Defining Deviation Down–Xtreme Weather, part 2

Would we recognize Xtreme Weather if it walked up and greeted us? Is our weather special?

Kevin Trenberth is trying (and to a large extent succeeding) to convince the world that today’s weather is a product of climate change–that what we see around is the weather we have to a large extent created. In particular, extreme events are more frequent and more extreme because we have both warmed the planet and added moisture to the atmosphere.

Is this true? I intend to spend some time looking through published statistics and will produce the inevitable mind-numbing series of charts both supporting and opposing the premise.

But what really is the premise? Is a normal day today different from a normal day of 75 years ago? (I say 75 because the modern warming laid to greenhouse gases is variously said to have commenced around 1945.)

Is Trenberth saying that a storm with X energy today would have been a storm of X-y% 75 years ago? That if we have 16 hurricanes in the Atlantic this year we would have had 14 75 years ago?

Now Trenberth writes that “If the problem is generalized to look at the entire probability distribution function (pdf) of the climate variables, then the biggest changes percentagewise occur in the tails of the distribution, where they can easily exceed several hundred percent (Trenberth 2011b). Accordingly, a change in climate is most likely to be perceived by encountering new “weather” and breaking records: changes in the extremes. Changes in certain extremes, such as higher temperatures and increases in heavy rains and droughts are expected with climate change (IPCC 2007; Trenberth 2011a).

But to me he’s saying just that extremes are easier to notice–not that only extremes are expected.

I’d really like to know what it would take to either prove or disprove his thesis. Many people have already addressed his claims regarding weather extremes, and although I’m happy to bring them all into one place with whatever commentary I can provide, I’m not sure how much value that adds. Can someone please plausibly explain to me:

What changes should we expect to see in both normal and extreme weather from 0.8C of warming and an additional 4% of moisture in the atmosphere, which he cites as observed phenomena?

Otherwise we get garbage like a post on Joe Romm’s weblog saying that the U.S. faces a $188 billion price tag from Xtreme Weather from 2011 to 2012. (The trick is that all damages from extreme weather are laid at global warming’s doorstep–because we never had extreme weather before global warming, obviously.)

I’m willing to look at variation in the length of growing seasons, variations in migration of species, precipitation records, drought records by region and storm records where they exist. But what if we find nothing unusual?

When all storms are special–remembering Dash’s preternaturally wise observation above–none will be.

54 responses to “Defining Deviation Down–Xtreme Weather, part 2

  1. You can reprint IPCC SREX if you want, it isn’t going to stop this meme from being discussed by the serious people with serious expressions on their faces on all the serious channels and serious blogs. This is serious stuff, ya know. Just look out the window! There are effects! Happening now!

    It’s propaganda by the formal definition:

    Propaganda is a form of communication that is aimed towards influencing the attitude of a community toward some cause or position by presenting only one side of an argument. Propaganda is usually repeated and dispersed over a wide variety of media in order to create the chosen result in audience attitudes.

    There are many at the NOAA, and authors involved with the SREX that aren’t buying into the extreme weather NOW meme. The data is easily available to contradict these claims. Conflating events now with the “evidence” of predicted increase in events later by models is a favorite rhetorical trick used, Thus the “consistent with” obfuscation.

    We all know what happens when activists are asked to choose between being effective and being accurate. Sandy is the most egregious example of this.

  2. “Kevin Trenberth is trying (and to a large extent succeeding) to convince the world that today’s weather is a product of climate change…”

    Today’s weather is a product of today’s climate… by the very definitions of the terms. Ergo, if the climate has changed (which ‘lukewarmers’ ostensibly accept) then yes… today’s weather is unquestionably a product of climate change. What else would it be? A product of the climate of 75 years ago?

    The real question is how much have the climate and weather changed and what can we expect going forward. To answer that I’d suggest looking at the portion of the Earth where AGW driven climate change is most pronounced… the Arctic. The carbon dioxide greenhouse effect, water vapor feedback effect, and ice-albedo feedback effect are all stronger in the Arctic than anywhere else on Earth… and consequently the Arctic has warmed faster than any other part of the planet.

    So what effect has this had on Arctic weather? Consider the way consistently higher temperatures are melting the ‘permafrost’ (which is no longer ‘perma’) and the sea ice. These are radical changes to the environment due to the consistently warmer weather. Then there are unique events like the massive storm over the Arctic ocean last summer… it would have been virtually impossible for a storm like that to happen 75 years ago because it would have been taking place over solid pack ice rather than open water and thin fractured ice and thus getting no heat and moisture fed in from the ocean. So here we see how the ‘minor’ long term increase in temperatures has significant impacts in and of itself AND how it also changes the nature or distribution of ‘extreme events’.

    Again, that’s in the Arctic with the most pronounced climate changes on the planet… but the same factors are evident everywhere. For example, the USDA has had to update its ‘plant hardiness zones’ as earlier thaws, later frosts, and generally higher temperatures have shifted the areas over which different plants can grow. Similarly tree burrowing beetles which used to die in cold winters have been able to shift their ranges northwards to the detriment of forests around the world. Et cetera. Again we see how the minor but ongoing temperature shifts have had major impacts. Now, drawing the connection between these changes and ‘extreme weather’ events is not so clear cut as it is in the Arctic because the changes at lower latitudes are not yet so profound that they are allowing weather which was previously ‘impossible’ or nearly so. However, the evidence tying extreme events to AGW in these regions is growing even if we ignore the underlying logic which states that climate change MUST cause weather change.

    • My questions are, is normal weather now different to normal weather 75 years ago? Are extreme weather events more frequent, more extreme, both or neither?

      • “My questions are, is normal weather now different to normal weather 75 years ago?”

        Clearly yes. Otherwise the permafrost would not be melting, the plant hardiness zones would not be shifting, et cetera.

        “Are extreme weather events more frequent, more extreme, both or neither?”

        WHICH events? Extreme warm events are more frequent and more extreme. Extreme cold events are less frequent and less extreme. Meteor strikes aren’t impacted at all. Then there are lots of other kinds of events (e.g. hurricanes, flooding, droughts) where the statistical evidence currently ranges from ‘no clear picture’ to ‘likely change but not incontrovertible’.

      • Are you saying that permafrost has not melted in the past and that plant hardiness zones have not shifted in the past? That doesn’t match my recollection.
        I know that growing seasons and the precession of seasons has happened frequently in the past. I know that animals have shifted their territory frequently in response to local conditions in the past. (I used to like to read almanacs.)

        You say extreme warm events are more frequent and more extreme. Are you sure of that? What sources do you use?

        And again, you say that extreme cold events are less frequent and less extreme. Can you point me to your source? That’s very different than simply saying there are less cold temperature records…

      • Tom, take a look at Pielke’s post on Sheppard. ” Are extreme weather events more frequent, more extreme, both or neither?” Nada, Nada

      • Thanks–I will. Probably means no evening post–lots of reading to do.

    • The arctic is the part of the world (along with the antarctic) in which we have the shortest and least robust records. We don’t even have a record of sea ice prior to 1980, except annecdotes about how much ice was piled up on what shore.

      ” if the climate has changed (which ‘lukewarmers’ ostensibly accept) then yes… today’s weather is unquestionably a product of climate change.”

      Lukewarmers don’t accept – at least not at face value – that the climate has changed. They accept that there has been a rise in global temperature. The question is, how does the rise in temperature affect climate? Even Trenberth will admit that temperature isn’t equal to climate.

      • Conrad Dunkerson

        “We don’t even have a record of sea ice prior to 1980, except annecdotes about how much ice was piled up on what shore.”

        We only have SATELLITE records back to 1980 (actually late 1979). However, we have fairly extensive records before that from submarines traveling under the ice and even further back from shipping records noting the edge beyond which the ice was impassable each year.

        “Lukewarmers don’t accept – at least not at face value – that the climate has changed. They accept that there has been a rise in global temperature.”

        These two sentences contradict each other. A “rise in global temperature” IS in and of itself a change in climate. Yes, temperature also has more complicated impacts on OTHER aspects of the climate, but even if it somehow magically had no side effects a global temperature change IS a global climate change.

        In any case, if you DO accept that there has been a rise in global temperatures is it not self-contradictory to then question whether “normal weather” has changed? Since when is temperature not part of the weather? Ditto if you truly understand and accept the carbon dioxide greenhouse effect… the content of CO2 in the atmosphere has increased, therefor the amount of radiation retained in the climate system has increased, therefor temperatures have increased, therefor “normal weather” now is different than “normal weather” before the change. How is this even a question?

      • Hi CBD, while I’m absorbing the information you linked to, here’s some of my own, scraped off of Roger Pielke Jr’s weblog. I understand he’s probably not your favorite source of information but as it’s directly relevant to the topic:

        In a nutshell here is the state of the science (here I focus on the US as Shepherd did):
        US floods have not increased over a century or longer (same globally).
        US hurricane landfall frequency or intensity have not increased (in US for over a century or longer).
        US intense hurricane landfalls are currently in the longest drought (7 years+) ever documented.
        US tornadoes, especially the strongest ones, have not increased since at least 1950.
        US drought has decreased since the middle of the past century.
        US East Cost Winter Storms show no trends (here also).
        Disaster losses normalized for societal changes show no residual trends (US, other regions or globally).
        Trends in the costs of disasters are not a proxy for trends in climate phenomena.
        Any presentation of the state of the science of extreme events and climate change that does not explicitly acknowledge the important conclusions above from the IPCC and other assessments is incomplete and potentially misleading.

        What did Shepherd say in his briefing?
        He said his spouse can see the impacts of recent extremes
        Weather is now on steroids, like baseball players
        Weather is like your mood, climate is like your personality
        He cites two scientists quoted in The Guardian saying that all weather is affected by climate change
        The impact of climate change can be seen in the price of Cheerios
        The recent Northeast blizzard is related to climate change
        Suggests that climate change is “loading the dice for extremes”

  3. The sophistry of ‘blaming’ the weather on AGW is that it allows faux scientists to assert that any event is due to the AGW they are profiting from so nicely.
    The reality- that the changes in climate are not distinguishable from the pre-AGW climate-is inconvenient to their schtick and must be avoided at all costs..

  4. “Are you saying that permafrost has not melted in the past and that plant hardiness zones have not shifted in the past?”

    I’m saying that, given that ‘climate’ is the ‘average’ or ‘normal’ weather of a region, ‘climate change’ inherently includes weather change. If the climate is changing then the weather HAS to be, because climate is just weather in aggregate. Claiming that you acknowledge climate change but question whether the weather is changing is like saying that you can see that a forest is on fire, but question whether any trees are burning.

    As to permafrost having thawed in the past… depends on how widely we are defining the term. There are areas which have always thawed and refrozen every few years… and then there are areas which had remained frozen for tens of thousands of years until recently. I was speaking of the (vast) area that used to be truly permanently frozen and now no longer is.

    Ditto the other stuff you throw out. Yes, plants change their ranges sometimes. That is different than ALL plant hardiness zones shifting northwards.

    For changes in extreme temperature events (hot and cold) see;

    http://www.pnas.org/content/early/2012/07/30/1205276109.full.pdf+html
    http://link.springer.com/article/10.1007%2Fs10584-012-0668-1
    http://onlinelibrary.wiley.com/doi/10.1029/2012GL052459/abstract
    https://www2.ucar.edu/atmosnews/news/1036/record-high-temperatures-far-outpace-record-lows-across-us

    Or, if you prefer, just this pretty graph;

    [video src="http://svs.gsfc.nasa.gov/vis/a000000/a003900/a003975/bell_final_comp.m4v" /]

    Essentially that shows a distribution of extreme hot (red) and cold (blue) events over time. The further from the zero line the hotter or colder the event. The baseline is fixed at the initial state where there was an even distribution of hot and cold events (because that distribution was used to DEFINE the baseline)… but then as time passes we see the formerly rare three standard deviations above normal temperature events becoming much more common and some events going up to a formerly unheard of five standard deviations above the baseline. Meanwhile extreme cold events can similarly be seen to be becoming less frequent and less cold.

    • Thanks for the links. I will read before responding.

    • Umm, CBD–I’m a little curious. Why did you offer a link to Hansen’s Perceptions of Climate Change when I’ve discussed it in two different posts and linked to it both times? Just curious.

      • Isn’t his/answer mis-leading? I think you are asking if the PDF of weather is changing as the climate warms. To say that acceptance of the latter requires acceptance of the former is not true. For the proposition to be true or of concern we need to know the point at which “extreme” is an issue. A shift in the mean within a distribution is not necessarily a problem.

      • Conrad Dunkerson

        “Why did you offer a link to Hansen’s Perceptions of Climate Change when I’ve discussed it in two different posts and linked to it both times?”

        Because you asked me to? Remember?

        You: “You say extreme warm events are more frequent and more extreme. Are you sure of that? What sources do you use?

        And again, you say that extreme cold events are less frequent and less extreme. Can you point me to your source?”

        That paper is one of the sources for both of those things.

      • Tom,
        AGW fanatics, like all fanatics, rely on circular reasoning implicitly.

    • What you state here is mostly true, but trivially true. In an environment with an increasing temperature trend, new records will have more highs than lows. This is obvious mathematically. Has the distribution truly widened? No.

      You are conflating an argument with whether the temperature has increased (it has) with whether extreme events have increased NOW. for the most part, there is very little evidence of this. If you want to take issue with the vaunted IPCC, then do so.

      Hansen’s report has been shown to be what it is, an activist’s report with carefully manipulated statistics for a political purpose. Comparing distributions the way he did was mathematically invalid, not to mention the areas and times chosen were carefully selected. He could have chosen to analyze global highs/lows with standard statistical measures, he did not. Ask yourself why?

      I don’t argue with the decision Hansen made to become an activist, but when he chose to do this, it taints his science. That’s the price to be paid. He knows his work will be carefully scrutinized, and yet he still puts out a weak case.

      You simply cannot manufacture strong science from weak evidence.

  5. In the mid-1970’s when I was in graduate school one of my friends did his thesis on pattern recognition forecasting. The idea was that if you could match the current weather (pressure) pattern with something in the past then you could forecast what would happen based on what happened. If you have the time and resources I think you could develop a pattern recognition system to look for an extreme event and then calculate the frequency of the event through time.

    There are data bases with pressure data that you could use for this kind of effort. For example: the European Centre for Medium-Range Weather Forecasts has a 40-year data base (1957 – 2002) http://www.ecmwf.int/products/data/archive/descriptions/e4/index.html, albeit if they call 1957 to 2002 forty years I wonder how usable it is and the Hadley Centre has a monthly historical gridded mean sea level pressure dataset covering 1850–2004 (http://journals.ametsoc.org/doi/pdf/10.1175/JCLI3937.1).

    I suspect if you used that data you would see other events like last summer’s polar storm have occurred in the past. Ultimately I have my doubts whether the extremes are caused by global warming hypothesis can ever be falsified enough so that it is not used by people and organizations with agendas.

    By the way Roger Pielke, Jr. has a relevant post on extremes today.

  6. Kevin Trenberth is trying (and to a large extent succeeding)

    His success is an artifact of all generations believing that they live in the worst of times. As we don’t have any concrete experience then our own generational experience we tend to believe the time we live in is the ‘worst of times’…which if course is nonsense.

  7. The primary energy behind weather comes from temperature differentials Since , according to current models, greenhouse gases cause warming at night and at the poles, they decrease the temperature gradient in both space and time. They.should, to some extent lower wind strength. Doesn’t anyone else remember the Russian paper 2 years ago on lowering of wind velocity in central Asia. There was an attempt to blame it on ghg’s before the author said that the likely cause was vegetative regrowth after collective farms were abandoned.
    Land cover changes generally increase temperature differentials and are the most likely culprit behind increases in thunderstorm severity and twisters.

  8. “I understand he’s [i.e. Roger Pielke Jr.] probably not your favorite source of information…”

    Not really. His doctorate in POLITICAL ‘science’ doesn’t really qualify him as an expert on ACTUAL science. That said, I find him often more reasonable than his father… who actually IS a highly qualified scientist.

    “In a nutshell here is the state of the science…”

    The state of the science… according to Pielke Jr., whose entire career has been about the politicization of science rather than actual research. BTW, I note that most of the links he provides to ‘support’ his points are to his own blog. A practice which you recently denounced. Double standard?

    That said, he limits his comments to the US, which has been relatively unscathed by the current climate save for recent years. However, even within that framework, none of what Pielke says contradicts my positions above… he doesn’t speak to temperatures at all. The things he does list are those that I described as falling into a range between ‘no clear picture’ to ‘likely change but not incontrovertible’. The fact is that there are studies which suggest many of those things ARE happening… and other research which disputes those findings. Pielke’s presentation is misleading in that he makes no mention of the fact that there is conflicting evidence.

    “He cites two scientists quoted in The Guardian saying that all weather is affected by climate change”

    This is part of a list of things Pielke indicates that Shepherd stated and at the bottom of the list he states, “this is horsemeat”. This goes beyond misleading into outright error. As I have explained above… all weather IS affected by climate change. That is simply and inescapably true because climate is weather in aggregate. The climate CANNOT change if the weather does not. This should be obvious.

    • CBD, I don’t think you’ve characterized Roger’s career very well at all. But to the extent that you have, what different words would you use to describe James Hansen?

      Many scientists focus on the U.S. because of the availability of data–didn’t Hansen? As for Pielke’s horsemeat comment, if you read below on his blog it’s easily explained by the Findus scandal in Europe. If lasagna represented as all-beef contains a little horsemeat, the fact that most of the lasagna is not horsemeat doesn’t help much.

      And I’m thinking about the Hansen paper. I’m trying to understand why there is an automatic assumption that looking for a frequentist distribution is so obvious to Hansen et al. It is not obvious to me. I’m trying to understand the correctness of his choice of a base period. It is not obvious to me. Once I get past that I’m sure I’ll see what he brings to the table. I’d really like to see the data that he used.

      • You can look here for the primary problem with Hansen’s math:

        http://tamino.wordpress.com/2012/07/21/increased-variability/

        Tamino is obviously not in the skeptic camp, so keep that in your “Open Mind”.

        Hansen is manufacturing this conclusion, probably not even intentionally. His op-ed’s this year at the NYT and WSJ based on this work are inexcusable though. At least they put it under “opinion”.

        But like Mann before him, it is not the math errors that are meaningful, this happens all the time, especially when new statistical methods are “invented”. It is how it is handled after the errors are pointed out. People are still running with these conclusions.

      • A reasonable comparison can be made between Hansen and Pielke SENIOR… but not Pielke Junior. All three are ‘activists’ but Hansen and senior have conducted extensive scientific research. Pielke Junior has not. He has a career in “science policy”, not science.

        “Many scientists focus on the U.S. because of the availability of data”

        And that’s fine. But you were citing Pielke’s statements about US events as supposed contradiction to my statements about global events.

        “I’d really like to see the data that he [i.e. Hansen] used.”

        What’s stopping you? It is linked in the paper;
        http://data.giss.nasa.gov/

        See the GISTEMP section.

      • CBDunkerson,
        “A reasonable comparison can be made between Hansen and Pielke SENIOR… but not Pielke Junior. All three are ‘activists’ but Hansen and senior have conducted extensive scientific research. Pielke Junior has not. He has a career in “science policy”, not science”.

        Is there something wrong with you. What do you mean that Pielke Jr. has no right to express his views? Just because he did not generate the raw data doesn’t mean he has no valid view of the data. Who the hell are you whereby you believe your opinion carries more weight than his? You sound like an arrogant fool.

      • CBD, that link goes to the GISS dataset home page. If that is all that Hansen et al offer as supporting data then I will unilaterally declare that his paper is insufficiently supported. That would be a true disgrace.

        Where is the data he used, the definitions of each level of ‘sigma’, in an archived dataset with the name of the paper on it?

        That’s as stupid as telling me the answers to my questions are in the IPCC AR4–or maybe in AR5…

        What happens if GISS adjusts, moves or deletes the data Hansen used?

        Please tell me Hansen is more rigorous than this.

    • Yes, we know, climate isn’t weather, and weather isn’t climate, unless you say so.

      Record cold snaps don’t count. Record hot streaks need to be written up and inferred to now happen 30x more often. 15 year global pauses are “short term fluctuations”, and last year’s record USA-48 temperature is an “extreme event” “consistent with” things to come.

      We get the picture.

      No trend is interpreted here as “no clear picture”. Actual no trend is *** no trend ***. No correlation. The picture is quite clear that there is no trend. It is not “lack of evidence”, it is a clear counter factual. The hypothesis of more extreme events NOW has been disproved. It is not murky.

      In honest science, the data matters.

      If the trends change later, well then the hypothesis can be revisited as the models suggest, but pretending the jury is still out, when it is not, is more propaganda then science. Our current and past tornado, hurricane, etc. counts are not going to change, we can do the more extreme events NOW assessment NOW. Obfuscating and directly contradicting the evidence for political purposes is unscientific.

      It isn’t that hard to examine the historical record. You don’t need a climate science degree. RPJ didn’t invent the data. He recently covers the US because of the statements *** others *** are making about US extreme weather. He regularly post global trends as well. Feel free to argue the data instead of attacking the source.

    • I guess it’s in the air. Judith Curry is running a thread on similar topics. This is from her post:

      In a twitter exchange (and also in his testimony), Shepherd refers to a paper by Trenberth that justifies his statements. The paper is titled Framing the Way to Relate Climate Extremes to Climate Change. Trenberth’s argument is basically: All weather events are affected by climate change because the environment in which they occur is warmer and moister than it used to be. I see two basic problems with Trenberth’s argument:

      There is no empirical evidence for a moister climate (see the paper by Von der Haar et al. [link]
      Warmer/moister does not necessarily imply more or more extreme weather events; weather events are controlled by atmospheric dynamics on scales from mesoscale to hemispheric.
      Shepherd’s and Trenberth’s arguments are not consistent with the recent IPCC Special Report on Extreme Events (SREX), which was discussed on this previous Climate Etc. thread. However, the U.S. National Assessment did highlight an increase in very heavy precipitation particularly in the northeast U.S. (this is also discussed in Wuebbles’ testimony); this is something that I have not looked at closely myself but I am now motivated to do so.

  9. Tom Sharf wrote: “You can look here for the primary problem with Hansen’s math:

    http://tamino.wordpress.com/2012/07/21/increased-variability/

    The writeup at that link questions the increased variability (i.e. ‘spread’) of temperature events shown in Hansen’s analysis… not the shift. That is, the early data showed a spread of +/- 3 sigma… for six total, but the late data ranged from -2.5 to 5… for 7.5 total. Tamino believes that the wide range in the late data is partially a product of using the early period as the baseline. He doesn’t question the shifting of the distribution (i.e. shifting away from cold events towards more and hotter heat events) at all. Thus, even if Tamino is correct, he wasn’t disputing the aspects of the Hansen paper which I am referring to.

    “But like Mann before him, it is not the math errors that are meaningful, this happens all the time, especially when new statistical methods are “invented”. It is how it is handled after the errors are pointed out. People are still running with these conclusions.”

    Of course people are still running with Mann’s conclusions… dozens of subsequent studies have proven that they were correct. The fact that he used a statistical analysis method which was poorly suited for showing non-recurring minor variations (but better suited for finding the ‘natural cycles’ of temperature change that Mann was actually looking for when he unexpectedly got the ‘hockey stick’ instead) had no impact on the long term results and the conclusions drawn from them. Similarly, the ‘flaw’ you think you have identified with the Hansen paper has nothing to do with the results under discussion. That temperatures have warmed, and thus decreased the incidence of extreme cold events while increasing the incidence of extreme warm events, is visible beyond question in the data. If Hansen et al are wrong about there ALSO being increased variability that’d be a good thing, but doesn’t change the other conclusions.

    • CBD:

      Is anyone arguing with you that temperatures have increased?

      If you think all Hansen was trying to show in his paper was that temperatures have increased in the last 50 years, then you are mistaken. He was making a case for wider distributions (i.e. more extreme extremes). This conclusion was invalid due to errant math he used.

      When people who are emotionally invested in an outcome invent new numerical methods, the result is often a mess. They invent new methods because the standard methods are not giving them the results they *** want ***. The media runs with these conclusions, and nobody ever reports the repudiation of said conclusion. It’s how the game is fixed.

      We don’t need to beat the dead horse of the Mann HS, it has been well flogged You can judge for yourself if the world is still running with the HS meme. In the end it is irrelevant where the climate has been, it matters where it is going. The murky view we get from tree rings isn’t very helpful in this aspect. The data is so noisy that you can pull a HS or just about any other graph shape you like by varying your statistical methods. Proxy science is the best we have, but it aint so great.

      • Conrad Dunkerson

        “Is anyone arguing with you that temperatures have increased?”

        Yes and no. They are arguing that weather has not changed while ostensibly accepting that temperatures have increased. The problem is that these two things are contradictory. Temperature IS weather.

        “He was making a case for wider distributions (i.e. more extreme extremes). This conclusion was invalid due to errant math he used.”

        And yet, the source you cited for this belief stated (in the link I provided) that Hansen’s conclusions about HOT extremes becoming more frequent and more extreme were CORRECT. If anything, he was arguing for fewer/less extreme COLD extremes than even Hansen’s results showed… which BTW would still be consistent with my point about more/hotter hot extremes and less could/fewer cold extremes which you are ostensibly disputing. Odd how you can draw the opposite conclusion of your supposed source and overlook that it has no impact on the argument you are trying to disprove.

        As to Mann and the hockey stick. Tom asked us not to discuss them, so I’ll just say that we’ll see how ‘dead’ the Hockey Stick is when the AR5 report comes out. :]

  10. Bob writes: “Is there something wrong with you. What do you mean that Pielke Jr. has no right to express his views?”

    Friend, you are clearly suffering from some kind of delusion… because I said nothing of the sort.

    • CBDunkerson, I repeat your comment,” “A reasonable comparison can be made between Hansen and Pielke SENIOR… but not Pielke Junior. All three are ‘activists’ but Hansen and senior have conducted extensive scientific research. Pielke Junior has not. He has a career in “science policy”, not science”.

      So, any resonable person would interpret this as, 1. Pielke Jr. is unreasonable, 2. Pielke Jr. has not conducted extensive research, and 3. Pielke Jr. is not a “true scientist”.

      • Item 2 there is an accurate representation of something I have said. Item 3 is a slanted extrapolation from something I have said. Item 1 has no basis in reality.

  11. Hi all. First, let’s work a little more on civility. It’s perfectly okay to have differing views on this subject. This is not 2 + 2 = 4. Or 5.

    Second, I think it in the best interests of this thread that we leave Michael Mann and discussion of the Hockey Stick out of the conversation. Too much of a lightning rod, in my view.

  12. “Where is the data he used, the definitions of each level of ‘sigma’, in an archived dataset with the name of the paper on it?”

    The base DATA he used for computing temperature variations IS the GISTEMP dataset. The methodology and calculations involved (what you seem to be asking for) are all explained in the paper itself and/or linked therein. The intermediate data results (i.e. values after calculations have been applied to the base temperature data) are included in one of those links here;

    http://www.columbia.edu/~mhs119/PerceptionsAndDice/

    • CBD, that is unsatisfactory. I have enough bones to pick with his Methods section already. The methodology and calculations are not explained in the paper itself. It’s equivalent to saying the answers I seek are in the library.

    • I would expect somewhere in a rigorous paper to see a tabular presentation of results:

      Definitions of weather events used: Drought metrics: Sigma 1 = value X on Palmer scale. Sigma 2 = Y and Sigma 3 = Z. Flood metrics: Sigma 1 = x acre feet of water in x period of time. Etc.

      Table X:

      Year No. Sigma 1 No. Sigma 2 No. Sigma 3
      1900
      1901
      1902

      • Tom, are we even talking about the same paper? Because the one I’m looking at has almost nothing to do with droughts or floods. It is almost entirely about temperature changes (i.e. the subject you asked me about).

        As to the explanation of the methodology in the paper being “unsatisfactory”…. sorry, but that’s more than a bit silly. This paper is, if anything, overly documented compared to most. If the documentation is lacking how have people been able to run their own analyses to confirm the conclusions on increasing high temperature extremes? For example;

        http://tamino.wordpress.com/2012/08/13/hansen-et-al-2012/

        I think the ‘problem’ may be that, like nearly ALL scientific research papers, this was intended for a scientific audience rather than the general public. It assumes a certain level of knowledge going in.

        For instance, the computation of standard deviations (i.e. ‘sigmas’) is not some esoteric new process which Hansen would explain to other experts, but rather something he would expect them all to know. You can read up on it here;

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

        Essentially, one standard deviation is the 95% confidence interval. So… if you take the temperatures for a period and plot their divergence from the mean you get a bell shaped curve. About two thirds are within one sigma/standard deviation, ~14% between one and two deviations on either side, et cetera. The further above and below average temperatures you go the less frequent the events are. Pure statistics. What Hansen has shown is that over time the distribution of events has shifted such that events which used to happen on less than 1% of the days recorded are now happening on 10%…and even warmer events which used to not happen at all are now happening ~1% of the time.

      • Hi CBD, Well, thanks. I’m an analyst. I am familiar with Gaussian distribution patterns–in fact I earlier mentioned that I am having trouble understanding why Hansen used a frequentist approach. I’m curious in fact as to when he adopted it–before or after he examined the data?

        And I write reports for a living. The customers for these reports include businesses, government offices, academia and non-governmental organizations. And they do include charts to visually represent data. But they also include tables. And they also include data.

        I have occasionally referred to outside data repositories such as Scirrus. When I do so I give the date of search, the search terms used, the number of results returned and the website, title of repository, and as much information as possible (not as much as I deem ‘necessary’) for people to do what I did.

        Can you in any way, shape or form deny that a table such as I describe above would help readers understand what he found? Because there is evidently a lack of understanding. And not just by me.

      • I should also add that I am unsatisfied with the explanation given for his choice of a base period.

      • BTW, those tabular results you asked about? They’re in the “unsatisfactory” link I provided previously. For example;

        http://www.columbia.edu/~mhs119/PerceptionsAndDice/FigP1A.txt

      • That’s data output of a calculation, not a table of results. That’s quibbling, of course, but I’m not going to reconstruct it given my time constraints. I still maintain that I do not see adequate explanation in the paper of what Hansen et al did, why they did it, when they did it and what the results were.

        And as it is clear from their explanation of methods that they looked at time periods before 1951, why not include the results? Surely James Hansen of all people would recognize the vulnerability of his selection of base periods to accusations of cherry picking.

      • Hansen writes that the distribution of temperature values about the mean is commonly assumed to be amenable to a Gaussian distribution. Is there discussion of this?

        More importantly, I’m looking for the actual numbers represented by the deviations from the mean what do they translate to in terms of temperatures or at the very least anomaly changes?

        And I’m starting another reading of this to see if I can understand the logic of his choice of time frame for the study and for his selection of a base period. Because so far I don’t get it.

      • Tom,

        You can dance around with it all you want. Hansen is obviously cherry picking the data to maximize the results. It’s what he does, he’s an activist. With Hansen, you have to assume this, he has lost the benefit of the doubt and the burden is on him to prove otherwise.

        One could pretend that he followed a traceable scientific procedure to avoid this conflict. He could have formulated his methods and choice of time periods before he examined or ran any data through it (similar to what was done with BEST). He could have presented other time periods, regions, numerical methods as a counter balance.

        But we all know this is extremely unlikely, or he would have stated it and documented it. What was very likely done was that he did a lot of data mining and found the largest variance he could.

        The cynics among us think he sees this criticism coming, does it anyway, and makes sure he gets all the media coverage possible before he is exposed. If you ask me, Hansen has become more of a liability than a benefit to his movement.

      • I think James Hansen is an honorable man and a hard working scientist who has contributed a lot to our understanding. I think he is now in an impossible position, only partly of his own making.

        I am not willing to abandon the idea of his making further contributions, but I really think that this paper won’t be counted among them.

        Classifying temperatures as merely deviations from a mean is fraught. I am unconvinced that a frequentist approach is best, despite his claim that Gaussian distribution is the common assumption.

        Deviation from a mean temperature in Samoa and 1, 2 and 3 sigma deviations may yield very large values. The same treatment of mean temperatures in Siberia may yield very small real temperature changes. What is the largest 3 sigma temperature deviation in real terms? Where did it occur? What is the smallest? Where did it occur? What proportion of 3 sigma events noted in his study are very small real temperature changes and what proportion are very large real temperature changes? It should be easy to plot that. Why didn’t he do so?

  13. Conrad Dunkerson

    Tom & Tom, frankly I tend to dismiss the kind of ‘he faked his data’ and ‘he did not provide enough information’ arguments you are making out of hand when the source is something like the Proceedings of the National Academy of Sciences. They’ve got a reputation as one of the most prestigious places to publish scientific research to maintain.

    Is it POSSIBLE that random people on internet blogs have discovered a massive problem that the PNAS reviewers overlooked? Possible, but not bloody likely… and if so… hey you said yourself, “It should be easy to plot that.” Great! You have all the same starting data! You could do it yourself and show all the things Hansen ‘left out’ and then we can see “Why didn’t he do so?”! Go on… disprove Hansen. It should be easy!

    Though… I did link to that post where Tamino actually did his own analysis and corroborated the more frequent and extreme warming / less frequent and extreme cooling findings. Oh well, no doubt he’s in on the conspiracy along with all those PNAS folks.

    “I am not willing to abandon the idea of his making further contributions, but I really think that this paper won’t be counted among them.”

    Too late. It already is. Which should also tell you something.

    • Hi CBD, as far as PNAS is concerned, well, Anderegg, Prall et al. ‘Nuff said.

      • Congratulations. I have no response to that level of self-referential ridiculousness.

      • There was a time when I had a similar level of respect for PNAS. But they trashed it by publishing an anti-science piece of baloney. See how trust can be thrown away? It’s almost a metaphor for a goodly portion of the climate debate. At the end of the day, I now do not automatically trust the output of PNAS.

        And I consider that a very sad event. I want to trust PNAS.

    • I didn’t say he faked the data. Hansen is too smart for this. I am saying he is presenting it in an “optimal” way for his particular viewpoint. Because he is an activist, then his papers fall into the same group of grey literature as a NGO IMO. It doesn’t make them false, it makes them suspect. He has an expressed agenda.

      I respect Hansen, he supports nuclear power, and he is way better than many scientists who are less than honest about their biases. He chose to come out of the closet about his activism, which is honorable. That doesn’t mean we should assume it doesn’t taint his views on the science.

      If there are easy to find trends in extreme events they would have been easily found already. The fact that Hansen is burying us with unlabeled graphs (sigma, very hot) with new numerical methods from specific times and regions, should make it clear that these conclusions are less than obvious.

      PNAS is pretty much pal review. Any member can publish and gets to select his own reviewers. Most peer review never checks the math anyway, so I’m not very accepting of this appeal to authority.

      “Too late. It already is. Which should also tell you something.”

      It does tell me something. Probably a lot different message than you got.

  14. Pingback: Knee in the Curve or Calm Before the Storm? | The Lukewarmer's Way

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s