So, are you complaining about the fact that science evolves over time? And, as for the plot in the 1990 report, that was basically a schematic plot not claimed to be based on any precise set of data. Furthermore, the graph only goes through around the middle of the 20th century and thus misses the entire rise that has occurred between ~1970 and now, which amounts to another ~0.5 to 0.6 C (e.g., here or here). In fact, if you look at the NASA GISS data, you can see that if that graph was supposed to represent just the northern hemisphere, which is traditionally what the proxy records have focussed on…or just land temperatures, which again has been the focus of proxy records, then the additional rise since ~1970 has been more like 0.8 to 1 C.
Read the RealClimate link that I posted. I particularly recommend comparing the sort of differences you get for 8-year trend lines (shown in the one graph on that page) vs. those you get for 15-year trend lines (link near the bottom of that article). That gives you some idea of how long a time period you have to do the calculation over to get trend lines that seem relatively robust to small changes in the start and end times.
brazil:
Furthermore i think you need to lay out your “little ice age” argument (or should i say Warren Meyer’s argument) in more detail. It seems entirely reasonable that we might have been in a “little ice age” during 16-1700. But an “iceage” and “coming out of it” has no explanatory power if you cannot establish an underlying mechanism at work as f. e. solar activity or something else.
My understanding is that research is not showing anything like solar activity being low 400 years ago and high now which would nicely harmonize with your argument of coming out of an iceage. as far as i am able to understand it - in actual reality no other explainations than AGW does as good a job of explaining the recent rise in GMT.
It doesn’t have much meaning because it is a truism. Everyone knows, and nobody disputes, that you can’t know temperatures from hundreds of years ago with 100% certainty. Nothing very interesting follows from this fact.
Lol. Kindly show me the four times you asked. Please use quote feature. Thanks in advance.
Do you agree that there has been a considerable change between the 2001 graph and the 2007 graph?
As far as I know, yes. My pal Warren says as follows:
Too bad you can’t be bothered to actually read my posts before jumping into this thread.
Let’s suppose for the sake of argument that Mann’s tools work poorly and the Little Ice Age was significantly colder than his reconstruction shows. One aspect of his conclusion – that current temperatures are unprecedentedly warm compared to the last 400 years – would still be correct.
On the other hand, if Mann’s tools work well – and he nailed the temperature reconstruction for the last 400 years – that aspect of his conclusion would still be correct. So it’s not a matter of how good his tools are. It’s a matter of what is being measured.
I don’t know how to make it any clearer. If you don’t understand, then so be it.
And it seems silly to me that you demand that I go digging through my posts to look for something that you could easily find for yourself. If you had actually read my posts in good faith, it would be just as easy for you to find the post in question as for me.
Essentially, yes. For example, if it were shown that Mann’s graph significantly overstates the stability of global temperatures over the past 1000 years (before the 20th century), then it has been discredited.
The difference is that there have been several additional proxy temperature studies so we now have many more plots…and, while there are some differences between them (and, in particular, some show more variability of the temperature over that period than Mann et al show), they all have several features in common…and particularly, they all have a similar warmth for the Medieval Warm Period that is lower than the late 20th century warmth.
Of course, this has not actually been shown. However, even if it were, that would not necessarily “discredit” it completely. Mann’s work was the first to embark on this multiproxy technique and the graph that he produced came along with some reasonably-sized error bars. So, at the very least, if the other graphs showed temperatures that mostly remained within the error bars of Mann’s graph then it would not be discredited at all. Even if they didn’t, it seems a little strong to say the work is “discredited” unless you use the term to mean “not completely correct in all respects” which would mean that almost all scientific papers would probably qualify.
It is also worth noting that the IPCC in its 2001 report made two specific claims based on the work of Mann et al. (and others). One was in regard to the temperatures in the late 20th century being higher than any previous comparable-length period in the last 1000 years and the other being in regard to the amount of rise in temperatures during the 20th century being greater than during any previous century over that time. Both of these statements were made using the term “probably” which is IPCC-speak for having an estimate probability of between 67 and 90% of being correct. Neither of these statements has been shown to be incorrect…and, in fact, the other proxy studies have all tended to lend support to at least the first claim.
Actually, I think you have overstated the case as I understand it. In fact, there is some very good evidence for “solar activity,” in the sense of sunspots, being lower around the time of the Maunder minimum…and there is some evidence (although less certain) that this actually corresponded to having lower solar luminosity and could provide an explanation for some of the cooling during the “Little Ice Age”.
However, there is no evidence that the changes in solar activity can account for the warming seen in the latter part of the 20th century into our 21st century…and this is the period over which we have the most certain measurements of solar luminance.
If you were to read and consider the original argument in it’s entirety you would see that i was actually describing my understanding of the general state of collective knowledge we actually do have. It is not a truism that we can’t know with absolute certainty the temperatures of the last 1000 years.
The instrumental record might have gone back that far. We may yet discover a proxy that always correspond to actual temperature. So we may know, to the extent that we may know anything.
Anyway this was not the central part of the argument, but groundwork. That does not equal meaningless. It is very meaningful and what it means is that we have to make do with what we can say knowing it is better than nothing, while at the same time aknowledging the uncertainty of our statements.
Correction, three times. The point about you being dodging and unresponsive, seemingly more interested in playing games than having a meaningful conversation stands though:
Yes! Insofar as the 2001 graph is illustrating one particular scientific paper by one particular scientist while the 2007 one is an aggregated graph showing data from several different papers by several different scientists, they are not the same. Therefore there is “change” between one and the next. That is a level of change that is to be expected in any two graphs taken from any two reports on any subject imaginable.
Also, because of that fact “change” is actually a utterly and completely meaningless statement to use in this context. Unless you would care to elaborate on what “change” you are refering to and how it has meaning in this context?
I will make you a proposition: What this graph is showing we can’t say much about, since we do not know what research it is based on, who made it, and what it is supposed to show, in contrast to the two previous graphs.
I would therefore trust the Mann graph and the aggregated one over this one.
To bad you can’t be bothered to actually read mine. As i already stated that i read through this thread. Actually i have stated that somewhere in the vicinity of 2-10 times by now. Once should be enough actually.
My point is this. We have tools to try to measure HGMT. They are imperfect. It might so happen that there is no hockey stick but a snail, a beetle or a bowl. It might also be that Mann’s tools did not work poorly and mismeasured The Little Ice Age, but that in fact there never were a Little Ice Age. Wrap your head around that idea if you can.
Now i am suspecting that you are vaguely and crypticly alluding to some distinction between “tools” understood by you as accuracy of measurement of something and in this case i suspect you think they are measuring some specific proxy. Then you are cryptically alluding to the fact that they may be accurately measuring the proxy but not accurately measuring the HGMT.
But when i make my argument saying that our tools are imperfect, that our knowledge is uncertain, then that includes both cases, i.e. for some proxies we may have trouble measuring the proxy back in time and at the same time one particular proxy or all of them might differ in significant ways from the actual HGMT.
That’s what i (implicitly perhaps) meant ALL ALONG. And you know what? The IPCC acknowledges this and given that being the case they try to survey all of the relevant research being done, weigh in the uncertainties as well as the certainties and then quantify what can be said and with what certainty it can be said.
And my understanding of the concensus they have reached and the implications thereof follows:
Perhaps i expect you to know your own definition off the top of your head or be more easily able to point it out since you once wrote it. As i myself would. Forget about it i found it now.
I think we are actually getting to the interesting part. Now Manns graph is showing a measurement by proxy. I don’t believe that Mann ever have claimed it to show THE Historical Global Mean Temperature. Only a measurement of it by proxy and statistical methods, that we can reasonably believe corresponds to the actual HGMT in such a way that it is meaningful.
If we had an instrumental record for the last 1000 years we would never be interested in making such studies as Manns.
Now, to “discredit” Mann given the above could possibly mean:
-
Showing that Mann had somehow manipulated the data so that what he presented did not correspond to the actual proxy he measured.
-
Showing that there are better proxies for measuring HGMT and that they actually show something completely different.
The latter i personally would not see as discrediting Manns work only as us getting better tools of measurement and science progressing. The conclusions about HGMT would be different though.
The former would discredit Mann though in my view. Is this your argument then that you believe Mann has been shown to be a dishonest scientist, manipulating his data?
Or is it the latter - that we now have better proxies and that they show that there is a Small Ice Age in your view disproving a “Hockey Stick”. Neither of the two seem waranted by the current state of research as described by the IPCC.
Absolutely not. However, the “evolution” of IPCC reports is an independent reason to have some skepticism towards what the IPCC is stating today.
It’s not a matter of explanatory power. It’s a matter of cherry picking. The cherry picking argument applies regardless of the cause of the Little Ice Age. Or even if there was no cause at all.
Yes it is. If you can’t understand why, then so be it.
You claimed to have “asked” me 4 times. From your quote, I see you asking me ONCE. And I answered promptly.
Here’s a truism for you: 1 != 4.
Ok, so you agree with my point.
Sure. The 1991 graph shows a pronounced MWP and LIA and great variation in temperatures before the 20th century. The 2001 graph shows little or no MWP and LIA and very little variation in temperatures before the 20th century. The 2007 graph is in the middle.
I would call that “considerable” change.
For purposes of my point, it doesn’t really matter which graph you trust and which graph you don’t trust.
And your statement was clearly incorrect. Because if you had read my posts in good faith, you would have seen and easily retrieved my definition. Which you ultimately did.
As far as your question about what it means to “discredit” the Hockey Stick I will give your question some thought.
Does that mean yes or no?
Perhaps not completely, but it’s a big problem. One of the essential aspects of the hockey stick is the “handle,” i.e. the assertion (or implication) that historic temperatures have not varied a lot over the past 1000 years (until recent times).
I suppose that we need to agree what it means for a claim to be discredited. At a minimum, I would propose that if it turns out (1) any significant aspect of the claim is wrong; (2) any significant aspect of the claim is not known with reasonable confidence; then the claim has been discredited.
By this standard, I don’t think that almost all scientific papers have been discredited.
Whether the IPCC has been discredited is a question for another thread.
And by the way, I am still waiting for cites for your claim that the “skeptics” are unwiling to modify their views in light of the evidence.
Or are you giving up on that claim?
I’m also waiting for answers to the questions I asked in Post #102.
I’m not exactly talking about discrediting Mann - I’m also talking about discrediting the hockey stick. Which isn’t the same thing. Because people have argued that even if Mann’s methods were flawed, most (?) of his conclusions were correct.
What do I mean by discrediting the hockey stick? I would say that the hockey stick has been discredited if any of the significant claims that make up the hockey stick is wrong or cannot be stated with reasonable confidence.
Brazil, you know what? Let’s agree to disagree and leave it at that.
Cheers
/ Sherwood
Suit yourself.
Do you still want to discuss splicing? It was way back in post #213, but I understand if you’re done with the thread.
Of course I do.
Brazil is making a couple of fundamental assumptions here that don’t work. Mainly through didacticism.
First off, his understanding of the hockey stick image rests on the idea that temperatures were largely stable, and now aren’t. That’s actually backwards, which if he wasn’t a layman he’d have the background to understand without being told. The point isn’t that the hockey stick was stable- it’s that temperatures fluctuations were far more gradual and less rapid than in the latter half of the 20th century. This needs to be explicit.
Secondly, the didactic argument represented by the discussion of the paramedics- or, shortly, the debate on divergence and splicing. Brazil maintains that if a spike in temperature fluctuation occurs in the historical record at the point where proxies are no longer necessary due to the presence of direct measurement, this proves a causal link between the two that “debunks” or “disproves” the measurement spike. Again, if not a layman, this would not need to be explained. What a historical simultaneity does is imply that there may be a link, which is why multiple proxies are used. As long as the majority of proxies show the same general trend, the conclusion that is then scientifically valid is that the measured spike in temperature is due to outside factors. Since the rise of improved direct measurement historically also corresponds to a great increase in human industry, this then would appear to be the causal link. It is an disingenuous argument, however, to solely use the historic closeness of the two events- the spike in mean temperatures and the introduction of accurate measuring devices- to claim that there is causal relationship. At best, on its own, it shows that there may be a causal relationship.
Thirdly, dealing with proxies- if proxies do not show current temperature trends, then the logical conclusion is not- and again, a non-layman should know this- to assume the proxy is invalid, but to investigate precisely how and why the proxy is not corresponding to temperature trends. If it is an issue of sensitivity, one gains a correlative factor and may in fact be more precise in using that proxy to go back in time. If it is an issue of the proxy being unreliable due to a specific facet or change in its nature- such as the change in wood-density noted post 1960, then any core sample must be checked for the same change in wood-density, and only trusted where that change does not exist. The point is, this does not render these measurements wrong or unusable- that is falsely didactic.
As to the claim that a motivated layman can understand an applied science- yes, theoretically. However, the argument was made most excellently above by implication, and I’ll make it explicit now: the primary problem is one of language and context. In any discipline or branch of a discipline, language is often very specific and used in an implied context, and the layman better be damned certain and careful not to mistake lay definitions for the ones being used in context in scientific papers.
As regards the discussion of the 100 glasses- brazil, statistically you would still see general trends in fluctuation of the data. If there is a systemic bias, then that same set of perspectives may be used on multiple drops, because intra-comparisons of fluctuations in the results should largely, if not entirely, remove any systemic bias from consideration of the system’s trends. Systemic bias may also be corrected for by the introduction of control tests to calibrate existing data.
These are all general critiques I felt of relevence to the thread.
I’m not sure why the distinction is that important. If I described the hockey stick using the word “stable,” feel free to substitute in the description “temperature fluctuations were far more gradual and less rapid.” I believe that all my points still stand.
But anyway, can you show me where I described the hockey stick as “stable”?
Again, can you show me where I made this claim? And please show me where I used the words “debunk” and “disprove” Please use the quote feature. Thank you.
I think you’re presenting a bit of a false dichotomy here. I don’t think those two options are mutually exclusive. The conclusion about the validity (or accuracy) of the proxy will necessarily be informed by the results of the investigation you describe.
If a proxy measurement diverges from the instrumental record, it raises a serious question about the accuracy of that proxy. If there is a really good answer to that question, it may be possible to rescue the proxy. So far, I haven’t seen any really good answers.
If you believe that I have misunderstood somebody’s use of language, please quote me where I do so and provide the correct definition for the word or phrase I have misunderstood. Thank you in advance for helping to educate me!!
I’m not exactly sure what you are trying to say here. But in the hypothetical I described with the 100 pieces of glass, there is absolutely no way to make an accurate measurement – not without some sort of additional test.
If you disagree, please describe – specifically – how you would do the experiment so as to “remove any [systematic] bias”
If I understand you correctly, that’s absolutely true. But irrelevant to the point that was being discussed. Sage Rat was basically arguing that you can measure something accurately by taking a large number of (possibly inaccurate) measurements and averaging them.
By the way, since you apparently feel you have superior understanding compared to a layman such as myself, would you mind describing your credentials? Thanks!!
Specifically, post 251, with repeated implications in the thread. You use ininuation as a form of argumentation quite frequently, and that is not honest debate.
I was summarizing your argument. It is extremely disingenuous to claim “well I never used the words ‘debunk’ or ‘disprove,’ ergo I did not say as much” when you have repeatedly asserted that the spike in temperature measurement occuring at the same time as the introduction of direct measurement devices shows that splicing the two data sets is false- this is not the case, it is only the case if such a causal link is proven (ie, you did not include the possibility that the introduction of modern measurement devices may independently correspond to a spike in temperature due to the simultaneous expansion of industrialization).
Because you have continuously focused on any insufficiency on the proxies as undermining the confidence of the data set. You’re also overly focused on semantic argument.
If the proxy measurement diverges, it raises the question, singular, of how well the proxy corresponds to the temperature record (how sensitive is the proxy?). As long as the proxy demonstrates the same temperature patterns as the measured data set, the data set extrapolated from the proxy is still completely valid in terms of trend measurement.
It is very difficult to not see this as deliberately disingenuous. The primary issue I’d bring up is the old debater’s trick of trying to force a yes or no answer to a false dichotomy, and then using a “denial to answer” as a form of argument.
“Would you kill 10 babies to cure AIDs?”
If they answer yes, they’re baby-killers.
If they answer no, they can be accused of not wanting to cure AIDs, or asked what bigotry they hold against people with AIDs.
If they don’t answer such a patently falsely framed question, they are attacked for being unwilling to take a stand, and then implicitly accused of having conceded the point the question was allegedly introduced to resolve.
There are times when there are only two options. However, rigid didacticism does not allow for increased or expanded understanding of the issue.
The hypothetical that Sage Rate described there is no way to make an absolute measurement. Data trends and relative measurements are completely possible. If you alledge to understand math, science, and statistics, this should be self-evident. Define a specific drop as the standard and use the same setup consistently for all other measurements. Systemic bias, unless there is a shift in the presence of the systemic bias (instrument drift) is thereby eliminated from any measurement or discussion of trends and changes in the data. Roughly speaking. One can always dream up scenarios where this wouldn’t hold in some specific way, but having read the thread, this is the issue that Sage Rat seemed to be trying to introduce. Honestly, many measurements we consider absolute have this kind of arbitrary reference assigned, and used as a comparison. Take the history of the meter as an example.
What that means is that while you can’t tell where the ball is without resorting to outside measurement, you can tell where the ball moves and with a fair degree of accuracy, increasing as you add multiple tests and determine internal relative bias of the testers, determine fluctuations in the rate of change of the ball.
Actually, in this case I would, simply because you are so insistent that you value form over function, and what are credentials but a form of function? I may change my mind later.
Sage Rat was arguing that can gain useful information by the various skewed measurements regarding changes in the data set. It’s simply not calibrated data. Are you familiar with the history of science? It’s this precise kind of fuzziness you find so wrong that led to the definition of absolute values for measurement today.
Whoops- reversed that. You insist you value function over form, and what are credentials but a function of form, that should have been