Saturday, February 6, 2010

Adventures Among the Alarmists

I've been spending far too much time reading 'climate change' websites and papers lately. Mostly I wanted to see if my skepticism about lack of model validation in climate science was well founded (it is, claims of impending doom are not supported by any sort of validated model predictions). I came across a good blog that I added to my reading list, written by a scientist in Japan working on climate modeling, and his frustrations with alarmism and appreciation of Bayes / Jaynes reflect my own (nice to see I'm not a complete crank), however his ensemble evaluation method which claims to 'do away with the truth-centered paradigm' is a bit troubling from a validation perspective (it seems like giving up on correcting any bias, might have a post on this after I read it again more carefully). So just for finding that site, the descent into the internet climate-swamp was worth it.

That's the extent of the good, here's some of the bad (or at least amusing and misinformed). This site is run by an NGO employee with a biology background. It has the sort of alarmist cheer-leading and appeal to authority that you would expect from a motivated, but uncritical social worker. One of the interesting things is the challenge page, where the challenge is to debunk 'anthropocentric global warming'. Now, I've never heard that particular turn of phrase (I've always heard it as 'anthropogenic'), and I thought the challenge was a little vague, so I asked them to clarify. The response I got was surprising, apparently 'anthropocentric' involves a philisophical / moral connotation as well as a scientific attribution one.  Here's the exchange:
jstults
that climate change is anthropocentric
Do you mean that the anthro forcing is the largest? That it is one of the biggest? That it is significant (an equal among peers)?
Would results showing that it was measurable but not practically significant meet your desiderata?

  • The meaning of anthropocentrism is more about how we dominate other species and generally overlook our duty towards others. Sadly, it has led to viewing the environment purely in terms of its value to humans; discussing climate change as if we are the only species impacted; and perceiving some members of our species as more important than others. As such, one can see that ‘anthropocentric’ has deeply philosophical and sociological meaning.
    I think you know that the ‘A’ in ‘AGW’ is usually taken to mean ‘anthropogenic’, as in ‘human caused’ — as in the overwhelming scientific evidence that the main driver of the current warming trend is C02 emissions from human activity.
    Did you have evidence to the contrary?
    “glaciers are melting… and not one entity that we trust with our money to look out for our interests as a species has any freaking clue as to why”
    There are entities that have a clue. They’re called climate scientists. They are not the entities that deal with your money.
    cheers

    • Martha said:
      The meaning of anthropocentrism is more about how we dominate other species and generally overlook our duty towards others
      I did not realize the philosophical connotations of that term. My personal ethical philosophy leans towards deontology, so I’m a bit familiar with the ethical concepts of ‘duty’. Would you say we have a ‘perfect duty’ towards the survival of other species?
      …as in the overwhelming scientific evidence that the main driver of the current warming trend is C02 emissions from human activity.
      I think the primary literature (and especially the recent findings about stratospheric water content) show that human CO2 emissions have effects with magnitudes similar to other forcings (at least on short time-scales, that’s one of the reasons it’s challenging to find the ’signal’ among the ‘noise’ in attribution studies). Can you point me at an attribution study that shows CO2 emissions are in-fact the primary driver?
      Thanks for addressing my questions and not flaming me.
      Cheers to you!

    • “glaciers are melting… and not one entity that we trust with our money to look out for our interests as a species has any freaking clue as to why”
      Nice work on finding that! I wasn’t sure what that was from until I stuck it in to Google. You realize I wasn’t the one who said that right? So what about the claims about the AGW-disaster connections? Any thoughts on that?

Hi,
1) Re. C02 Science. What primary literature are you speaking of? The overwhelming majority of published climate science says C02 is the main driver of the current warming trend. This science is linked and discussed all over this site and many other credible science sites. You aren’t the first visitor I’ve seen ignore it and I’m sure you won’t be the last.
I see you depend alot on Pielke. I encourage you to know that Pielke has been extensively debunked by competent scientists.
2) Re recent ‘findings’ regarding stratospheric water vapour. Is it Susan Solomon’s research you’re talking about?
http://www.realclimate.org/index.php/archives/2010/01/the-wisdom-of-solomon/
Romps and Kuang?
http://www.sciencedaily.com/releases/2009/04/090420121421.htm
These scientists are not challenging the core science of AGW.
Solomon, for example, wonders if changes in stratospheric water vapour might have helped keep the warming from being even worse in the past decade.
You know, no one is suggesting that all feedbacks are known or understood. Global climate and the carbon uptake cycle is a complex system. Climate knowledge is an evolving science. However, we don’t have to know everything to know alot, and the overwhelming majority of climate scientists say the evidence continues to agree with AGW.
The question is why you think they are wrong.
If you have evidence that they are wrong, please post it. This site has a thread for Challenging the Core Science. :-)

 Of course, they wouldn't publish my reply!  So here's what you would have seen if they weren't such wieners:
C02 is the main driver of the current warming trend
I guess I'd still appreciate a pointer to that study; what I've been able to find so far is that radiative 'forcing' from CO2 has an effect with a size comparable to lots of other stuff (hence the problem of 'natural variability' sometimes masking our ability to measure the effect), and estimating the climate sensitivity to CO2 is still an open research problem.
...you seem to rely on Pielke a lot...
Well, when it has to do with his area of expertise (policy and disasters), and how his research was misrepresented, I do (you brought it up btw). I'm aware of 'name calling' on the internet directed at Pielke, but not any published rebuttals of his work (again, pointers to the lit appreciated).
Solomon, for example, wonders if changes in stratospheric water vapour might have helped keep the warming from being even worse in the past decade.
That was kind of my point, that effect was one of similar magnitude and of opposite direction to the CO2 forcing, so they 'canceled' (at least in the short term up to present).
the overwhelming majority of climate scientists say the evidence continues to agree with AGW
If by AGW you mean 'humans are affecting climate', then most 'skeptics' would agree too. If by AGW you mean some sort of moral prescription (you didn't answer my duty question btw) or particular policy package, then I'm not sure what science has to say about that.
...extensively debunked by competent scientists...overwhelming majority of climate scientists say...
Appeal to authority much? I wouldn't mind if you link to a search or two on scholar.google.com, and anyone that comes along latter would benefit from it too.
The question is why you think they are wrong.
You are assuming an awful lot about what I think. My question was about clarifying the meaning of 'anthropocentric' (I hadn't heard that particular phrasing before), and what exactly would 'count' in answering the challenge on this post. I wasn't trying to tilt at some windmill or pick a fight ('Challenging the Core Science' as you so reverently put it).

I thought, surely this was an easy slow one over the plate for someone to answer.  Link to the attribution study that shows CO2 is the 'main driver' (the other questions about duty and disasters were just for poking fun).  So I googled for 'quantifying anthropogenic forcing', here's the papers Google found.

The most recent is from 2006, and it is far from a slam dunk (I was surprised, I expected this to be a very straight-forward result based on what I'd seen on alarmist websites and the IPCC). The attribution method is simplistic (multiple linear regression optimal fingerprinting), and they scale the model outputs so that they fit the observations! Are you calibrating or attributing!? The other thing I thought was interesting was this snippet:
It is misleading and ultimately fruitless to suggest there can be some kind of ‘‘global’’ detection and attribution analysis capable of summarising the climate-related information-content of a dataset in a single estimation procedure.
I am a skeptic (which I was able to figure out by consulting this resource), and I am more optimistic about the fruitfulness of newer / better methods than that. I think the only way forward is a fully integrated Bayesian approach where you tackle the whole big uncertainty quantification / parameter fitting problem in a coherent manner. [Update: I re-read that paragraph and thought it needed a little more explanation. Attribution studies and model validation are closely linked to uncertainty quantification and data assimilation, the whole process is an inverse problem much like tomography, unfortunately those sorts of problems can be quite sensitive to the regularization (assumptions), so out-of-sample validation becomes even more critical. This is the sort of thing that leaves them open to Lindzen's 'prosecutor's fallacy' criticism:
Most importantly, Hasselmann (1997) noted that, for any formal detection and attribution procedure to get started, it is necessary to confine attention a priori to a relatively small number of competing explanations for observed climate change. If the number of allowed model-simulated signals, m, is too large, then it becomes increasingly likely that at least one signal will closely resemble a linear combination of the others, leading to a so-called ‘‘degenerate’’ estimation problem in which the data are insufficient to constrain the bi. We will consider cases up to m = 4 in this paper, which allows us to cover the main known drivers of recent near-surface temperature change: greenhouse gases, anthropogenic aerosols, solar variability and volcanic activity. We will find that, even with m = 3 or m = 4, many results become ill-constrained by the kind of large-scale data considered in this paper.
See, it's an inverse problem (with noise): Bayes leads the way! I won't touch their assumption about linear superposition of forcing effects (but, but it seems to work over this parameter range), this update is already too long.]

Any way, that's my trip report. If you have any interesting attribution studies that I missed link 'em in the comments. Thanks.

5 comments:

  1. Martha said: "I encourage you to know that Pielke has been extensively debunked by competent scientists."

    Yes, so thoroughly debunked...

    Poke your head out of the echo chamber people!

    ReplyDelete
  2. This cracks me up:
    Bob Ward, the policy director of the Grantham Research Institute on Climate Change at the London School of Economics, said: "A lot of the climate sceptic arguments are being made by people with demonstrable right-wing ideology which is based on opposition to any environmental regulation of the market, and they are clearly being given money that allows them to disseminate their views more widely than would be the case if they didn't have oil company funding."
    --Think-tanks take oil money and use it to fund climate deniers

    Really? How expensive is a blog? Upload on arXiv? Any idiot with an opinion can be read by the world (hey, stop laughing at me!). No need for an oil baron backer (a man can dream...).

    ReplyDelete
  3. I guess my question about attribution was (by pure dumb luck on my part) a good one (though it shouldn't be surprising when you consider establishing causation):
    There is no scientific merit to be found in the Executive Summary. The presentation sounds lke something put together by Greenpeace activists and their legal department. The points being made are made arbitrarily with legal sounding caveats without having established any foundation or basis in fact. The Executive Summary seems to be a political statement that is only designed to annoy greenhouse skeptics. Wasn't the IPCC Assessment Report intended to be a scientific document that would merit solid backing from the climate science community -- instead of forcing many climate scientists into having to agree with greenhouse skeptic criticisms that this is indeed a report with a clear and obvious political agenda. Attribution can not happen until understanding has been clearly demonstrated. Once the facts of climate change have been established and understood, attribution will become self-evident to all. The Executive Summary as it stands is beyond redemption and should simply be deleted.
    -- Andrew Lacis

    NYT has more details

    ReplyDelete
  4. More on the topic of establishing causation (aka optimal fingerprinting): How persuasive is a good fit? A comment on theory testing

    This is particularly troublesome when it comes to theories that involve the solution of PDEs, because you can force them to give you any solution you want!

    ReplyDelete
  5. I found the 'skeptic checklist' handy; here's a denialism checklist based on the five characteristics presented in this paper. (Verbatim snippets in italics)
    1. Identification of conspiracies While conspiracy theories cannot simply be dismissed because conspiracies do occur,8 it beggars belief that they can encompass entire scientific communities.
    2. Use of fake experts
    3. Selective use of primary literature selectivity, drawing on isolated papers that challenge the dominant consensus or highlighting the flaws in the weakest papers among those that support it as a means of discrediting the entire field
    4. Expecting impossible standards of proof
    5. Misrepresentation and logical fallacies

    ReplyDelete