Climate Change Models

[HedgeWizard]

Thought I would post this here;

 

http://onlinelibrary.wiley.com/doi/10.1002/2015EA000154/epdf

 

New publication regarding questioning the largely proposed sensitivity of the climate's equilibrium.

 

Thought it would be good sauce for those who can parse it. Maybe a good guy to bring on the show in between the immigration blitzes.

[TheAuger]

The problem with "climate science" is that they are comparing observed temperatures with these climate models (the IPCC has funded over 400 of them).  This is not empiricism, and therefore, it isn't science.

[Jsbrads]

To quote Richard Feynman, how you create your theory doesn't really matter only if it is supported by the data.

[jimofflorida]

If you have ever spent any time working with large scale simulations/models (which pretty much excludes almost everyone)  there is only one thing that comes to mind regarding such models...

George Box "All models are wrong, but some are useful"

For those that are "sure" scientists have climate all figured out, feel free to watch the results of a dozen billion dollar models predict the track of a hurricane 24hours into the future.

each type of model is "designed" to push the agenda that the senior model advocates think is the right way. These are all scientists that are sure they are right, each has convinced somebody to invest millions of dollars that they are right. Here comes the tricky part, they cannot "all" be right. The climate of planet earth has gone thru warm spells, cold spells, huge grass fires that burned entire continents, volcanic eruptions that darkened the skies; and yet here we are.

People are so wrapped up in the need to be right, they have forgotten how to pursue what is truth. Do not think that science is this emotionless vacuum in pure pursuit of knowledge. The only good thing it had, well "used to have" going for it was that people could freely exchange those ideas, defend them, attack with logical arguments until eventually that kernel of truth did pop up. Today we have lost the ability to have an open and lively debate about important topics. NASA most certainly DOES adjust the data when it comes to things like temperature, there are reasons why you need to do such a thing. The trouble is, when the reasons switch from science to political agenda. Now science has discovered they get a lot more funding if they produce results that match what the funding organization are expecting. 

and here is where it all falls apart, most of the populous (and all of Hollywood) have lost the ability for even a little logical thinking. They get all spun up about 0.002% change in something because media tells them they should. They have not looked at "climate" over the past 50,000 years. We live on a changing hunk of rock that is always evolving. Do people freak out over the Sahara desert today, I guess I should not mention it was not always like that!!! oh no Global warming;  AHHHHHH, run!

 

science does NOT mean everything said is inherently truthful...it means the ability to logically pursue truth is something you possess and are willing to do.

[ofd]

Quote

For those that are "sure" scientists have climate all figured out, feel free to watch the results of a dozen billion dollar models predict the track of a hurricane 24hours into the future.

How is that related to climate change?

 

Quote

They have not looked at "climate" over the past 50,000 years.

More like 400 000 years. 

 

Quote

They get all spun up about 0.002% change in something because media tells them they should.

Preindustrial C02 levels were 280 ppm, now they are around 400 ppm. That's a 40% increase. And if CO2 and temperature are correlated this means it may get warmer in the future overall.

[Erwin]

12 minutes ago, ofd said:

How is that related to climate change?

Because you can make a model do whatever you want it to do, and can easily be influenced by the developers' biases (which is very lefty).

14 minutes ago, ofd said:

Preindustrial C02 levels were 280 ppm, now they are around 400 ppm. That's a 40% increase. And if CO2 and temperature are correlated this means it may get warmer in the future overall.

If the correlation were causal, yes.

But the increasing rate of temperature increase took place a century prior to the fossil fuel industry. A can't be the cause of B if B came first.

[ofd]

Quote

Because you can make a model do whatever you want it to do, and can easily be influenced by the developers' biases (which is very lefty).

How is the inability to track a hurricane over 24 hours related to climate change?
 

 

Quote

 

But the increasing rate of temperature increase took place a century prior to the fossil fuel industry. A can't be the cause of B if B came first.

 

Not if there are many factors influencing climate.

 

[Erwin]

6 hours ago, ofd said:

How is the inability to track a hurricane over 24 hours related to climate change.

You were talking about climate change research / predictions and its reliance on models. I can show you a simplified example of how they do it if you like.

Jim was making the case that the models they use are basically crap. The financial models are just an example of that.

6 hours ago, ofd said:

Not if there are many factors influencing climate.

Perhaps I misunderstood, but I are you proposing that the number factors can reverse the order of causality?

[ofd]

Quote

You were talking about climate change research / predictions and its reliance on models.

Sure, but the models are different. And models change, get more precise take different data into account.
 

 

Quote

 

Perhaps I misunderstood, but I are you proposing that the number factors can reverse the order of causality?

 

I propose that the temperatures rose prior to the industrial age to their 'normal' levels because they returned to the mean after having been down in the medieval little ice age. Which was most likely caused by a combination volcano eruptions and sun spots. After the effects of those went down, the temperature returned to the mean. The temperature increased compared to a little ice age and got back to 'normal'. Now it increases propably because of CO2.

[Erwin]

9 hours ago, ofd said:

Sure, but the models are different. And models change, get more precise take different data into account.

Exactly my point. Who feeds the data? That is subject to the researcher or developers.

9 hours ago, ofd said:

Now it increases propably because of CO2.

CO2 is a gas, and all gases are subject to Avogadro's Law (AL): Pressure * Volume = molarity * Temperature * Rydberg's Constant

When the the temperature of CO2 or any gas rises, pressure increases. The Law of thermodynamics states that high pressure fluids move to low pressure fluids. In our atmosphere, the low pressure can be found in high altitudes, so that's where heated air moves. At low pressures, gases increase in Volume. When a gas increases in volume (aka decompression), Temperature decreases. Where the gas move up in temperature, the pressure approaches 0 kpa. That is such a low pressure that the particles never bounce off of each other to even have a measurable temperature. That's how cold the gas gets.

If you still don't believe me, consider that gas decompression is a central component of your refrigerator. The cold decompressed air is run through the pipes allowing heat transfer from your food to the gas, which heats the gas and cools the food.

So... Ya. It's not the CO2. The whole reason they say it's CO2 is because the experiments were run by heating air in a tank. As in, an environment where gas cannot expand, and thus, preventing the air to cool and normalize.

IMO, that experiment is so stupid that only an academian would fall for it.

[ofd]

 

Quote

 

IMO, that experiment is so stupid that only an academian would fall for it.

 

 

Or because atmospheric physics is a lot more complicated than you think. There are feedback loops, clouds play an important and so on. Because of that, the temperature gradient doesn't simply change constantly the higher you go up. Depending on where you are, it may get hotter the higher you are. Say at a height of 50 km, the temperature may just be as high as 3 km above ground. https://upload.wikimedia.org/wikipedia/commons/9/9d/Comparison_US_standard_atmosphere_1962.svg
 

Quote

The whole reason they say it's CO2 is because the experiments were run by heating air in a tank.

The reason CO2 and other greenhouse gases are though responsible is because they re-emit infrared radiation coming from the earth's surface to all directions, including the earth.

[Jsbrads]

Yes CO2 absorbs sunlight that reflects off the planet. This warms the atmosphere. However what percentage of the reflected light was the CO2 absorbing when the atmosphere was 280ppm? 1%? 13%? 87%? It makes a difference. 

Great source: http://clivebest.com/blog/?p=1169

[Erwin]

9 hours ago, ofd said:

Or because atmospheric physics is a lot more complicated than you think. There are feedback loops, clouds play an important and so on. Because of that, the temperature gradient doesn't simply change constantly the higher you go up. Depending on where you are, it may get hotter the higher you are. Say at a height of 50 km, the temperature may just be as high as 3 km above ground. https://upload.wikimedia.org/wikipedia/commons/9/9d/Comparison_US_standard_atmosphere_1962.svg

Gas tanks don't have clouds or feedback loops or temperature gradients; they are also simple enough to be used as examples for high school physics. What does a gas tank have to do with the atmosphere?

9 hours ago, ofd said:

The reason CO2 and other greenhouse gases are though responsible is because they re-emit infrared radiation coming from the earth's surface to all directions, including the earth.

 

7 hours ago, Jsbrads said:

Yes CO2 absorbs sunlight that reflects off the planet. This warms the atmosphere. However what percentage of the reflected light was the CO2 absorbing when the atmosphere was 280ppm? 1%? 13%? 87%? It makes a difference. 

Great source: http://clivebest.com/blog/?p=1169

My argument still stands. The atmosphere is not a gas tank. The greater the heat, the greater the pressure, and thus the faster that the gas moves up which cools the air via decompression.

[AccuTron]

Back up to the starting line:  all this alarmism was brought up to begin with by...drum roll...

-falsified data

-illegal math and physics

--HUGE money trail

So don't sweat (pun) any of it.  The climate is doing what it always has.  So are humans (see also "forked tongue").

[ofd]

Quote

What does a gas tank have to do with the atmosphere?

I have no idea. You brought up the gas tank as a refutation to my argument. 

 

Quote

The atmosphere is not a gas tank. The greater the heat, the greater the pressure, and thus the faster that the gas moves up which cools the air via decompression.

Not really. Read up the summaries or have a look at the graph I posted.

[Erwin]

1 hour ago, ofd said:

I have no idea. You brought up the gas tank as a refutation to my argument. 

Yes. Because the "experiments" they ran were about heating gas in a tank. So the results cannot be extrapolated  to the atmosphere.

1 hour ago, ofd said:

Not really. Read up the summaries or have a look at the graph I posted.

I did. Again, this tracks temperature vs CO2 concentration. Notice how they make no mention of volume? That's because they assume volume is constant, aka a tank. So your graph has no relation to the environment, but they don't mention it. That's their sleight of hand. Very dishonest of them if you ask me.

In the atmosphere, higher pressure air expands (volume increases), which cools the air down again. 

[Dylan Lawrence Moore]

My favorite part of the whole "climate debate" is that water vapor acts for about 90% of the greenhouse affect.

Second favorite is that the warming effects of CO2 are logarithmic.

Third favorite is that all the models for like the last 13 years have been wildly wrong at predicting the future climate.

Fourth favorite is religious reliance on said models. I did some computational chemistry my junior year at college, and we had a hard enough time predicting how 100 water molecules would act with each other under different parameters, let alone the weather on the entire goddamn planet. We were specifically told that the purpose of the models was to get an idea, which one could then go and do empirical tests on (for example, figuring out different surface areas of zeolites to see which would make the best catalyst). The models themselves meant nothing without further testing.

[jimofflorida]

and it looks like we have gotten around to the "faith" part of the thread that also occur with regard to climate change hehe

The science behind simulations/models is very complicated; as a system engineer I have designed massive system models and the effort it takes to "avoid" introducing bias is considerable. You can easily see what you want rather than accept that what the model is telling you is useless (working for months/years only to have to accept you have, well, nothing useful is tough to do). When I would create such system models (these were not network simulations, they were behavioral models) the most important factor to capture is the one that is almost always overlooked...model output confidence. This is why I brought up the current weather models which are MUCH simpler than global climate models and yet, they are extremely poor with regards to their degree of confidence in a prediction. It is not that the science is wrong, it is that today it is wrapped up in the politics of funding. If you received $20M (which is not that much with large scale models; go hunt up NCEPs budget) for a model and at next years budget review are you going to stand up and state "well, we now have 8% confidence" or will you say "we now have refined our model such that we have established an 80% confidence interval in just what you politicians thought".. which means little and trust me with how much time is spent on PowerPoint slides to not tell a lie but let people walk themselves right off the logic cliff. Do not feel bad or misled, it is the nature of the beast. The CO2 levels on this planet have often change by a large amount, in many cases there were NO people on the planet and yet there goes the CO2 levels for climbing for 50,000 years before it settles back down. The tiny spike that people get flustered about now is just soooo small compared to the global climate picture. Always keep in mind that when headlines spout stuff like "<XYZ> is the <lowest/greatest> it has been in recorded history...recorded history is such a small slice of time. The ice sheet gas analysis does give us insight into things at the 100,000 year+ range which is good, but understanding that, why it moves the way it does, how do sun cycles impact the earth... it is a very difficult problem and instead of letting pure science do what it should, we have gotten all wrapped up in sensationalism. Ok, I rambled on enough...so I  raise my glass to the dearly departed soul of science, I will miss you! 

If you want to panic forget CO2... fukushima and the poisoning of the pacific... THAT gives me nightmares.

 

[Erwin]

yes. People think of these models as something like a derived mathematical identity (a true-by-definition relationship like d = v * t). In reality it is a set of instructions. Who chooses those instructions those instructions? The devs.

50 minutes ago, jimofflorida said:

as a system engineer

interesting, are you referring to the System's engineer college major, or literally the engineer that interfaces the techs with each other?

[jimofflorida]

degrees in EE and computer science worked my way up from developer to chief system engineer (some of my staff had their masters in computer simulations); along the way I did a fair amount of simulation work and even some model based system engineering (sysml stuff, although that really is just another flavor of the year/decade methodology that only government contractors will ever afford because its insanely expensive). last tool I used for smaller jobs was extendsim which was not bad but they all suffer from scalability issues as complexity increased. larger jobs had to be coded up with calls to matlab as needed and yes, fortran is still alive and well hehe. I would often have sims running for days or even weeks and then EIT would deploy a new power saving plan to all computers, oh, every few months that would graciously power down the computers when the room motion sensors would turn off the lights **poof** there goes weeks of work...glad I retired

[Erwin]

sysml, Fortran, AND extendsim.... 

 

It's very interesting to speak with more experienced engineers. Sometimes I find that the wider the age gap gets, the less of a common language / tech we use. 

[ofd]

 

 

Quote

 

Yes. Because the "experiments" they ran were about heating gas in a tank. So the results cannot be extrapolated  to the atmosphere.

 

I am fairly sure they ran experiments in the atmosphere too.
 

Quote

Notice how they make no mention of volume? That's because they assume volume is constant, aka a tank.

Have a look at the density graph.

[jimofflorida]

11 hours ago, Erwin said:

sysml, Fortran, AND extendsim.... 

 

It's very interesting to speak with more experienced engineers. Sometimes I find that the wider the age gap gets, the less of a common language / tech we use. 

 

Erwin, 

as an FYI, sysml is still the modeling language of choice for MBSE, and fortran is still (yes,even today) THE primary language for all the science algorithms that process the satellite weather algorithms!!!

and before you ask the senior scientists are well aware that software languages have moved on and their response is "*&^$%*&$%" you we use FORTRAN. So for those that want to extend/fix the algorithms what happens is that they get the current science baseline algorithms and then... re-code them, usually into C++... now improvements are often made to the algorithms, but then have to be re-coded into the fortran baseline. This is not an example of politicians interfering with science, but it is an example of how scientists can definitely let their "personal" opinions drive how they do their work.

If you wish to learn more about this topic, the center of much of the weather satellite imagery science is the university of Wisconsin.

 

[Erwin]

1 hour ago, jimofflorida said:

sysml is still the modeling language of choice for MBSE, and fortran is still (yes,even today) THE primary language for all the science algorithms that process the satellite weather algorithms!!

I see. I'm just a soft dev so I've never had to use sysml directly, but I'm guessing the domain model UIs run it on the backend...

As for Fortran, the only time a dev (even the profs) ever mentioned Fortran to me was either as a quick history lesson or the butt of a joke. Is Fortran used to avoid having to re-write the existing codebase into C / C++? Or is it actually preferred to the more modern techs?

[Erwin]

6 hours ago, ofd said:

I am fairly sure they ran experiments in the atmosphere too.

The only atmospheric experiment I can think of is where they send a balloon up in the air. The balloon expands (cooling the gas) to the point where it eventually pops.

6 hours ago, ofd said:

Have a look at the density graph.

I did. It refers to flux density (the rate of energy transfer), not CO2 density. Has nothing to do with gas volume.

[ofd]

Quote

The only atmospheric experiment I can think of is where they send a balloon up in the air. The balloon expands (cooling the gas) to the point where it eventually pops.

There are a lot more.
 

 

Quote

 

I did. It refers to flux density (the rate of energy transfer), not CO2 density. Has nothing to do with gas volume.

 

I am sure pressure is correlated with volume.

[Erwin]

1 minute ago, ofd said:

I am sure pressure is correlated with volume.

Yes, CO2 pressure. Not flux density.

[ofd]

Quote

I did. It refers to flux density (the rate of energy transfer), not CO2 density

Quote

Yes, CO2 pressure. Not flux density.

Can you pick a coherent position?

[Erwin]

Just now, ofd said:

Can you pick a coherent position?

Flux density (which is what the graph tracks) has nothing to do with volume.

Pressure has everything to do with volume.

I think you are confusing flux density with pressure.

[ofd]

Follow the green line young Padawan.

 

[Erwin]

4 minutes ago, ofd said:

Follow the green line young Padawan.

 

Idk what that means

[jimofflorida]

11 hours ago, Erwin said:

I see. I'm just a soft dev so I've never had to use sysml directly, but I'm guessing the domain model UIs run it on the backend...

As for Fortran, the only time a dev (even the profs) ever mentioned Fortran to me was either as a quick history lesson or the butt of a joke. Is Fortran used to avoid having to re-write the existing codebase into C / C++? Or is it actually preferred to the more modern techs?

the scientist will not change what they are used to, its a culture thing. Industry/governments that want to use the theoretical algorithm baseline code that is used actually always has to take what the scientists produce for their software and the first step is to produce a reference translation that given the same input produces the exact same output...and it happens time and again (scientists do not care). You can offer your code make but they will golf clap at your efforts and say no thanks hehe

Fortran does have some language constructs that make working with complex matrix processing a little easier and, oh, about 20 yrs ago the compilers did produce more efficient code; but those days are long gone. So if you want to move into a fundamental global science research, remember to keep a working knowledge of Fortran! ;-)

for system engineering the methodology of the year is driven by the accrediting group that popped up some time ago call INCOSE. If you are interested in moving into system engineering, visit their site and consider getting accredited as the process is long and starting early can look good on the resume.

[Erwin]

45 minutes ago, jimofflorida said:

the compilers did produce more efficient code; but those days are long gone.

Wait the compilers got less efficient??? Are you sure it's the compilers' problem? It seems to me that us young'uns develop at a way higher level abstraction than the old-schoolers (increasingly object-heavy languages / frameworks)

For e.g. where I work, our codebase has a an actual text snippet object (manages string sequences like putting a puzzle together as opposed to individual strings), i.e. we work with our Snippet class more often than we work with native strings, and mind you even native strings are wrappers around c-strings... The point being that there is a performance cost as the abstraction layer increases.

How can you be sure that it isn't the code itself but rather the compiler that is responsible for the inefficient code? Did you actually compare the assembly output?

55 minutes ago, jimofflorida said:

for system engineering the methodology of the year is driven by the accrediting group that popped up some time ago call INCOSE. If you are interested in moving into system engineering, visit their site and consider getting accredited as the process is long and starting early can look good on the resume.

I'm extremely worried about this new trend of progressive credentialism. For a long time, I've wanted to delve into robotics as a future venture. Robotics pieces many different hard and soft techs together. I'm worried that I might run into trouble with the various accrediting groups, if this trend keeps up. Actually, this is the reason I'm in soft dev. It's extremely hard to regulate something so concept-based because concepts evolve very quickly. By the time a law is enacted, who knows how quickly it will become either unapplicable or defunct...

[ofd]

The green line that says pressure and the text that defines what pressure is.

[Erwin]

5 hours ago, ofd said:

The green line that says pressure and the text that defines what pressure is.

I did. It proves my point.

At the highest altitude for a gas, temperature dropped. 220K < 280K.