Monckton: part two

[Essan]

Ah well, in that case the IPCC must biased after all :unsure:

[Admiral_Bobski]

Just a quick note to say that not all temperature measurement sites show a warming, and some even show a cooling (the oft quoted one being the interior of the Antarctic continent). GW theory doesn't say that warming will occur evenly throughout the globe, but accurate thermometer readings from the majority of the globe would be useful data to have. Some would argue that we have satellite measurements which clearly show a warming trend, but it is always worth cross-referencing data from one source with data from another, for accuracy's sake. And, even then, this doesn't prove that humans are causing the temperature to rise, only that the temperature is rising!

Let me quickly say again, for the record, that I don't deny an apparent global warming trend but I question the degree to which mankind is affecting the trend.

:unsure:

C-Bob

PS - As for the IPCC report, it's only a couple of months away so let's wait and see...!

[parmenides3]

If what you are questioning is the degree to which humanity contributes to GW, then what we are dealing with here is the issue of attribution. Attribution studies are an important part of climate science and can be the cause of some controversy and disagreement. There is a section in the new AR4 on detection and attribution; some of the material which will have been used in its compilation is readable here: http://www-pcmdi.llnl.gov/ipcc/subproject_publications.php

Unfortunately, some of the papers can only be read as abstracts, for copyright reasons.

Given that you appear to accept that there is some anthropogenic contribution to global warming/ the recent warming trend, can I ask you to suggest a proportion/percentage which you think might be attributable to us, from a total which includes natural forcings such as solar variation and volcanicity?

Do you think it is 30%? 50%? More? Less? This would be helpful, as it gives us a potential starting-point for the next stage of the discussion.

Suggestions from all readers are welcomed.

:)P

[Admiral_Bobski]

If what you are questioning is the degree to which humanity contributes to GW, then what we are dealing with here is the issue of attribution. Attribution studies are an important part of climate science and can be the cause of some controversy and disagreement. There is a section in the new AR4 on detection and attribution; some of the material which will have been used in its compilation is readable here: http://www-pcmdi.llnl.gov/ipcc/subproject_publications.php

Unfortunately, some of the papers can only be read as abstracts, for copyright reasons.

Given that you appear to accept that there is some anthropogenic contribution to global warming/ the recent warming trend, can I ask you to suggest a proportion/percentage which you think might be attributable to us, from a total which includes natural forcings such as solar variation and volcanicity?

Do you think it is 30%? 50%? More? Less? This would be helpful, as it gives us a potential starting-point for the next stage of the discussion.

Suggestions from all readers are welcomed.

:)P

Hi P3 - I see you have posted on two threads, so I had to decide which one to answer first! I'll take this one as your other post requires a bit more time to respond to.

I have said from the outset (way back in my first post in the environment change boards, I think) that I don't deny there appears to be a certain degree of waming going on, and that my issue is with humanity's contribution to it, if any. Now, don't get me wrong, I also accept that we put a certain amount of CO2, methane and other "GHGs" and volatiles into the air. I question the exact cause of the current warming (I know the Industrial Revolution is often blamed, but it could be nothing more than a convenient scapegoat), I think not enough credit is given to the fact that the Earth can take care of itself, and I question how much warming the "GHGs" actually cause in the dynamic scenario of global climate. I think too many assumptions are made on the one hand, and I think not enough thought is given to certain factors on the other.

I would not be so bold as to try to put values to our "contribution" to global warming - I doubt the figure is particularly high, and I wouldn't be surprised if it is hovering somewhere around the 0% mark (yes, that's not a typo - that's zero percent). I have not yet seen any incontrovertible evidence that global warming can be directly attributed to human activity. Start there! :unsure:

C-Bob

[parmenides3]

Please explain why this is not convincing?:

:)P

[Admiral_Bobski]

Please explain why this is not convincing?:

:)P

I think I've already made my feelings on graphs such as this clear. CO2 levels are rising - no argument from me. Temperatures are rising (approximately 1 degree Fahrenheit in 50 years) - again, no argument. The link between them? Inconclusive. Also, at present I object to the sudden switch from proxy data to physical readings - explanations to follow when I have more time...

C-Bob

Edited December 16, 2006 by Captain_Bobski

[parmenides3]

I think I've already made my feelings on graphs such as this clear. CO2 levels are rising - no argument from me. Temperatures are rising (approximately 1 degree Fahrenheit in 50 years) - again, no argument. The link between them? Inconclusive. Also, at present I object to the sudden switch from proxy data to physical readings - explanations to follow when I have more time...

C-Bob

The point of this graph is that the link between CO2 and temperature looks quite clear. I understand that it is possible to plot the temperature against any data which shows a rise since 1880 [e.g. average weight of British children], and this does not necessarily prove a correlation. But, in the case of the global temperature, you cannot so this for any forcing, such as Solar, and get a similar result [without 'fixing' the data].

Is it wrong to add measured data to proxy data? This is one of the central objections to another graph, the 'hockey stick'. It isn't bad science - the NAS, for example, concluded that this was a perfectly acceptable methodology. It does make sense to use the best available data, after all, and this is what is done here. There are no direct measurements of CO2 prior to 1950, so proxy data is the only available source, and direct measurement is more accurate than proxy data, so that is why it has been added.

You'll find a lot of technical, but interesting, objections to a range of proxy measurements on www.climateaudit.com . Now, what proportion of that ~1F increase since 1950 is attributable to human activity (GHGs). As you seem reluctant to commit to a figure, I'll suggest one; the amount of temperature increase attributable to increased atmospheric CO2 is most likely to be in the range of 30-50%. Any objection to this number?

:)P

[Admiral_Bobski]

The point of this graph is that the link between CO2 and temperature looks quite clear. I understand that it is possible to plot the temperature against any data which shows a rise since 1880 [e.g. average weight of British children], and this does not necessarily prove a correlation. But, in the case of the global temperature, you cannot so this for any forcing, such as Solar, and get a similar result [without 'fixing' the data].

We know there appears to be some correlation between temp and CO2, but this graph doesn't prove what that correlation might be.

Is it wrong to add measured data to proxy data? This is one of the central objections to another graph, the 'hockey stick'. It isn't bad science - the NAS, for example, concluded that this was a perfectly acceptable methodology. It does make sense to use the best available data, after all, and this is what is done here. There are no direct measurements of CO2 prior to 1950, so proxy data is the only available source, and direct measurement is more accurate than proxy data, so that is why it has been added.

So if you accept that the proxy data is less accurate than the direct data then why do you not also accept that the two data sets really shouldn't be sharing a set of axes?

You'll find a lot of technical, but interesting, objections to a range of proxy measurements on www.climateaudit.com . Now, what proportion of that ~1F increase since 1950 is attributable to human activity (GHGs). As you seem reluctant to commit to a figure, I'll suggest one; the amount of temperature increase attributable to increased atmospheric CO2 is most likely to be in the range of 30-50%. Any objection to this number?

Yes, my objection to this number is that 1) it is unprovable, and 2) it's a big range. This stresses two of the problems with global warming theory - firstly there is no proof and secondly the estimated ranges given are often so large as to be meaningless.

Rather than producing pretty graphs and drawing tenuous links between data sets, perhaps climatologists should be applying themselves to making a computer model that can actually be tested against observations.

C-Bob

Edited December 16, 2006 by Captain_Bobski

[parmenides3]

On the relationship between temperature change and CO2, as I said before, we are dealing with an attribution issue. The only way in which attribution can be done is through some kind of modelling. This is because it is otherwise impossible to separate the effects of a range of forcings and feedbacks. This paper, by Stott et. al., is a formal attribution study: http://funnel.sfsu.edu/courses/gm310/artic...nturyCauses.pdf

Hansen et. al. (2005) attributes 1.5 W/m2 of added atmospheric radiation to CO2, more than methane, the other GHGs, solar forcing, biomass burning [which appears surprisingly small in effect by their measurements] put together. To relate this to the changes in temperature, we also have to take into account the negative forcings. Both studies conclude that CO2 is the largest [by a considerable amount] single factor effecting recent warming.

I don't understand what your objection is to placing proxy data and measured data on the same axis, if they are measures of the same thing. The relative accuracy of the proxy data is poorer than the measured data, but this does not mean it is inaccurate, nor that it is somehow wrong, just that there is less (scientific) certainty about the numbers. Regardless of this, the trend still looks the same, whatever you do with the error bars, on both datasets.

Avoiding the semantics of the term 'proof', which should properly only be used to describe mathematical formulae or legal cases, we have to accept that the only way to estimate the influence of CO2 is through modelling (see above), and that there has been a huge amount of work done on detection and attribution in the last twenty years, which almost universally reaches the same conclusions about both the causes of recent temperature changes and the dominance of CO2. If you look at the sort of papers illustrated above, you should be able to see that issues of bias and scientific rigour are taken seriously, and every effort is made not to put the cart before the horse when compiling data. To suggest an unconscious 'bias' implicit in either the science or the findings is to ignore the insistence on scientific method, which takes such matters into account.

The range I proposed is my own suggestion, not one from another source; it is wide, so as to give you an opportunity to answer the question I keep posing about how much you think might be attributable to us humans. Individual attribution studies tend to be quite precise about the numbers they come up with though, of course, there is variety from paper to paper. So are you willing to put a number to human contributions, or not?

Climatologists do not waste their time producing pretty graphs, nor do they draw tenuous links between data sets; they seek and make use of the best available data in all cases to provide the most accurate models they can. All of the models can be tested against observations. This is hindcasting. They have got quite good at this. The problem [as you have pointed out] often lies in the relatively short timespans of measured data for most metrics; 1880 is about as far back as many measurements go, the CET being a notable exception.

I don't think you are giving full credit to the work that climate scientists do, or the models that they use. The complexity and detail is quite stunning, the science is rigorous and much tested, and the uncertainties have been decreasing over time, as more information becomes available. One interesting point of note is that, in their most recent study, COP12, the Hadley Centre is confident enough to place a single value on the estimated temperature increase to 2100. Of course, this is a function of the modelling exercise, but it does suggest that they are becoming increasingly clear about the most likely scenario for climate change in the coming century.

You have said you are willing to accept that humans have contributed to recent temperature increases. If this is the case, we must also be able to contribute to future slowing of temperature increase. Given the possibility that failing to do so would have a negative impact on human (and all life) on Earth, surely it then follows that we should make the effort to do this.

:)P

[Admiral_Bobski]

On the relationship between temperature change and CO2, as I said before, we are dealing with an attribution issue. The only way in which attribution can be done is through some kind of modelling...Both studies conclude that CO2 is the largest [by a considerable amount] single factor effecting recent warming.

There are a couple of issues with the study you linked to that need to be highlighted. There is always some degree of "tweaking" required to make these models fit in with observations (a fact they allude to by declaring that this is an issue with "Simplified Energy Balance Models", but to some extent it is an issue with all computer models). If a model requires tweaking then its fundamental basis is obviously flawed, as its intial output does not take all necessary information into account. Only by making adjustments without necessarily fully understanding them can the models be made to fit in with actual data, thereby exposing a distinct gap in our knowledge of the climate system.

Towards the end of the report they say "There are considerable uncertainties in some of the forcings used in this analysis. The model simulates roughly half...of [the] sulfate aerosol observed over Europe, and will therefore underestimate their direct cooling effect. However, there may be some compensation from omitting the warming effect of black carbon aerosols." They go on to address other inaccuracies in the model, but this example will do for this point. Leaving out pertinent details of cooling- and warming-forcings, and assuming that they will cancel each other out, is not sufficient to provide a definitive scientific conclusion. The fact that the results of the experiment show 20th century warming is neither here nor there since they have tweaked the models to make sure, in effect, that this is what they will show. You could argue that the "tweaking" takes into account this discrepencies, but even that is not a valid argument since those subtleties have not been modelled and we don't know what effect these other forcings were likely to have had in the long term.

There are too many factors working together in the actual global climate system to be able to construct a sufficiently detailed computer model - something is always going to get left out, and without that "something" being included we can't genuinely say that we know what effect it would have had.

I don't understand what your objection is to placing proxy data and measured data on the same axis, if they are measures of the same thing. The relative accuracy of the proxy data is poorer than the measured data, but this does not mean it is inaccurate, nor that it is somehow wrong, just that there is less (scientific) certainty about the numbers. Regardless of this, the trend still looks the same, whatever you do with the error bars, on both datasets.

Allow me to post a couple of graphs - the first is a graph from the IPCC website which shows temperature variation over the last 1000 years or so, using proxy data (blue line) at first then being substituted with actual measurements (red) in the relevant period. The grey area around the blue line shows the uncertainty in the data and, naturally, this grey area becomes smaller as time goes on until it is virtually gone by the time proxy data is replaced with actual data.

Since this grey area shows legitimate thresholds for the retrieved proxy data, the graph could look like this:

(Forgive the terrible shoddiness of the line, but it gives the basic idea.) The blue line shows the "direct" temperature as determined from the proxy source, but this "direct data" is subject to inaccuracies. Therefore it is legitimate to show the maximum potential for the temperature data on a graph. Suddenly 20th Century warming doesn't look so bad.

Avoiding the semantics of the term 'proof', which should properly only be used to describe mathematical formulae or legal cases, we have to accept that the only way to estimate the influence of CO2 is through modelling (see above), and that there has been a huge amount of work done on detection and attribution in the last twenty years, which almost universally reaches the same conclusions about both the causes of recent temperature changes and the dominance of CO2. If you look at the sort of papers illustrated above, you should be able to see that issues of bias and scientific rigour are taken seriously, and every effort is made not to put the cart before the horse when compiling data. To suggest an unconscious 'bias' implicit in either the science or the findings is to ignore the insistence on scientific method, which takes such matters into account.

When I have used the word "proof" I have been intending it in the mathematical/scientific sense, not as some arbitrary concept of "correctness". What we have are scientific assumptions and suppositions. The fact that we do not know all of the ins-and-outs of the carbon cycle, that what we do know is subject to inaccuracies and tolerances, and that the climate models are based on our imperfect understanding of these cycles (in addition to the fact that the carbon cycle interacts with other cycles too) quite clearly shows that our computer models are not going to be anything more than approximations - rough approxiomations, at that, and ones that are as likely wrong as right. As for eliminating bias, there are still groups who follow the entire process through from original proposition to conclusion without the input of any third party - not good scientific practice for eliminating bias.

The range I proposed is my own suggestion, not one from another source; it is wide, so as to give you an opportunity to answer the question I keep posing about how much you think might be attributable to us humans. Individual attribution studies tend to be quite precise about the numbers they come up with though, of course, there is variety from paper to paper. So are you willing to put a number to human contributions, or not?

Your suggested range is completely meaningless, then. It is an arbitrary range based upon your beliefs. You keep posing the question to me and I keep responding that I will not put a range on it - I do not claim that we are absolutely not contributing to global warming, but nor is there sufficient evidence (to my mind) that we are. Therefore it is totally meaningless, not to mention pointless, for me to proffer a randomly-selected figure from out of the air. Our contribution may be 50%, or our contribution may be 0% (it could even be that, in some way we haven't even considered yet, we are preventing GW from happening at a "natural" rate, and that our contribution could be -50%). The point is, without being able to accurately define how certain elements of the climate system contribute to the overall climate, percentage attributions of "blame" are completely without scientific foundation and, as such, without merit.

You have said you are willing to accept that humans have contributed to recent temperature increases. If this is the case, we must also be able to contribute to future slowing of temperature increase. Given the possibility that failing to do so would have a negative impact on human (and all life) on Earth, surely it then follows that we should make the effort to do this.

I shall say one last time, in despair, that it would be irrational to do something that could have serious repurcussions for us and the planet. Here's a wild scenario for you: we follow Kyoto and its subsequent treaties for the next 100 years. As a result, we cripple ourselves economically and the human race starts to go into a decline. Before you know it, and much to everyone's surprise and horror, we end up, both economically and technologically, in the dark ages. The human race continues to survive and eventually goes through its second Industrial Revolution until, finally, we are back at square one. What was the point? Surely it is better to stabilise the economy, channel funding into cleaner fuel sources and sort ourselves out first, our CO2 emissions dropping all the time as we develop, switch to, and use cleaner and cleaner forms of fuel?

Okay, so this scenario is a hideous dystopia that goes nowhere compared to an unlikely utopia that brings us back to the Garden of Eden - yes, it is ridiculously exaggerated - but this extreme example shows a fundamental difference in approach. Why is it that the basic premise of "no regrets" is somehow "bad for the environment"? The whole point is that this approach eradicates the need to cut CO2 emissions because CO2 emissions fall naturally as a result of the approach. If CO2 emissions really do have a significant effect on global temperatures then we've done a good job. If not then at least we have cleaner (and more efficient) forms of fuel. With Kyoto, if CO2 emissions really do have a significant effect on global temperatures then we have saved the planet but royally screwed ourselves up by knobbling ourselves economically and stunting our development. If CO2 emissions don't have an effect on global temperatures then we have achieved nothing but knobbling ourselves. Great. It seems that it is all a matter of perspective. Forcing ourselves to cut down on emissions is not as good an idea as finding other sources that come with reduced emissions. I cannot understand why people can't see this.

Anyway, that was a ridiculously long post by me, so I'll leave it for now, let you mull it over and see what you have to say.

Later!

C-Bob

[parmenides3]

Two points about the climate models. There is currently a hot debate about whether climate modelling is an initial value problem or a boundary problem. My instincts would tend to go - as yours seem to do - with the initial value approach, but I'll be honest here, I don't know enough to have a strong opinion on this. This is important because if it is an initial value problem, then we face the difficulties you have pointed out; that some degree of uncertainty is going to be present in any model, and, if it is an uncertainty over the initial values of a forcing or feedback, it has a huge knock-on effect over time.

As I understand it, the models use principle component analysis to eliminate, or fix, variables which are known to have a limited or non-existent effect on the outcome of model runs, and concentrate on the variables which have the most uncertainty attached to them. Through a process of thousands of runs, both backwards and forwards in time, the modellers reach a best-fit set of formulae/values, which are tested to destruction under a range of scenarios. Ensemble runs are often used at this stage to reach a 'most likely' outcome. The modellers will argue that, even with the uncertainties involved, this procedure provides a good overall [large-scale] picture of what has, is and will happen in the climate system. The more often model runs come up with the same patternsd or trends, the more confidence the modellers can have in the formulae or variables which they have used.

Trouble can set in at this point, when journalists, politicians or, in a few cases, science faculty PR departments, start to discuss the findings of the models in terms of prediction or certainty. All climate modelling is, almost by definition, about trends and probabilities, not about facts and certainties. Some interesting work has been done recently on using Bayesian methods to evaluate model output, and these seem to be bearing some fruit, You will never, however, be satisfied if you seek the kind of certainty you are implying in your posts; it is a function of the science that this is not possible.

The second point is that you have misrepresented what the scientists do with the models, probably unconsciously. They don't 'tweak' to make the models show 20th century warming, they 'tweak' the values of forcings to get the models to replicate historic measured data. The observation of warming in the 20th century is irrefutable; even the most hardened critic of climate science will acknowledge that the global temperature has increased. The point of the paper was to improve our understanding of which forcings have most impact in this change, either positively or negatively, not to 'prove' that the climate is in a warming phase.

Using the error bars/ variability spreads in graphs to show that the unreliability of data makes conclusions unreliable is to misrepresent the function of error bars/ variability spreads. The variability is a function of the degree of certainty, not of the likely or probable range of temperatures. So, whilst it is theoretically possible that historic temperature values are underestimated in almost all of the recent analyses, in real terms, it is highly unlikely. The values which are currently accepted as the most likely temperatures prior to the instrumental record have been subjected to huge quantities of analysis, counter-referencing and statistical/probabilistic evaluation. It is true that sometimes papers appear which cast new light on historic values; these are sometimes significant enough to cause an adjustment to the records and to the models. Some of these adjustments are positive, some negative. One example is a paper this year which concluded that the Medieval Warm Period may not have actually been as warm as has been assumed; it suggests lowering the values for the global historic temperature for that period, gives good reasons for so doing, and is scientifically sound. As and when the paper has been subjected to substantial scrutiny, its experimental basis replicated and its findings verified (which might take some years), it may well be that the 'Medieval Warm Period' will become a term no longer used. {I'm not advocating this; it's just an example]. But there is quite substantial agreement about the relative warmth of the current climate in comparison to the past.

You seem to contradict yourself on the question of attribution. In an earlier post, you said that what concerned you was the amount of warming attributable to human activity; in this last post, you seem to say that this is meaningless and irrelevant. In doing so, you are still characterising the state of knowledge about the climate and the skill of climate models in terms of extreme uncertainty and inaccuracy. This runs counter to what I have found in my own research. Without for a moment denying that there are issues, questions and problems, the state of our knowledge of the climate is good, if not extremely good. We do know what forcings and feedbacks effect the global temperature with a large degree of confidence. We have a good range of probabilities for the likeliest values for these affecting agents. If you wish me to accept that there are 'certain elements of the climate system', the effects of which we have a limited understanding, I will have to ask you to be more specific. which elements of the climate system are we talking about here?

I am not going to dispute the political issues with you. I will say, though, that your phrasing might let some readers think that I am, as a person who contends that AGW is real, therefore to be seen as some kind of ecological alarmist. I promise you, the two are not coterminous. AGW does not necessarily imply any specific attitude on the part of the person who advocates it. I very carefully avoided implying any acceptance of Kyoto or any similar agreements in my summary comment. I also have not suggested anything involving economic hardship. The question was framed in logical, not political terms; the conclusion is meant to follow from the premise. And I certainly wouldn't advocate adopting policies which might have the kind of repercussions you suggest. But, putting aside the issue of whether your assumption that adopting Kyoto [or any other agreement] would cause financial hardship is tenable, what method do we have for 'stabilising the economy'? (is this the UK economy, or the global economy?). What cleaner fuel source are we going to use? How do we fund research except out of taxes?

Here's a counter-dystopia: following the refusal of the USA, China and India to reach any agreement about emissions, Kyoto is abandoned, as is any attempt to make any new agreement. Business continues as usual. Emissions continue to rise. As a result, the global temperature continues to go up and the climate becomes increasingly unstable and extreme. As a result, we cripple ourselves economically, climatically and physically. Before you know it, much to everyone's

surprise, there are a billion fewer people on the planet, not enough water or energy to meet global demand, a further billion climate refugees moving from the developing World into the developed World, and oceans and ecosystems acidified beyond viability. The human race enters a new spiral of conflicts and insurrections which result in some countries returning to the dark ages, whilst others use their wealth to develop exclusive technologies which allow them to dominate the increasingly scarce resources through occupation of land or possession of those resources... It goes on.

(And, by the way, why would making ourselves do without a new fridge every couple of years send us back to the dark ages? This is a non-sequitur.)

Once again, you seem to muddle climate concerns with ecological ones. 'No regrets' advocates deregulation, tax reduction for businesses and the use of free market forces to self-regulate. This is errant folly and has no impact whatsoever on what we emit or what happens to our climate, unless it is destructive. I think everyone can see that finding a source of energy which comes with fewer emissions would be wonderful. How long should we wait for someone to make the breakthrough discovery that makes this possible? The objection to 'no regrets' is that it has no effect, not that it is harmful to the environment (which it might also be).

Finally, back to the CO2 issue. CO2 at any level in the atmosphere has a greenhouse effect. Without CO2 in the atmosphere, Earth would be uninhabitable. More CO2 in the atmosphere increases the greenhouse effect, trapping more heat in the system. There is a delayed response in the system to the increased heat; it is effectively 'stored up', to be released at a later date.

Thousands of scientists around the world, over twenty or more years, have tried to understand the processes and implications of these facts, in the light of the other fact, that our burning of fossil fuels is adding more and more CO2 to the atmosphere and is likely to continue doing so for some years to come. Thousands of scientists are very confident that the global temperature is going to continue to rise for the foreseeable future. Some scientists disagree, some don't like the political element which now surrounds climate issues and climate science. Some think that the IPCC is flawed, some think that the language which is used to discuss climate change has infected the community too much. But most, by a large majority most, agree that the climate is warming, that CO2 is responsible for more of this than any other causal agent, and that the consequences of this will be damaging to the World's environment and the human population. Why do they think this? It is neither a mystical belief nor a necessary condition of gaining employment; it is a conclusion based on the principles of science, of reason, observation, analysis and deduction.

I started my personal 'quest for truth' with all the usual assumptions about the climate and global warming. Realising very quickly that my assumptions were poorly-founded and questionable, I did the research, read the papers, argued the points... For a while, I had real doubts that AGW was happening at all. But in the end, I couldn't avoid the recognition that the people best placed to inform me of the issues were the experts, scientists and climatologists who make a study of the climate their life's work. The simple truth is, they know and understand more than I do, better than I do. If they are confident of the facts outlined above, in addition to the study I did for myself, then who the hell am I to disagree with them?

I'm waffling again. Time for bed.

:)P

[Admiral_Bobski]

I think this is going to have to be my last post for a while - I write my posts, I read through them and they seem to make perfect sense and then others (not just you, P3!) come along and completely misinterpret what I am saying.

You will never...be satisfied if you seek the kind of certainty you are implying in your posts; it is a function of the science that this is not possible. The second point is that you have misrepresented what the scientists do with the models, probably unconsciously. They don't 'tweak' to make the models show 20th century warming, they 'tweak' the values of forcings to get the models to replicate historic measured data. The observation of warming in the 20th century is irrefutable; even the most hardened critic of climate science will acknowledge that the global temperature has increased. The point of the paper was to improve our understanding of which forcings have most impact in this change, either positively or negatively, not to 'prove' that the climate is in a warming phase.

I am not looking for "certainty" in the models at all! I am looking for a model that can be programmed, have historical data entered into it and come out with something roughly akin to current climate. You will argue that they have done this, but the very fact that they have tweaked the forcings shows a lack of understanding of what the forcings are. (I never said that they tweaked any data, which is what you imply that I said - I quite clearly said that they tweaked the models which is an entirely different kettle of fish).

I have also never denied that there has been some apparent warming in the 20th Century, and if you think that's my stance then you're not reading my posts properly. I never suggested that the paper was intended as a "proof of current warming" - I was highlighting the inaccuracies in the climate models. Due to the very nature of climate, even the smallest of inaccuracies can make a big difference over a long period of time (but that statement depends on whether you agree with the opinion that climate is inherently chaotic, which you do not).

Using the error bars/ variability spreads in graphs to show that the unreliability of data makes conclusions unreliable is to misrepresent the function of error bars/ variability spreads.

It absolutely is not!! The whole point of the "error bars" is to show the range of possible values that can be derived from the extrapolated data. The fact that the blue line on the graph takes, in effect, the average value does not deny the possibility that any one data point could be at either the top or the bottom of the error margin. There is even a finite probability that every one of those data points is wrong by the same degree, and that the line is actually in completely the wrong place. Hence it is entirely legitimate to redraw the line at the top of the error margin to emphasise the possibility that the data is not accurate.

You seem to contradict yourself on the question of attribution. In an earlier post, you said that what concerned you was the amount of warming attributable to human activity; in this last post, you seem to say that this is meaningless and irrelevant.

That is not what I said at all, and a gross misrepresentation of what I actually did say. I have said all along that I have never seen any conclusive evidence that suggests that makind is contributing to the warming of the planet. Conclusive evidence, that is. I have said all along that I am not denying that there is a possibility that we may be contributing to warming. But the distinct uncertainty in this latter statement should be sufficient to quell your desire for me to apply a percentage figure to human activity. I have never said that we absolutely are not contributing to warming, but I have not seen conclusive evidence to suggest that we are. Therefore to suggest a figure attributable to human activity is meaningless and irrelevant, since I have no solid information on which to base that figure. To pluck a number out of the air does nobody any favours. No contradiction there.

In doing so, you are still characterising the state of knowledge about the climate and the skill of climate models in terms of extreme uncertainty and inaccuracy.

Absolutely. Climate scientists know a great deal about their own particular aspect of climate science - very few claim to know it all, and I am suggesting that there are gaps in the understanding that, perhaps, they are not even aware of. Processes that haven't been considered yet, for example. Once again I shall say that the Earth's ecosystem is a highly complex beast, with many subtleties and surprises, and I cannot believe that we are able to model this system to any degree of accuracy with our current understanding of it.

I am not going to dispute the political issues with you. I will say, though, that your phrasing might let some readers think that I am, as a person who contends that AGW is real, therefore to be seen as some kind of ecological alarmist. I promise you, the two are not coterminous.

I'm sorry but I don't think I have ever suggested that you were an alarmist at all, P3 - what phrasing in particular were you unhappy with? I have referred to alarmists on occasion because sometimes a good way to make a point is to take an extreme example, but I don't recall ever having lumped you into the "alarmist" category, and in the past I have specifically excluded you from it.

what method do we have for 'stabilising the economy'? (is this the UK economy, or the global economy?). What cleaner fuel source are we going to use? How do we fund research except out of taxes?

Well, let's consider just the UK economy - how have we stabilised it in the past? I'm not suggesting any new technique, just sensible economic policy that doesn't involve throwing vast sums of money away on a kneejerk reaction to an unsatisfactorily defined global "problem". What cleaner fuel source? We have to develop them first, which is what I have said before. In fact the whole point was that by funding research into technological studies, the fuel sources would come as part and parcel of that process. Research is already funded out of taxes (and by, gasp, industry), but my point was that throwing money into treaties such as Kyoto will take money away from that kind of research or, if research funding is kept up, destabilise the economy. But let's face it - when times are hard scientific research is one of the first things to suffer. That's basically why the Apollo program was finally scrapped.

Your dystopia, by the way, is as valid as mine was - it depends on your point of view. The difference is that I gave a comparable utopia. (And, by the way, Kyoto implies more than just "doing without a fridge every couple of years".)

Once again, you seem to muddle climate concerns with ecological ones. 'No regrets' advocates deregulation, tax reduction for businesses and the use of free market forces to self-regulate. This is errant folly and has no impact whatsoever on what we emit or what happens to our climate, unless it is destructive. I think everyone can see that finding a source of energy which comes with fewer emissions would be wonderful. How long should we wait for someone to make the breakthrough discovery that makes this possible? The objection to 'no regrets' is that it has no effect, not that it is harmful to the environment (which it might also be).

I have not said that "No regrets" is without its flaws, but that it is a good foundation upon which to build a sensible policy. With the correct restrictions in place (for example, fining businesses who do not pursue alternative fuel sources, or demanding that a proportion of their research goes into finding cleaner alternatives) it could work very well. Since the best Kyoto has done for us so far is introduce the concept of "Carbon Trading", which has no benefits for the environment whatsoever in the long run, what use is it?

I'm not going to go through the CO2 issue again, because I really am just repeating myself now. I see no point in continuing this debate if you are consistently misinterpreting and misrepresenting my points. I have far better things to do with my spare time than go back over the same ground time and time again in the vain hope that, one day, one of my points will actually sink in.

I will continue to browse the boards, but I shall leave any serious debating until 2007.

B)

C-Bob

[parmenides3]

I am sorry if you feel that I am misinterpreting your posts; it is completely unintentional. I am really trying to get at what your doubts are, and trying to offer responses, as you seem to be doing with mine.

I am not looking for "certainty" in the models at all! I am looking for a model that can be programmed, have historical data entered into it and come out with something roughly akin to current climate. You will argue that they have done this, but the very fact that they have tweaked the forcings shows a lack of understanding of what the forcings are. (I never said that they tweaked any data, which is what you imply that I said - I quite clearly said that they tweaked the models which is an entirely different kettle of fish).

You are right, I am arguing that this is what is currently happening. The tweaking of the forcings is one of the main points of running the models. The only way we currently have of calculating the forcings at work is through modelling. The models dtart with the data, ad the known variables and those with a high degree of confidence attached to them, then add the other variables. If the model output does not match the observed data, then the forcing values are tweaked to see how the model run adjusts. This is done many times over, until the model run replicates the data as closely as possible. At this point, the various focings are tweaked once again, to see how much difference they make to the model output. When the process is done enough times, patterns emerge which indicate to the scientists what the limit range of carious forcings are, and what the best match values are. For some time now, the values for the principal forcings have been narrowing as more observations and more intercomparisons of different models is done, until we are now at the stage where the models can be set to hindcast without the data inputs, and a close match can be made with the historic data.

Your point about the inaccuracies, which resembles the point about the 'butterfly effect' - only applies if climate modelling is an initial value problem. As neither you nor I can be certain whether it is or not, and as there is still some dispute about this in climatology, I am happy to concede that this is a potential problem with the models, but not that it is a actual problem, if you will accept the distinction. On this matter, I choose to trust the experts.

I did not intend to demand once again a number, as you have made clear that you think this is pointless (to some extent, I agree with you; it was meant as a semantic exercise). I also didn't mean to imply that you deny any human influence, or any warming; my response was based on the language of your comments, which did seem to contain these implications. I am at a loss, though, as to what might count as conclusive evidence, if the material I have posted still fails to meet your criteria. You asked for experimental verification of the effect of CO2 on temperature, and I provided that; you asked for an illustration of the correlation between CO2 and temperature historically, and I provided that. That the CO2 in the atmosphere comes from human fossil-fuel burning has been verified by physical analysis of the atmosphere; the carbon burned as fossil fuel has a specific composition which is distinguishable from other carbon from other sources. The paper I posted elsewhere, by Karoly and Stott, concludes that the CET changes are linked to human activity. All the other papers analysing climate trends which I have read reach the same conclusions, with more or less force (allowing for the peculiar exceptions such as Baliunas and Soon, or Svensmark, which have been discussed before). So what is missing? Why is this not sufficient evidence? What makes it inconclusive?

Climate scientists know a great deal about their own particular aspect of climate science - very few claim to know it all, and I am suggesting that there are gaps in the understanding that, perhaps, they are not even aware of. Processes that haven't been considered yet, for example. Once again I shall say that the Earth's ecosystem is a highly complex beast, with many subtleties and surprises, and I cannot believe that we are able to model this system to any degree of accuracy with our current understanding of it.

I am wondering if this is the core of your uncertainty (but not saying that it is, only that it might be). If you are unable to believe that we are able to model the Earth's climate system with any degree of accuracy, then nothing I can say will sway that belief. If this is the bottom-line of your doubt about climate science (and it is not an unreasonable doubt, by any means), then you will reject any argument I can place before you, as it will rely on the opposite assumption, that we can model the climate with sufficient accuracy to make meaningful statements about the future climate. This probably represents a genuine impasse which will limit the extent to which we can make progress on this issue.

Please don't think that your points don't sink in; there is much of importance in your doubts and the reasons for them, but, given the impasse above, you, too, will find it hard to convince me that I should revise my own point of view; we simply don't agree.

I am going to skip the political discussion entirely; this strand isn't really about what we should do any more (if it ever was). Originally, I suppose it was about whether or not Monckton's articles represented a challenge to those who claim AGW. I would say not, with supporting evidence from those scientists who have made the effort to respond online, showing the errors in Monckton's articles. This does not mean that the AGW issue is cut and dried, however. It also doesn't mean that there can be no alternative explanation for the recent observed changes in climate. But, in common with the other counter-arguments I have read elsewhere, Monckton's science fails because it is bad science - faulty - whereas, whilst the science of AGW is not perfect, it is rigorous and it has been tested and found to be sound; it is also improving constantly.

Thank you for your interesting and challenging posts. I hope you have an excellent Christmas break, and look forward to another opportunity to consider the issues surrounding climate change in the new year.

Best wishes,

:)P

[Admiral_Bobski]

Thanks for the reply P3 - apologies if my post seemed overly grouchy (woke up with a seriously bad back and not feeling on top of the world!). I appreciate the time you took to reply. I think you are right that we seem to have reached an impasse. Perhaps with a couple of weeks' break I can come up with an alternative mode of "attack" - at the end of the day the models are important elements in the debate but not the be-all and end-all of it, so perhaps there is another way to proceed. Something to mull over with the mulled wine!

I wish you (and everyone on the boards) a peaceful Christmas and a Happy New Year, and hope to return reinvigorated in 2007.

Signing off for now,

C-Bob

:unsure:

[parmenides3]

Today's browsing has taken me to the CCSP (Climate Change Science Program) publication: 'Temperature trends in the lower atmosphere...'

The CCSP was asked by the Bush administration (not AGW fans) to clarify the state of knowledge of climate change, so as to better inform policy. The report was published in May of this year. It sums up a great deal of the most recent measurement, analysis and model work of the past few years. In its executive summary, it compares the modelled temperature change in the Tropics and globally, and looks at the lower stratosphere, upper and lower troposphere and surface temperatures, with the observed measurements of these zones.

The findings show how complex it is to model the climate, and brings out some important limitations fo the climate models, especially with respect to the tropics. It contains clear graphics correlating the models' output (49 different models) with actual measurements. These show that the range of actual measurements tend to be much narrower (unsurprisingly) than the range from the models, but within the spectrum of the models' output.

This is not a document 'championing' the 'AGW cause'; in fact, it seems to err on the side of caution. In terms of what it says about the state of climate modelling and the causes of recent warming, it should be of interest to those of you who have been following the arguments/discussions on this thread.

http://www.climatescience.gov/Library/sap/...-1/finalreport/ I recommend the 14 page executive summary as a decent starting point.

If you don't feel inclined to read it, the report reaches the conclusions (amongst others) that there has been a warming trend in global temperatures at the surface of 0.12C/decade since 1958, and 0.16C/decade since 1979. It also concludes that the troposphere has warmed, and the stratosphere has cooled. Though there are inconsistencies in the models, and some are clearly 'better' than others, it points out that the measurements are consistent with the model output.

The report also concludes that there has been a substantial human impact on these global temperature increases (though it doesn't put a number to this).

With reference to C-Bob's issue on 'tweaking', the report is full of examples of the kind of tweaking that is done, with reasoning why it has been done, for example, the adjustments made to the radiosonde data from the late '50s-'60s.

As far as I am able to ascertain, the report is about as neutral, factual and unbiased as it can be. It is not influenced by the interests of individual scientists, faculties or government institutions, nor by industrial interests. I hope this counts as sufficient evidence to agree that a) the global climate is warming, B) human activity contributes 'substantially' to this and c) the climate models have some skill at modelling real climate when measured against observed data [with the observation that, as per the recommendations of the report, they still need improving].

:)P

[parmenides3]

A quick supplement. C-Bob; many of the uncertainties you have appear to be addressed by this publication, by the Royal Society:

http://www.royalsoc.ac.uk/downloaddoc.asp?id=1630

I hope you and others find it useful.

:)P

[Admiral_Bobski]

Hi, I've moved back over to this thread since the CO2 debate is a bit off-topic over on the other one. So here's P3's last message (the part that is relevant to the CO2 debate):

I think you have still not taken into account the solar forcing which is thought to be the principle cause of deglaciation. If the graphs showed only the response of temperature to CO2, you would have a point, but they do not. For the reasons stated above, therefore, there is nothing wrong with the correlation.

We are not around a plateau on the Milankovitch cycles. According to this: http://en.wikipedia.org/wiki/Milankovitch_cycles we are in an intermediate/rising phase, though, as you rightly point out, this is unclear at this time. I am uncertain how the Milankovitch forcings are supposed to work in relation to deglaciation, but one imagines that there would have to be a period of increased insolation before the effects kicked in, followed by an increase in life and a decrease in albedo, producing a feedback into the system. It is fairly clear that the recent warming phase cannot be attributed to Milankovitch forcings; this is one of the alternative hypotheses to AGW which has been studied in detail.

So basically what you are saying is that the Vostok ice cores don't give an adequate representation of the actual CO2/Temperature correlation. If this is the case then this graph is utterly useless in the CO2/Temp argument and should not be brought up again in this respect, which is a shame as it is one of the few long-term historical records we have...

As for Milankovitch cycles - these cycles are "an alternative...to AGW that has been studied in detail"...and yet we still don't really understand them. Perhaps more research into this area is needed then? In the wikipedia article under the heading "The Future" there is the following quote:

An often-cited 1980 study by Imbrie and Imbrie determined that "Ignoring anthropogenic and other possible sources of variation acting at frequencies higher than one cycle per 19,000 years, this model predicts that the long-term cooling trend which began some 6,000 years ago will continue for the next 23,000 years."[7]More recent work by Berger and Loutre suggests that the current warm climate may last another 50,000 years.[8]

So the relative warmth of present day is expected to continue (with a general downwards trend) for between 23,000 and 50,000 years. During this time there will be plenty of natural variation, for example the short-lived upwards trend that caused the Mediaevel Warm Period, the short-lived downwards trend that caused the Little Ice Age, and the current upwards trend. I fail to see any reason to presume that the current upwards trend will continue ad infinitum.

Enough for now.

C-Bob

PS - P3, in your post above this line:

The report also concludes that there has been a substantial human impact on these global temperature increases (though it doesn't put a number to this).

...should really read as...

"The report also concludes that the models suggest a substantial human impact on these global temperature increases (though it doesn't put a number to this)."

"Factoring in" human activity seems to make the models "work" better.

Edited December 22, 2006 by Captain_Bobski