What makes humans unique among life forms on this planet is the ability to take large and meaningful actions to thwart the natural order. Otherwise, we would be subject to the same constant waxings and wanings (of temperatures, of rain, of rivers, lakes and oceans, of places free of volcanic activity or mile-high glaciers) and random cataclysms that make life miserable and short for most creatures, at most times, in the planet’s history.

Yep. We call them glacial potatoes because they are well-rounded (from glacial polishing) and mostly about the size of potatoes.:-)

probably wont reach the end…
lynching is too good for the likes of me..

It ALSO depends on what KIND of simulation…

the kind of simulation that is used in things like design, or graphics, is basically finite modeling and tons of code for figuring out whats in front of, whats in back of, how to render transparency, etc… these models can be VERY accurate, and given the right code behind them, you can actually “run the model”. that is, make an engine or some device, describe the materials, and flows, and actually model the engine running and get some darn good guestimates as to the final outcomes.

what is used in Global Warming and certain other simulations is something called Cellular Automata… which goes all the way back to my youth and Conways game of life.

The Game of Life, also known simply as Life, is a cellular automaton devised by the British mathematician John Horton Conway in 1970

The game is a zero-player game, meaning that its evolution is determined by its initial state, requiring no further input. One interacts with the Game of Life by creating an initial configuration and observing how it evolves, or, for advanced players, by creating patterns with particular properties.

the funny thing is that this software shows things like Gentification are natural biological self organizing processes.

https://en.wikipedia.org/wiki/Conway%27s_Game_of_Life

this format of simulation has many variations, and not just the simple one that conway came up with, or even ones that degrade to “mathematical beauty”…

in a Global Warming simulation, if you were to write one,
would first start with the grid…
the representation of the thing you want to model.

this is quite equivalent to RESOLUTION..
the more grid squares, the more accurate your model CAN be (but not guaranteed to be)

IF you ask the average person, after explaingin a bit of this, what size they think a grid square should be in a simulation of the earth, as an intuitive concept… most of them actually pick pretty good resolutions… in terms of accuracy.. .

that is, without regard to limits, and the other structural quirks, they tend to pick a good fudge

that most common answer in my journies… 1 meter… some say 1 mile… or 1 kilometer..
they like ones…

the answers they give depend on their perception of the size of the earth in their reality…

so.. 196.9 million mi² [196,900,000]

so your model at 1 mile resolution would have 38,769,610,000,000,000

thats a problem.. your starting to hit something called the Tyrany of Numbers
at least that was what it used to be called before we lost it…
(i still remember it but can never find it)

38 Quadrillion cells in your spreadsheet is a bit much, eh?
so now you start cutting it down..

how far down? well that depends on what resources your talking about..
given the global warming people are not model experts and dont really want such (you can tell cant you? look up who the greatest people in computer modeling and you wont find them flocking to global warming centers)

if you make each grid square 400 miles on a side, your now talking

492,250 X 492,250

so now we are talking 242,310,062,500 grid squares..

if each square held one value, the above is how many 64 bit bytes
you would be using (32 bits integer, 32 bits decimal)
unless you extended the floating point bit size.

the computations follow conways life…
the square in question is computed along with the 8 squares around it to create the next iterative state.

so that means your memory usage just doubled..
one to hold current state, one to hold next state (which then becomes current state)

with one value, as in conways most simple life…
you get 9 values to compute, with the rounding error for all of them

your now about to see how the numbers blow off the table and become insurmountable

so now you have your 64 bit word, your 400 mile grid size, and your two arrays in memory
484,620,125,000 bytes each containing ONE VALUE

lets say that value is temperature… ok… we have the temperature of each square…

while the model is 2D, its actually 3D, the square you are modeling i the real world is not
400×400, and an infinitely think slice
its 400x400x400

now tell me what the temperature of that unit from sea level to 400 miles up is

ok.. so we have to use surface temperature…
we dont have 242,310,062,500 drones to have hover 200 miles up and 400 miles apart.
do we?

your going to find that your going to have to accept many layers of delusion to keep going..
ie. we dont have the computing power yet to do this right…
even IF we have all the data..
even if we could somehow compute the accurate average temperature of a 400 mile unit square

but you say, the planet is not smooth like a marble it has many heights and levels and mountains and more..

well, if your cubes are 400 miles on a side, your planet is a marble thats real smooth

8,850 meters is the height of everets… thats just above 1.3 miles with 398.7 to go…

but you say, temperature is not enough, you need more.
yeah, you need to add things like temperature, gas make up, flow, and so on

and the models grow like rube goldberg devices…

and they dont work, because their basis theory of fixing the error propagations of the various limits is fake
YES FAKE

They dont tell you, but if you program simulations, and are good, you would notice that their idea of fixing this is to keep aligning the data of the past till the model works and so the data in the future is right

might work for a finite element model (probably would)
but not this form of model…

so… now… lets say your going to add this.. and now you have about 1k of variables..

484,620,125,000 X 1000 = 484,620,125,000,000

and if you just Add them up and average them…
thats 8008 operations per square.

thats 3,880,837,961,000,000,000 operations, each with rounding errors and for one iteration of averaging.
not more complicated than that.

if you had one processor running at 10gig FPO a second…
it would take 388083796 seconds to do the calculations (while ignoring writing it to memory takes even longer)
6468063 mins
107801 hours
4491 days
12 years

so on one computer that fast, it will take 12 years to do one iteration..
if you need a thousand of them… you can take 1200 years to see the result
oh, but of course you want to run this many more times, right?
good thing cellular automata are highly able to be paralleled…

but note… the next question is.. how big a time tick?
1 measure a day? no, that ignores night vs day difference
2 ticks..
how about 1 second? 86400 units for a day, 31536000 for a year
and we are going to do what, 100 years?

and again… we hit a wall in computation time and so on..

and it gets WORSE..

next.. if people want to hear it…

understanding chaos theory makes their work even more impossible given conditions

This happened in the English midlands a few years ago when flood control channels had been allowed to become overgrown for the same reasons.

What Every Computer Scientist Should Know About Floating-Point Arithmetic by David Goldberg, published in the March, 1991 issue of Computing Surveys
https://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html

Floating-point arithmetic is considered an esoteric subject by many people. This is rather surprising because floating-point is ubiquitous in computer systems. Almost every language has a floating-point datatype; computers from PCs to supercomputers have floating-point accelerators; most compilers will be called upon to compile floating-point algorithms from time to time; and virtually every operating system must respond to floating-point exceptions such as overflow.

Floating-Point Computation by Pat Sterbenz, is long out of print….

Rounding Error

Squeezing infinitely many real numbers into a finite number of bits requires an approximate representation. Although there are infinitely many integers, in most programs the result of integer computations can be stored in 32 bits. In contrast, given any fixed number of bits, most calculations with real numbers will produce quantities that cannot be exactly represented using that many bits. Therefore the result of a floating-point calculation must often be rounded in order to fit back into its finite representation. This rounding error is the characteristic feature of floating-point computation.

This does NOT mean a simulation is wrong…

But it sets limits on certain KINDS of simulation types.

these kinds of models tend to use a cellular automata format.
and THAT format means its an iterative model not a finite (element) model.

this kind of model is most affected by the rounding problem and the BEST you can do is push the issue off in various ways…

i will explain that in my next post.. (and hopefully will be tolerated for those who actually want to learn enough conceptually from some detail and gedanken (a la einstein), to be able to understand what is going on!)

an iterative model is a loop model.. the iterations are ticks of time
now… does time tick? no… time does NOT tick…at least at our size
in fact, depending on whether your on a plane or not, time varies

In an empty universe time has to be infinite IF when you add matter, time slows. 😉
[call that Artfldgrs Theorem of the Void… ]

but for many simulations, and games which are forms of simulations
you need time ticks, so that is one DIMENSION of RESOLUTION (and there are many)

[please note that these terms may or may not match terms in the art, these terms are put forth here so that average people with average vocabulary and so forth can understand the concepts without having to translate a complicated term – both einstein and richard feynman subscribed to the model that if you really understand it, you can share it in common terms]

the fewer ticks, the less accurate, the less time it takes to run, and fewer ‘structural’ errors are injected.

the more ticks, the more accurate, the more time it takes to run, and more ‘structural’ errors are injected.

the above is foundational and often ignored..
its considered esoteric, many people who take up modeling never think of this stuff as its really really bordering on philosphophical, evne more so if you know the physics in detail (like i do) from top to bottom… ie. from Planck sizes to astronomical sizes, to dimensionals…

from what i see, these researchers in global warming are on par with the feminists in the social sciences.

they both produce obvious self serving wrong crap… often VERY wrong crap.
but they create it for power in policy not validity..

with the understanding above…

Round-off Errors and Computer Arithmetic
https://www3.nd.edu/~zxu2/acms40390F15/Lec-1.2.pdf

In a computer model, a memory storage unit “word” is used to store a number.
•A word has only a finite number of bits.
•These facts imply:

1. Only a small set of real numbers (rational numbers) can be accurately represented on computers.
2. (Rounding) errors are inevitable when computer memory is used to represent real, infinite precision numbers.
3. Small rounding errors can be amplified with careless treatment.

do not be surprised that (9.4)10= (1001.0110)2 can not be represented exactly on computers.

so what does a global warming simulation do with this?
well, it doesnt tell the public its limitations, that the more iterations the lower the accuracy
the more detail and more iterations the more errors are injected over the time its run..

this is why we cant do a good weather report from today out to next year

this is a foundational fundemental limit on simulations…
YES… you CAN make your “words” larger… our systems went from 8bits, to 16, to 32, and now we are at 64..

but the larger the word the slower the calculations…

this is why simulations are not easy, each move in benefit costs you someplace else
why?

well, information follows the same laws as conservation of energy and thermodynamics!!!
actually…energy = information
its how we know black holes evaporate

ultimately one of these foundational issues would be enough to limit things
but guess what, there are a lot more and they span the smallest realms of reality to the largest

meaning they are baked into the pie and you deal with them, but cant remove them
[not when time, and energy are limited]

Same here in New England. Each spring we have a new crop of rocks in our yard as the frozen ground in winter pushes up the detritus of the what the glaciers left behind. Like weeds we have to pull them, out and then get a new crop the next spring 🙂

I’ve said for years that, particularly since we and our preferred foodstuffs are adapted to a not particularly warm interglacial, what we need to be doing is making sure we and our systems are tough enough for a period of NO ICE.

Whether or not global warming is happening is not as important as the results. More droughts, in some areas, often with more fires. More floods, in some areas.
Droughts and floods are subject to technological “fixes” to minimize the problem. Like the NYT in Feb 2018 warned about too much underbrush in CA was a fire disaster waiting to happen — but CA gov’t, pushed by “keep it all natural” enviro fanatics, didn’t take the steps to minimize fire risk in a drought.

More recently in flooding around the Missouri, there were prior plans to use the many man-made diversions of the water in a way to minimize flood risk, and flood damage. These plans were changed so as to help the environment remain “more natural”, putting a higher priority on bio-diversity based theories of limited interventions. Again, the gov’t policies making the flooding worse.

There should be more alarmism about droughts, fires, and floods — and how to minimize damage from them, including more dams, reservoirs, and flood control plains.

If there really WAS a crisis:
a) the UN would be holding all “climate change” meetings by internet video — many companies already do this for real work, not party going – ending “jet travel” for UN workers
b) building more nuclear power generators. India and China are both doing this, and the US should, too.

Also, none of the models have been able to explain the little ice age of the 13th and 14th centuries.

And finally, we know that massive volcanic explosions that send dust into the stratosphere can temporarily cool down the climate. If we felt it wise, we could always recreate this effect with hydrogen fission explosions.

Bottom line: I’m not losing any sleep worrying about it.