The Holocene Deniers

[Gracchus]

I've been trying to figure something out about Glenn. His apparent inability to differentiate between the 95% confidence interval and the standard deviation.

...


So that means no less than 6 times this point has been explained for Glenn yet he doesn’t seem to ever note it. He ignores it and then touts his general ignorance of this crucial difference as somehow "superior knowledge".

This is what interests me in this forum. How an why do people become immune to reason. The phenomenom is especially apparent in politics and religion.

It seems to be a structural problem.
The human brain has clusters of neurons, modules if you will. Sensory input is routed to several modules, which then have a sort of committee meeting to interpret what was seen, heard, felt, tasted, or smelled. The committee makes a decision and fires a signal to another module and which produces a feeling of confidence, or doubt.

The committee votes and passe the decision on to another committee of modules which decide how to interpret the perception and what, if anything to do about it. There may not really be enough information to make a rational decision but we are wired to reach a decision, to reach a rewarding feeling of confidence, because while being wrong is dangerous, it is very often more dangerous to do nothing.

So our brain makes a decision. Then and only then, does our conscious mind become aware of the decision. Then our conscious mind begins to rationalize that decision. If we don't know the why the subconscious mind reached the decision it did, and that it used incomplete or erroneous information, we will make something up.

Now, remember, sometimes we have made our subconcious decision based on insufficient or erroneous sensory data. But we already know the right answer, and the more wrong decisions we make based on this erroneous or insufficient data, the more we have reinforced our certainty. When we begin to doubt ourselves, doubt makes us afraid, and we don't like that. We're not supposed to like it.

Scientists are trained to check their decisions against the external world. That is, they get others with different biases to go over their procedures and reasoning.

Religion and politics are based on reasoning that is very hard to check. People can be very certain of these things without much reference to facts, experience and reason. They postulate spirits and souls and deities and undemonstrable entities, and there is no way to check, there is no way to falsify, there is no way to verify. These unverifiable entities allow us to avoid a clash between reality and belief. They allow us to keep our cognitive dissonances from destroying our minds. But this means the part of our mind that examines facts rationally, is disconnectd from the decision making process. Reason doesn't get a vote on the committee.

So in religion, we have spirit, deities, and mysteries that keep us from destroying our minds by stripping our gears in clashes between reality and our perceptions of it.

In politics we have concepts like freedom, patriotism, loyalty, duty, that enable us to commit the most stupid and appalling atrocities and still keep a good opinion of ourselves.

See for instance: On Being Certain: Believing You Are Right Even When You're Not .

It is not a philosophy text, it is about neuroscience.

Most of you are not going to read it. If you read it you won't understand it. If you understand it you won't dare to believe it.

There are hundreds of Jewish sects based, they all claim on the Torah. There are tens of thousands of Christian sects all based, so they claim, on the Bible. There are thousands of Hindu sects all based on the Vedas. There are dozens of Buddhist sects, possibly hundreds, based upon irreconsilable differences in doctrine. Muslims will leave off fighting and killing each other only long enough to turn on non-Muslims.

But science gives us answers, albeit tentative answers, that are agreed upon by Jews, Christians, Muslims, Hindus, Buddhists and others. Scientists do disagree about science, it is true. But there are so many viewpoints, and scientists are so argumentative that they have reached consensus on a good many things that have proven useful.

Religious ideas, are useful because the keep us from realizing how out of touch we are with reality. If we realized how crazy we were we couldn't live with ourselves. Science is the struggle to acquaint ourselves with what is real. Science is our method of attaining sanity, and not just security.

 

[grmorton]

But, given that you do not trust the temperature data, talking to you about low and high temperature records is pretty pointless.

it is not if YOU believe in the temperature record. By believing that the temperature records are good, then you have to explain the ridiculous situtation that the world has not warmed enough to carry us away from century old record lows.

You all say I should believe the experts predictions of future temperature rise. But, it hasn't worked as they say it will if one looks back to the past to see if their models could have predicted the earth's temperature starting long ago.

This is from my blog on June 2. It shows how the IPCC predictions of future temperature rise don't work if applied to the past.

A Backward Look at IPCC predictions of Temperature Rise
Today we are going to take a backward look at the IPCC's predictions about global warming to see if the past 40 years have behaved as they claim the next 40 years will behave.

The IPCC says this

"The equilibrium climate sensitivity is a measure of the
climate system response to sustained radiative forcing.
It is not a projection but is defined as the global average
surface warming following a doubling of carbon
dioxide concentrations. It is likely to be in the range
2°C to 4.5°C with a best estimate of about 3°C, and is
very unlikely to be less than 1.5°C. Values substantially
higher than 4.5°C cannot be excluded, but agreement
of models with observations is not as good for those
values."
IPCC, 2007: Summary for Policymakers. In: Climate Change 2007: The Physical Science Basis.
Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change
[Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and H.L. Miller (eds.) Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA., P. 12


They say that the temperature rise will go up 2 to 4.5 degrees for every doubling of CO2. The temperature rise due to CO2 is logarithmic and the math works out as follows:

T = 2 x ln(ending CO2 ppm/starting CO2 ppm)/ln(2x starting CO2 ppm/starting ppm)
(ppm is parts per million of CO2)


Let's first see how this warming compares to the satellite data. The Huntsville Satellite data started in 1979 and the starting ppm, in 1979, was 335 ppm. From that, and the rise in CO2, one can derive what temperature rise we should have seen. I use the famous Keeling curve as the source for the CO2 rise in this calculation. It is the most often cited CO2 curve and is often used to scare the hell out of all of us. that is the picture below.

I want to emphasize that in the pictures above and below all the curves are TEMPERATURE not CO2. For the satellite data they are the predicted temperature with the starting CO2 being the CO2 ppm in 1979, (335 ppm) and the ending being the CO2 in 2007 (388 ppm). The 2x starting is 670 ppm, so the curves are what we should expect for the temperature rise since 1979. Clearly we have been way below the expected.

The equation since 1979 is 2 x ln(current year CO2ppm/335)/ln(670/335) or
5 x ln(current year co2ppm/335)/ln(670/335).

One can clearly see in the expected temperature rise expected due to CO2 far far out ran the observed rise as measured by the satellites. This clearly indicates that the above quotation from the IPCC is utterly flawed. The world's temperature is NOT rising at either 2 or 4.5 or even 5 (as some older sources declared). And it shows that it is NOT going to rise in the future at the scary rate they claim.

Now lets look at the global temperature anomaly derived from NOAA's Global Climate at a Glance. I plotted the global temperature anomaly since 1958, when the Keeling Curve, the most often cited CO2 curve. In this case the starting CO2 is 315 ppm in 1958. Once again, the world is not warming as predicted by the IPCC. The temperature rise is far smaller than what would be expected.

This, to me, says that the IPCC is nothing but a scaremongering organization. Their predictions of the effect of CO2 has not worked out in the past but we are expected to believe that it will work out as they say in the future. Why should we believe them? Surely someone somewhere has plotted the data as I just plotted it to see if the past matches the predictions these guys are making. That would just be good science. But I have never ever seen anyone do what I just did. Is it possible that the IPCC isn't publishing things like this, which might make people think that the world isn't behaving as they say it will as the 'temperature kills civilization?

 

[grmorton]

Actually Glenn, it looks to me like the main thermometer "correction" occurs in 1989 with the switch over from LiG to MMTS, but I believe there are a number of studies related to both the variety of temperature measurements and the instrumentation

The switch over was not all at once across the country.

This is plainly visible in the second graph I showed which shows the incremental effects of the various changes. Note the little jump in the red line at about 1989.

There are a number of studies in the Peterson 2003 paper in section d (again I have to mention that) that discuss this very detailed topic at great length.



I cannot, and I stress, I cannot recommend enough reading it again.

I have read it and I would ask you again to please explain why we should trust the work when rural stations in 1989 to the present have as many heat sources nearby (as I showed with the pictures) as do the urban stations???? The entire paper is based upon a false presupposition that the rural stations are today not affected by modern heat sources.

The thing that you seem not to realize is that the switch to mmts required an electrical connection to the instrument. That brought the stations CLOSER to buildings than the old stations were, which increases the heat bias.

Further, there's a side-by-side study (apparently quite thorough) comparing the Liquid-in-glass thermometers with the MMTS systems by Quayle in 1991.

But the problem is that when they changed from the manually read stevenson screens to the MMTS, the move towards buildings cause an addition of heat to the system that isn't taken into account.

I will have to dig that paper up since it comes up over and over again. A fact anyone reading these Peterson papers closely would be unable to avoid seeing over and over again.

I didn't raise it, so don't do it on my account.

 

[grmorton]

I've been trying to figure something out about Glenn. His apparent inability to differentiate between the 95% confidence interval and the standard deviation.

Now, on a number of occasions he has stated something to the effect that for something to be 95% confident it has to be within 3 standard deviations.

Indeed this is technically correct in that 95% of the population data itself is within 2 standard deviations (which is within 3, by definition). This would be the same as saying "the 95% confidence interval is somewhere within the data set between the extremes."

But that isn't the same as the 95% confidence interval. This is often, as I've stated repeatedly, defined quite differently and assuming one has more than 5 data points it will be within 1 standard deviation of the mean. That's just how the math goes.

So I went back through and took a look at how many times this has been explained to Glenn:

In Post #233 Glenn was told what the 95% confidence interval on a mean was. It was explicitly stated

In post # 266 Glenn states:


Now this was 33 posts after the confidence interval equation was explicitly posted. I assume Glenn knows how fractions work and what the impact of increasing a denominator is, so I will prefer to assume he simply ignored it.

So he was corrected, again, in Post # 270

In post # 307 again Glenn states:


Again, in Post #308 I had to remind him how the confidence interval works.

In Post # 340 Glenn again states:


YET AGAIN he was corrected in Post #341


In post #488 Glenn writes:


Glenn was corrected in Post # 496

Now, as I said earlier, to be technically fair, Glenn was correct that the 95% confidence interval is somewhere inside of 3 standard deviations, but again, the key is that it is also within 1 standard deviation. (assuming more than about 5 data points). This was a point I initially blew off since it was apparent he was unable to differentiate between these two rather different concepts, and for that I was a bit hasty. But indeed unnecessarily limiting a confidence interval by confusing it with a standard deviation was what was more apparent by his point. But again, indeed the 95% confidence interval is, indeed, somewhere within 3 standard deviations. It is also within 1 standard deviation if you have 5 or more data points. (I was wrong to simply say he was "factually incorrect" when indeed the 95% confidence interval is somewhere within the distribution. But Glenn unnecessarily broadened that out to 3 standard deviations when, in fact, it is within 1 standard deviation if you have more than 5 data points!)

This is explained GRAPHICALLY in post # 501


He then, later, says this in Post #542:


Again, the difference between a confidence interval on the mean and a standard deviation is explained in Post # 546


So that means no less than 6 times this point has been explained for Glenn yet he doesn’t seem to ever note it. He ignores it and then touts his general ignorance of this crucial difference as somehow "superior knowledge".

Not getting lost in your pedantry, Thau, as I did last time. It ain't gonna happen.

But I am glad that AT LAST you finally admit that my criticism of the confidence intervals applied to the global temperature is correct. It sure took long enough to drag that out of you.

Back to presenting data. And Thau, if you can't understand physics, that is simply too bad for you, but I told you right up front I wasn't going to let you get me into the weeds where no one besides you and I understand things. This was your strategy (as is evidenced by what you did at the start of this thread), is your strategy now, and will continue to be your strategy. I won't bite

Below is a picture where I shiftend the land temperature anomaly to match the ocean temperature anomaly for 1880. Note that the land warms much faster than the oceans (one can't put an air conditioner in the ocean next to those thermometers) and the rise since 1970 is noticable as the world cemented over and added air conditioning.

 

[grmorton]

This is what interests me in this forum. How an why do people become immune to reason. The phenomenom is especially apparent in politics and religion.

Reason? You think it is reasonable to have air conditioners next to thermometers?

The rest of your point doesn't matter until you answer the above question. I would look at you and wonder why you are so unreasonable?

Do you think this is a great place to measure the temperature???? I don't and will never agree that it is a great place. If that makes me unreasonable, so be it.

 

[David Gould]

grmorton,

Unfortunately, I cannot see your images for some reason.

However, I have plotted the data myself. From 1958 to 2008, I get a figure of 2.75 degrees per doubling for the GISS data (which is similar to the NOAA figures) well within the projected range.

For the UAH data, I get 1.96 degrees per doubling, again well within the range. However, to be fair, this should be reduced to 1.57 degrees to get surface estimated warming from the lower atmosphere data. It is still within the predicted range.

In addition, the predicted range that you quoted is for equilibrium climate sensitivity. Thus, doubling CO2 does not instantly give a set rise. You need to wait for a few years for the system to come into equilibrium. Prior to that, the temperature increase will be smaller. So, at present the temperature increases have been well within the predicted values. (which is to be expected, since this is actually well-established physics).

 

[grmorton]

grmorton,

Unfortunately, I cannot see your images for some reason.

However, I have plotted the data myself. From 1958 to 2008, I get a figure of 2.75 degrees per doubling for the GISS data (which is similar to the NOAA figures) well within the projected range.

For the UAH data, I get 1.96 degrees per doubling, again well within the range. However, to be fair, this should be reduced to 1.57 degrees to get surface estimated warming from the lower atmosphere data. It is still within the predicted range.

In addition, the predicted range that you quoted is for equilibrium climate sensitivity. Thus, doubling CO2 does not instantly give a set rise. You need to wait for a few years for the system to come into equilibrium. Prior to that, the temperature increase will be smaller. So, at present the temperature increases have been well within the predicted values. (which is to be expected, since this is actually well-established physics).

See, this is the problem. CO2 works like the windows on your car. It traps radiative heat in the IR range. It SHOULD work as rapidly as that because radiation moves at the speed of light. So, you are quite wrong, physically, on how rapidly CO2 should impact the temperature.

 

[David Gould]

grmorton,

Of course the land warms much more quickly than the oceans. The heat capacity of the oceans is an order of magnitude more than the atmosphere. It takes around 800 years for the oceans to match surface warming.

While the thermocline barrier means that in the short term we can discount the deep ocean and just measure the upper ocean, it still takes an awful long time to warm that part of the ocean as much as the earth's surface - decades, in fact. So we would expect the difference between ocean and surface temperature anomalies to increase over time, until atmospheric CO2 stabilises.

 

[David Gould]

See, this is the problem. CO2 works like the windows on your car. It traps radiative heat in the IR range. It SHOULD work as rapidly as that because radiation moves at the speed of light. So, you are quite wrong, physically, on how rapidly CO2 should impact the temperature.

Seriously, this is so badly wrong that I do not know how to address it.

The fact that radiation moves at the speed of light has absolutely nothing to do with how rapidly the earth responds in terms of temperature change.

And I note that you did not address the fact that the graphs for GISS and UAH show that the temperature increases that we have observed all fall within the predicted ranges regardless of our disagreement over how rapidly the earth should respond.

 

[grmorton]

Seriously, this is so badly wrong that I do not know how to address it.

The fact that radiation moves at the speed of light has absolutely nothing to do with how rapidly the earth responds in terms of temperature change.

Oh yes it is right. Do you have a physics degree? I do. And I have published in atmospheric radiative transfer. Now, if the temperature doesn't rise as rapidly as the radiation equations would require, then there are feedback loops which prevent it from happening.

Let me ask you David, AND Gracchus, do you all think this station is a grand way to measure temperature?

Why is it nutty for me to think that an air conditioner blowing on a thermometer isn't a really good idea?

And I note that you did not address the fact that the graphs for GISS and UAH show that the temperature increases that we have observed all fall within the predicted ranges regardless of our disagreement over how rapidly the earth should respond.

The only graphs I have seen are the ones I posted. You didn't post any. My graphs show the rise in temperature being less than the IPCC would have predicted. When you present graphs and tell me where you got your data I will comment on your graphs.

 

[grmorton]

grmorton,

Of course the land warms much more quickly than the oceans. The heat capacity of the oceans is an order of magnitude more than the atmosphere. It takes around 800 years for the oceans to match surface warming.

While I will agree with part of the above, the land will cool at night much quicker than the oceas because of the same difference in heat capacity. The heat capacity difference is not as important when averaged over a century worth of warming. We are not speaking here of daily rises and falls in temperature. We are speaking of the rise in temeperature over the century. Therefore your critique doesn't apply.

While the thermocline barrier means that in the short term we can discount the deep ocean and just measure the upper ocean, it still takes an awful long time to warm that part of the ocean as much as the earth's surface - decades, in fact. So we would expect the difference between ocean and surface temperature anomalies to increase over time, until atmospheric CO2 stabilises.

I am delighted that you are aware of the thermocline and we agree that most of the heat is in the very upper part of the ocean, it is still the case that the cementation of the world in the 20th century has had a major impact on the temperature record. Some of the class 1 rural stations, stations far from heat sources, have not risen over the course of the century, unless the editors at GISS got to the station and made it rise in temperature via editing.

 

[grmorton]

This is what interests me in this forum. How an why do people become immune to reason.

Hey Gracchus, do you think it is reasonable for the climatologists to measure temperature with this station?

Do you think 20 air conditioners around this thermometer will affect the temperature? Yes or no? (One can't capture all the air conditioners in one picture)

Please explain to me why it is I who am unreasonable to think that this station is nutso.

 

[David Gould]

I told you where I got my data from: GISS and UAH. As I cannot see your graphs, can you tell me what temperature increase you get for a doubling of CO2? Is it within the predicted range?

How do you post picture?

As to a physics degree, no, I do not have one. No degree at all, actually. But temperature response is not instantaneous, not even in a car.

As to you being 'nutty' I do not think I have called you that. The thing is, though, that NOAA and GISS and Hadley and so forth are all aware of these issues. This is why they do not simply use raw temperature data.

 

[David Gould]

Here is the UAH versus LN CO2 graph:

LN CO2 v UAH

 

[David Gould]

Here is the GISS versus LN CO2 graph:

ln C02 v GISS

 

[David Gould]

Note that you will also need to multiply the UAH slope by a further .8 to get a better idea of the surface warming that the UAH data, which is low atmosphere data, is showing.

 

[David Gould]

grmorton,

Regarding GISS editing the data to show warming, it is a little odd that you resort to conspiracy theorism. I would have thought that, while you disagree with their methods and their conclusions, you would start with the assumption that these people are honest.

 

[David Gould]

grmorton,

Regarding heat capacity difference not being important over a century's worth of warming, yes, it is. The ocean will continue to lag the atmosphere - indeed, it may well increasingly lag the atmosphere* - until atmospheric CO2 is stablised.

*to clarify, this might not always be the case - it depends on the rate of CO2 increase and the rate of tranfer of energy to the oceans. I have discovered that in the last decade or so more energy has entered the oceans than the atmosphere due to some la ninas. However, it has only been in the ratio of 4 to 1, not enough to have ocean temperatures rise by the same amount as atmospheric temperatures. However, the upper ocean has warmed by around .1 of a degree, about the same as the atmosphere.

 

[David Gould]

Basically, in the warming scenarios that I can think of apart from massive ocean heating due to a superplume going off under water, the atmosphere will warm first. The oceans will warm later.

 

[thaumaturgy]

Not getting lost in your pedantry, Thau, as I did last time. It ain't gonna happen.

OK.

But I am glad that AT LAST you finally admit that my criticism of the confidence intervals applied to the global temperature is correct.

Actually I'm not. I'm not saying that at all.

What I did say was you clearly don't understand the difference between a confidence interval on the mean and a standard deviation.

Here's the illustration of the point:

(I have penciled in approximately 3 standard deviations).

NOTE how the 95% confidence interval on the mean (the green lines) is very much more narrow than merely 3 standard deviations (the orange lines).

That was my point. You keep talking about 3 standard deviations as if it were the same as the 95% confidence interval on the mean.

While the 95% confidence interval on the mean is indeed within the 3 standard deviation lines, it is very much smaller. And a rather different concept. (Now you were correct to say that the 95% confidence interval was within 3 standard deviations, but it is waaaay within 3 standard deviations! It is even within 1 standard deviation if you have >5 samples!) So I'm merely being pedantic to "give you the point".

Back to presenting data. And Thau, if you can't understand physics, that is simply too bad for you

And if you can't understand how data is dealt with, too bad for you. Because we are talking about what? DATA!

, but I told you right up front I wasn't going to let you get me into the weeds where no one besides you and I understand things.

Actually I'm not so certain you are among that list, Glenn.

This was your strategy (as is evidenced by what you did at the start of this thread), is your strategy now, and will continue to be your strategy. I won't bite

If you read the Peterson paper (or any of the papers related to dealing with temperature data you would know how crucial statistics is to this discussion.

I don't have respect for scientists who treat anecdotal data as if it were the whole story.

That's why people believe in quack medicine.

The fact you call statistics in a data discussion a "diversion", or "in the weeds" is far more telling about your science than anything I could possibly post.

Below is a picture where I shiftend the land temperature anomaly to match the ocean temperature anomaly for 1880. Note that the land warms much faster than the oceans (one can't put an air conditioner in the ocean next to those thermometers) and the rise since 1970 is noticable as the world cemented over and added air conditioning.

I don't mean to oversimplify the system, but doesn't water have a higher heat capacity than rock? (LINKY)

Ergo one expects water to be able to absorb more energy before increasing temperature than rock.

q = mCdT

q = heat
m = mass
C = specific heat
dT = change in temperature

q/(mC) = dT

if C[sub]land[/sub] < C[sub]sea[/sub] then more energy is required to be pumped into the seas than the land to give a given rise in temperature (dT), correct?

(I know, I know, this is a gross oversimplification, but that's the way we like to play on this thread isn't it?)