While I'm not a "denier" per se, the IPCC has made some monumental errors (hockey stick, anyone?) in the past that have really hampered their ability to be a source of non-political "good science".

> The test in science is whether findings can be replicated using different data and methods. More than two dozen scientific papers, using various statistical methods and combinations of proxy records, have produced reconstructions supporting the broad consensus shown in the original 1998 hockey-stick graph, with variations in how flat the pre-20th century "shaft" appears.[12][13] The 2007 IPCC Fourth Assessment Report cited 14 reconstructions, 10 of which covered 1,000 years or longer, to support its strengthened conclusion that it was likely that Northern Hemisphere temperatures during the 20th century were the highest in at least the past 1,300 years.[14] Ten or more subsequent reconstructions, including Mann et al. 2008, have supported these general conclusions.

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

The collected historical temperature data series are messy. You plot them and it looks like noise. A simple average will tell you nothing. The regression analysis that goes on in these reconstructions tends to follow something like this algorithm:

1) Compare your collected data series to each other to determine a "reliability" index for each data series.

2) Construct a series as a weighted average on this reliability index.

3) Iterate, using your new series to calculate the reliability index.

There are substantial issues with the above approach, having to do with sensitivity to particular kinds of signals. For a while, the Climate Audit folks got a lot of mileage over a data set that was put in "upside down" in one of the reconstructions. Rather than claiming "data error," the author defended it based on the idea that the above process would just give it a negative weighting!

Regardless, you should see why the above could easily be susceptible to issues. Further, the disagreement between these reconstructions and what our limited historical sources tell us about the MWP, etc., makes them further suspect.

This is a problem for the global warming model because it means that we don't know what the historical standard deviation for temperature is. So we can't say whether the last 3 decades warming is "within normal deviations" or "unprecedented."

Your critique sounds interesting but I don't know enough to evaluate it--I'm not a scientist myself.

I would just point out that scientific debate occurs through published papers and conference presentations, not programming message boards. I hope that you are publishing in those channels if you feel you have substantive contributions to this kind of research.

Nope, that's not actually how science works. Rather, that is "formally" how science work. Actual practice is far more messy. Personalities matter, personal bureaucratic clout matters.

And happily, informal communication and message boards matter more than you would think!

Also, the published papers reporting these studies are all easily available. I highly advise reading them for yourself.

snowwrestler, it may vary by field, but my (non-climate) field relied almost entirely on informal communications, presentations, write ups being shared but not published.

I don't have direct experience with climatology, but from what I understand it's somewhat similar: conferences like AGU are really important to climatologists for that reason.

To present (even a poster) at a conference like AGU, you have to submit an abstract and it must be accepted by a committee of other scientists. It's not a paper, but still counts as a form of review and publication.

Edit to acknowledge: obviously a lot of work gets done informally during conferences as well.

But it's in the service of publishable research, is it not? I don't think many scientists are going to say, "I'm changing my research program based on something I read on Hacker News yesterday." There's a bar for serious consideration.

No offense, but those are the sorts of things that people say when their own research cannot pass muster for publication.

edit: misspelling

You just claimed not to be a scientist, yet you claim the high ground in knowing how science is conducted?

I made a jerk comment above.

I stand by the importance of published work in scientific discourse, but I apologize to schoper for taking a personal shot that was not called for.

What's your critique of the hockey stick graph? In what sense was it an error, let alone a "monumental" one?

The flaws are numerous, but the easiest to explain is that they employ a post-hoc screening bias. Imagine that you gather a few dozen time-series together. They go back two thousand years. Then you regress them against the (very short) instrumental temperature record in the period where they overlap. You keep the ones that pass and throw the rest away. You form some sort of linear weighted average of the selected proxies (i.e. PCA for example).

The thing about this process is, if you feed in completely random red-noise time series you will get a hockey stick. The reason is simple, the selection period will mirror the instrumental temperature (mental check: it has to, after all we only selected the proxies that matched the instrumental temperature). You have your blade. But none of those red noise proxies actually reflect past temperature do they? They just happened to be selected by random chance. So in the time before the selection period, the noise cancels out and you have a flat line. There is your stick.

This is the core of how these "proxy reconstructions" work, and it is nonsense.

Outside of that, it gets even better. The stories are comical and numerous. Mann famously fed a contaminated Finnish lake sediment series into his algo, which happily regressed it against temperature and found that it passed significance if it was used upside down (in the opposite orientation to that suggested by the author in the original source paper). So in it went. GIGO. Or how about the question of which temperature series to regress against? You'd think you'd regress against local temperature right? Or maybe at least the same continent? Oh ye optimistic fool, go and google "teleconnection" and "non-local statistical relationship". This is what passes as "climate science".

A National Academy of Sciences inquiry found that the graph was generally correct. The graph has been validated by 12 reconstructions that produced similar results.

If I provided multiple independent reconstructions of historical temperatures, would you change your position and accept the scientific consensus?

Does it sound by my post that I am ignorant of these "multiple" reconstructions? They all use the same proxies and similar techniques, Esper, Briffa, bristle cones, etc. I have read a good number of them. They are all awful.

Paleoclimatology isn't any worse than some other soft sciences I guess - the difference is that there isn't so much riding on them being right.

For the interested, check out Wahl and Ammann 2007 that addresses the simplistic criticism that gd1 offers.

Wahl and Ammann 2007 does not appear to adequately address gd1's points. It just applies a slightly different statistical approach using the same flawed premise.

There are countless arguments and rebuttals to be found on both sides of the debate, but I've found the ones below to be particularly helpful:

Pro Hockey Stick:

[1] Original MBH98 paper (PDF): http://stephenschneider.stanford.edu/Publications/PDF_Papers...

[2] Wahl and Ammann 2007 (PDF): http://www.cgd.ucar.edu/ccr/ammann/millennium/refs/Wahl_Clim...

[3] Wikipedia article: http://en.wikipedia.org/wiki/Hockey_stick_controversy

[4] Real Climate Hockey Stick "Dummies Guide" (PDF): http://www.realclimate.org/dummies.pdf

Skeptical of Hockey Stick:

[5] Skeptic blog post "Casper and the Jesus Paper": http://bishophill.squarespace.com/blog/2008/8/11/caspar-and-...

[6] "The Hockey Stick Illusions" book (same skeptic author): http://www.amazon.com/The-Hockey-Stick-Illusion-Climategate/...

[7] McIntyre & McKitrick 2003 (PDF): http://climateaudit.files.wordpress.com/2005/09/mcintyre.mck...

Thanks. It would definitely have been better for those listening to this discussion to hear your counterpoints rather than me having to go look it up in a sea of words. Especially since I'm not really interested in the field. Also, you called his criticism simplistic without explaining why. Not very good manners.

The grandparent offered specific criticisms against said reconstruction, and you countered with, essentially, "others got the same result". Scientific questions are not decided by a majority vote. Do you have more concrete arguments?

By the way, are you a climate scientist?

The question was, would he accept independent reconstructions from other sources to change his opinions, since he finds PCA so objectionable. There are obviously lots of concrete arguments--I could come up with a pretty list of papers--but my time is valuable and I was curious if he was genuinely interested in the answer. (Apparently not.)

I don't even get what you're getting at: scientific arguments are also not decided by someone spouting off something they read off of Climate Audit and pretending to be an expert when they don't even have a passing familiarity with the scientific literature. Asking an open question "well, what would make you change your mind" is a perfectly reasonable response, especially when the answer is, expectedly, "nothing."

No, I'm not a climate scientist, though an immediate family member is.

Actually his answer was that he is familiar with several independent reconstructions and that they suffer from similarly problematic flaws, not that he is not "interested in the answer".

Your second paragraph is an ad hominem, where you make various assumptions about the OP. Yet there is still not a single coherent argument why the OP was actually wrong.

If you start with 100 random time series and then select only those that match the recent instrument record, you no longer have a random selection of data. So why would there be reason to believe that they will consistently "cancel out" to a flat line?

If you ran this experiment numerous times I would expect to see the "average" line prior to the instrument record vary as a random walk.

If the series don't really correlate to temperature (and the red noise doesn't) even though they appear to during the calibration period, then how it moves before the calibration period will be random. So when you take your weighted average of this randomness they will tend to cancel out (or at least tend to be flatter than the reconstruction during the calibration period). The greater the number of series, the more pronounced the effect. The point being that a hockey stick shape is an artefact of the method used, whether or not the proxies actually are good indicators of past temperature. I am trying to keep this somewhat simplified.

Of course you don't have to take my word for all this, thanks to Climategate we have all the behind-the-scenes dirty laundry:

http://foia2011.org/index.php?id=4191

Here Rob Wilson does just the test I'm describing and seems a little concerned about the results.

Just for funsies: a cousin comment to yours argues that the actual hockey stick issue is that they're no longer sticks if you extend them past 500 years ago. This doesn't jive at all with what your analysis would suggest: if there's no power above noise provided by the proxies, then wouldn't we expect the shafts to be infinitely long?

The reconstructions by proxy referred to by gd1 leads to a hockey stick as far as back as you can go. Expect of course were there is very sparse proxy data at which point the trends will be completely random.

The cousin refers to the obvious long-term temperature variability. Nobody doubts the major ice ages. But the long-term reconstructions use very different techniques.

The main issue with any "Hockey Stick" as presented is the length of the shaft.

The stick all but vanishes when you look at a long term climate projection, rather than the last 500 years or so.

[citation needed]