeling the weather is an initial value problem, while modeling
>>> >the climate is a boundary value problem.  Well posed boundary value
>>> >models don't diverge wildly.
>>> >
>>> >-dan
>>> 
>>> What is the "boundary" in the "boundary value problem?"
>>> 
>>> If it is the edge of weather behaviour for the current climate in one
>>> location, then "extreme weather" (so-called for the old climate, but
>>> it is "normal" weather for the emerging climate) is moving the
>>> boundary ... AKA, "climate change."
>>> 
>>> Mash a bunch of adjacent, interacting, two-dimensional "boundary value
>>> problems" together, and what do you have that is worthy of use to
>>> justify civilization-killing policies?
>>
>>just pointing out this ^^^ is argument by fear mongering,
>>AKA the F in FUD.
>
>Actually, it was emphasizing a question.
>
>Let's develop it further but leave out the bit you commented on
>without addressing the question: 
>
>Mash a bunch of adjacent, interacting, multi-factored,
>two-dimensional, incomplete "boundary value problems" together, and
>what do you have that is acurate enough over time scales of longer
>than decades to justify any policy that would impose new rules on a
>society to forestall (instead of possibly spending less later to adapt
>to) consequences that *might* arise decades hence?
>
>How well has back-casting validated said models?
>
>
>>> 
>>> What is a "well posed boundary value model" in the context of climate
>>> models, where the boundaries are moving?
>>
>>I have never built a climate model, 
>
>Neither have I.
>
>Has Dan?

There seems to be no evidence here that he has.

[toc] | [prev] | [next] | [standalone]

#711637

On Friday, November 2, 2018 at 8:31:17 AM UTC-4, Wally W. wrote:
> On Wed, 31 Oct 2018 10:22:13 -0700 (PDT), Edward Prochak wrote:
> 
> >In response to sergio's comments and my own goal of trying to
> >talk about physics in this group, I am starting this thread.
> 
> Climate models are applied computer science to mash together a
> selection of partially understood principles from physics.
> 
> You might have picked a better subject in a spearhead attempt to get
> momentum for your "goal of trying to talk about physics in this group"
> 
> Just sayin'.
> 
> BTW: notice the absence of kensi in this thread with the stated goal
> and is linked to the climate science she seems to worship.
> 
> 
> >So the goal is to investigate the limits of climate models.
> 
> Though the context there was (so-called by sloppy thinkers or
> propagandists) "modeling experiments," from one of my posts in
> alt.global-warming:
> Subject: Re: Climate change provided high octane fuel for Hurricane
> Date: Thu, 25 Oct 2018 06:25:50 -0400
> Message-ID: <7563tdhb91ph8g8derr8mi9nu4ipm52kui@4ax.com>
> 
> A modeling "experiment" observes a computer program. Said computer
> program embodies an approximation of an incomplete understanding of a
> limited number of the relevant interacting natural processes, to a
> finite number of decimal places that can't track even one chaotic
> process very far before the program's output will diverge wildly from
> how Nature actually behaves.
> 
> 
thanks Wally for the comments. Glad to see you are willing to
participate.  Your comment about modeling experiments is a useful
one. This is why IIUC the conclusions drawn by the climatologists
are based on multiple models in the same way that multiple weather
models are used to predict hurricane strength and movement.

So while any one model can diverge, predictions of multiple models
based on different assumptions and/or using different computational
methods essentially provides multiple samples, giving better insight.

So that defects in any one models is not justification to doubt
the conclusions.

Ed

[toc] | [prev] | [next] | [standalone]

#711100

Edward Prochak <edprochak@gmail.com> writes:
> In response to sergio's comments and my own goal of trying to
> talk about physics in this group, I am starting this thread.
>
> So the goal is to investigate the limits of climate models.

One place to look is this article:

https://journals.ametsoc.org/doi/10.1175/BAMS-D-15-00135.1

"The Art and Science of Climate Model Tuning"

From the intro:

"While the fundamental physics of climate is generally well established,
submodels or parameterizations are approximate, either because of
numerical cost issues (limitations in grid resolution, acceleration of
radiative transfer computation) or, more fundamentally, because they try
to summarize complex and multiscale processes through an idealized and
approximate representation. Each parameterization relies on a set of
internal equations and often depends on parameters, the values of which
are often poorly constrained by observations. The process of estimating
these uncertain parameters in order to reduce the mismatch between
specific observations and model results is usually referred to as tuning
in the climate modeling community."

The need for approximate parameterizations that aren't derived directly
from first principles is one of the chief sources of uncertainty in
modeling, so understanding why and how the models are tuned is one way
to approach the limits of models.

-dan

[toc] | [prev] | [next] | [standalone]

#711639

On Friday, November 2, 2018 at 5:08:06 PM UTC-4, Dan Riley wrote:
> Edward Prochak <edprochak@gmail.com> writes:
> > In response to sergio's comments and my own goal of trying to
> > talk about physics in this group, I am starting this thread.
> >
> > So the goal is to investigate the limits of climate models.
> 
> One place to look is this article:
> 
> https://journals.ametsoc.org/doi/10.1175/BAMS-D-15-00135.1
> 
> "The Art and Science of Climate Model Tuning"
> 
> From the intro:
> 
> "While the fundamental physics of climate is generally well established,
> submodels or parameterizations are approximate, either because of
> numerical cost issues (limitations in grid resolution, acceleration of
> radiative transfer computation) or, more fundamentally, because they try
> to summarize complex and multiscale processes through an idealized and
> approximate representation. Each parameterization relies on a set of
> internal equations and often depends on parameters, the values of which
> are often poorly constrained by observations. The process of estimating
> these uncertain parameters in order to reduce the mismatch between
> specific observations and model results is usually referred to as tuning
> in the climate modeling community."
> 
> The need for approximate parameterizations that aren't derived directly
> from first principles is one of the chief sources of uncertainty in
> modeling, so understanding why and how the models are tuned is one way
> to approach the limits of models.
> 
> -dan

Cool. Thanks Dan. Just the kind of resource we need.

Ed

[toc] | [prev] | [next] | [standalone]

#711660

On 11/5/2018 10:23 AM, Edward Prochak wrote:
> On Friday, November 2, 2018 at 5:08:06 PM UTC-4, Dan Riley wrote:
>> Edward Prochak <edprochak@gmail.com> writes:
>>> In response to sergio's comments and my own goal of trying to
>>> talk about physics in this group, I am starting this thread.
>>>
>>> So the goal is to investigate the limits of climate models.
>>
>> One place to look is this article:
>>
>> https://journals.ametsoc.org/doi/10.1175/BAMS-D-15-00135.1
>>
>> "The Art and Science of Climate Model Tuning"
>>
>> From the intro:
>>
>> "While the fundamental physics of climate is generally well established,
>> submodels or parameterizations are approximate, either because of
>> numerical cost issues (limitations in grid resolution, acceleration of
>> radiative transfer computation) or, more fundamentally, because they try
>> to summarize complex and multiscale processes through an idealized and
>> approximate representation. Each parameterization relies on a set of
>> internal equations and often depends on parameters, the values of which
>> are often poorly constrained by observations. The process of estimating
>> these uncertain parameters in order to reduce the mismatch between
>> specific observations and model results is usually referred to as tuning
>> in the climate modeling community."
>>
>> The need for approximate parameterizations that aren't derived directly
>> from first principles is one of the chief sources of uncertainty in
>> modeling, so understanding why and how the models are tuned is one way
>> to approach the limits of models.
>>
>> -dan
> 
> Cool. Thanks Dan. Just the kind of resource we need.
> 
> Ed
> 

Great!   looks like a good paper, have not read it yet, and no paywall.

[toc] | [prev] | [next] | [standalone]

#711661

On 11/5/2018 1:00 PM, Sergio wrote:
> On 11/5/2018 10:23 AM, Edward Prochak wrote:
>> On Friday, November 2, 2018 at 5:08:06 PM UTC-4, Dan Riley wrote:
>>> Edward Prochak <edprochak@gmail.com> writes:
>>>> In response to sergio's comments and my own goal of trying to
>>>> talk about physics in this group, I am starting this thread.
>>>>
>>>> So the goal is to investigate the limits of climate models.
>>>
>>> One place to look is this article:
>>>
>>> https://journals.ametsoc.org/doi/10.1175/BAMS-D-15-00135.1
>>>
>>> "The Art and Science of Climate Model Tuning"
>>>
>>> From the intro:
>>>
>>> "While the fundamental physics of climate is generally well established,
>>> submodels or parameterizations are approximate, either because of
>>> numerical cost issues (limitations in grid resolution, acceleration of
>>> radiative transfer computation) or, more fundamentally, because they try
>>> to summarize complex and multiscale processes through an idealized and
>>> approximate representation. Each parameterization relies on a set of
>>> internal equations and often depends on parameters, the values of which
>>> are often poorly constrained by observations. The process of estimating
>>> these uncertain parameters in order to reduce the mismatch between
>>> specific observations and model results is usually referred to as tuning
>>> in the climate modeling community."
>>>
>>> The need for approximate parameterizations that aren't derived directly
>>> from first principles is one of the chief sources of uncertainty in
>>> modeling, so understanding why and how the models are tuned is one way
>>> to approach the limits of models.
>>>
>>> -dan
>>
>> Cool. Thanks Dan. Just the kind of resource we need.
>>
>> Ed
>>
> 
> Great!   looks like a good paper, have not read it yet, and no paywall.
> 

there is an analysis (engineering tool) with modeling or in control
systems  called "Sensitivity Analysis"
where you have a blackbox  with many inputs and one or several outputs,
and you very each input by itself say +_10% and monitor how much and
which out put changes.

[toc] | [prev] | [next] | [standalone]

#711679

On 11/5/2018 2:04 PM, Sergio wrote:
> On 11/5/2018 1:00 PM, Sergio wrote:
>> On 11/5/2018 10:23 AM, Edward Prochak wrote:
>>> On Friday, November 2, 2018 at 5:08:06 PM UTC-4, Dan Riley wrote:
>>>> Edward Prochak <edprochak@gmail.com> writes:
>>>>> In response to sergio's comments and my own goal of trying to
>>>>> talk about physics in this group, I am starting this thread.
>>>>>
>>>>> So the goal is to investigate the limits of climate models.
>>>>
>>>> One place to look is this article:
>>>>
>>>> https://journals.ametsoc.org/doi/10.1175/BAMS-D-15-00135.1
>>>>
>>>> "The Art and Science of Climate Model Tuning"
>>>>
>>>>  From the intro:
>>>>
>>>> "While the fundamental physics of climate is generally well established,
>>>> submodels or parameterizations are approximate, either because of
>>>> numerical cost issues (limitations in grid resolution, acceleration of
>>>> radiative transfer computation) or, more fundamentally, because they try
>>>> to summarize complex and multiscale processes through an idealized and
>>>> approximate representation. Each parameterization relies on a set of
>>>> internal equations and often depends on parameters, the values of which
>>>> are often poorly constrained by observations. The process of estimating
>>>> these uncertain parameters in order to reduce the mismatch between
>>>> specific observations and model results is usually referred to as tuning
>>>> in the climate modeling community."
>>>>
>>>> The need for approximate parameterizations that aren't derived directly
>>>> from first principles is one of the chief sources of uncertainty in
>>>> modeling, so understanding why and how the models are tuned is one way
>>>> to approach the limits of models.
>>>>
>>>> -dan
>>>
>>> Cool. Thanks Dan. Just the kind of resource we need.
>>>
>>> Ed
>>>
>>
>> Great!   looks like a good paper, have not read it yet, and no paywall.
>>
> 
> there is an analysis (engineering tool) with modeling or in control
> systems  called "Sensitivity Analysis"
> where you have a blackbox  with many inputs and one or several outputs,
> and you very each input by itself say +_10% and monitor how much and
> which out put changes.
> 
I use that method here in these newsgroups all the time!

[toc] | [prev] | [next] | [standalone]

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

Back to top | Article view | sci.physics

csiph-web