So what? There’s divergence during the overlap period.
Take a look at the spliced graph on Wikipedia and you will see. Again, the bottom line is very simple: No splicing = no hockey stick.
The “switch” was straight from your heart rate analogy. If you are now saying that your analogy is flawed, fine. It was your analogy - not mine.
If he discards those methods for the entire reconstruction, that’s fine. If not, it’s a problem.
No, this is not clear; it appears to contradict your comments in post #194. How about if we start over, completely ignore post #194, and you describe what you think happened just using your own words?
Come on, don’t play coy – this isn’t a Rodgers and Hammerstein musical and I’m not going to chase you into the barn so I can pin you down in the hayloft and kiss you. 
Instead, I’ll just have to steal your thunder.
From: p. 752 in Briffa et al. 2002. Tree-ring width and density data around the Northern Hemisphere: Part 1, local and regional climate signals. The Holocene, v. 12, 6 pp. 737-757.
Is that where you got the quote? It’s exactly the same phrase you quoted in post # 185.
How is it non-responsive? Let’s look at your question:
And the quote:
From: Hughes, M. 2002. Dendrochronology in climatology – the state of the art. Dendrochonologia v. 20.
Translated:
How does that not respond to your question? It is evidence related to the density of tree rings responding to snow melt times and snowfall amount.
So what? Why does the demand that the proxy work over the 20th century trump the demand that the proxy work over the wood densities over which it can be established to work?
Let’s look at these two possibilities:
Con: The 20th century is arbitrary - the wood density doesn’t know or care which century it’s in.
Pro: The wood density is a physical factor known to react to the temperature and seasonal fluctuations of its environment.
Pro: It’s really hard to argue that the wood densities aren’t tracking temperature when the correlation is 0.84 when used properly.
What would you say are the pros for #1 and the cons for #2?
Which graph? Can I assume that you mean that this graph is spliced?
What about this graph? Which ones are spliced and which ones aren’t?
No it doesn’t. And my post was perfectly clear. You simply tried to obfuscate matters by accusing me of claiming that there was a “conspiracy.”
How about you simply withdraw your strawman argument? It doesn’t matter if the IPCC’s alleged omission amounts to a “conspiracy” or not. Did they do it or not? And if they did it, was it appropriate or not?
Two extremely simple questions.
Whatever. You seemed to be criticizing the fact that I neglected to provide a link for that quote. Well, how about asking for one first?
No. I got it here:
http://www.climate-skeptic.com/2007/09/chapter-4-skept.html
It’s non-responsive because it doesn’t answer my question. Duh.
Does your quote mention insect herbivory at all? Simple yes or no question.
Does your quote mention 20th century insect herbivory being unique to the most recent 50-100 years? Simple yes or no question.
Does your quote mention any factor as being unique to the most recent 50-100 years? Simple yes or no question.
Look back at the argument you were making: You were basically saying that splicing is not a problem if there is overlap. I am saying that this is true only if there is little or no divergence. If there is divergence, then my argument about splicing applies with full force.
Getting back to your heart beat analogy, let’s suppose that they measure your heartbeat by hand for two hours, and it’s always pretty close to 72 beats per minute. However, in the last 15 minutes of the two hours, you are also hooked up to a monitor which shows a much more erratic heart rate over the last 12 minutes.
In that case, a “spliced” graph might cause people to reach the likely erroneous conclusion that your heart rate started becoming erratic only in the last 12 minutes. And the fact that the simpler measurement technique diverges undermines such a conclusion. Despite the fact that there is overlap.
I don’t understand your point well enough to say. Are you saying that the period after 1960 is known to be unique (in the last 1000 years) in terms of wood density?
Both are spliced. I was referring to the second one.
I agree with you that this graph is spliced. It is, after all, a multiproxy.
But you say that this graph is spliced:
That graph contains 10 different reconstructions, not all of which use multiple methodologies (in other words, not every proxy on that graph is a multiproxy.) You apparently could not distinguish this, and since you can’t tell the difference between a spliced mulitproxy and a non-spliced single proxy, how would you be able to tell if a spliced multiproxy had problems where the splicing occurred?
In case you’re curious – check out the sources cited in this page.
The dark red line is not spliced – it only uses glacial records. The light green line is not spliced – I haven’t read the paper, but the abstract only mentions tree rings and no other reconstruction. The very light blue line is also not spliced – it uses only tree ring records.
The last link doesn’t work, at least not for me.
-XT
None of the links works for me, and I’m a little reluctant to guess what your point is.
However, I will note that both of the graphs you asked me about before clearly contain both proxy and instrumental records.
My apologies, the links have an extra “http://” in them.
They’re both links that have been posted in this thread before.
Just because the black line and the dark red line are shown in the same figure doesn’t make them spliced.
It depends how you define “spliced.” I never formally defined what I mean by “spliced,” but I think I’ve described the problem pretty well – to say the least. Suffice it to say that if you take the black line out of the graph, there’s no more hockey stick. Or much less of a hockey stick.
How are you defining “spliced”?
The light green line has been discussed at length in this thread already. If you look at it alone (without any splicing), there’s clearly no hockey stick whatsoever.
I believe that’s the study that showed an embarrassing drop in temperature in recent years and had to be truncated by the IPCC. i.e. clearly not a hockey stick. Are you claiming otherwise?
The series from different data sets would have to be mathematically combined to be spliced, not just depicted along the same axes on the web or on paper.
Are you talking about the Briffa (light blue) reconstruction here or the Esper reconstruction? The light green line is Esper, the Briffa is the one you’ve been saying ‘had to be truncated.’
But that black line is the instrumental data. As in direct measurements using thermometers. Are you saying you don’t trust it?
Or are you saying the temperature has been just as high in the past millennium? If so my question would be: so what? What does that prove?
The important question is whether there is currently a warming trend, and if so, what is causing it. If it’s the same cause as other warming periods in the past few millennia, then there’s no problem - the forcing will most likely go away on its own, or has done so already. But if the current warming is anthropogenic, then we must make it go away.
I’m not sure what you mean by “mathematicallly combined,” but I am using the word “splice” to include situations such as that in the linked spaghetti graph. Where proxy records are depicted right along with the instrumental record.
The graph comes with a little legend:
Not in connection with the arguments I have made so far in this thread.
Not exactly – I am saying (among other things) that based on the proxy and instrumental records, we cannot conclude that current temperatures are unprecedentedly high.
Let me answer your question with a question: If it were known with certainty that temperatures today were higher than at any time in the last 1000 years, would that mean anything to you?
It depends on what level of confidence we’re talking about. Of course we can’t say with 100% confidence that the current temperature is the highest in 1000 years.
Yes, it means we should investigate and find out what’s causing it. Then we try to figure out if the warming trend is likely to continue, and decide whether we should do anything to mitigate it or to deal with it.
But the temperature record by itself doesn’t prove anything. It certainly doesn’t prove it’s anthropogenic.
My position is that we can’t say so with any (reasonable) degree of confidence. Not on the basis of the proxy records, anyway.
Do you think that perhaps the “hockey stick” should have been omitted from the IPCC Third Assessment Report?
What about this chart, which was (apparently) also used by the IPCC?
If not, why not?
What’s a “reasonable degree of confidence”?
Most proxies show that the temperature over the past millennium was lower than now, so even if there are problems with each, there is basis for claiming it’s the warmest it’s been in 1000 years. Looks to me like it’s at least a 2-sigma confidence, but that’s just a WAG based on those plots alone. Even if it’s not the warmest, it’s certainly one of the very few warm periods.
I think it should be there, because it puts the current warming trend into context. And of course the temperature record is important in understanding the natural forcing effects - how the temperature correlates with sunspot numbers, volcanic activity, etc. That’s partly why we think natural forcings cannot explain the global warming in the 2nd half of the 20th century.
But the hockey stick by itself is not a proof of anthropogenic global warming.
What do you mean by “apparently”?
No reasonable person could honestly state that they are confident that recent temperatures are unprecedentedly high.
See the earlier discussion about systematic error.
Ok, so now you’ve answered your own question. If past temperatures have been as high as (or higher than) current temperatures, it puts current temperatures “into context.” (whatever that means).
The chart came from an advocacy web site and did not link to the actual IPCC report.
When a scientist makes a “claim,” that doesn’t mean 100% confident. He/she is just reporting the best hypothesis that seems to fit the available data.
As for systematic errors, I guess I missed the discussion about how systematic error can affect all the different proxies in the same manner - could you summarize or direct me to a specific post?
It means it gives us some idea of the timescale and magnitude of natural temperature variabilities, and how the current warming trend compares to them.
Scientific theories aren’t often proved or disproved by a single definitive “proof.” It’s usually a preponderance of many different observations that guide us towards the conclusion. Therefore it’s a very robust process, and one erroneous result (or even several) is unlikely to lead everyone into the wrong conclusion. For anthropogenic global warming, just off the top of my head we have:
[ul]
[li]Temperature records that show definite warming trends in the 2nd half of the 20ch century[/li][li]Proxy records that seem to imply this warming is fairly unusual[/li][li]The fact that we’ve dumped a lot of CO2 into the atmosphere[/li][li]Observed increase in atmospheric CO2 concentration[/li][li]The fact that CO2 acts as greenhouse gas[/li][li]The lack of a notable increase in solar output or other natural causes of the observed global warming trend[/li][li]Decent correlation between known natural forcings and temperature proxy data up to mid-20ch century (implying we have decent understanding of those natural forcings)[/li][/ul]
And of course each bullet point is a consensus of dozens or hundreds of independent research papers. You could discredit some individual ones but I think we’d still be left with the same overall picture. And I’m not really sure what alternative picture you’re trying to make us see. Supposing the temperature proxies are all incorrect, what does that imply?
So what?
A lot of the proxies are based on similar measures. Also, when you look at spaghetti charts like that from Wikipedia, you have no way of knowing that they weren’t cherry picked. Not only that, but as you basically admit, some of the proxy studies DON’T show that recent temperatures are unprecedentedly high.
Fine, so if “context” is important, then you’ve answered your own question.
Did I ever claim otherwise?
Anyway, if somebody argues that Claim X is established by the preponderance of facts 1 through 10; and it turns out that fact 1 is wrong, then it tends to undermine Claim X.
I personally believe that catastrophic AGW theory is essentially a hoax. However that is an issue for another thread. This thread is more narrow.
