A Note: Global Warming Threads

[Trippy]

It is.
But the SD that he shows for the entire period is that cooling is a mirror image of warming.
But the preceeding period of 1880 to 1980 was one of predominate warming. Clearly the projection is based on a virtually flat record for the period shown, thus you get equal 2 sigma warming and cooling trends (no idea why there would be an inflection going forward in time though)
Arthur

I can in part agree with this - the statement about it being flat, and it seems like, in some respects at least, it's giving ground to critics, because it implies that the average temperature for the 1950-1980 period was normal, which I suppose by definition it was, given that's what the average was (presumably) calculated based on.

Again, i'm not sure it's what I would have done.

To the other, predicting that the temperature could be up to 0.3c cooler in 2020 is not neccessarily the same thing as predicting a cooling trend between now and then. Consider, for exampe, a major eruption let's say comparable to Tambora, or Pinatubo, the effectively follow a biased coin tosing model, and the further in the future you look, the more likely they become, remembering that Tambora caused a (or is thought to have caused) a ground temperature anomaly of -0.5c in 1816 (and yes, there is evidence suggesting a major unrecorded eruption in 1810 which might have contributed to the cooling, but this only really strengthens my point).

Consider this - there have been (that I can find information on) 6 eruptions since 1450 that have caused some degree of known temperature anomaly. The average temperature anomaly is -0.5c, and the average time between them is 88 years. I would suggest that this suggests that the probability of such an eruption occuring in any given year is somewhere around 1%.

Based on this, I wouldn't have much compunction suggesting that in a 40 year period, that the chances of such an event occuring are around 33%. From here there are a couple of paths we could take.

We could calculate an expected cooling based on the average ground temperature anomaly caused by these events, this gives us 0.17c.
We could calculate an expected cooling based on the maximum ground temperature anomaly, this gives us 0.26c


The point here being that considering one factor only I have no problems accepting the possibility of a 0.3c drop in average ground temperature at some point in the next 40 years without it forming part of a trend.

[adoucette]

Since they are using 5 year smoothed averages, a year with an Eruption, even a big one, won't cause much impact to the temp trend.

[Trippy]

Since they are using 5 year smoothed averages, a year with an Eruption, even a big one, won't cause much impact to the temp trend.

I'm inclined to disagree for a number of reasons, one of which is effectively an appeal to authority, but another of which is if you consider Tambora, Tambora caused a negative temperature anomaly for three years running.

Addendum:

It occurs to me I could have made this clearer, I'm not disagreeing with you on the point that using running averages smooths the data, it's one of the reasons they get used, I'm disagreeing with you in that it is possible to upset them, either by one off extremes, or extended deviations.

[adoucette]

Except Tambor (1815) wasn't within that record period, so there would be no statistical reason to include it based on data derived from the time period chosen.

In the 1950 to 1978 period they are using there were no eruptions that caused a measurable climate anomaly, thus you would not expect one in the statistical projections going forward

Arthur

[Trippy]

Except Tambor (1815) wasn't within that record period, so there would be no statistical reason to include it based on data derived from the time period chosen.

In the 1950 to 1978 period they are using there were no eruptions that caused a measurable climate anomaly, thus you would not expect one in the statistical projections going forward

Arthur

But I'm not suggesting it should be, am I?

What I am suggesting, basically, is that with a 1 in 3 chance of occuring in any 40 year period, and a virtually 1 in 2 chance of going >60 years without occuring, and the odds of their having been one by 2020 of around 2 in 3, that it should be expected that there would be one in period that the forecast covers.

And look what happened since 1978 - Pinatubo, 1991, caused a negative anomaly of 0.5c

Also, keep in mind that if you do not understand the cause of the 1950-1960 cooling, then one can not rule out th possibility of another similar episode.

[adoucette]

What I am suggesting, basically, is that with a 1 in 3 chance of occuring in any 40 year period, and a virtually 1 in 2 chance of going >60 years without occuring, and the odds of their having been one by 2020 of around 2 in 3, that it should be expected that there would be one in period that the forecast covers.

That's all true, but the 2 sigma line was not created using that kind of intelligent analysis. It's simply a derived based on the std deviation of the chosen record.

In any case, I think I've made my point, it really is an example of cherry picking climatic data because even back then climatic data to 1880 was available, but not used.

The fact that you've now linked it the report at least lets NEW posters see the context of that graph, which is a good thing.

Arthur

[iceaura]

But, as I've learned from McIntyres site, when climate scientists start working with statistics all sorts of strange things happen.

Another valuable lesson from that site is what can happen when statistics and mathematical models are manipulated by people completely ignorant of the physical science behind the numbers (and quite possibly possessed of a dubious agenda or viewpoint).

All climate data is cherry picked in a sense, from some scaling pov or another. The question is whether the choice of time period was reasonable or not - whether it reasonably supports the argument being made.

[adoucette]

All climate data is cherry picked in a sense, from some scaling pov or another. The question is whether the choice of time period was reasonable or not - whether it reasonably supports the argument being made.

NO NO NO

That's NOT the definition of Cherry Picking data.

http://en.wikipedia.org/wiki/Cherry_picking

Cherry picking is the act of pointing at individual cases or data that seem to confirm a particular position, while ignoring a significant portion of related cases or data that may contradict that position

If the choice is reasonable, by definition, it's NOT cherry picked.

Arthur

[Trippy]

That's all true, but the 2 sigma line was not created using that kind of intelligent analysis. It's simply a derived based on the std deviation of the chosen record.

It doesn't (or shouldn't) require that level of intelligent analysis to consider to consider a large volcanic eruption as being almost a certainty.

In any case, I think I've made my point, it really is an example of cherry picking climatic data because even back then climatic data to 1880 was available, but not used.

Still not sure I agree with this, I can think of several 'reasonable' reasons why one might choose to scale the graph as it was, most of which are practicle considerations to do with print scaling, among other things.

[adoucette]

Print scaling?

LOL

[Trippy]

Print scaling?

LOL

Laugh all you want, it's something that anyone who has ever actually done statistics in a professional setting should know.

When you're selecting you axes you have to take legibility into account. What are you trying to show with your graph? What is the best way to display the required data, and the best balance between context an legibility?

It doesn't neccessarily constitute cherry picking.

[RenaissanceMan]

"It is of course, the global warming scam, with the (literally) trillions of dollars driving it, that has corrupted so many scientists, and has carried APS before it like a rogue wave. It is the greatest and most successful pseudoscientific fraud I have seen in my long life as a physicist. (Emphasis added.) Anyone who has the faintest doubt that this is so should force himself to read the ClimateGate documents, which lay it bare. (Montford's book organizes the facts very well.) I don't believe that any real physicist, nay scientist, can read that stuff without revulsion. I would almost make that revulsion a definition of the word scientist.

So what has the APS, as an organization, done in the face of this challenge? It has accepted the corruption as the norm, and gone along with it." - Hal Lewis, emeritus professor of physics, University of California, Santa Barbara

Now consider this graph of atmospheric carbon dioxide. It is intended to spark fear because it SHOWS a DOCUMENTED INCREASE in CO2. While it is completely accurate, it is also completely misleading. This is how to lie with facts.

3.bp.blogspot.com/_UeZlUJSTKEk/TLZIj_7z1hI/AAAAAAAABtU/fwDGhaxAnEo/s1600/Al+Gore%27s+Graph.jpg

Note that the Scary Graph, above, spans only 70 parts per million.

Now we proceed to the same data presented with a zero base and showing the anthropogenic (man-made) component of carbon dioxide, estimated at ~3.4% of the total:

2.bp.blogspot.com/_UeZlUJSTKEk/TLZLm-zOigI/AAAAAAAABtc/zewmAKI4Ag8/s1600/Al+Gore%27s+Graph1.jpg

That little red line at the bottom doesn't appear nearly as threatening, does it.

Finally, let's add just one other greenhouse gas, water vapor. Water vapor constitutes roughly 1.5% of the atmosphere by weight. But by number of molecules, or ppm, it is almost 2.1% or 20,864 parts per million.

How does the 11 to 13 ppm of anthropogenic CO2 look compared to 21,000 ppm? The graph can't do it justice. On any reasonable graph, say 8 1/2" by 11", the man-made carbon dioxide is indistinguishable from the 0 base line.

And for this, the Luddites want you to cut your energy use 80%? Let's see them do it first.
They're all flying to Cancun for the next big conference, where they'll lie in the sun for a few days, and dine on steak and lobster, at government expense.

Thanks, suckers.

Just Google "environmental conferences" and click on the top link. It will show you conferences ALL OVER THE WORLD, to be attended by these hypocrites. They can't videoconference. That would be practicing what they preach.

[Trippy]

"It is of course, the global warming scam, with the (literally) trillions of dollars driving it, that has corrupted so many scientists, and has carried APS before it like a rogue wave. It is the greatest and most successful pseudoscientific fraud I have seen in my long life as a physicist. (Emphasis added.) Anyone who has the faintest doubt that this is so should force himself to read the ClimateGate documents, which lay it bare. (Montford's book organizes the facts very well.) I don't believe that any real physicist, nay scientist, can read that stuff without revulsion. I would almost make that revulsion a definition of the word scientist.

Not much of a physicist then if he/you can't even understand the Beer-Lambert law, or Simple Harmonic Motion.

[RenaissanceMan]

Not much of a physicist then if he/you can't even understand the Beer-Lambert law, or Simple Harmonic Motion.

"Moderators" talk like this, pretending to know MUCH more than a university physicist?

Whatever is this world coming to.

You should moderate more, and post a very great deal less.

[Trippy]

Let's deal with this post that you edited out first (i'm subscribed by email).

You're number 2 on my Ignore List.

Anthropogenic carbon dioxide concentration in earth's atmosphere is emphatically not related to simple harmonic motion.

ciao

This isn't the claim I made.

If you understood the finer points of my statement, you'd understand just how farcical this is (maybe that's why you retracted the statement).

The point, being made, is that the observation that increasing partial pressure of CO₂ in the atmosphere leads to increased absorption of IR in bands corresponding to vibro-rotary modes is a matter of simple harmonic motion - masses on springs.

If you were half the physicist you're pretending to be you'd also know that the same predictions can be made using Quantum Mechanics.

The only real question then becomes 'what are the effects of the added heat energy?'

Unless your going to argue that being in a ro-vibrational state other than the ground state is something other than heat energy? Or unless you want to talk about Blackbody/greybody raditaion, atmospheric emissivity, and the effective temperature of the earth...

Or the Goff-Gratch equation, or the Arden Buck equation, or even, once again, the Beer Lambert law and what it means in respect of the role of Water vapour.

Among other things.

As for this:

"Moderators" talk like this, pretending to know MUCH more than a university physicist?

Whatever is this world coming to.

You should moderate more, and post a very great deal less.

You are, of course, welcome to your opinion.

[Dredd]

Almost to a person, those Republicans who were the victors, taking over leadership of the U.S. House of Representatives, are global climate change / global warming deniers.

Two recent studies (this month) have debunked the foundation of their play pretend reality engendered by Karl Christian Rove.

[iceaura]

All climate data is cherry picked in a sense, from some scaling pov or another. The question is whether the choice of time period was reasonable or not - whether it reasonably supports the argument being made.
”
NO NO NO

That's NOT the definition of Cherry Picking data.

http://en.wikipedia.org/wiki/Cherry_picking

“
Cherry picking is the act of pointing at individual cases or data that seem to confirm a particular position, while ignoring a significant portion of related cases or data that may contradict that position
”
If the choice is reasonable, by definition, it's NOT cherry picked.

? You seem to be repeating my point for me, but you are unhappy about something. What is it?

Do you have a problem with my use of "some scaling pov or another" in place of "significant portion of related cases"? How would you establish the necessary relationship among the cases, without reference to the argument being made?

How does the 11 to 13 ppm of anthropogenic CO2 look compared to 21,000 ppm? The graph can't do it justice. On any reasonable graph, say 8 1/2" by 11", the man-made carbon dioxide is indistinguishable from the 0 base line.

The actual anthro boost in CO2 is much larger than that,

according to lots of people whom you apparently think should be believed because they are university physicists, and the rest of us tend to believe because their work has been peer-reviewed and seems solidly reasoned from adequate data.

But it's still very tiny, on your apparently intuitive scale of significance. Which brings up the actual argument you are making: the concentration of CO2 in the air is a small number, and looks very tiny on graphs that make it look very tiny, therefore a small increase in it is nothing to worry about. You will perhaps pardon those of us who don't take that kind of argument seriously? Try it out on, say, the cyanide concentration in your blood - let us know what happened.

[adoucette]

? You seem to be repeating my point for me, but you are unhappy about something. What is it?

Your statement "All climate data is cherry picked in a sense".

Is false.

Good climate data is decidedly NOT cherry picked.

Again, see the definition of what it means to "cherry pick" your data points.

Arthur

[iceaura]

Your statement "All climate data is cherry picked in a sense".

Is false.

Good climate data is decidedly NOT cherry picked.

That's not all I said. You are carefully selecting part of a sentence to change its meaning.

There are scalings and povs generating assertions, from which any timespan of climate data is cherry picked. That doesn't make it bad data. The question is one of reasonable support for the argument or observation being made. You can't accuse a climate scientist of cherry picking without showing how their criteria of data selection were unreasonable. Just showing that they selected this or that timespan, this or that data set, is not adequate.

[adoucette]

That's not all I said. You are carefully selecting part of a sentence to change its meaning.

There are scalings and povs generating assertions, from which any timespan of climate data is cherry picked. That doesn't make it bad data. The question is one of reasonable support for the argument or observation being made. You can't accuse a climate scientist of cherry picking without showing how their criteria of data selection were unreasonable. Just showing that they selected this or that timespan, this or that data set, is not adequate.

Again, you are simply reducing the definition of "cherry picking" to any use of selected timespans, and that's NOT what cherry picking means.

Cherry picking ALSO includes this important caveat:

while ignoring a significant portion of related cases or data that may contradict that position

So if a scientist provides data for period B - C to make a point, but if by simply changing the period's endpoint to either A or D the data now contradicts that point, then one can make the assertion that use of B - C vs say A - C is cherry picking the data.

Not all climate data need be plotted over the entire course of climatic data, but on the same token, the inference from the data should not rest upon the specific time endpoint chosen.

A simple example is the oft seen example of claiming that GW has stopped by showing a negative slope of temperatures when using 1998 as the starting point.

The negative slope is highly dependent on use of the very hot El Nino year of 1998 as the endpoint.
Simply move the starting point back just a year and the negative slope disappears.
Move it back around 2 years and the slope is once again positive.

Arthur