Loading ...
Sorry, an error occurred while loading the content.

Re: News: Research finds Greenland glacier melting faster than expected

Expand Messages
  • Martin Sewell
    Alex, et al. ... I have a pink shirt, but have never worn it. I suspect that one needs to be extrememly macho to carry it off. ... Bayesian analysis can t
    Message 1 of 24 , Sep 1, 2011
    • 0 Attachment
      Alex, et al.

      --- In climate-change-forum@yahoogroups.com, Alex Harvey <alexharv074@...> wrote:
      > I'm amused because as it happens my shirt is blue but it turns out
      > I have two pink shirts. I have a black shirt but never wear it, and
      > I don't have any white or grey shirts.

      I have a pink shirt, but have never worn it. I suspect that one needs to be extrememly macho to carry it off.

      > Next question: using Bayesian uncertainty analysis tell me how many
      > feedbacks operate in the climate system that haven't been discovered
      > yet.

      Bayesian analysis can't tell you anything about the real world, but simply informs us how to update our degrees of belief in the light of new evidence. The problem of 'unknown unknowns' can be dealt with by putting a prior on 'any other'.

      > And of these, how strong are they likely to be?

      This depends on how well existing models explain the data.

      I would be interested to know if you were the editor of the next IPCC report, what would you do differently?

      Cheers

      Martin
    • Alex Harvey
      ... Give it a try! Bayesian analysis can t tell you anything about the real world, but simply ... It sounds that you are admitting that you know full well that
      Message 2 of 24 , Sep 2, 2011
      • 0 Attachment
        On 1 September 2011 18:51, Martin Sewell <mvs25@...> wrote:

        --- In climate-change-forum@yahoogroups.com, Alex Harvey <alexharv074@...> wrote:
        > I'm amused because as it happens my shirt is blue but it turns out
        > I have two pink shirts. I have a black shirt but never wear it, and
        > I don't have any white or grey shirts.

        I have a pink shirt, but have never worn it. I suspect that one needs to be extrememly macho to carry it off.

        Give it a try!

        Bayesian analysis can't tell you anything about the real world, but simply informs us how to update our degrees of belief in the light of new evidence. The problem of 'unknown unknowns' can be dealt with by putting a prior on 'any other'.

        It sounds that you are admitting that you know full well that "99.2% certainty" is a subjective quantity probably fudged to be as close to 100% as possible.

        If 'unknown unknowns' can be "dealt with by putting a prior on 'any other'" can the IPCC now tell me if God exists?
         
        I would be interested to know if you were the editor of the next IPCC report, what would you do differently?

        My view is that the IPCC shouldn't exist. But if it must exist, and if I was a dispassionate, competent version of Rajendra Pachauri, I would immediately make the following changes:

        1) scientists who are also climate change advocates would not be selected as lead authors as they would be deemed to have a conflict of interest.
        2) scientists who would be put in a position where they would have to review their own work would not be selected.
        3) scientists who had any connection with renewable energy interests would not be selected.

        Alex
      • Martin Sewell
        ... The figure 99.8% is the proportion of the pdf of total radiative forcing from anthropogenic agents that is greater than zero according to the research
        Message 3 of 24 , Sep 2, 2011
        • 0 Attachment
          --- In climate-change-forum@yahoogroups.com, Alex Harvey <alexharv074@...> wrote:
          > It sounds that you are admitting that you know full well that "99.2%
          > certainty" is a subjective quantity probably fudged to be as close
          > to 100% as possible.

          The figure 99.8% is the proportion of the pdf of total radiative forcing from anthropogenic agents that is greater than zero according to the research shown here: <http://www.ipcc.ch/publications_and_data/ar4/wg1/en/figure-2-20.html>. It is an artefact of the sum of the individual forcings. Science necessarily involves making assumptions.

          > If 'unknown unknowns' can be "dealt with by putting a prior on 'any
          > other'" can the IPCC now tell me if God exists?

          The scientific case against the existence of a god is that the evidence is adequately explained by a simpler hypothesis, and simpler hypotheses have a stronger pedigree in this universe.

          > 1) scientists who are also climate change advocates would not be
          > selected as lead authors as they would be deemed to have a conflict
          > of interest.

          Climate change advocacy is a product of the science, so is not a conflict of interest.

          > 2) scientists who would be put in a position where they would have
          > to review their own work would not be selected.

          Certainly, peer review, not self-review, should be used.

          > 3) scientists who had any connection with renewable energy
          > interests would not be selected.

          This sounds reasonable, but you should of course also include the (much larger) fossil fuel and nuclear industries too. However, relevant research by anyone that has passed peer review in credible journals should be considered.

          Regards

          Martin
        • Alex Harvey
          ... It sounds like you know full well that 99.2% certainty is *subjective *- not a conclusion derived from a scientific study of the climate system itself.
          Message 4 of 24 , Sep 2, 2011
          • 0 Attachment
            On 2 September 2011 22:01, Alex Harvey <alexharv074@...> wrote:

            It sounds that you are admitting that you know full well that "99.2% certainty" is a subjective quantity probably fudged to be as close to 100% as possible.

            And sorry - this was very badly worded. What I wanted to say is:

            It sounds like you know full well that "99.2% certainty" is subjective - not a conclusion derived from a scientific study of the climate system itself. Instead of measuring the climate, this number is derived from a survey of conclusions of peer reviewed papers. So it puts numbers around what is essentially a circular appeal to authority. It's the same as arguing "because Naomi Oreskes showed that 100% of published papers support the AGW theory, therefore it's right". I would add that the number has probably been fudged to be as close to 100% as possible.

            Alex
          • Martin Sewell
            Alex, et al. ... Using expert witness via peer review is not the same as an appeal to authority. In fact this is ultimately how science progresses. Why do
            Message 5 of 24 , Sep 3, 2011
            • 0 Attachment
              Alex, et al.

              --- In climate-change-forum@yahoogroups.com, Alex Harvey <alexharv074@...> wrote:
              > It sounds like you know full well that "99.2% certainty" is
              > *subjective *- not a conclusion derived from a scientific study of
              > the climate system itself. Instead of measuring the climate, this
              > number is derived from a survey of conclusions of peer reviewed
              > papers. So it puts numbers around what is essentially a circular
              > appeal to authority. It's the same as arguing "because Naomi Oreskes
              > showed that 100% of published papers support the AGW
              > theory, therefore it's right". *I would add* that the number has
              > probably been fudged to be as close to 100% as possible.

              Using 'expert witness' via peer review is not the same as an appeal to authority. In fact this is ultimately how science progresses. Why do most people believe the Out-of-Africa model of human origins? Because most scientists believe it to be true.

              In fact one could even argue that this is the case with advanced maths. How do we know that Andrew Wiles proved Fermat's Last Theorem? Ultimately it was when other mathematicians finally agreed that he had.

              Regards

              Martin
            • Alex Harvey
              ... believe implies that the believer not a scientist. The scientists themselves are not sheep believing things simply because everyone else does. You seem to
              Message 6 of 24 , Sep 3, 2011
              • 0 Attachment
                On 3 September 2011 21:20, Martin Sewell <mvs25@...> wrote:

                Using 'expert witness' via peer review is not the same as an appeal to authority. In fact this is ultimately how science progresses. Why do most people believe the Out-of-Africa model of human origins? Because most scientists believe it to be true.

                Believing something is true simply because it is what most scientists believe implies that the believer not a scientist. The scientists themselves are not sheep believing things simply because everyone else does.

                You seem to admit that the assertion "99.8% certain" is a subjective analysis of 'expert witness' via peer review. Without those papers as inputs, it would be impossible to derive this figure. Do you admit this?

                Do you also admit that it is impossible to lay out the raw data of climate science, apply objective mathematical analysis to the data, and derive the assertion "an enhanced greenhouse effect is 99.8% certain"?

                Alex
              • Martin Sewell
                ... Did you personally derive Newton s laws of motion, Einstein s theory of general relativity, the physics of quantum mechanics and the theory of evolution?
                Message 7 of 24 , Sep 3, 2011
                • 0 Attachment
                  --- In climate-change-forum@yahoogroups.com, Alex Harvey <alexharv074@...> wrote:
                  > Believing something is true simply because it is what most
                  > scientists believe implies that the believer not a scientist. The
                  > scientists themselves are not sheep believing things simply because
                  > everyone else does.

                  Did you personally derive Newton's laws of motion, Einstein's theory of general relativity, the physics of quantum mechanics and the theory of evolution? Or perhaps, like most of us, you put some faith in the science that has gone before us. For example, if we were born a few centuries ago, we would likely believe in god.

                  > You seem to admit that the assertion "99.8% certain" is a subjectiv
                  > analysis of 'expert witness' via peer review. Without those papers
                  > as inputs, it would be impossible to derive this figure. Do you
                  > admit this?

                  The figure is derived via Monte Carlo simulations conducted by Boucher and Haywood (2001) <http://www.springerlink.com/content/nkmx3fvlvd9yj1x9/>.

                  Hume (1740) pointed out that 'even after the observation of the frequent or constant conjunction of objects, we have no reason to draw any inference concerning any object beyond those of which we have had experience'. In other words, science without making subjective assumptions is literally impossible.

                  > Do you also admit that it is impossible to lay out the raw data of
                  > climate science, apply objective mathematical analysis to the data,
                  > and derive the assertion "an enhanced greenhouse effect is 99.8%
                  > certain"?

                  As above, an element of subjectivity is always necessary, what's important is that it is kept to a minimum, and the avoidance of any systematic bias or ideology.

                  Regards

                  Martin
                • Alex Harvey
                  ... Thank you for the paper. You have prompted me to read it again. See below. ... It is a question of how many subjective assumptions are reasonable before
                  Message 8 of 24 , Sep 3, 2011
                  • 0 Attachment
                    On 4 September 2011 02:44, Martin Sewell <mvs25@...> wrote:
                     
                    > You seem to admit that the assertion "99.8% certain" is a subjectiv
                    > analysis of 'expert witness' via peer review. Without those papers
                    > as inputs, it would be impossible to derive this figure. Do you
                    > admit this?

                    The figure is derived via Monte Carlo simulations conducted by Boucher and Haywood (2001) <http://www.springerlink.com/conttnt/nkmx3fvlvd9yj1x9/>.

                    Thank you for the paper. You have prompted me to read it again. See below.

                    Hume (1740) pointed out that 'even after the observation of the frequent or constant conjunction of objects, we have no reason to draw any inference concerning any object beyond those of which we have had experience'. In other words, science without making subjective assumptions is literally impossible.

                    It is a question of how many subjective assumptions are reasonable before your analysis becomes a joke. I suspect Hume would turn in his grave if he read the IPCC AR4.


                    > Do you also admit that it is impossible to lay out the raw data of
                    > climate science, apply objective mathematical analysis to the data,
                    > and derive the assertion "an enhanced greenhouse effect is 99.8%
                    > certain"?

                    As above, an element of subjectivity is always necessary, what's important is that it is kept to a minimum, and the avoidance of any systematic bias or ideology.

                    In this case the IPCC has presented a thoroughly circular argument.

                    Boucher and Haywood (2001), p. 299:

                    Different scenarios are set up where one of the three PDFs is
                    assigned to each RF component shown in Fig. 1. The normal and
                    log-normal PDFs are set up so that the range given by Ramaswamy
                    et al. (2001) encompasses the mean +- x standard deviations
                    (Table 1). As Ramaswamy et al. (2001) state that the range is based
                    on values in the published literature, our calculations assume x
                    values of 1.0, 1.5, and 2.0.


                    So the values in the published literature are derived from models which are known to be flawed. To the extent that they are flawed, different models are likely to be flawed in the same places. For example, they all get the radiation right because that part of the science is understood and can be modelled. But they all get the clouds wrong because that part is not understood at all.

                    Next, using different flawed models, different flawed values appear in the published literature for the RF of clouds. From these, the PDF function is chosen to represent the cloud RF.

                    Finally, summing all PDFs to get the TPDF, you conclude the probability of positive total radiative forcing is 99.8%. (Curiously Boucher and Haywood got 75%-97% so again I suspect that the IPCC have fudged their analysis to get it so close to 100%. But that's just another matter.)

                    In any case, it is a completely bogus, circular argument. I argue that the models are likely to be flawed, and therefore all flawed in similar ways. You say that's very unlikely because they all give the same answer. It is a deliberate deception that is being inflicted on the public.

                    Alex


                  • Martin Sewell
                    ... The IPCC s figure of 0.2% is the more recent, see their explanation here: . ... The
                    Message 9 of 24 , Sep 4, 2011
                    • 0 Attachment
                      --- In climate-change-forum@yahoogroups.com, Alex Harvey <alexharv074@...> wrote:
                      > Finally, summing all PDFs to get the TPDF, you conclude the
                      > probability of positive total radiative forcing is 99.8%.
                      > (Curiously Boucher and Haywood got 75%-97% so again I suspect that
                      > the IPCC have fudged their analysis to get it so close to 100%. But
                      > that's just another matter.)

                      The IPCC's figure of 0.2% is the more recent, see their explanation here: <http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch2s2-9-2.html>.

                      > In any case, it is a completely bogus, circular argument. I argue
                      > that the models are likely to be flawed, and therefore all flawed
                      > in similar ways. You say that's very unlikely because they all give
                      > the same answer. It is a deliberate deception that is being
                      > inflicted on the public.

                      The existence of model uncertainty does not imply deliberate deception, all models are wrong.

                      Martin
                    Your message has been successfully submitted and would be delivered to recipients shortly.