Essay by Eric Worrall
Because when your models can’t predict the climate, what you really need is a bigger computer, right?
Supercomputer Gadi crunches climate change
By Marion Rae
Updated June 23 2022 – 6:26pm, first published 6:23pm
New funding will unlock powerful Australian simulators to give the world’s scientists a better chance of tackling climate change.
…
The inaugural director of the new facility, Andy Hogg, said cutting-edge computer simulations and models will crack open climate change, extreme weather events, and past and future Earth systems.
“This will not only mean more powerful and insightful research, but hopefully better decisions for the pressing challenges and acute stresses our nation and world face,” Professor Hogg said.
But he says computer models are only as good as the people behind them.
“We are creating an open source weather, climate and Earth system modelling powerhouse that anyone across the globe will be able to access,” Prof Hogg said.
Based at the Australian National University, the Australian Community Climate and Earth System Simulator can calculate predicted weather and climate conditions from a few hours to many decades in the future.
…
Read more: https://www.canberratimes.com.au/story/7793124/supercomputer-gadi-crunches-climate-change/
Open source sounds promising. But perhaps Professor Hogg should have also mentioned, climate modelling systems are also only as good as the data they use. As Anthony Watts and many others have documented, much of the world’s climate data could charitably be described as junk.
A better GIGO machine.
The real problem is knowing the right question to ask:
How many angel can dance at the head of pin?
or
How Unicorns reproduce?
are
Wrong questions.
Wrongest of all questions there, even as in the premise of philosophically theoretical one:
“How many multiverses are there?”
cheers
Freedoms Must Be Surrendered – I Robot.
If not worth it, then what!?
If I may. Pardon me, but got to do it.
“Eagles – Hotel California (Lyrics) – YouTube”
nice one
As long as they continue to abide by the IPCC directive to assume that CO2 is the control knob for atmospheric temperature as their starting point (the GI part), all they will ever get will be faster GO.
Yeah. They will be able to get from GI to GO a lot faster. Applause everyone.
Maybe a super duper computer will explain why overlaps collapse climate sensitivity. As if we did not know already…
I would like the 1976 U. S. Standard Atmosphere (Some places). Without looking up anything, the Mississippi River had exceptional floods from 1973 though 1975, 1974 down some. Atmosphere dropped so much rain to make 1973 and 1927 floods equivalent. Looking up I found that each put 125 cubic nautical miles of water in the Gulf.
125 cubic nautical miles of water is a lot of water.
That’s one hell of an amount for us mere mortals to contemplate.
And to think, some people get all hot under the collar if you don’t keep the water turned off while brushing your teeth, then turn it on when you are done brushing.
They should contemplate just one single cubic nautical mile of water if that was what they had available to use to brush 3 times a day. (Hint: That is one HECK of a lot of water. I don’t care if you’re Aquaman or not.)
The volume of water in the Gulf is about 580,000 cubic miles, or 380,600 cubic nautical miles (a rather unusual unit). 125 cubic nautical miles is thus 0.033% of the Gulf’s volume. If the Gulf were enclosed the extra water would have raised its level by 1.6 feet. That may seem like a lot, but compare it to the average depth of the Gulf, which is 5,300 feet – suddenly it doesn’t seem all that impressive.
E. S…good stuff….
IMO Probably a cloud cover that gets closer to a real Earthlike upward flux of 240 would be more representative, but it is hard to achieve in the online UChicago possible selections. Anyway, your point is valid, cloud cover, of which Earth has about 65%, renders CO2 forcing irrelevant in comparison…You can run it yourself as you have for the 800 ppm doubling case, you will get about 2 C warming if CO2 increases to 3200 ppm. To get to 3200 ppm CO2, fossil fuels will have been depleted twice over. The “triple amplification” by water vapor is already included by running Modtran with fixed RH.
There is no valid cloud scenario in modtran to represent an approximation of a global average. And in hitran you have it neither. Yet there are a couple of more or less linear relations that help to triangulate a valid approximation. And we are only discussing the LW side here.
The respective CRE (cloud radiative effect) depends on altitude, lapse rate and optical thickness. The latter point is why a “cloud cover of 65%”, including a lot of low optical depth clouds, is not really helpful. For any given CRE overlaps with GHGs, especially vapor, will be the larger, the lower the cloud is.
With these simple considerations you can pull out a lot of information even from the simple modtran.
Doing so you get 2W/m2 for 2xCO2 and only about 0.65W/m2 for WV feedback. And that is only good enough for a climate sensitivity of 0.67K. Yet this excludes other feedbacks, of which the most notable is “lapse rate feedback”, which is not a feedback all, but negative anyway. It is roughly a 30% discount on surface warming vs. troposphere warming. So that is only 0.47K for doubling CO2, plus any further positive feedbacks. Not that they would even matter at this point.
Ahhhh, yes, the historic data. Nail and head Eric.
Garbage-in, garbage-out holds true no matter how super your computer.
Just produces more garbage faster….
We need a volunteer to unplug Gabi. Problem solved. Next?
The \wind will stop blowing and the power grid will collapse hence shutting down Gabi no volunteer needed
Open Source, these days means, no actual shutting down possible.
🙂
Sorry guys there is no need for another climate model, so far the Russian model INM-CM is doing fine.
What we need is an ultra-MAGA Russian colluuuuuuuuuusion model. Probably Simon has one he can lend us.
But Chris, if you keep promoting a Russian science just because they got something right you’ll be accused of collusion, and arrested by the FBI…
Somehow, I’m sure that the Australian super computer will come to the conclusion desired by the Australian government of the day. CAGW isn’t science, it is socioeconomic vaudeville. Song and dance theatre as science. P.T. Barnum is roiling in his grave.
A “super computer” can do far much better politics than Australian government, especially an ‘open source’ one.
🙂
It’s already been established scientifically that there is no crisis.
So what are they going to solve exactly?
The political crisis, the ‘electrolyte crisis’ perhaps!
Hard, but hopefully solvable.
A wee bit harder than solving what a ‘woman’ is, but hopefully solvable.
🙂
cheers
If that Ozzie computer could solve the climate crisis, it would be remarkable feat given that there is no genuine climate crisis, just a manufactured one.
Who needs a new computer. The Science is settled, Al Gore said so..
They gave him the Nobel prise, so it must be true.
Michael VK5ELL
Righty oh, Nik! When supercomputer Gadi “cracks open climate change” will 2+2 sum to 4 within, or will just another simulation pleasing to its funders’ and operators’ established interests be coaxed forth from its ultra-quick new math? Honest koalas, wombats, platypuses, kangas and their joeys breathlessly await the answer.
Well see, it has worked already and it hasn’t even started yet. Crisis Solved!
/sarc
Where can I find data for climate crisis?
You should just watch the ECS & cloud feedback symposium sessions on the youtubes posted by dresser. Even Schmidt’s given em schmidt throughout. It’s great fun. https://www.youtube.com/user/dessler2/featured
Lorenz discovered that the climate, being chaotic, can not be successfully modeled. I just stumbled over the following quote:
link
But the Future proves the Past, in so many regards.
Like for example, the future of applied communism around the world, proved beyond any doubt the past of Marxism. Proved how silly, stupid and criminally delusional Marxism is.
And it is a given for high intelligent and highly conscious self aware beings, like us humans, to straggle at their best, for even the slightest of chances to have some kinda of the smallest ‘vision’ of the future to be.
cheers
42, all you need to know.
Any time now.
Life, the universe and everything.
So long and thanks for all the fish!
Would not be surprised if that the final answer, as in some way is stands also ‘clearly’ for;
“LOL” … Laughing Out Loud.
🙂
cheers
“42” = ASCII for “*”. “*” is of course a place holder for some meaning to be determined later. Or in a loose interpretation: “Whatever you make of it.”
Lost Adams far too soon.
via Roman numbers
LII, (L2), (<2) , <<-LL, <>-LII, LII-42-<>.
<> is the full inverse-opposite in meaning of v v – (v.v).
v-5, v v-5 5 (5.5).
v-5 danger, pestilence, big problem. v v-5 5 loud and clear pestilence.
just a silly exercise outside the ASCII 🙂
cheers
Got to say this;
very much probable, that my previous reply makes no sense at all there.
What so ever. 🙂
Oh well… no parapet shooters there, at all!
cheers
Now they can get the wrong answer faster and with more significant digits.
It’s much, much worse than we thought!
Ha ha ha
But how else would you pillage billions of taxpayers dollars?
ever changing climate is as good of an excuse and a permanent black hole as any.
You can spend a lot of money trying to change the weather. Someone gets it so it’s economic stimulus.
Yes a shinier Ouiji board, a bigger crystal ball and a bushel of tea leaves are all we need to predict what’s coming. They worked so well in the past didn’t they? This is the problem. People seeking and disseminating the truth will try any method and change methods as the need arises. People telling lies just stick with the same old routine over and over because the output is predetermined and the methods are irrelevant.
This should be pretty simple for the supercomputer. After all it only has to make a choice between two well known possible outcomes.
We have already had one “great thinker” chime in about the promise offered by supercomputer Gadi:
“To see what can be, unburdened by what has been. And then to make the possible actually happen.”
— Kamala Harris
Gordon:
Both you, and the Australians, need to read NASA’s Fact Sheet “Atmospheric Aerosols: What Are They, and Why Are They So Important?
Just Google the Title.
Have you forgotten that I found no need for further discourse with you regarding your SO2 aerosol claims not all that long ago?
Gordon A. Dressler
NASA states that SO2 aerosols are REFLECTIVE, and therefore causes cooling of the lower atmosphere, and the Earth’s surface..
Like a sunshade.
Their ppm .concentration in the atmosphere is meaningless. QED.
Totally off topic.
I ask that you quit your childish pestering of me with your garbage.
Gordon A. Dressler:
Fine.
Continue wallowing in your ignorance.
You really can’t do that quote without the choreography.
You forgot to include where she giggles uncontrollably. Every time she does that, she reminds me of Rosanne Barr.
Kamala Harris is the new Yogi Berra
I need a laughing emoji.
Computer models…an article in Nature yesterday (https://www.nature.com/articles/d41586-022-01668-1) gives us yet another example of the anti-correlation between models and observations.
“Some researchers argue that the record is simply too sparse to show clearly what is going on, or that there is too much natural variability in the system for researchers to spot long-term trends. But it could also be that the IPCC models are missing something big, says L’Heureux, “which is a more serious issue”. Seager thinks the models are indeed wrong, and that the planet will experience more La Niña-like patterns in future”.
They then go on to propose a Rube Goldberg-type convoluted way the supposed melting of Greenland could somehow affect ENSO:
“England and his colleagues model how an AMOC collapse would leave an excess of heat in the tropical South Atlantic, which would trigger a series of air-pressure changes that ultimately strengthen the Pacific trade winds. These winds push warm water to the west, thus creating more La Niña-like conditions. But England says that the current IPCC models don’t reflect this trend because they don’t include the complex interactions between ice-sheet melt, freshwater injections, ocean currents and atmospheric circulation. “We keep adding bells and whistles to these models. But we need to add in the ice sheets,” he says”.
You can’t make this up…
You mean that The Science isn’t yet settled?
So now they’ll be able to get the wrong answer quicker?
Does the supercomputer run exclusively on renewable energy?
Is it carbon neutral?
What happens to the supercomputer when the wind stops blowing in the middle of a model run?
Faster super computers simply spit out garbage quicker; they don’t “solve” anything,….
GIGO RULES!
Someone here rightly claimed “garbage in garbage out”, and size/power of the computer makes NO DIFFERENCE. So true…
“[…] charitably described as junk.”
That charity could stand for a lot more donations.
The *ahem* adjusted temperature series of various sorts lead one to conclude…………… anything you want.
They can’t even predict tomorrow’s weather.
The new Supercomputer Gadi will churn out junk results from junk data, exactly in line with government junk policy statements, faster than ever seen hereto.
What if the new computer doesn’t give the answer that Professor Hogg wants?
If your program is garbage, a faster computer just gets you the wrong answer sooner.
+10000.
This is the problem precisely. A faster computer is pointless if the program is wrong. The climate is so complex that it might not be possible to ever model it accurately. What is absolutely clear is that computer climate models are not improving at all despite running on faster and faster machines. This means that the assumptions built into the computer models are wrong.
Right, now if we can just complete replacement Earth, we can use it to compute a real model after it finds the question for which “42” is the answer, of course. That might have enough computational power to do the calculations properly. Analog computers are always best (for single tasks that is). (Do I need /sarc for obvious Hitchhiker’s guide references?)
Bill Toland:
Yes, the major assumption built into the models (that greenhouse gases are the Control Knob of Earth’s Climate) is totally false. They have little, or climatic effect.
The actual Control Knob for our climate is simply the amount of SO2 aerosols circulating in our atmosphere, primarily from random volcanic eruptions. And unless they can learn when the next VEI4 or larger eruption will occur, they will never be able to predict future temperatures.
Unfortunately, the cause of our changing climate is so simple that most people refuse to accept it, even though its premise is falsifiable (empirically testable), and has been tested and validated hundreds of times.
“little or no” climatic effect”,
Might it not be better to program it to solve the problem of how to make the grid cope with solar cutting out, heaters being turned on and EVs being plugged in during early evenings? If the bloody thing puts out any result that isn’t ” its worse than we thought” they will get a bigger one, and we are headed there no matter what you would find if science just shut up and did it’s job.
This is false.
A reliable analytical model only needs fundamental constants to produce a useful result.
For example, ocean surface cannot sustain a temperature above 30C and ocean water cannot exist below -2C. So I predict, to the first order, that the average global surface temperature will be 14C; the average of the two extremes.
This result will be proven more accurate than any current climate model. In fact, if you look at warm ocean regions like Nino34, the CIMP3 models of 2000 vintage are already provably wrong per attached with forecast 2020 temperate averaging around 30C and sustaining above 30C for months already. The actual Nino34 temperature has a very slight negative trend of minus 0.16C per century over the satellite era.
A structural engineer starts with fundamental properties like Young’s modulus and then use analytical models to determine stresses based on loads and then checks against yield strength to see if the structure holds up. Ability to actually determine the load cases comes with experience. For example, a dam spillway design will be based on some probable flood data and that data can be highly variable.
Empirical relationships are useful where the system is not well understood but beware system non-linearities. For example, the albedo of the clouds above oceans has quite a sharp minimum at 27C.
You are correct with regard to structural models. But the global Climatr Change models are trying to solve an initial-value problem, so those initial values are critical toward getting to a correct-ish answer.
No – the climate system is self-regulating. It has stood the test of time with massive internal and external changes as well as severe shocks.
A climate model should be able to start from any conditions and end up stable. That is the first step. If it cannot do that then it is bunkum.
I can confidently predict that no open ocean water can or will sustain a temperature above 30C. It is physically impossible with the current atmospheric mass. That should be a hard limit in all climate models. In fact, if they were useful analytical tools, they would arrive at that value no matter the starting conditions.
Yet tropical waters are above 30C in summer.
Yes for maybe August. Just like the Arabian Sea and Bay of Bengal in April.. But they cannot sustain above 30C and if you are looking at waters around Florida they are not open ocean. The surrounding land causes mid-level air divergence from the Gulf of Mexico.
You silly people post your arcana and make your little jokes while the leftists run roughshod over the real world. The “climate change “ scam is winning while you debate the number of angels on a pinhead. Speak up, grow a pair and engage!
Please enlighten us on your successes. What? Letter to the editor? Contact your local member? Harangue people on street corners?
Thanks for asking, Alexy. First, stop treating this as a scientific matter. It’s a political matter. Second, make your voice heard where it will make a difference, not here. Third, be clear and strong about the truth. “Climate change” is a lie, pure and simple, call it out as such and don’t equivocate. Finally, understand who the enemy is and fight them on every front. The enemy is the international left, the UN, the WEF, leftist oligarchs like Soros and Fink and Gates. Instead of blithering on about albedo and ENSO, understand you are in a war and start fighting.
It is a scientific matter. That is the basis of climate change rhetoric.
How do you propose I fight these entities? Wave my fist? Buy a pitchfork?
I will reask you the question I asked Patrick.
What have you done that has been successful?
Please provide (at least) three concrete examples of these “places where your voice will make a difference”.
Please provide templates for how we should phrase our contributions in those places to ensure that they “will” indeed make a desired “difference”.
Please provide answers to the question actually asked.
The only way to change the warmist’s propaganda is to make the argument that warmer is better. Most people do not care about the science, mainly because it is too far over their heads so they instead react emotionally. Sales 101. It will be much easier for people to accept warming as good rather than to teach them science that they have no desire to learn.
I couldn’t agree more.
We can disagree on whether there is any enhanced greenhouse effect due to elevated CO2, we can disagree on whether fossil fuel burning has raised the concentration of CO2 in the atmosphere in the first place. We can bicker about whether aerosols are a minor or determinative factor to observed temperature rise. We can dispute whether the observations are being validly collected such that maybe there hasn’t actually been any real rise.
We don’t need consensus on these points. If we agree that there is no climate emergency, and any minor warming from whatever cause is likely to be beneficial, that should be the main message. The “cure” threatens to k!ll us, and the sickness, is imaginary.
Garbage in, garbage out.
Maybe they can let it loose on the power question here in Aust. 1st the sun does not shine at night therefore no Solar Power. 2nd the wind does not always blow, today is the almost perfect example not a breathe of wind, therefore no wind power either. You work it out, where is our power coming from. Could it be coal.
It will all be fine once they replace coal with magic…
Thorium Liquid Salts Cooled Reactors https://www.copenhagenatomics.com
A PC in the hands of an excellent programmer can overachieve. A supercomputer in the hands of a charlatan will always underachieve but does at least get twelve hour clock time right twice a day.
What if nature doesn’t want to follow our human logic because randomness offers her so much more freedom and excitement?
Another team of programmers overestimating their ability to simulate the real world.
Quote:“Professor Andy Hogg, Chief Investigator of the ARC Centre of Excellence for Climate Extremes.
Investigator?
…….not researcher or scientist?
Centre of Excellence?
Climate Extremes?
They really do lay it on thick – just who exactly are they trying to convince?
Australia National University?
Super Computer
Are they:
A small start but at least they got Climate Extremes correct, with Australia being a desert that’s the only sort of weather and climate you will ever get there
## As John McEnroe famously raved at Wimbledon a long time ago…
You Can Not Be Serious?
I would remind Prof Hogg, “Pig swill in, Pig swill out” or something like that.
No-one know better than Andrew McC. Hogg that carbon dioxide cannot be a signficant driver of atmospheric temperature (Glacial cycles and carbon dioxide: A conceptual model, Geophysical Research Letters, 2008) but, fortunately for is career, he doesn’t know it.
good at what?????
and computer models can predict climate..well as people behind them?
“A computer lets you make more mistakes faster than any other invention … with the possible exceptions of handguns and Tequilla.” — Mitch Ratcliffe
PS : Slashdot’s “Quote (/ Quip) for the day” about a month ago, which made me LOL …
Australia? what is the “CO2” content of electric energy there?
What are the induced “emissions”?
Good luck with this project, I wish them well. It can help clear the forest from the trees.
Processing faulty models faster and in more detail will make them better.
The goal of the supercomputer is to impress people.
The prediction is decided in advance (a coming climate crisis)
Seems to impress people, coming from a supercomputer.
They could make the same wrong prediction on the back of an envelope.
But most people wouldn’t believe that.
Could it solve the American inflation?
If you don’t know the natural rules , a supercomputer is useless !
Computers give us answers but do they solve problems? I am aware they can solve math problems but that isn’t what I am talking about. They give us information but just because you have more information that doesn’t solve the problem.
The problem is not the computer. It is the programme. The programme is useless. Why? Because there is not enough data to make it work, and they have designed it to need that data, instead of writing a programme that can work with the data available.
The programme, as a result, has to simulate data, and then feed it into itself, to pretend to work. Unsurprisingly, the programmer of the simulation always creates exactly the data needed to have a climate catastrophe as the output.
I am not sure if the better analogy is a self-unplugging robot or a clockwork orange.
It would probably help a lot if they actually started using models instead of linear progressions of unrelated data.
can calculate predicted weather and climate conditions from a few hours to many decades in the future
A few hours? I think we can agree most adults can already do that.
“you think it will rain later?”
“Nope”
I thought that the science was settled. So why do they need more calculations?