This was an eye opener - cont.

[Rossini]

Nailed.

 

[ScooterTDI]

<a href=showthread.php?s=&postid=15587540#post15587540 target=_blank>Originally posted</a> by greenbean36191
Stu, after multiple very long threads, and having it pointed out to you multiple times you're STILL arguing against a point about past climate change that no one has tried to make.

Find a single scientist who has ever claimed that ALL climate change (now or in the past) has been due to humans. I think you'll be pretty hard pressed to even find any who are surprised that CO2 wasn't the original cause of warming 19,000 years ago, given that it's been known for quite a while that the CO2 didn't start climbing until well after the warming. The more important question though is how are the points your trying to make even relevant to modern warming? Once again, the fact that a different sequence of events occurred in the past is no argument about what is occurring now.

Here is a good question that your comments bring up:

How can we base models of future climate change upon past climate changes that have entirely different time course and cause/effect relationships?

Scott

 

[stugray]

greenbean36191,

"The original claim was that of the 1000+ articles about global warming in scientific journals, all are in agreement that the planet is in an accelerated warming trend. Finding a list of scientists who disagree with that stance is not a refutation because it doesn't address the claim."

Yes it does.

If he had stated that he had read 1000 articles and they all agreed then that COULD be a true statement.

However since he mentioned 1000+ , that is an assertion that he was talking about ALL of the articles published in scientific journals.
Therefore it is a statement that is easily proven false and one that is hard to prove true without examining ALL articles ever published on the subject.


"Stu, after multiple very long threads, and having it pointed out to you multiple times you're STILL arguing against a point about past climate change that no one has tried to make."



Not exactly - the atlantis statement was really meant as a joke.

I merely pointed out that article as yet another example of how the scientists had it wrong. The article even states that.

The Earth axis tilt issue is another disturbance to the "system" that was not modeled properly.
It is similar to my previous mention about how "outside influences to the model" can invalidate the model if not accounted for ( similar to a large asteroid strike causing volcanic activity ).

Some of those disturbance sources CAN be modeled if we had thought of them ( earth axis wobble) and some CANNOT be modeled because we cannot predict them ( asteroids, volcanoes, tsunamis, etc ).


No-one commented on one of my previous examples of an un-modeled disturbance: The fact that Cosmic radiation flux varies wrt the Earth over periods of millions to 10s of millions of years as the earth orbits around the galactic center.
By some estimates it may have varied enough to cause noticeable warming.
It is also a theory that could explain huge jumps in evolution of the species - extra comic radiation causes spikes in mutations.

Stu

 

[Rossini]

:lol:

 

[HornetMech242]

Wasnt there a time where great minds agreed that the earth was flat?

Point is global warming caused by man true or not, there are people that will jump on board to make them selves look more credible.

 

[greenbean36191]

How can we base models of future climate change upon past climate changes that have entirely different time course and cause/effect relationships?

GCMs are based on physical relationships, not temporal relationships, so which came first doesn't matter. If permafrost melts at 33 degrees it doesn't matter whether that temperature came from solar variation or increased CO2. When that thawed vegetation rots it yields a certain amount of methane regardless of why it melted. That methane breaks down into a certain amount of CO2 regardless of its source and that amount of CO2 has a physically constrained radiative forcing regardless of where it came from or how fast you add it. The sun doesn't know or care if that CO2 was a feedback or the original forcing.

If he had stated that he had read 1000 articles and they all agreed then that COULD be a true statement.

Well he was clearly paraphrasing Naomi Oreskes' conclusion which was essentially that she read almost 1000 abstracts and they all agreed.

Therefore it is a statement that is easily proven false and one that is hard to prove true without examining ALL articles ever published on the subject.

Give it a shot. I think you will find it a lot harder than you think. I'll give you a hint though: look in the gray literature in non-ISI journals.

I merely pointed out that article as yet another example of how the scientists had it wrong. The article even states that.

The Earth axis tilt issue is another disturbance to the "system" that was not modeled properly.

I read the article and never saw this conclusion. Orbital wobbles and procession on the axis as the cause for ice ages has been the mainstream view since the early 1920s. The heyday of the idea that they were caused by changes in CO2 predates that. I don't see anything in the article suggesting that the new paper makes any case other than that what had already been hypothesized is true. As the article itself says, "they confirmed a theory that was first developed more than 50 years ago that pointed to small but definable changes in Earth's rotation as the trigger for ice ages."

No-one commented on one of my previous examples of an un-modeled disturbance: The fact that Cosmic radiation flux varies wrt the Earth over periods of millions to 10s of millions of years as the earth orbits around the galactic center.
By some estimates it may have varied enough to cause noticeable warming.

No one commented on it because it's nothing more than an untested hypothesis. It has no legs and at this point there's not much to discuss about it. Sure there are estimates that it may have resulted by noticeable warming, but they're unsupported and brought to you by some of the same people who claim that we have been cooling since 1998. There is no known mechanism by which GCMs affect climate on the decadal to centennial scale. There is no measured effect. For the last 40 years when most of the observed warming has occurred, GCMs don't even show correlation with temperature. The best proposed mechanism is that they cause increased cloud formation, but the real world patterns don't fit the predictions of that hypothesis. You don't see a greater effect at high latitudes where GCM flux is greatest and cloud formation lags GCMs by several months (when the lag should be days at the most if the proposed mechanism is right) before the correlation breaks down completely about 15 years ago.

So on one hand you have a hypothesis with a well-defined and demonstrated mechanism that gives you the right magnitude, timing, AND pattern of change based on well-measured changes in forcings. On the other hand you have a hypothesized, but unmeasurable forcing that has no demonstrated mechanism and doesn't fit the pattern or timing predicted, and probably doesn't fit the magnitude of warming. I say probably not because in order for GCMs to replace CO2 as the primary driver of warming you would not only have to demonstrate a mechanism and measure the effects, but you would have to show why everything we know about CO2 as a forcing is wrong but still fits the expected pattern. That's a REALLY tall order.

Wasnt there a time where great minds agreed that the earth was flat?

Sure, if you go back to the early Greeks, back before the basis of science as a discipline was even established.

 

[ScooterTDI]

<a href=showthread.php?s=&postid=15602398#post15602398 target=_blank>Originally posted</a> by greenbean36191
GCMs are based on physical relationships, not temporal relationships, so which came first doesn't matter. If permafrost melts at 33 degrees it doesn't matter whether that temperature came from solar variation or increased CO2. When that thawed vegetation rots it yields a certain amount of methane regardless of why it melted. That methane breaks down into a certain amount of CO2 regardless of its source and that amount of CO2 has a physically constrained radiative forcing regardless of where it came from or how fast you add it. The sun doesn't know or care if that CO2 was a feedback or the original forcing.

I disagree. You can't really validate a model by replicating one effect if the desired use of the model is for prediction of an entirely different effect.

You can include all the mechnisms in a model that are sufficient to reproduce a particular effect under a particular set of conditions, but this included set of mechanisms may not be sufficient/appropriate to accurately model another effect.

I think you are over-estimating the physical reality of these models. As I have said before, you can examine each little piece of a model in isolation and each little piece of the model will seem logical enough, but when lumped all together, you can arrive at something much different than reality, especially in a chaotic system.

Scott

 

[ScooterTDI]

I have a good (and simple) example of what I mean whaen I say that individual parts can be logical, but when lumped together they drift from reality:

Lets say that I wanted to model the physiological effects of exposure to a substance. I develop a model of physiological effects versus dose and the time over which a dose was aquired using a ton of available data. I develop another model for a different substance in the same way.

Now, what happens when a person is exposed to both substances? I have models for each substance that seem reasonable enough, so lets say that I assume that the effects are linearly additive.

My modeling results are very likely to be in great error. Why? Perhaps there is a complicated synergistic interaction between the substances and the body that can't be modeled with the available data of each substance seperately.

This example shows how the whole can differ from the sum of the parts and how you can't validate a model for one effect or set of conditions and expect it it to be valid for all other effects of sets of conditions

Scott

 

[greenbean36191]

My modeling results are very likely to be in great error.

Yes, and that error would be apparent when you go to validate the model and it has little skill in replicating reality.

No one takes valid bits and assume that the whole must therefore be valid.

 

[ScooterTDI]

<a href=showthread.php?s=&postid=15603613#post15603613 target=_blank>Originally posted</a> by greenbean36191
Yes, and that error would be apparent when you go to validate the model and it has little skill in replicating reality.

No one takes valid bits and assume that the whole must therefore be valid.

Again, I would disagree. From what I have read in the literature they are validating there models against historical data going back thousands of years. In the historical data, the CO2 changes lag the temperature changes. Therefore, they are validating the models against scenarios that are not appropriate for a proper validation. They aren't appropriate for validation because there are entirely different system of processes/interactions occuring than anthropogenic warming. In the previous example I gave on the physiological effects model, you could validate the model by performing some experiments to see how the two substances interact. With the climate change models how can they validate the model against anthropogenic warming data? Given the scale of global temperature changes and variations, I don't think they can adequately validate their model based upon data in the past 50-100 years or so.

As an example, perhaps warming occured sometime in the past due to orbital effects. Again hypothetically, lets say that the oceans released CO2 into the atmosphere because of this warming of the oceans and the lowering of gaseous solubilities. In this situation, you have a net flux of CO2 out of the ocean. In the current anthropogenic warming, we would expect a net flux of CO2 into the oceans. Being able to recreate a flux out of the ocean may involve different processes than recreating a flux into the ocean. If the processes governing the flux into the ocean differ from those governing flux out of the oceans, then the validation using historical data is meaningless with respect to using the model to predict anthropogenic warming.

On the second point, I'd disagree again. Modelers do it that sort of thing all the time. You yourself just said in a previous post that the models are based upon physical principles and thats why historical data that does not represent anthropogenic climate change can be used to validate them for use at predicting anthropogenic climate change. In fact, not everything in those models is a hard physical law. When you start considering all of the assumptions, estimations, approximations, etc. within these models it is much harder to have any faith in the validity of the model as a whole simply based upon the validity of the pieces.


Scott

 

[Tang Salad]

^ That reads very much like a Postmodernist critique. It's a little odd (to me) to see one in a discussion of science.

 

[ScooterTDI]

Let me reiterate that modeling is not science.

Furthermore, the validation of a model is the single most important part of model development and should be closely scrutinized.

Scott

 

[stugray]

greenbean36191

"No one commented on it because it's nothing more than an untested hypothesis. It has no legs and at this point there's not much to discuss about it."

I wouldnt say NO legs:

http://www.space.com/scienceastronomy/090827-cosmic-rays.html


"From the iron-60's location and concentration, the German group later calculated that the putative supernova went off 2.8 million years ago at a distance of about 100 light years away. Fields believes this was probably too far away to have caused an extinction-level event.

"I'd call it a near miss," he says.

The cosmic rays from this supernova may have had an effect on the climate, but to cause serious biological damage, a supernova would need to explode within about 30 light years of Earth. "

Yet another example of "outside disturbances" to our climate that cannot be modeled.

And what I was talking about when I mentioned the cyclic nature of higher background cosmic radiation:

"In a similar vein, Melott and his colleagues found a possible link between the bobbing of our Sun up and down in the galactic plane and a 63-million-year cycle in fossil biodiversity. The hypothesis is that our solar system is exposed to more cosmic rays every time the solar system peaks out of one side of the galaxy. "


I have NO Idea if that addresses Rossini's elegantly worded comment ( or lack thereof ) above.

Stu

 

[stugray]

I suppose Rossini will just dismiss this article by 'Don J. Easterbrook: Professor Emeritus of Geology at Western Washington University' as another "single suspect maverick scientist":

http://www.globalresearch.ca/index.php?context=va&aid=10783

Here's a great quote:

"Global climate changes have been far more intense (12 to 20 times as intense in some cases) than the global warming of the past century, and they took place in as little as 20"“100 years. Global warming of the past century (0.8° C) is virtually insignificant when compared to the magnitude of at least 10 global climate changes in the past 15,000 years."



So the whole planet is getting their panties in a bunch over "virtually insignificant" Global Warming.


Here's a quote from his conclusion:

"Global warming (i.e, the warming since 1977) is over. The minute increase of anthropogenic CO2 in the atmosphere (0.008%) was not the cause of the warming"”it was a continuation of natural cycles that occurred over the past 500 years."


Sounds kind of like something someone on this forum has been trying to say......who was that again?....


Here's a list of some of his other publications:

http://www.ac.wwu.edu/~dbunny/pubs.htm#global

I think he knows a little bit.

Stu

 

[stugray]

Another interesting read:

http://www.climatephysics.com/GlobalWarming/GWScience.htm


But we cant trust an article that uses logic & the scientific method now can we?

These 130 scientists must all be KOOKs too!

http://www.climatephysics.com/GlobalWarming/LetterGerman.html

A quote:
"A real comprehensive study, whose value would have been absolutely essential, would have shown, even before the IPCC was founded, that humans have had no measurable effect on global warming through CO2 emissions. Instead the temperature fluctuations have been within normal ranges and are due to natural cycles. Indeed the atmosphere has not warmed since 1998 "“ more than 10 years, and the global temperature has even dropped significantly since 2003."



Here's a little speech by some Kook, geologist, senator that walked on the moon:

http://www.climatephysics.com/JackSchmitt.html

Stu

 

[stugray]

I Guess Rossini must be right!
I think it is true that ALL scientists believe in GW.

We cant trust ANY of these 54 scientists, especially:

Ivar Giaever
Institute Professor, School of Engineering and School of Science
Rensselaer Polytechnic Institute
Nobel Prize in Physics 1973


Here's quote from their letter to the American Physical Society:

"Current climate models appear insufficiently reliable to properly account for natural and anthropogenic contributions to past climate change, much less project future climate."

Another point that has been made ( or attempted to be made ) about the validity of climate models earlier in this forum.

http://www.climatephysics.com/GlobalWarming/APS.htm

Apparently we cant trust a bunch of members of the APS either.

Stu

 

[greenbean36191]

From what I have read in the literature they are validating there models against historical data going back thousands of years. In the historical data, the CO2 changes lag the temperature changes. Therefore, they are validating the models against scenarios that are not appropriate for a proper validation. They aren't appropriate for validation because there are entirely different system of processes/interactions occuring than anthropogenic warming.

They're validated against direct observations and against reconstructions of the distant past. They have to pass both tests to be considered robust and useful for forecasting. The model only knows the initial conditions, the change in each parameter over time, and the relationships between parameters. It doesn't know or care whether it's starting a run from 800 BC, 1889, or 2009.

Your CO2 flux example is actually a good one for why the relationship is what matters, not the order of events. You can create simple models of CO2 flux based on temperature, alkalinity, and partial pressure of CO2 that are excellent at reproducing reality based on the relationship of those three parameters. It doesn't matter whether the system is absorbing or releasing CO2, heating up or cooling down, or alkalinity is rising or falling. The same model still works.

The only case where the order of events matters if you use a naive model based purely on correlations rather than physical interactions. That's not how climate models work. Here is an example of one such model that is very good at matching observational data, but might not be useful for other scenarios- http://tamino.wordpress.com/2009/08/17/not-computer-models/

You yourself just said in a previous post that the models are based upon physical principles and thats why historical data that does not represent anthropogenic climate change can be used to validate them for use at predicting anthropogenic climate change. In fact, not everything in those models is a hard physical law. When you start considering all of the assumptions, estimations, approximations, etc. within these models it is much harder to have any faith in the validity of the model as a whole simply based upon the validity of the pieces.

No, that's not at all what I said. The models are made up of assumptions, which are either justified by direct observations or physical relationships. The model still has to be validated as a whole.

Let me reiterate that modeling is not science.

You're free to reiterate that, but I think most scientists would disagree with you. Modeling is used to take observations and make and test predictions. That's the basis of science. Models are nothing more than another kind of experiment. They're fundamental in just about every branch of science where actual controlled lab tests are too expensive, impractical, or impossible- ecology, molecular biology, astronomy, geology, oceanography, physics, chemistry, etc.

 

[greenbean36191]

I wouldnt say NO legs:

In relation to climate change, it has no legs. The fact that the earth is bombarded by GCRs is well established. The link to climate change, especially as a significant driver is not established.

Yet another example of "outside disturbances" to our climate that cannot be modeled.

Nor would they be useful to model. Models are interested in trends. Rare occurrences with no trend or cyclicity on relevant timescales can be ignored, especially if they render the question posed by the model moot.

In planning for retirement do you take into account that civilizations periodically collapse and their money becomes worthless? No, because it's unlikely to happen and unpredictable. Even if you assumed it would happen in the near future it would not help you make any decision about where to put your money.

I suppose Rossini will just dismiss this article by 'Don J. Easterbrook: Professor Emeritus of Geology at Western Washington University' as another "single suspect maverick scientist":

Stu, you have a very serious problem with assessing the credibility of sources. We've already addressed this exact article before, too.

First of all, it's published by a think tank. They almost always have political agendas and are never peer reviewed.

The guy also only cites his own publications, the vast majority of which are not peer reviewed and none of which are highly relevant to any of the claims he makes. It's an appeal to authority and nothing else. We are supposed to believe what he says because he said it and he has a PhD.

Besides that, several of his key points are uncited, including his claim that there have been multiple periods of climate change in the past 15,000 yrs that were much more dramatic than the current episode.

He also makes the logical fallacies of claiming that the cause of previous climate changes is most likely the same cause of current change and that more severe changes occurred before humans were present, therefore the current changes will not be catastrophic to humans.

The biggest red flag of all though should have been when he made the claim that global temperatures have cooled recently. There is absolutely no statistically valid way of extracting a cooling trend from any of the global temperature records over the time period he talks about. That's not an honest mistake anyone with any training in statistics would make either. It's an outright and intentional lie.

But we cant trust an article that uses logic & the scientific method now can we?

Not if the logic it uses is horrible.

The article starts off by citing a paper in E&E, which is not a scholarly journal. The conclusion, with much dramatic emphasis, is that humans cannot be responsible for increased CO2, which ignores the fact that isotopic analysis shows that most of the increase in CO2 bares the signature of fossil fuel combustion and that budget estimates put emissions in the range of what we actually see. I probably should have stopped reading right there.

The article goes on making lots of accusations of delusion and fraud, and committing logical fallacies itself. I stopped reading when I got to the disproving of the 4 predictions of global warming, all of which are straw men. Refuting them does nothing to damage the credibility of AGW because they aren't predictions made by the theory.

These 130 scientists must all be KOOKs too!

Again, global cooling. Not just any cooling though, but significant cooling. Yes, they are kooks AND liars if they read the letter and signed their names to it.

 

[ScooterTDI]

Ok lets go through these points one by one.

<a href=showthread.php?s=&postid=15631094#post15631094 target=_blank>Originally posted</a> by greenbean36191
They're validated against direct observations and against reconstructions of the distant past. They have to pass both tests to be considered robust and useful for forecasting. The model only knows the initial conditions, the change in each parameter over time, and the relationships between parameters. It doesn't know or care whether it's starting a run from 800 BC, 1889, or 2009.

Ok lets go through these points one by one.

First, the model are being validated against a VERY short period of anthropogenic warming (i.e. the last 50-100 years). I fail to see how they can produce a reliable validation based upon data from such a short period time. Validation against non-anthropogenic climate changes (i.e. the reconstructions) doesn't really bear much significance when the processes relating to anthropogenic climate change differ from those relating to non-anthropogenic climate change.

<a href=showthread.php?s=&postid=15631094#post15631094 target=_blank>Originally posted</a> by greenbean36191
Your CO2 flux example is actually a good one for why the relationship is what matters, not the order of events. You can create simple models of CO2 flux based on temperature, alkalinity, and partial pressure of CO2 that are excellent at reproducing reality based on the relationship of those three parameters. It doesn't matter whether the system is absorbing or releasing CO2, heating up or cooling down, or alkalinity is rising or falling. The same model still works..[/B]

That is not necessarily true at all. These complex climate processes can and often are non-reversible along the same pathway. Different processes can be encountered going in different direction between two points. The pathway in one direction is not always the same as the pathway in the opposite direction.

My example was intended to explain that the pathways being modeled were different. Perhaps, I didn't explain the example well enough. Yes, solubility of gasses is reversible along the same path, but there is more going on in reality than just the solubility of gases.

In my example, the non-anthropogenic warming released CO2 from the ocean due to reduced solubility. That reduces the availibility of CO2 for marine environments and increases it for terrestrial environments (among other effects, like pH). In the anthropogenic warming case in my example, CO2 availibity is increased for both terrestrial and marine environments. These are not the same situations and different processes are governing the dynamics of these two different scenarios.

<a href=showthread.php?s=&postid=15631094#post15631094 target=_blank>Originally posted</a> by greenbean36191
No, that's not at all what I said. The models are made up of assumptions, which are either justified by direct observations or physical relationships. The model still has to be validated as a whole..[/B]

Well, maybe I misunderstood then.

How do you make a direct observation of all of the processes that play a role in climate change? You can't. They are forced to make assumptions and estimations due to a lack of data. If they had all the data they needed, they wouldn't have to make assumptions and estimations. Due to this lack of data, these assumptions are "justified" by logic rather than direct observation.

I just read in the latest issue of "New Scientist" two articles in which some process that was previously thought to have a negligible impact was discovered to actually be one of the primary causes for a particular effect. I only remember one right now. It had to do with ocean mixing and it turns out that copepods and other marine life are actually a very important driver of the mixing within the oceans. It turns out that this may actually have a significant impact on climate and is not addressed in current climate models.

<a href=showthread.php?s=&postid=15631094#post15631094 target=_blank>Originally posted</a> by greenbean36191
You're free to reiterate that, but I think most scientists would disagree with you. Modeling is used to take observations and make and test predictions. That's the basis of science. Models are nothing more than another kind of experiment. They're fundamental in just about every branch of science where actual controlled lab tests are too expensive, impractical, or impossible- ecology, molecular biology, astronomy, geology, oceanography, physics, chemistry, etc. [/B]

I would disagree. In my experience, most scientists are skeptical of the reliability of models, as they should be. Models are not another kind of experiment. An experiment, by nature, MUST have a basis in physical reality, whereas modeling does not have to fulfill this requirement. I think most (good) scientists see modeling as a tool to be used in the exploration of the interactions going on within complex systems, but not as some kind of experiment. Experiments, again by nature, are able to confirm or refute a hypothesis. Models have no ability to do this.

As a chemist, I would say that I do not consider them a fundamental part of chemistry. In fact, computational chemistry is still very weak in its abilities. Computational chemistry can't do much unless you have a lot of reliable experimental data on which to form the model. It is mainly used for trying to determine the exact mechanisms of a reaction. They basically fiddle with the reaction mechanisms until they can get something that comes pretty close to the experimental data. Without the detailed experimental data, they have no way of knowing which mechanism is correct. In this way, computational chemistry like this is a pseudoscience because it does not have the ability to test hypotheses.

Scott

 

[Meisen]

I know I have commented somewhere back a ways but I'll say it again. Flawed or unflawed climate models, political agendas and the nefarious plans of mad scientists everywhere aside, look at this from a very simplistic view.

It took the earth a long time to store up the amount of sequestered carbon. Like millions of years. Maybe a little less, maybe a little more. There is a lot of it stored. Its clearly a mechanism for storing carbon ie everyone agrees there is a lot of carbon tied up in the world's oil, coal and natural gas deposits, right? We found a lot of it, some say as much as half of it already. We released (and continue) releasing as much of that carbon as we could in less than 150 years. In fact the profit motive means there is a strong drive to not only release as much as we can of it but to keep increasing (well, global economic meltdown notwithstanding) the rate at which we burn it. We'll keep doing that unless heed is paid. The effects are probably complicated, perhaps beyond our ability to figure out. But some basic ones we can count on are increased CO2 levels in the atmosphere and the ocean.

Isnt it just the slightest bit concerning to anyone that we should release a significant portion of millions of years of carbon sequestering in a geological blink of an eye? That we cant imagine a way to replicate that process in any kind of human time scale. And all this delay to action because of a small but vocal minority of the earth's inhabitants seem to think it's a hoax, or a polictical agenda or whatever (or, my favorite, bad for business). Seems to me anyone who would argue against being concerned, reducing consumption, taking some steps even, has a stake in the status quo (or thinks they do).







Maybe I'm just paranoid.