Modeling and Simulation
Earth Day 22 April
For a while I worked at NAVAIR’s Warfare Analysis division, where modeling and simulation ruled. A PhD there told me, “All models lie. Some are useful.”
I quickly learned in modeling and sim the results are only as good as the input data.
Poetrooper sends.
An Earth Day reckoning? Greenies panicking over the discrediting of computer models predicting disaster
By Thomas Lifson
We have learned the hard way that the scientists who produce mathematical models predicting disaster are not to be granted the presumption of infallibility. Nor are their demands that we impoverish ourselves in order to avert a disaster a matter of “settled science.” The coronavirus doom that the Cassandras of prestigious scientific organs predicted is nowhere to be found, and it turns out that models are based on assumptions and can be spectacularly wrong.
Americans have sacrificed trillions of dollars and turned our lives upside-down based on predictions of millions of deaths if we didn’t follow radically change our way of life, based on the pandemic predictions of modelers in the U.K. and USA.
How embarrassing for the modelers that tell us we have to make even deeper sacrifices for a global warming disaster that is decades away, if it ever develops. You can smell their panic in this planned propaganda blitz from the Olympian heights of the media establishment. Chris White of the Daily Caller News Foundation writes:
A project co-founded by the Columbia Journalism Review is asking hundreds of news outlets to focus their reporting on climate change on Earth Day as journalists focus primarily on coronavirus coverage.
The journalist heading the effort believes the media should be devoting the same level of attention to global warming as they do to a virus that has killed more than 100,000 people worldwide. (snip)
The founders behind Covering Climate Now are asking their network of more than 400 media outlets to blanket the airwaves with stories about climate change during the week of Earth Day. Reuters, Bloomberg News, and The Daily Beast are among the biggest names listed as partners of the project.
The project’s co-founder, Mark Hertsgaard, has said news outlets should be devoting the same amount of time to climate change as they do COVID-19.
Paraphrasing the Wizard of Oz, “pay no attention to the modeler behind that curtain!”
Read the rest of the article here: American Thinker
GIGO still applies. Thanks, Poe.
Category: General Whackos, Guest Link
Shout it from the highest tree tops, BULLSHIT!!!
BULLSHIT!!!
Anyone that can read will understand that our weather is controlled by cycles of the sun and nothing more. There have been many cycles of earth warming, climate change, whatever you want to call it for centuries, before man was ever here we had global warming! Notice we are still here and nobody died. Now go back and watch CNN>
My idea of proper modeling comes from Victoria showing off her Secrets.
Maybe we should grab all of the Chicken Littles that are running around screaming that the sky is falling and put them in the dumplin’ pot. Now the Chinesecommunist Originated Virus Infecting Disease of 2019 (COVID19) has infiltrated the Chinese owned Smithfield Meat Packing Plants causing it to shut down. There goes the price of bacon.
I used to be into modeling especially hotrods.
But then the glueheads ruined it and you couldn’t
buy glue anymore without a permit.
The same glueheads are now ruining the country.
Here’s a neat model:
https://www.amazon.com/dp/B000GKU2IO?tag=duckduckgo-ffab-20&linkCode=osi&th=1&psc=1
I still have the Guillows Giant scale B-17
in the box waiting to be built to match my
dads Fortress. Someday.
https://www.guillow.com/b-17gflyingfortress.aspx
Need glue.
1/28 scale? Holy balls! I generally stick to 1/48 for my planes. Big enough to see what’s going on, but doesn’t take up half the room, and most kits are affordable. 1/35 for ground-based stuff and 1/350 for ships for mainly the same reasons, also anything above destroyer-size gets spendy in that scale.
Once the Chinese Bat Flu—both the panic and the bug itself—subside, I need to get that pinup picture of her drawn up so I can get rolling on her B-17. I don’t suppose anybody here on TAH does that kind of artwork?
My dads nose art was a rabid wolf drooling
small bombs from it’s fangs.
The unit motto was “death and destruction”.
Your examples get me thinking about it.
Keep posting them.
There’s an outfit called Bedlam Creations that prints custom waterslide decals. Send them an image and tell them what dimensions it should be. That’s the plan for my wife’s B-17.
Nice! VP-40, the Fighting Marlins, stationed at NAS Whidbey Island, WA. Last squadron to fly the P-3 Orion, still active flying the P-8A Poseidon.
I knew that would keep you from
reminding me to stay on topic.
Damnit ‘beans, ya done gone and put them wings on bottomakwards. And the motors too. And sucking up to ‘Ed using Nauticalese will get you nowheres.
Yeahhhhh nobody saw that comment coming, KoB!
Shots fired!
?!?Whut??!
So maybe… if we pull the wings off at the root… swap them port and starboard… invert them and reattach… the engines would be on the right side of the wings.
Modeling says epic fail, the mechanics said no fuckin’ way!
Not really on topic. But maybe it is.
Heard a good way of looking at this mess a day or two ago. Don’t know who said it or even if it was from TV or radio. Anywho, the analogy is still good even if I can’t properly credit the idea.
When you start out to build a skyscraper, you don’t wait until all the materials are on site to begin the process. No, you start digging the foundation first with what is needed for that, then take delivery of materials for the rest of the building as you need it.
That is how we should be looking at ending this hiding from a virus situation. Yes, some folks should remain at home for the foreseeable future. Others need to go back to work so that all of us, wherever we are, can eat, for instance. Work on the foundation first. The top floor will come later – if you store the beams for it on site, they will just be in the way.
Now THAT is a model that makes sense.
That’s how our essential manufacturing business has been working…the foundation is our production team and we’ve been here at the plant working. Our office personnel are working from home, our VP of operations is here to physically move their production tickets and change requests throughout the departments as needed. We minimize risk with masking protocols and distance protocols. Not every business can duplicate that as easily, we are fortunate that way and thus far no one here has become ill….(knock wood).
Everyone else is home, we have been able to keep everyone working and no one has been laid off yet.
Now we need to let other businesses come back on line as you state and start the process to get people back to earning a living.
Twenty three million people instantly removed from their health insurance is a rather large risk pool on currently under or uninsured, a topic for a different threat but one that should give everyone pause….the government can take away everything while offering $1200 as compensation…that hardly seems adequate to me…
Time to get government out of the way and let people get back to work before the cure becomes worse than the disease if it hasn’t already.
Regarding the cost, I frankly don’t care at all what it costs. Not a single bean counter was worried about the cost of nearly two decades of war that made some companies an exceedingly good amount of money while costing the taxpayers a bundle…we never seem to worry about the cost of the Common Defense, but everyone becomes a fiscal conservative when we discuss the General Welfare…probably because caring for the average citizen doesn’t make any of the government’s owners a bundle of money.
All true.
Testify!
How much debt of ours does China own? it should be zero. Use that money to help repair our economy. Hopefully, the rest of the world would follow suit.
Okay … now show us your PERT chart!
Hmmmm…. modeling says that we should be having 50 to 60F here in my kingdom, but because there are snowfields up north, with the wind blowing southward over them, it’s in the 30s at night and I have to put out mealworms for the birds because the bugs haven’t emerged from hibernation just yet.
Modeling… it always assumes a steady state when the Chaos factor actually rules everything.
I’ve always loved the “ceteris paribus” assumption that I learned in Economics; “I know there are lots of other factors involved but for purposes of trying to sell my point I am going to ignore them all”.
Chaos is fun, too. Change the fifth decimal place on your input and your output goes through a wormhole and winds up in another universe.
That’s why real sciencers always specify the type and model of the hardware involved, the configurations, the language, software, and algorithm used, all those pesky little details. And the data, of course. Unlike some hockey-stick “scientists” who keep things secret.
No country on earth spends (read: wastes) more money than these United States trying to save the world from the evil scourge that is humanity.
Don’t know about y’all, but I’m tired of being treated like a virus or a liability.
Might be time to rise up and teach the technocrats why they issue mere suggestions, not rules/policy/law/decrees.
Ymmv.
There’s a more fundamental problem with using climate modeling to predict long-term climate change: model validation.
An unvalidated model is simply someone’s guess regarding the underlying math guiding the behavior of a system. That guess may be spot-on across the board; it may be correct for some range of data but not others; or it may be total crap. Validation – that is, the comparison of the model’s prediction(s) against reliable real-world observations – allows you to assess just how accurately the model represents and predicts reality.
We simply don’t have enough long-term accurate climate data to do that. And until we do, we can’t validate a climate model’s accuracy over the long term.
We can predict with some accuracy what will happen weather-wise in a few days. Over a few decades? No, we can’t – and that’s precisely what we need to know from a climate model if we’re going to use it to make the kind of decisions the environmentalist “doom and gloom” crowd want made.
I’ve said it before, but I’ll repeat it: show me a model that can predict – accurately – from the Medieval Warming Period through the Little Ice Age (1600s) through today, and I’ll listen. But I won’t hold my breath. And I’ll want to see the raw data used to validate it, too.
We already know Mann’s hockey stick model is total bunk, as he fudged and cherry picked his data.
“We simply don’t have enough long-term accurate climate data to do that”
Amen to that. Tree rings for crying out loud. Oh, excuse me, “Dendrochronology” (now it’s science).
Scientists can get a bit farther back then that.
Record-shattering 2.7-million-year-old ice core reveals start of the ice ages
By Paul Voosen
Aug. 15, 2017
Science Mag Link
So, do those ice cores give the weekly temperature hi/low for each year? Do they specify the precise location at which the readings were taken?
Rhetorical questions? Yes. But remember: for most of the planet, we don’t even have daily (or even weekly) hi/lo temperature data for anything other than maybe the last 50 to 70 years. Ditto rainfall, barometric pressure, and wind data.
I’m relatively certainly we need far more than we have to validate a model that’s going to be used to predict global climate 200 years from now with any accuracy whatsoever. That’s especially true if you’re literally talking about making resource use decisions that will affect the livelihood (and possibly the life and death) of literally billions of humans.
We didn’t even have accurate maps of most of the earth’s land surface until after WWII,to say nothing of that other 3/5 of the earth’s surface. Temperature and weather data? Laughable.
But we have a sample of what happened at one teeny weeny point of the earth’s surface 2.7 million years ago! Now that’s something to build a theory (and, evidently, a career) on!
Flown in a jet lately? Driven in a newer car? Used a newer cellphone? Pretty much everything you do these days has a computational model behind the design of it.
And just as we can’t use a computational fluid dynamics (CFD) used to examine how a plane will fly to recreate a flight based off some imprecise telemetry, we can’t use a climate model to reproduce past climates based off imprecise inputs. This is for a variety of reasons, including the lack of input data, the simplifications the models take, the resolution of the mathematical approximations, etc.
But, just like the use of CFD to test plane designs, we can see how the modeled environment -be it a plane, or the climate- responds to some new input parameters. For a plane, that might be higher wind shear, speed, or some other key variable, and for climate models that might be changing CO2, ice albedo, solar radiation, etc. It’s a tool to experiment, and understand, “What if?”
This is a fundamental misunderstanding of how they’re used. It’s like how some people are willfully ignorant of purchasing power when talking about military budgets – yes, the US DOD budget is ~4x that of China, but due to purchasing power differences, it’s far closer than that.
Finally, the other thing about model accuracy is how you measure it. If a virus model predicts 80K dead in the US, and only 60K die, did the model overestimate deaths by 33% (20K more than 60K), or did it overestimate deaths by 0.006% of the population (20K of 330M)?
Finally, accuracy isn’t a binary thing. If a doctor performing brain surgery is off by an inch, that’s a problem. If a Tomahawk cruise missile ‘misses’ its target by an inch, it’s not.
no, they don’t. Jets, cell phones, etc are all systems that are in existence to day and the designs are VALIDATED to work.
Models for future designs are based on VALIDATED designs.
For climate, novel virus, etc, the models are based on ASSUMPTIONS, not validated and working designs or data.
ASSUMPTIONS are often based on nothing more than ASSERTIONS or GUESSES.
Last, yes, if the model forecasts 200K dead but less than 25% of that occurs, the model AND the underlying ASSUMPTIONS are wrong, wrong, wrong.
It is actually unfortunate that we use the word model for both forecasts of assumptive concepts and testing of validated data and designs.
But I can tell you that I do not model ANYTHING for IT design that I cannot intimately define with validated engineering or material data beforehand.
Unlike climate science where an assumed level of “forcing” is simply input and data simply changed to match the output. Such as the sudden cooling of the warmest era in US history (1920–1950) to “prove” the models are correct.
Reminds me of how people jump on “theory” in the theory of evolution… or “militia” in another context. Linguistic imprecision.
I should also point out that models are the basis of sims, and do not in themselves create outputs–only assumptions.
But sims are notorious for producing garbage results due to cognitively challenged morons creating them.
Which is where we prove the saying that “in theory, it works, but in practice not so much”
Hey now, easy on the models and sims!! My Functional Area is Modeling and Simulations (FA57).
Of course I’ve seen how it can be manipulated too.
Fort Knox 2005 CCTT, suicide bomber in the scenario blew up multiple M1As Abrams….
Believe it or not, the physics behind climate models are all validated as well – not once, but continually. Yes, there are approximations… just as there are in the models behind jet planes, cell antennas, etc.
The virus models are quite a different beast from the climate models – the virus models are far simpler. They’re generally a single equation, using a few factors, all of which are approximated. The climate models are generally using well-understood physics, well validated by numerical models. Yes, approximations abound, and getting exact predictions is not possible – but getting differential analysis based off changes in variables certainly is.
A plane that looks good as a model but fails in the wind tunnel doesn’t fly. Even some that pass the wind tunnel still don’t work that well when they finally leave the ground. Reference the F-104 Starfighter, AKA the Lawn Dart, about which the Germans said you could have one of your very own (condition notwithstanding) if you just buy a random plot of land and wait.
Not a single one of these climate models has even survived the wind tunnel stage.
Modern climate models survive the ‘wind tunnel’ stage on a daily basis; individual physical processes are validated in direct ways, and the complex systems are validated via statistical measures.
I think comparing model technology from the 60’s with model technology today is impossible. The cell phone in your pocket is more capable than the biggest supercomputers back then, and can do things -like being a cell phone, or translating languages on the fly, or recording video- that someone back in the 60’s would not have thought remotely possible for a phone.
You may be underestimating what we dreamed of during the 60’s. What some of you young’uns too often fail to recognize is that it is we oldsters who dreamed the dreams that led to the designs which set things in motion to develop the technology so many now take for granted. Of course we were taught systems and science by dreamers even older than we were who passed along how to make dreams happen.
That’s fair enough – there were clearly big dreamers, even if it was mostly seen in things like television (Star Trek) or science-fiction (Asimov or Clarke). And some of that filtered into common culture. Maybe a better way of looking at it, at least to me, is the staggering advances in computer graphics over the years.
One need only look at the progress from black and white, low-resolution CRT displays to high-definition, millions-of-colors displays common nowadays, to the recent birth of AR/VR devices that put you ‘in’ scenes. Could it be predicted? Sure, people did. But comparing a high-definition augmented reality display with a 5-inch black and white raster tube showing still images back in the 60s really drives home the vast differences in capabilities of modern hardware.
Those same differences apply to modeling. In fact, these days the top supercomputers in the US use graphics cards originally designed for video games to run large calculations!
Climate models do NOT explain why the Carboniferous atmosphere was 30% oxygen (and had massive fires as a result, and before that, the Silurian period also at 30% O2) and supported Very Large Dragonflies and centipedes that were up to SIX feet long, and we now have a 20% O2 atmosphere, instead.
Models for a nonstatic and chaotic medium like a planetary atmosphere are pure hogwash. They do NOT work. They are NOT related to reality, period.
And yeah, dragonflies can grow much larger in one generation, if they are raised in an atmosphere with a higher level of 02 than the current 20%.
I know models; I don’t know why the carboniferous atmosphere was 30% oxygen, and why climate models apparently (according to you) can’t explain that. My guess is that lacking a lot of the input conditions -since we lacked full-spectrum sensors in the Carboniferous- has something to do with that.
More importantly, they do pretty good at predicting modern climate on decadal scales. Which is probably more important than going back 300M years or so.
I’ll deal with the chaos aspect in reply to Hondo below.
Oh, it’s simple, LC. Plants were the dominant life form in the Carboniferous epoch. Animals were mostly insects and arachnids on land and some aquatic critters in the waters. The Carboniferous period is also known for continental drift.
Surely, you’re aware of Meganeura, the giant dragonfly??? Are you remotely aware that certain insects like dragonflies can increase exponentially in size in ONE generation alone by simply increasing their oxygen intake? That’s already been proven in the lab.
Considering that the Carboniferous period began approximately 358.9 million years ago and ended 298.9 million years ago, and that the 30 percent O2 content of the atmosphere allowed massive fires to blossom, and our current atmospheric O2 level is 20%, we’re lucky to be alive at all.
The Carboniferous period is best known for its plant life, which later became coal veins over a very long, long period of time. I have a few fossils from that time. Its duration of approximately 60 million years makes it the longest period of the Paleozoic Era and the second longest period of the Phanerozoic Eon.
The difference between then and now is the oxygen level. If we were sent back in time to the Carboniferous period, we’d be subject to oxygen poisoning, also known as hyperoxemia.
And I have no idea why you would expect a climate model of any kind to explain why dense vegetation adds high levels of O2 to the atmosphere. That is just silly.
LC, the money quote from your comment:
“It’s a tool to experiment, and understand, ‘What if?'”
And therein lies the rub–global climate change proponents don’t use modeling to say “What if?” but rather to say “What is,” what clearly and definitively IS.
When it ain’t…
Well, I would be inclined to disagree, or at least edit that somewhat – sure, some people, who generally aren’t experts, are using them to say what definitively is… but most scientists who use them know that’s not the case.
Now, if you do a large set of simulations, varying multiple parameters (erasing away some of the effects of chaos, as I’ll elaborate on below), and looking at one change, like an increase in CO2, and 98% of the results show a warming effect… well, that’s statistically significant. Close enough to say, “We’re pretty damn sure this is the likely result.”
Can it be wrong? Sure. If you were a betting person, you wouldn’t take those odds, though.
Yep. Models that have been validated by decades of experience using and refining them. And even then they build prototypes to check out the real world results of the models, sometimes with surprising, even fatal, results; specially instrumented “test” aircraft flown by “test” pilots on “test” flights.
If the same rigor was used in the pseudo-science of climatology perhaps they would get believable results and a little more respect. Until then a bit more humility is in order.
The CFD models used by the aerospace industry have been extensively validated using real-world wind tunnel and flight test data, LC. Over the ranges for which they have been validated, they are known to produce accurate outputs. Aerospace design engineers can thus use them with confidence in their outputs, provided they stay within the models known regions of validity.
Climate models, to date, have not been similarly validated – at least not over a significantly long period (e.g., a period of a couple of centuries or more). That degree of validation is required if we’re going to use them to predict climate changes over the next century. Without that type of validation, we simply don’t have a clue whether or not the model output is anywhere close to accurate. Anyone selling such output as “settled science” is IMO selling snake oil.
And IMO climate models never will be adequately validated during our lifetime.
Why? Because as you pointed out we simply don’t have accurate historical climate data that goes back far enough from enough locations. For starters, the thermometer wasn’t even invented until 1709 – which is after a couple of climate events that still aren’t fully understood (the “little ice age” of the 1600s and the Medieval warm period). Any climate model to be used to predict future climate that can’t retroactively replicate those accurately – as well as events since – simply isn’t “ready for prime time”.
Finally, input data accuracy can indeed be a problem when modeling a system like climate that likely has a chaotic component (early weather modeling was found to have a chaotic component, and it’s a good bet that climate modeling does as well). Lorentz found that out when he did some early predictive weather modeling – and round-off error or trunctation (in the 6th or 7th significant figure of his input data, if I remember correctly) was found to dramatically affect the results from run to run on what otherwise appeared to be the same input data. The result was the development of chaos theory.
Could we say that the “models” used back in the early 1970’s were repudiated? Seems like the same people who were claiming that we all would soon be dead because of global cooling then were repudiating their own “data” and “models” when they claimed a few years later that we were all doomed because of “global warming.” When that didn’t work out either, they simply changed the “data” and the “models” to predict “climate change.” And somehow also claim that those of us who see their hypocrisy are “climate deniers.” (As if anyone actually denies climate or climate change, but that’s another topic.)
There are entire fields in Math and Comp Sci devoted to studying the effects of just those things you mentioned. And the study of error is ubiquitous; they even have a term for it (heh).
Error is baked in the cake. To name just one, the value of Pi; pick a value, you’re wrong!
Back in high school physics we had to recite the value of Pi out to some astronomical number of digits past the decimal point. It was brutal. And we were expected to use that one to do our calculations. Without a computer. Well, we had “computers.” They were what would now be called a calculator. We used slide rules.
Then there was proving the inaccuracy of the formula E = mc squared. Don’t ask. Please.
Having also had to recite pi to way too many decimals, here’s a neat article by NASA about just how many you need to be ‘accurate’. It’s a lot fewer than you’d think:
https://www.jpl.nasa.gov/edu/news/2016/3/16/how-many-decimals-of-pi-do-we-really-need/
Tried to remember how many digits we typically used back in that physics class before reading the article. Just can’t remember, but doubt that it was 15. Seems like maybe 7. Or not. Dunno.
Data point, the TI-85 graphing calculator goes to 9 decimal places.
I didn’t have to go far, it lives almost exclusively on the nightstand.
The climate models have been too, to the extent that it’s possible. You have to first understand that there is no coverage where you can say a model is validated under every possible configuration. That’s not true for the CFD models behind planes, which often use idealized conditions instead of turbulent simulations. And even when they do turbulent simulations, that pesky thing that you refer to called chaos implies they can’t test every condition.
Yet, they still validate, because with enough statistical coverage of a phenomena, you can be reasonably sure of how to simulate it.
Validating climate models for centuries is obviously not possible; that’d be like asking that a vaccine for COVID be tested for 100 years before we give it to make sure no age-related effects happen. We can’t do that.. and we don’t need to do that. Because we know enough about the individual components of these things to validate them in simpler ways. Want to validate your albedo effect calculations? We don’t need hundreds of years when we have countless different angles, surfaces and telemetry we can test against.
As for chaos, yes, exact calculations are impossible – but that’s already true in the finite precision computers give us, yet models still prevail. This is partly because we’ve learned to do statistical analysis of complex systems, including chaotic ones, where instead of getting a singular correct answer, you get a probability density function, for example. And it’s also partly because chaos is limited within certain constraints – both physical and mathematical.
Look at the structure of the common examples of a Lorenz attractor for one example of that, or for a more intuitive grasp of the limits of chaos, look at how with our ‘chaotic’ weather system, we don’t have temperatures swing out to, say, 200-degrees F. This are physical constraints by the physics of the system; the same physical constraints modeled by the simulations.
Chaos theory is often waved around as misunderstood limit on things much like quantum mechanics is. In reality, both are subject to analysis, error estimation, and validation.
The Chicom Virus models all pointed to a disaster beyond the scope of the Black Plague through the Middle Ages and that we all had to go around and flagellate ourselves to become worthy of the cure provided by the Spirits of the On High !!!!!
Yeah, well….
Now we see it even more clear, the climate change bull shit saying that this virus is all because of “Globullshit Warming”, to quote the Pope, and that we have to get right with the powers of a one world government to become worthy enough to be able to survive the coming climate catastrophe…
Yeah, well, on another account….
Neither of these things are #1 true and #2 it never will be true, and finally #3, it is NOT an extinction level event on either case.
I learned from my Grandfather another three key7 points of dealing with the media or life in general.
1) There is always a crisis.
2) It’s NEVER as bad as they say it will be.
3) The cure is ALWAYS worse than the disease in the first place…
But it’s a cold weather bug, Thunderstixx. It doesn’t like warm weather. It will very likely be back next fall, when we least expect it.
Some of these viruses, like the flu viruses that change every year, prefer cold weather. Is that because they know that people like to rub their noses after they sneeze?
We’re all DOOMED!!!!!!
(Nobody’s said that for a while. It needs to be said occasionally.)
OK, then!
Doom on you! Doom on you! Doom on you!
Well. Well, well, well. Just for that, think I’ll eat the last piece of chocolate cake. Neener, neener, nee-ner.
I have to ask why farmers who normally plant corn and soybeans are not planting popcorn this time?
It seems to me that popcorn, which is a grain commodity, might have a better price on the commodities markets than field corn or soybeans, especially with shipping being somehow semi-interrupted.
I think I will dig out my Whirlypopper and make a bowl of popcorn tonight….
Summer will be here, some day.
” It doesn’t like warm weather”
Neither do I, but I persevere. Color me skeptical until I can find out exactly why it can’t survive at body temperature or less, outside the body.
Don’t know, of course, but would suspect it has something to do with fresh air, Vit D from the sun, exercise, and a few other things of that sort which one tends to get more of outside than inside.
The most salient point from the above quote, “news outlets should be devoting the same amount of time to climate change as they do COVID-19.”
The talking heads have driven the panic of this pandemic more than the actual virus. Incessant reporting of the climbing – cumulative – death told without concurrent reporting of the – cumulative – number of those recovered is sensationalist journalist at its finest. Or worst, depending on your perspective.
That quote above proves they are aware of the impact of their fear-mongering. Worse, they want to do it again for their next agenda item.
Latest figures scream “800,000 infected world wide!” Means that to date 7,000,000,000 uninfected.
That’s why people go into journalism;
To change the world”. Kind of arrogant for a degree that is basically four years of freshman English.
If your predictions are always wrong, then your models and simulations are fantasy. These “experts” that can’t predict their way out of a paper bag are the modern version of Gypsy snake oil salesmen and tarot card readers. Gaze into the tea leaves and you’ll see that cow farts are destroying the planet.
The difference is that engineering is mechanical and can be tested, and modeling/simulations mostly work.
Anything in a chaos system like a planetary atmosphere is chaos, which does not respond to modeling or simulations and forecasting a steady state environment is baloney. There is no such thing in chaos.
Precisely!
Perhaps the wrong descriptive for a completely imprecise system.
Heh…
Somewhere in my vast library, POetrooper, I have a very good book on chaos theory. It is mind boggling. Now I have to dig that out of the chaos in my bookshelves.
Chaos absolutely responds to modeling – it just does it in a statistical manner. If somebody were to run a single instance of a climate model, and say, “Aha! The temperature in 50 years will be X in Spokane, Washington on July 30th, 2070!”, they’re an idiot.
If they run a (sufficiently large) set of experiments, they can gather statistical data from those runs, and build a probability map. They can then say (for example), “84% of our simulations pointed to a net increase of at least 2 degrees in July in Spokane Washington in 2070.” And that’s valid.
This is how studies of turbulence, another chaotic system, are done all the time. This effects everything from chemical mixing in any turbulent flow (air, water, whatever) to planes flying the sky.
Uh, no it does NOT respond to modeling. If it did, then weather forecasting would be far more accurate than it is, LC. Let me know when that sinks in.
Last week, snow was supposed to go 35 to 50 miles south of me. Instead, it all went from 5 miles south to the Canadian border and I had to shovel the confounded sidewalk, because the weather forecasters do NOT take the chaos factor into account.
Their forecasting software and systems are far more accurate than 20 years ago, but they STILL can’t forecast anything more than a BEST GUESS!
The wind can shift directions with no notice and turn a bright sunny day into a tornadic train wreck. Hail is forecast as rain – not even as freezing rain, just as rain.
This is chaos at work, and there is NO way to predict weather with real accuracy. Never has been, unless you count wet sidewalk (rain) and dry sidewalk (no rain).
They can’t even predict when and where lightning will strike, or a beastly cold stream of air coming straight down Lake Michigan in the middle of July, requiring all those without coats to go into stores and buy sweaters and coats to wear home before they freeze to death. Yeah, I’ve been caught in that kind of thing!
It does. I can send you plenty of articles about statistical analysis of chaotic systems – turbulence is a big one, as I mentioned below. Your belief that chaos is why US weather prediction is so poor is what’s wrong.
And yes, I said US weather prediction – other places have fewer issues, like the UK Met Office, or the European center. Here’s one article that explains a little bit about it, and gets into NOAA’s plans to improve:
https://observer.com/2019/08/weather-forecast-noaa-prediction-models-accuracy/
What that article (in skimming it) didn’t get into is the fundamental differences in the models, too – the US models are only beginning to pull real-time data into the simulations, a process called ‘data assimilation’. The UK model has this integrated now, and I imagine the European center’s model does too, but I know less about it.
By some estimates, the US ‘forecast skill’ is some decades behind that of the UK model, largely because of this. Then you have issues, per the article above, about a lack of computing time. Want a higher fidelity forecast? You need more data points in the simulation, but that requires more hardware or time, or both. We lag on that. So much so that the US Air Force contracts out its weather forecasting to the UK Met Office.
And while you’re correct that you can’t tell exactly where wind will shift, you can do statistical studies like I mentioned above – which is exactly what all studies of chaotic systems do. It’s not ‘a’ simulation of weather, it’s an ensemble of simulations from which statistics are calculated, and ideally observational data is incorporated.
Again, chaos is not some unconstrained force – it’s open to analysis. If you still doubt this, just google ‘analysis of turbulent systems’ or something similar, and read away. As for lightning strikes, funnily enough they’re getting better at predicting those – here’s one such article on that. There’s still a long ways to go, and yes, limits to the capability, but it’s getting better:
https://www.popularmechanics.com/science/environment/a29703282/artificial-intelligence-lightning/
Your response ignores chaos factors completely. Tornadoes are the most chaotic weather events possible. The 2011 super outbreak was NOT predicted, but killed 324 people and injured 2,400 more. https://www.ustornadoes.com/2012/04/25/april-25-26-27-28-2011-super-outbreak/
The 1974 super outbreak was NOT predicted, started as a single storm and split into 148 tornadoes – NOT predicted – and killed 335 people, injured over 6,000 more, when three weather systems collided and spawned the super outbreak.
If statistics are SO accurate for weather predictions, how come these events weren’t forecast???? Get over yourself. Weather is a system built on chaos, NOT mathematical models, whether you like it or not. Snowstorms change course at will and peter out of turn into blizzards, tornadoes come out of nowhere with no forecast included, etc., etc., etc.
If these things were REALLY subject to statistical stuff, then how come they always surprise the meteorologists?
No, it really doesn’t. Go read the Observer link – if chaos meant there were fundamental limits to forecast ability, and we reached them, then how is it that UK forecasts are much more accurate than US ones?
And one of the main factors in forecasts is limited computing resources – especially for things like tornadoes, which require high resolution simulations. Think of it this way – if you want someone to be a marksman, and give them 5 bullets to shoot, they’re gonna be less skilled than someone who has 5000.
And to put things in perspective, if you’re unfamiliar with the effects of “Moore’s Law”, the biggest supercomputer of April 2011 had a whopping 1.7% of the capability of today’s biggest system. We don’t even have numbers as far back as 1974, but today’s biggest system compared to 1999’s biggest system is 62 thousand times more capable. Go back another 20 years and it’s on the same order of magnitude.
Comparing the capabilities we had back then with what we have now is like going into space, pulling out an old spyglass with a 2x magnification and poor optics, and proclaiming that modern NRO satellites can’t see shit because you, with your trusty spyglass, can’t. These things are orders-of-magnitude more capable than they were just a few short years ago.
Yes, weather is chaotic. Nobody disputes that. But you don’t seem to understand that statistical analysis of chaotic systems is routinely done, and that chaos is not the same as non-deterministic, or random.
We really need those climate change-ologists to step up with their models! I saw a story “on the weather channel” about three Cat 4 Tornados, and one Cat 3 that moved through Mississippi in less than two weeks, it must be due to the rising see levels, and temperatures ’cause they only happen along that place “known as tornado alley” like…each year in the spring and fall…never mind, same stuff different day!
“…dah polar bears..” -Al Gore
“How dare you!” -Some autistic teenager
“We choose truth over facts” -Creepy Uncle Joe
It IS happening, they can FEEL it!
I heard a State Trooper braging about how his boss said he was a model officer. Then he looked up model and saw it was a smaller imitation of the real thing.
My father spent over 40 years working on flight simulators for the Marine Corps. Quote Its a simulator not a duplicator Unquote.
Genuine “error” is evenly distributed. If you keep having to adjust a model in one direction, that isn’t “error”, it is ” fudging”.
If even most adjustment is one way, fudging.
Statistical Process Control is one way to expose tinkering, or just that you are not really controlling a process.
Repeated “Adjust the data” is blatant proof of a false theory or model.
“You can’t be allowed to see raw data” is also proof of falsehood.
Models that cannot accurately predict the past are worthless in predicting the future.
Funny how the current damn-panic requires finding “new” casualties to count, while handwaving the oddly empty hospitals.
And the trend is clearly down now. Despite Cuomo’s “peak?” chart with the “peak?” Arrow way on the down-slope.
That is fudging, folks. And the proggies are wrecking your lives to gain political advantage.
Again.
Progs always tout anything that supports their political objectives whether it makes any rational sense or not. They are totally dishonest and ethics free. Exhibit “A” is our own Squidturd.