Skip to main content

TR Memescape

  • TalkRational: Supernaut is less gay than 90% of the men on TR.

Topic: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial (Read 432 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #25
Here we see a cameo of everything that has gone so horribly wrong with climate science.  Now we codify a model in a deeply non-linear system with ill-understood or unknown feedbacks, don't demonstrate any predictive skill yet claim it has some relevance. 

The model used in this paper was not created to make predictions or forecasts,

:facepalm:

Quote
Despite the reduced horizontal resolution, the model has a stable and realistic control climatology and has demonstrated utility for simulating the past, present and future evolution of the climate system

and the paper does not offer any predictions or forecasts.

The whole fucking paper is a modeled prediction starting from the last glacial melt water pulse you halfwit.  Now I understand Lamian's old sig.  You are a truly implacable waste of groceries.

  • F X
  • The one and only
Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #26
 :parrot:
"If you pick up a starving dog and make him prosperous he will not bite you. This is the principal difference between a dog and man."
― Mark Twain 🔭

Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #27
Here we see a cameo of everything that has gone so horribly wrong with climate science.  Now we codify a model in a deeply non-linear system with ill-understood or unknown feedbacks, don't demonstrate any predictive skill yet claim it has some relevance. 

The model used in this paper was not created to make predictions or forecasts,

:facepalm:

Quote
Despite the reduced horizontal resolution, the model has a stable and realistic control climatology and has demonstrated utility for simulating the past, present and future evolution of the climate system

and the paper does not offer any predictions or forecasts.

The whole fucking paper is a modeled prediction starting from the last glacial melt water pulse you halfwit.  Now I understand Lamian's old sig.  You are a truly implacable waste of groceries.
Evolution of the system is about tipping points. It's not telling people weather reports, it's looking at feedback loops that affect attractor basins. i.e. stable patterns we associate with climate in this case. Those models are concerned with boundary conditions and state flip. At least, that's what it means to predict the evolution of a dynamic system. You can't predict future points even on a Lorenz attractor, let alone an open dynamic system. That's a different kind of predictive value.
Love is like a magic penny
 if you hold it tight you won't have any
if you give it away you'll have so many
they'll be rolling all over the floor

Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #28
Evolution of the system is about tipping points. It's not telling people weather reports, it's looking at feedback loops that affect attractor basins. i.e. stable patterns we associate with climate in this case.

You're preaching to the wrong choir.  I have a deal of sympathy for your position here but it's the modelers you need to be berating.

Those models are concerned with boundary conditions and state flip.

Ah, no, the model used in the OP paper is a General Circulation Model (GCM).

Quote
Modelling the impact of a Southern Ocean freshwater pulse
Our simulations used the Commonwealth Scientific and Industrial Research Organisation Mark version 3L (CSIRO Mk3L) climate system model version 1.2, comprising fully interactive ocean, atmosphere, land and sea ice sub-models22, 63, 64. CSIRO Mk3L employs a reduced horizontal resolution and is designed for millennial-scale climate simulations. The ocean model is a 'rigid lid' model (i.e., the surface is fixed), with a horizontal resolution of 1.6˚ latitude × 2.8˚ longitude and 21 vertical levels, while the atmosphere model has a horizontal resolution of 3.2˚ latitude × 5.6˚ longitude and 18 vertical levels. A smoothed version of the 2-Minute Gridded Global Relief Data (ETOPO2v2) topography (https://www.ngdc.noaa.gov/mgg/global/etopo2.html) was used for the simulations (Supplementary Fig. 7). Despite the reduced horizontal resolution, the model has a stable and realistic control climatology and has demonstrated utility for simulating the past, present and future evolution of the climate system63,64,65.

This particular GCM is described here https://eprints.utas.edu.au/2278/3/03CHAPTE.PDF

Quote
The development of Mk3L from these components is described in detail by Phipps (2006). The resulting model is computationally efficient, portable across a wide range of computer architectures, and suitable for studying climate variability and change on millennial timescales.

And

Quote
The model physics is described in Section 2.2,

And while revealing this to be a very impressive beast of a GCM, nowhere is there any mention of attractor basins or state flips.  These models operate by solving the relevant equations - or what are taken to be the relevant equations - across grid boundaries using finite element and/or finite difference techniques.  It's a brute computational force solution.

At least, that's what it means to predict the evolution of a dynamic system. You can't predict future points even on a Lorenz attractor, let alone an open dynamic system. That's a different kind of predictive value.

I also have deep reservations about the likelihood of success of this kind of modelling effort given the nature of the coupled dynamical system under consideration.

  • F X
  • The one and only
Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #29
Here we see a cameo of everything that has gone so horribly wrong with climate science.  Now we codify a model in a deeply non-linear system with ill-understood or unknown feedbacks, don't demonstrate any predictive skill yet claim it has some relevance. 

The model used in this paper was not created to make predictions or forecasts,

:facepalm:

Quote
Despite the reduced horizontal resolution, the model has a stable and realistic control climatology and has demonstrated utility for simulating the past, present and future evolution of the climate system

and the paper does not offer any predictions or forecasts.

The whole fucking paper is a modeled prediction starting from the last glacial melt water pulse you halfwit.  Now I understand Lamian's old sig.  You are a truly implacable waste of groceries.
Do you actually think, for just a moment, that an alarmists will be honest about this?  Or anything else?
"If you pick up a starving dog and make him prosperous he will not bite you. This is the principal difference between a dog and man."
― Mark Twain 🔭

  • F X
  • The one and only
Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #30
I pretty much gave up hope for honesty from climate alarmists back in 2010
"If you pick up a starving dog and make him prosperous he will not bite you. This is the principal difference between a dog and man."
― Mark Twain 🔭

Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #31
Evolution of the system is about tipping points. It's not telling people weather reports, it's looking at feedback loops that affect attractor basins. i.e. stable patterns we associate with climate in this case.

You're preaching to the wrong choir.  I have a deal of sympathy for your position here but it's the modelers you need to be berating.

Those models are concerned with boundary conditions and state flip.

Ah, no, the model used in the OP paper is a General Circulation Model (GCM).

Quote
Modelling the impact of a Southern Ocean freshwater pulse
Our simulations used the Commonwealth Scientific and Industrial Research Organisation Mark version 3L (CSIRO Mk3L) climate system model version 1.2, comprising fully interactive ocean, atmosphere, land and sea ice sub-models22, 63, 64. CSIRO Mk3L employs a reduced horizontal resolution and is designed for millennial-scale climate simulations. The ocean model is a 'rigid lid' model (i.e., the surface is fixed), with a horizontal resolution of 1.6˚ latitude × 2.8˚ longitude and 21 vertical levels, while the atmosphere model has a horizontal resolution of 3.2˚ latitude × 5.6˚ longitude and 18 vertical levels. A smoothed version of the 2-Minute Gridded Global Relief Data (ETOPO2v2) topography (https://www.ngdc.noaa.gov/mgg/global/etopo2.html) was used for the simulations (Supplementary Fig. 7). Despite the reduced horizontal resolution, the model has a stable and realistic control climatology and has demonstrated utility for simulating the past, present and future evolution of the climate system63,64,65.
it is implied by the systems model format. The modelers may well be misusing the type of model to say something like "in the future at x date, the total heat balance of the linked atmosphere ocean system will be y". That would be a misuse of that sort of model. What they can say though, is something like, "with a change of x in factor y, the regulating system z will move to a different pattern of stability or will simply break up and disappear. Or potentially, will no longer be able to regulate some positive feedback a such that ot becomes a runaway positive feedback loop" . In the last case there is a series of subtleties that would matter but they are largely case specific.

This particular GCM is described here https://eprints.utas.edu.au/2278/3/03CHAPTE.PDF

Quote
The development of Mk3L from these components is described in detail by Phipps (2006). The resulting model is computationally efficient, portable across a wide range of computer architectures, and suitable for studying climate variability and change on millennial timescales.
yeah, that's not saying what conclusions can be drawn from that sort of studying. But it is implied in the fact that it is a dynamic systems model that it can only provide systemic predictive utility.

And

Quote
The model physics is described in Section 2.2,

And while revealing this to be a very impressive beast of a GCM, nowhere is there any mention of attractor basins or state flips.  These models operate by solving the relevant equations - or what are taken to be the relevant equations - across grid boundaries using finite element and/or finite difference techniques.  It's a brute computational force solution.

At least, that's what it means to predict the evolution of a dynamic system. You can't predict future points even on a Lorenz attractor, let alone an open dynamic system. That's a different kind of predictive value.

I also have deep reservations about the likelihood of success of this kind of modelling effort given the nature of the coupled dynamical system under consideration.
That's a separate issue. A dynamic systems model (for an open system which is all of them except for engineering applications which are actually highly linearized regarding initial conditions) is always going to miss the butterfly effects for butterflies that were averaged in to regional generalizations. This matters for long term models quite a bit. More importantly for this situation, the open system here has quite a few subsystems and missing one entirely would ruin the model.
Love is like a magic penny
 if you hold it tight you won't have any
if you give it away you'll have so many
they'll be rolling all over the floor

  • F X
  • The one and only
Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #32
http://rsta.royalsocietypublishing.org/content/365/1857/2053

In general, because the true climate system is highly complex, it remains fundamentally impossible to describe all its processes in a climate model, no matter how complex the model itself is.
"If you pick up a starving dog and make him prosperous he will not bite you. This is the principal difference between a dog and man."
― Mark Twain 🔭

  • F X
  • The one and only
Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #33
Nobody can actually model a single thunderstorm either.  Hard as it is to believe, we don't actually know exactly what happens in a thunderstorm. 

And there are around 760 of them each hour.  Every hour, every day.  Just recently there is some quality video evidence of some of the activity above the storms,  But no model comes close to modeling how it is happening.

https://www.youtube.com/watch?v=E14OZ95hqGA

https://www.youtube.com/watch?v=pnAAB0z2o-k

Since thunderstorms are a constant and major factor for weather, they matter for climate models. (along with larger scale storms)

Not being able to model storms, as well as clouds, is why no climate model actually represents the real world. 

"If you pick up a starving dog and make him prosperous he will not bite you. This is the principal difference between a dog and man."
― Mark Twain 🔭

Re: Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial
Reply #34
http://rsta.royalsocietypublishing.org/content/365/1857/2053

In general, because the true climate system is highly complex, it remains fundamentally impossible to describe all its processes in a climate model, no matter how complex the model itself is.
this is true but kind of irrelevant unless you understand dynamic nonlinear models of open systems to predict point specific outcomes. They can't. But they can provide system state outcomes from specific feedback loops.
Love is like a magic penny
 if you hold it tight you won't have any
if you give it away you'll have so many
they'll be rolling all over the floor