Merged Recent climate observations disagreement with projections

mhaze said:
With a system that has multiple components interacting, such as the climate, lower correlations have meaning.

For example, the correlation of Co2 to temperature, depending on how it is done, ranges from r=0.21 to r=0.44. Correlations of the PDO to world temperature are r=0.79, IIRC.

At this state of the art many of the causes and interrelationships are poorly understood and cannot be accurately separated out.
Click to expand...

So here you have r=0.79 then a few posts later you changed your mind to r=0.85 for some reason and still no source. Even though I guess you must have checked your source to modify your "IIRC" to your new value for r, right?. Unless you aren't so certain about that value either, in which case...
 
mhaze said:
Regarding your colorful graph, may I direct you back to the subject header of the OP (Original post,thread title).

"Recent climate observations disagree with projections".
Click to expand...
What did happen to you now? A bit claustrophobic? When somebody ask questions and you should both answer and show the extent and variety of your knowledge you become a hermit crab? Remembering the OP in post #1000 once you wandered where you liked without leaving any place to visit? What will be your subject? "Recent climate observations disagree with projections"? What does it mean? "It rained today when the forecast one week ago had said 'sunny' " so cold fronts, low-pressure areas, jet stream and cumulus nimbus -don't forget that, specially cumulus mediocris- don't exist?

I'm sorry, I'll continue with Georgieva's and its colourful title and abstract that doesn't match well the content and the discussion it it (maybe a condition to publish it?).

And don't follow your path of 15 years, now 10 years and by September a "if you don't show the Earth warmed in the past 10 minutes then is global cooling!". Your famous "30 years is climate, less is weather" -an ill interpretation of what the climatic normals mean- contradicts loudly with your each-day-shorter periods you claim you need to falsify AGW -mighty mHaze-. And my oldest nephew's height correlates quite well with 10 years moving average world temperature (and the number of sun spots in recent past correlated with his pimples) better than PDO and global temperature, and what does it prove? (Wow! A temperature correlates with another temperature within the same system! First time ever, stop the presses!!!

[My regards to Steve]
 
recalling to OP

Poptech said:
Computer Climate Model Validation has again failed despite the best attempts at mathematical deception by its proponents.
Click to expand...

given the physical effects that we know are not included in our current global models, and the systematic errors in seasonal/decadal forecasts, where is the downside in saying that current global models give us "merely" insight and understanding? that we do not expect predictive accuracy, nor does any lack of predictive verification on such small space-time scales have any significant impact on the value/relevance/importance of basic insights from climate science?
 
aleCcowaN said:
What did happen to you now? A bit claustrophobic? When somebody ask questions and you should both answer and show the extent and variety of your knowledge you become a hermit crab? Remembering the OP in post #1000 once you wandered where you liked without leaving any place to visit? What will be your subject? "Recent climate observations disagree with projections"? What does it mean?.....
Click to expand...
I've rebutted your Meehl 2003 chart with ahh.....this thread's discussion of empirical data....

Couldn't resist.

You've done a bit of work in independently analyzing the time series for aK and temperature. Do you have a conclusion that you can make based on that work? State your conclusion clearly, then I'll be happy to comment on it.

fsol said:
So here you have r=0.79 then a few posts later you changed your mind to r=0.85 for some reason and still no source. Even though I guess you must have checked your source to modify your "IIRC" to your new value for r, right?. Unless you aren't so certain about that value either, in which case...
Click to expand...
Right. I haven't found the reference where I got the 0.85 from. Doesn't mean I won't, but I'm not going to spend hours hunting for it either.

Is the exact number relevant?

What I posed was a Null Hypothesis.

Pixel42 said:
I didn't ask a question.
Click to expand...
Good Non Question.

lenny said:
recalling to OP

given the physical effects that we know are not included in our current global models, and the systematic errors in seasonal/decadal forecasts, where is the downside in saying that current global models give us "merely" insight and understanding? that we do not expect predictive accuracy, nor does any lack of predictive verification on such small space-time scales have any significant impact on the value/relevance/importance of basic insights from climate science?
Click to expand...
That's reasonable.

They are not.
 
Last edited:
mhaze said:
Right. I haven't found the reference where I got the 0.85 from. Doesn't mean I won't, but I'm not going to spend hours hunting for it either.

Is the exact number relevant?

What I posed was a Null Hypothesis.
Click to expand...

Let's see.

If you state something to be a fact to make some kind of point and it later turns out that you are just plucking numbers from thin air, then yes I think the "exact number" is relevant.

If you state something as a fact that you can't support but use it in various places to support your argument, then yes I think the "exact number" is relevant.

If you state "IIRC" then you are asking us to trust your recall is accurate. If then you can't turn up a source, then yes I think the "exact number" is relevant.

If you are trying to convince someone that you have the slightest idea of what you are talking about, then yes I think the "exact number" is relevant.

If you even want to *pretend* to discuss something in good faith, then yes I think using the "exact number" is relevant.

Keeping track of sources really isn't that difficult.
 
lenny said:
recalling to OP


given the physical effects that we know are not included in our current global models, and the systematic errors in seasonal/decadal forecasts, where is the downside in saying that current global models give us "merely" insight and understanding? that we do not expect predictive accuracy, nor does any lack of predictive verification on such small space-time scales have any significant impact on the value/relevance/importance of basic insights from climate science?
Click to expand...

The issue is really: given the physical effects that we don't know are not included in our current models. The"unknown unknowns" that led us up to the Global Financial Crisis.

It will be the view one holds on that uncertainty that would determine how you would interpret and use the insights the models give us and the weight we would place on them in real world policy.

The point relevant to the OP is that everyone will take their assessment of information avaiable and decide the ramifications of the predictive failings of these models over the last decade. Some will decide that it points to significant unknown missing knowledge, others will decide it is down to significant known missing knowledge.

Both sides of that debate can be rationally held.
 
Last edited:
fsol said:
Let's see.

If you state something to be a fact to make some kind of point and it later turns out that you are just plucking numbers from thin air, then yes I think the "exact number" is relevant. .....
Click to expand...

Yes, of course. But I have not stated something to be a fact to make some kind of point -

Sure. For purposes of discussion only I'll propose the following. Consider this the null hypothesis. Let's be clear there is no discussion of PDO in the Gergieva paper, so we are just playing with the idea. And we're interested in seeing how and if Warmers can actually engage in a simple, everyday bit of scientific discourse.
  1. PDO affects world temperature, R=0.85.
  2. AK index is correlated with world temperature up to year 2000+, R=0.85 or higher.
  3. There is a completely circumstantial fact that the AK index, the world temperature indices, and the PDO waveform for the 20th century are very similar. In other words, an increasing AK index does not cause an increasing PDO and a decreasing AK index does not cause a decreasing PDO.
I'm open for modifications of the hypothesis. The facts to make a point are yet to be introduced and the point is yet to be made.

Adjust or correct #1 as you like, or we can use D'Aleo's number that you brought into the conversation. That PDO affects world temperature is blatantly obvious, as global temperatures are often detrended for PDO effects.
 
Geckko said:
The point relevant to the OP is that everyone will take their assessment of information avaiable and decide the ramifications of the predictive failings of these models over the last decade.
Click to expand...


In point of fact the last decade is well within the range predicted by computer models, so proposing people assess the imagined predictive failure is intellectually dishonest to say the least.


Geckko said:
Both sides of that debate can be rationally held.
Click to expand...

How is possible to rationally discuss the implications of something that is a figment of someone’s imagination? One may as well propose we can hold a rational discussion regarding who is on Santa Claus’s naughty list.
 
Last edited:
[I'll answer any message address to me tomorrow, now I want to end my analysis Georgieva's paper]

Many things astonished me in Georgieva's paper:

The accuracy of the hypothesis supposedly tested once each variable is analyzed.

The temperature series have an error margin, specially in more than 80 years data, that cannot be inherited to the p=0.01 proposed. It had to be simply ignored. One among various origins for temperature margins is the incompleteness of the grid data collected. In the following figure gray areas mean lack of data:

temperature_anomaly_hansen.gif


I made my own calculation for 1880 figure based on crutem3 dataset and I obtained -0.29°C which is consistent with the 1951-80 base on Hansen's and 1961-1990 used by Hadley (a difference of 0.07-0.08°C). For 1880, on 1296 5°x5° cells in the grid for each hemisphere, there are data for 300 to 500 -it varies month to month- that is about 12 to 20% of Earth's surface, mainly land temperatures. The emerging error margin is low enough to document the raising tendency in world temperature, but it has to be dealt with care when hypothesis like r=0.85 with p=0.01 are laid.

The following figure shows the temperature trend and its error margins. I added an exponential curve not to explain any tendency at all. It's a pseudotendency drawn just there to explain later why the paper looks like it makes sense.

temperatures-errors-pseudotendency.gif

For the aK-aa index, the paper assumes aa index to be useful as it is proved that it correlates quite well with other indexes that take a whole bunch of observatories. The aa index takes daily data from two antipodal observatories (from 1868 on, one in England and the other one in Australia). Georgieva's paper takes a previous work that extend the values for 1844-1867 based on one sole station, because it is proved that values from the same observatory correlates pretty well with aa index. Then, the ak-aa index correlates quite well with aa index that correlates quite well with other geomagnetic indexes that correlates quite well with solar phenomenons linked -not 100% understood how- to variations in total solar activity. Besides, the aK index is mathematically more related to dB scale or Richter scale that a linear.

How this correlates-with-correlates-etc. and a little blurred temperatures have any real relation that come to a r=0.85 with p=0.01, well, it's a miracle. Somebody must have had an epiphany.

When I completed the series to this month's values using information from http://www.geomag.bgs.ac.uk/cgi-bin/aaindex and analyzed about half last century, I got this figure (ak-aa: orange, world temperatures: green):

last50years-akindex-vs-hadcrut2.gif


[Values for 1970 take the average for 1960-1970 / Year 2009 is accounted as a complete year in spite data end in July and August]

I took the solar cycle (11 years) to smooth the tendencies of both aK-aa and temperatures (although all the values are correct, there is an intrinsic error in doing it and I hope somebody will comment because the conclusions will be even more interesting).

How well "correlated" looks it now?

The main problem here is the supposed relation between world temperatures (70% oceans with an splendid temperature inertia) and a constantly jumping index somehow related to variations in solar activity. Let me put it this way: theres a high correlation between a derivative of what causes the Earth to be a pretty warm world and the variable that integrates energy fluxes. Again, there is a high correlation between a value and something related to its second derivative (!?). There are no time delays either. Incredible!

All this happen because of the pseudotendency: when exponential functions are involved you may mix up the function and its derivative. Here the pseudo-exponential curve for 1850-1960 for both temperatures and aK provides the suspicion both variables are related, and r seems to confirm it. An analysis for the last 50 years wipes out any relation.

The abstract and title, and the forced hypothesis test seems to me to be a load of tosh, though other parts might be meritable.
 
Last edited:
mhaze said:
Yes, of course. But I have not stated something to be a fact to make some kind of point -

Sure. For purposes of discussion only I'll propose the following. Consider this the null hypothesis. Let's be clear there is no discussion of PDO in the Gergieva paper, so we are just playing with the idea. And we're interested in seeing how and if Warmers can actually engage in a simple, everyday bit of scientific discourse.
  1. PDO affects world temperature, R=0.85.
  2. AK index is correlated with world temperature up to year 2000+, R=0.85 or higher.
  3. There is a completely circumstantial fact that the AK index, the world temperature indices, and the PDO waveform for the 20th century are very similar. In other words, an increasing AK index does not cause an increasing PDO and a decreasing AK index does not cause a decreasing PDO.
I'm open for modifications of the hypothesis. The facts to make a point are yet to be introduced and the point is yet to be made.

Adjust or correct #1 as you like, or we can use D'Aleo's number that you brought into the conversation. That PDO affects world temperature is blatantly obvious, as global temperatures are often detrended for PDO effects.
Click to expand...

You are the guy who claims a source for the figure. Put up or just shut up. You might as well ask, "well what if the sea was made of chocolate?" "what if candy was made of moon beams?"

The fact that you seem to think that your hypothesis, based on...umm...seemingly numbers you pulled from your arse, even deserves discussing...
 
PDO affects world temperature, R=0.85.
Click to expand...

wrong...PDO has no impact on the total energy budget of the planet...it reflects local concentrations of heat energy.

It's exactly equivalent to saying it's 30 outside now....it's just a sizeable anomaly that impacts various locals...your 30 degree back yard is an anomaly that affects a smaller locale.

CO2 concentrations, methane, orbital distance, affect world temperatures as does output from the sun and aerosols or albedo changes that reflect incoming radiation.

THOSE affect world temperature.

It will show up in the PDO at some point.....:garfield:
 
macdoc said:
wrong...PDO has no impact on the total energy budget of the planet...it reflects local concentrations of heat energy.
Click to expand...

True but...

I guess you can say that the PDO (and ENSO) can kinda effect global temps if they move heat from where it is normally 'hidden' (the deep ocean) to where it can be seen (the ocean surface and subsequently the lower atmosphere).

The total energy doesn't change but the temps do (briefly, the upper ocean loses heat fast).
 
DogB said:
True but...

I guess you can say that the PDO (and ENSO) can kinda effect global temps if they move heat from where it is normally 'hidden' (the deep ocean) to where it can be seen (the ocean surface and subsequently the lower atmosphere).

The total energy doesn't change but the temps do (briefly, the upper ocean loses heat fast).
Click to expand...
Briefly?

Well, sure, if 60-80 years is "briefly".
 
mhaze said:
the cgi_bin in the url indicates that the page contents came from a server response, strip that and what came after

http://www.geomag.bgs.ac.uk
Click to expand...

Yeah, I found that...I was trying to find the whole data set, but could only find a query for single data points...that would take too long.

I found the aa-index data elsewhere, and it seems to match what Alec uses.

If I have time, I will present some data comparisons of my own tomorrow.
 
mhaze said:
Yes, of course. But I have not stated something to be a fact to make some kind of point -

Sure. For purposes of discussion only I'll propose the following. Consider this the null hypothesis. Let's be clear there is no discussion of PDO in the Gergieva paper, so we are just playing with the idea. And we're interested in seeing how and if Warmers can actually engage in a simple, everyday bit of scientific discourse.

  1. PDO affects world temperature, R=0.85.
  2. AK index is correlated with world temperature up to year 2000+, R=0.85 or higher.
  3. There is a completely circumstantial fact that the AK index, the world temperature indices, and the PDO waveform for the 20th century are very similar. In other words, an increasing AK index does not cause an increasing PDO and a decreasing AK index does not cause a decreasing PDO.
I'm open for modifications of the hypothesis. The facts to make a point are yet to be introduced and the point is yet to be made.

Adjust or correct #1 as you like, or we can use D'Aleo's number that you brought into the conversation. That PDO affects world temperature is blatantly obvious, as global temperatures are often detrended for PDO effects.
Click to expand...

You fell for their bait and switch. You are confusing the year to year variance with the long term trend.
 
You must log in or register to reply here.

Back
Top Bottom