Climate Dissent

By Keith Kloor | October 7, 2011 11:34 am

When Al Gore said last week that scientists now have “clear proof that climate change is directly responsible for the extreme and devastating floods, storms and droughts that displaced millions of people this year,” my heart sank.

That would be the heart of Myles Allen, an author of this recent paper. His sentiment is candidly expressed in this Guardian piece published today, which is headlined:

Al Gore is doing a disservice to science by overplaying the link between climate change and weather.

This is not the first time that Allen, a card-carrying member of the climate science fraternity, has pushed back on what he considers an overhyping of the climate/extreme weather link not just by Gore, but by some climate scientists as well. Then again,  John Nielsen-Gammon contends:

Leading climate scientists clearly have different opinions on relating global climate to current or future local events. I don’t think it’s a matter of overhyping, just an honest difference of opinion. On the other hand, I don’t think the public is generally aware of this difference of opinion.

It looks like Myles Allen is intent on raising that awareness. In doing so, I think he’s going to give climate advocates heartburn.

UPDATE: In the comment thread at his Guardian column, Allen takes issue with the subhead:

Apologies, I’ve been doing my day-job all day (as one does), so only just seen the secondary headline, which (as OPatrick picked up) is wrong. It should read (like the article) “To say that we are causing weather events that could not have occurred without human influence is just plain wrong.” As you probably know, contributors don’t get to write the headlines.

 

CATEGORIZED UNDER: climate change, climate science
  • NewYorkJ

    Allen: Given the paucity of reliable records and bias in climate models, it is quite impossible to say whether an observed event could have happened in a hypothetical pristine climate. Our research focuses on quantifying how risks have changed

    The first sentence is where the “difference of opinion” primarily is – attributing individual events.  Allen’s research deals with quantifying the risks of certain types of events occurring, a very different proposition.

    The precise magnitude of the anthropogenic contribution remains uncertain, but in nine out of ten cases our model results indicate that twentieth-century anthropogenic greenhouse gas emissions increased the risk of floods occurring in England and Wales in autumn 2000 by more than 20%, and in two out of three cases by more than 90%.

  • NewYorkJ

    RC’s post on 2 recent papers on the topic, including Allen’s:

    http://www.realclimate.org/index.php/archives/2011/02/going-to-extremes/

     

  • Harry

    Myles Allen called him on this, noting that as climate change takes effect, statistically one would expect that about half of all extreme events should become worse and that the other half should become better. In other words, when an extreme event happens, it could just as easily have been made less severe by climate change than more severe.
    Trenberth responded with the example of record highs and lows. The United States climate has now warmed enough that 2/3 of all temperature records being set are record warm temperatures. So at least in that case, climate change is already favoring one kind of extreme event.
    http://blog.chron.com/climateabyss/2011/01/ams-2011-extremes/
    I think Trenberth is more correct here than Allen, although both points are valid and have to be considered.  I would say Gore is taking the Trenberth rather Allen point of view.
    People do think that there is no debate among scientists, when of course there is.  I do think the more conservative Allen point of view is not going to help clarify the issue of, what should we be doing now.  By doing nothing now, because we don’t have all the answers, means that by the time we do have them, it will be too late anyway.


  • kdk33

    Of course, there is no need for climate models when considering this question.  The planet has cooled and warmed and sat static for various periods since about 1940.  Reasonable meteorological data should be availabe over that time period. 

    It would be childs play to plot the extreme weather event of your choice as a function of temperature – floods, droughts, storms, whatever. 

    If any of the “people who want to do something about climate” (I hope that isn’t offensive), could point me to where that has been done, I would be grateful.

    ps.  I would find a plot of temperautre as a function of temperature un-
    interesting.

  • Fred

    Allen’s line of research is apparently computer models that account for past climactic events.  The accuracy of climate models in general is dubious.  See:

    http://www.c3headlines.com/climate-models/

    I wonder if Myles Allen has ever constructed a climate model of any demonstrated predictive accuracy?  If not, then Dr. Allen may suffer from a milder version of the obviously out-of-touch quality displayed by Mr. Gore.

    Speaking of an out-of-touch quality, Dr. Allen, in writing about a fellow researcher’s study who used Allen’s data, has the line “since springtime floods in the UK tend to result from melting snow, and thanks to greenhouse warming there is less snow around.”

    Maybe he should have thought about last winter in the UK.  It was:
    “…a weather event which brought heavy snowfalls, record low temperatures, travel chaos and school disruption to Great Britain and Ireland. It was referred to as The Big Freeze by national media and it was the coldest winter in Britain for 31 years with an average temperature of 1.51C.”  This is from:

    http://en.wikipedia.org/wiki/Winter_of_2010%E2%80%932011_in_Great_Britain_and_Ireland

    Did Dr. Allen’s models account for the UK winter of 2010-10?
     

  • EdG

    Having Gore as a spokesman for AGW is like having Madoff defending Wall Street.

    And as this describes, Gore is just getting worse. Now he is getting as emphatically ‘certain’ as a religious fundamentalist – which is the opposite of real scientific thinking.

    Is there nobody on the AGW Team that can silence this ongoing PR disaster? That would surely help their cause.

    In the meantime, I must laugh at some comments here attacking the messenger – Allen – when the message is so obvious.

    And N-G’s suggestion that the ‘public’ is not aware of the overhyping is also amusing. Very amusing. That wolf crying is probably the main cause of the rapidly melting credibility of the AGW Team.

    On the bright side, it is great to see so many AGW proponents now  asking penetrating questions in the post-Climategate Era. So much better than the pre-Climategate ‘Dark Age.’

  • hunter

    No, Myles Allen is going to give alarmists heartburn.
    Dr. N-G, whom I greatly respect, is in my humble opinion granting too much good will to the alarmists.
    The public is well aware of what hype smells like. And the AGW community is permeated with hype.
    People know when self-serving fear mongers are acting like self-serving fear mongers, no matter how sciencey their spiel.

  • Keith Kloor

    EdG,
    Your acidic comments are more a reflection of you than the those you vilify. Your analogy is absurd. Gore hasn’t defrauded anyone. I happen to think he’s counterproductive to the goals of the climate movement, partially because he’s become such a lightning rod, partially because he “overplays” the climate links, as Allen says, but also because he’s got an outmoded view of the obstacles to policy and political progress.

    You also mirror Gore in all your repetitive talking points (every comment seems to mention “crying wolf” and in your overstating (Morano-style) of the supposed “melting credibility of the AGW team.”

    I guess if, like Morano, you say it enough, you can convince yourself of it.  

  • Keith Kloor

    Hunter (7),

    Conversely, is the public “well aware of what hype smells like” when climate skeptics demonize Al Gore, James Hansen, Michael Mann et al and dismiss a whole science as a hoax, etc, etc?

     

  • Tom Fuller

    Keith re Hunter, I think the answer is clearly yes. In poll after poll the public reaffirms willingness to address both broad environmental issues and climate change. They broadly agree with the consensus and express broad faith in science and scientists. 

    Clearly the Morano brigade has not succeeded in influencing the wider public. It is a networking tool for the converted. Morano’s (or his employer’s) big mistake was thinking it had wider appeal than it does, which explains its recent decline. 

  • Fred

    Keith:

    You don’t like it when “a whole science is dismissed as a hoax.”  But, is climate science really a science that is contributing to knowledge or is it just a make-work program for the technically-trained?

    I have to hand it to climate scientists.  They get paid for constructing computer models for past events and for events so far out in the future that there is no “out-of-sample” testing possible.  As my link above shows, when the models are used to predict contemporary events, they are wrong.

    Constructing models of difficult-to-predict phenomena that are run real-time on new data and have to be accurate are a whole different animal.  As someone who knows the difference between these two types of computer models, its hard to give any credence to climate modeling of the type Dr. Allen is doing as worthwhile “science.”  Neither Dr. Allen nor Al Gore are saying anything that adds to the store of human knowledge.

     

  • Keith Kloor

    Fred (11),

    It’s not that I don’t like it; I find the characterization absurd. You also have a very narrow idea of what constitutes climate science if you think it’s all about computer models.

    I really have no idea if you say things like this out of ignorance or a partisan desire to caricature a profession. 

  • kdk33

    Am I to understand that noone has bothered to plot real observed meteorological data versus temperature and all of this “extreme weather getting worse” is from computer model?

    Really?

  • Fred

    Keith:

    I have no interest in denigrating the “profession” of climate scientist.

    Someday, in my dotage, I might like to shift fields into climate science.  There, I can apparently get paid to construct non-value adding computer models that “account for” past events, that are wrong, or predict conditions too far into the future to be tested. 

  • BobN

    I think Myles Allen was right to chide Al Gore for so emphatically stating the link between anthropogenic climate change and recent extreme events, particularly since to Gore anthropogenic climate change is really only GHG-induced climate change.

    Is it possible that anthropogenic forcings (GHG, aerosols, land-use changes) had some influence on the magnitude of recent events? Absolutely! (and in some cases, that influence may have reduced the severity of the event rather than exacerbated it)  Problem is there is no “clear proof” that anthropogenic climate change is “directly responsible” for such events. 

    The other problem is that Gore’s language suggests that without anthropogenic climate change, we wouldn’t have had any of the recent droughts, floods, heat waves etc., which clearly is not the case.  So yes, Dr Allen is right to critcize Gore.

  • Fred

    kdk33 has a point when he questions whether there is any real data to support the idea that extreme weather events are getting worse or increasing in frequency.  Dr. Allen states that “…the impacts of climate change are overwhelmingly felt through changing risks of extreme weather.”

    The data says there has been no increased risk of extreme weather events.  See the graphs and text from a post showing that the number of strong to violent (F3-F5) tornadoes has decreased over the past several decades, hurricane frequency is down, and there has been no systematic change in the frequency nor magnitude of floods.  See:

    http://wattsupwiththat.com/2011/09/15/24-hours-of-climate-reality-gore-a-thon-hour-13/

    It looks like Dr. Allen is no more with it than is Al Gore. 

  • steven mosher

    In brand marketing when you are asked to pick a spokesmodel a whole series of disaster’s go through your mind. Anita Bryant and Orange Juice. OJ  and Hertz. Tiger woods and whatever crap he pushed. The other choice is to create a virtual spokesperson.. mr clean, col, sanders, etc. They never get caught in ethical lapses ( hehe, thats a fun idea for a TV show, col. sanders and Aunt Jamima shacking up! with Mr clean and betty crocker as their neighbors.. but  I digress).

    We know that spokespeople are paid endorsers. That doesnt bother us. And nobody stays away from hertz because of OJ.  There is a little period of pain and then you rebuild the brand.

    Why is the gore situation different?

    Discuss, can the AGW movement afford Gore as a spokesperson? what does his credibility have to do with the credibility of the science? can the movement survive his departure or is the brand forever tainted?

    Option B:  if you had to build a virtual spokesperson for AGW what would they look like and why?  get as specific as you can.

    rough drafts due next week. take your time 

  • Nullius in Verba

    Option B: He’d probably look like Klip O’Toole.

  • http://initforthegold.blogspot.com Michael Tobis

    13: not remotely.

    GCMs do a bad job with extreme events and a much better job with typical seasonal weather. Other classes of global model can’t even represent extreme events.

    Climate models are, so far, mute on this question.
     

  • kdk33

    MT,
    Very interesting.  Can you point me to the data I requested in #4?”

  • RickA

    Keith:

    What is a “climate advocate”?

    Is this a person who advocates for the climate?

    Seriously, I don’t know what this label means. 

  • steven mosher

    kdk33. what data are you looking for? 

  • kdk33

    @22,
    I would like to see where someone has plotted observed extreme weather variables against temperature.

    It seems to me that we’ve had, since about 1940, periods of cooling and warming so have a range of tempseratures.  We should also have reasonable meteorological data over that period.  The thesis here is that warming increasing extreme weather events.  I’d like to see this plot to know if there is at least correlation.

    If it hasn’t been done, I’ll gladly settle for data in some kind of format I can easily pull into Excel, and I’ll plot it myself.

  • Eric Adler

    Fred @5,
    The reasons for the inability of climate models to make exact predictions of future events are well understood.  They don’t have to do exact prediction to make them useful.  They are used to provide a statistical understanding of what is likely to happen in the future. This is certainly valuable as a forecasting tool to determine future trends.
     

  • steven mosher

    WRT 23.  Since 1940  is 70 years of data. looking at the record you dont have much variability in warm and cold, you clearly dont have the kind of increase expected in the next 90 years. By definition extreme events are rare, so you might be lucky to find a couple extreme events in 1940-2010, by that definition of extreme.  So you have the temperature, but you dont have a big enough variation in temp to say anything meaningful about the warming that is coming, maybe you’l find a trend, maybe not, before I even started down that route I’d question the power of such an approach. a better approach is to understand the governing dynamics of extreme events first.  or at least define what you mean by extreme.. 

  • Nullius in Verba

    #23,
    I don’t know of any data in a usable format. NOAA have some records on climate extremes, and there are various databases of gridded temperature.

    I’d suggest you have a think about exactly what question you’re asking, though. We do of course have a vast range of temperatures – sunny to cloudy, day to night, summer to winter, equator to poles. Temperatures can vary from about -90 C in the Antarctic winter to about +60 C in the North African desert. Presumably you want the temperature anomaly instead.
    What counts as extreme for one place is normal weather for another. If the Arctic get’s warmer, is its weather getting more or less extreme? And the day-to-day temperature at a particular location is far more variable than when averaged over a month, a gridcell, or over the globe. Do you count weather moving from one place to another? Hurricanes are well known to form when the sea-air temperature difference exceeds a particular value (usually when the sea temperature is more than 30 C) but the same doesn’t apply to the places the hurricane hits. Some correlations are obvious – droughts tend to be when it’s hot, blizzards tend to be when it’s cold – but that’s probably not as informative as one might hope. It’s a very messy problem.

    A simple thing to try might be to find some place with similar geography a bit closer to the equator, where the average temperature is on average a few degrees warmer than where you are. If the weather you got there was transplanted, would it count as “extreme”?

    However, I’d be very dubious about the results. Climate isn’t one-dimensional. It’s influenced by all sorts of factors besides temperature in a very non-linear way, and the effect of warming in one location might be completely different to its effect in another. The idea that you can push up one slider and get a single simple overall effect for any type of weather is a bit crazy. Don’t you think?

  • http://www.veteransfreedomfarm.org steven mosher

    Niv.
     agreed, the notion that one can look at the 20th century ( say 100 years of record) and somehow apply that to the problem seems wrong headed.
     

  • Dean

    The phrase ‘directly responsible’ in the article Keith linked to is not in quotes. We don’t know if Gore used that phrase or not. I’d like to see exactly what Gore said in context before deciding whether he overstated the case or not.

  • Dean

    Gore is quoted with this:

    “The environment in which all storms are formed has changed. It’s influence is now present according to the leading scientists in all storms, and they speak of relative causation.”

    I’m not sure what relative causation means, but this quote is far from what most here are assuming Gore said.

     

  • http://initforthegold.blogspot.com Michael Tobis

    Dammit Mosher is right again. I hate that.

    The trouble is that we don’t have much in the way of theory or simulation on the question either. We simply have hope that severe events will not get drastically more common than a smooth monotonic shift in climate would indicate, and the beginnings of evidence that the hope is not going to be fulfilled.

  • NewYorkJ

    Dean (#28),

    That was my thought too, but I haven’t been able to find audio or video of the speech that’s being covered to confirm anything.  I found a 5-minute “opening remarks” audio, but it doesn’t cover the issue. 

    Myles Allen appears to be making the mistake of presenting it as a direct quote from Gore before attacking it.

    I did find this clip of a recent presentation of Gore on the topic, indicating global warming leading to increases in extreme precipitation, droughts, and wildfire activity.  Brief, but pretty good (could have used more references, such as the Westerling study he presented on wildfires).  At any rate, the Guardian paraphrase wouldn’t be an accurate representation of anything from this presentation.

    http://www.youtube.com/watch?v=SY2sBLKU2x8&feature=player_embedded

  • Chuck Kaplan

    Steve Mosher,

    Just for the record Colonel Sanders was a real man,not a colonel though. As somone who worked in Louisville for KFC.

  • kdk33

    This gets more interesting all the time…

    So nobody has bothered to plot observed “extreme” data as a function of temperature.  Mosher and NiV speculate that this is a waste of time (I absolutely disagree, and consider this to be the minimum required first step).  

    The question, of course, isn’t how do I define extreme, it is how do “the people who want to do something about climate” define extreme.  I (we) are repeatedly told that “extreme” weather will increase.  (If anyone recalls, in an earlier thread I asked NYJ for the extreme weather variable of his choice – so I’m quite flexible in this regard)

    Yet, nobody has looked at how extreme events have changed during the warming since about 1975 (no that’s not a waste of time; it is the minimum requried first step).  MT says the climate models are mute on the question; MT later says we don’t have much theory on the question.

    Pray tell, on what basis does this claim about exteme weather rest.

  • kdk33

    The trouble is that we don’t have much in the way of theory or simulation [or observational data, apparently] on the question either. We simply have hope that severe events will not get drastically more common than a smooth monotonic shift in climate would indicate.

    I’m floored.  I don’t know if this is a joke or staggering imcompetnce.  Again, the claim that extreme weather will increase, juxtaposed to an admission that this is not supported by thoery, simulation, or observation.

    So, I call BS.

    ps.  is this the kind of challenge we can post on the new members only web site, or am I just confused.

  • kdk33

    The notion that one can look at the 20th century ( say 100 years of record) and somehow apply that to the problem seems wrong headed.

    Really.

    OK, Mosher, so then, according to you, there has been no significant warming.  None.  And the 100 years or so of realiable weather data we do have is insufficient to detect trends.

    See, you can’t have it both ways.  Either:

    It is warming enough to be plainly detectable and enough to attribute some (never a defined percentage, of course) of that warming to humans, and we understand enough from observation to make predictions about the consequences.  In which case it is perfectly reasonable to look at how that warming has affected weather.

    Or there hasn’t been any warming significant enough to affect weather, in which case the warming simply isn’t significant. 

    Which is it?

  • kdk33

    The idea that you can push up one slider and get a single simple overall effect for any type of weather is a bit crazy. Don’t you think?

    As a matter of fact, I agree with this, but not the way you are thinking about it.

  • NewYorkJ

    Many of us just ignore your stuff at this point, kdk, since you have a tendency to both distort and/or dismiss evidence presented to you.  Not worth the time.

    http://blogs.discovermagazine.com/2011/10/03/moranos-faux-indignation/#comment-79664

    You could even learn something from Al Gore.  His presentation in #31 indicates he’s ahead of you on this issue.

  • http://www.veteransfreedomfarm.org steven mosher

    Kdk. what you have said makes no sense. I award you no points.


    Really.
    OK, Mosher, so then, according to you, there has been no significant warming.  None.  And the 100 years or so of realiable weather data we do have is insufficient to detect trends.”
    #######
      wrong, you can detect trends in the temperature. Thats not the problem.

    It is warming enough to be plainly detectable and enough to attribute some (never a defined percentage, of course) of that warming to humans, and we understand enough from observation to make predictions about the consequences.  In which case it is perfectly reasonable to look at how that warming has affected weather.
    Or there hasn’t been any warming significant enough to affect weather, in which case the warming simply isn’t significant.
    Which is it?
    ####
       we are talking about “extreme” events. rare events. Lets take a simple example. 100 year floods. So, if you only have 100 years to work with you’ve got 50 years to establish a  frequency and the last 50 to detect the increase. But the last 50 years sees about 1/4 of the warming we expect in the next 100, and its the next 100 we are concerned about. Basically, with 100 years of data, you’ve got a problem cause estblishing tight CIs on your frequency of EXTREME events. Like I said, this depends upon what you mean by extreme. I mean rare.

  • Nullius in Verba

    Kdk33,
    I agree with you, I’m just saying the question is even harder for anyone to answer than simply plotting one against the other.
    What sort of extreme do we actually care about? If nice weather is very rare, and becomes more common, is that an increase in an “extreme” – “extremely nice” – and if so, is it bad news? To talk about extremes as dangerous and damaging weather one moment, and merely unusual weather the next is misleading.

    The climate where I am is fairly cold and wet. If you head a few degrees of latitude south, the temperature is warmer. We often go there on our holidays because it’s so nice. The people living there are a lot less bothered about weather than we are (and conversely, those to the north of us more often have to take precautions against bad weather) so on the most simplistic level I’d say that weather would get less “extreme” (in the sense of uncomfortable/inconvenient) as the climate warmed. Is that something I can extrapolate to the globe?

    Defining it as “rare” events gives a completely different answer. If extreme weather is defined as weather that only occurs once every hundred years, then obviously after the warming, extreme weather will still only occur once every hundred years. Modify the definition to mean weather that only occurs every hundred years on a particular date, and it becomes a question about what happens to the tails of the distribution as the temperature rises. Does the whole thing shift uniformly – in which case it depends on how thick each tail is – or does the spread, skew, or other property change? Warming by increasing the minimum and not the maximum might be expected to narrow the spread, giving fewer extremes at the cold end, for example.
    That sort of effect is likely to be local. In some places the minimum might move more, in others the maximum, in others the whole thing will shift, in yet others you’ll get more skew in the middle with no effect at the extremes. Some places will cool on average. Some places the tails will increase and in others they will decrease, and the only way to tell is to do regional climate forecasts for every part of the globe and add all the pluses and minuses up, which of course climate models are still unreliable for. They can’t even model the middle of the distribution at a regional level, let alone the precise shape and weight of the tails.
    It’s an impossible question to answer at the moment, and is a lot more complicated than simply plotting weather extremes against temperature. Climate isn’t one-dimensional.

    By the way, it’s very easy to count up 100-year events, because they happen all the time. If you stand in one place, then the probability of it experiencing a 100-year flood sometime this year is 1% (assuming the probability isn’t changing over time). If you take 100 places (far enough apart for the weather to be independent), and ask what the probability of any of them experiencing a 100-year flood this year is, you get 1-0.99^100 which is 63%. Take 500 places, and it’s more than 99% certain one of them will experience a 100-year flood this year.
    And of course, for a lot of places a 100-year flood would be nothing to worry about. “Rare” doesn’t mean the same thing as “bad”.

  • kdk33

    Well, perhaps we are making progress…

    First, I reiterate:  The question, of course, isn’t how do I define extreme, it is how do “the people who want to do something about climate” define extreme. 

    I think warmists use “extreme” to mean both rare (nowadays) and bad (otherwise, why worry!) but, as NiV hints, the same arguments apply to rare and good.  As I said above, I’m quite flexible on definition.  Shortly, I’ll explain why this isn’t actually very important.

    Now, to Steven Mosher. 

    I take it then that, according to you, observational data suggests no correlation between warming and extreme weather events; or, at a minimum, is mute. 

    Do you now agree with me that there is no basis for the claim that warming will increase extreme events?  If not, why?

    I understand your argument about extreme events being rare -hence justification for observation data being mute rather than refuting – but it is an incorrect one as I will explain.  At a minimum, you are claiming these events can’t be measured, which opens up a huge can of worms,

    Now to NiV

    Take 500 places, and it’s more than 99% certain one of them will experience a 100-year flood this year.

    Exactly.  This is reason #1 that Steven Mosher’s non-measurable claim is incorrect.  It is also, I think, where climate science goes wrong.  If you believe extreme events are cuased by warming, then you find avidence everywhere; but, if you can’t detect a trend, then you have no scientific basis for the claim.

    Does the whole thing shift uniformly ““ in which case it depends on how thick each tail is ““ or does the spread, skew, or other property change?

    It really doesn’t matter (though I’m not sure what “other property” you might invoke), as long as we agree that the distribution is continuous and doesn’t have multiple local maxima (and if we disagree, we have an entirely different argument on our hands, so let me know) then we can look at trends of less-rare events as indicators of increased likelihood of extreme, and the trends versus temperature would be the basis for forecasting – after all, it is this causal relationshipe that is claimed.

    To explain more.  If we define extreme as a 500 year event, and if the distribution is continuous, then an increase in 500 year events will be acompanied by an increease in 400 year events, and 300 year events, etc. etc.  So, the max, min, mean, mode, variance would all change, and these are measureable and trendable with temperature.

    A workable definition of extreme weather would not be that difficult to construct.  I’ve assumed wamists had a working definition in place; if not, how can they claim these (undefined?! and unmeasurable?!) events will increase.
    ===============================================

    In summary, we seem to all agree that:  there is no observational basis for the claim extreme weather will get worse with warming.  Climate models cannot adress the questions, and there is no good theory on why this should happeen. 

    So, do we all agree that the claim is anti-science?

  • Fred

    Eric @#24,

    Myles Allen created a retrospective model for a specific event, the UK flood of 2000.  He assigned a 1 in 10 chance that AGW increased the risk of that event by 20% or less.  Since there has been no overall increase in floods or other extreme weather events as CO2 has increased (see my link in #16) it seems unlikely that CO2 variations play any role in causing floods.  Furthermore, creating models that “account for” portions of past events with no ability to forecast seems to add no value.

    Climate models that have attempted to predict future conditions have had major problems with accuracy (see my first link in #5).  Models, like climate models, where all the drivers are not known nor accurately quantified, where you have only a small number of years with “good” data, and where there has been little, if any successful out-of-sample testing are worthless IMHO.  Judith Curry just recently stated that there is:

    “…deep uncertainty associated with our reliance on projections from climate models, which are loaded with uncertainties and do not adequately treat natural variability.  Further substantial areas of ignorance remain in our basic understanding of some of the relevant physical, chemical, and dynamical processes.”
    From:
    http://judithcurry.com/2011/10/08/usgcrp-draft-strategic-plan/#more-5207

    Given this context, it is perfectly reasonable to ask how Myles Allen’s modeling would explain the cold and snowy UK winter of 2010-11.  It appears that Dr. Allen is guilty to a lesser extent of the same sin of Al Gore of excessive claims (fantasies) of how extreme weather events are caused by increasing CO2 levels.


     

  • hunter

    KK,
    Your comment regarding my skepticism of the AGW movement misses the point, which is not unusual for members of the AGW community.
    You conflate your interpretation of climate science with science, and dismiss any criticism of your interpretation as an attack on climate science as a whole.
    That is clearly not the case with the vast majority of those who disagree with you,and particularly for me.
    You are in effect a defender of an orthodoxy and have decided that any disagreement with your view of the faith is a heresy and rejection of the entire faith.

  • Fred

    hunter, @42,

    Don’t be too hard on Keith.  Keep in mind he is just about the only warmist blogger who will allow those who do not believe in AGW to post their thoughts unhindered. 

  • NewYorkJ

    Fred (#41),
    You’re incorrect to assert that extreme events haven’t increased, but your source (wuwt) is a common source of confusion.
    http://www.skepticalscience.com/extreme-weather-global-warming.htm

    High precipitation days have also increased globally.

    http://www.knmi.nl/publications/fulltexts/global_extremes_for_jgr.pdf

    Attributed to human activities

    http://www.nature.com/nature/journal/v470/n7334/full/nature09763.html

    The false presumption you’re making is that rainfall is the only important factor in determining net impacts on flooding.  Higher temperatures can dry out soils in some areas and seasons, buffering the effects of the heavy precipitation increases during those times.

    http://www.skepticalscience.com/Does-more-extreme-rainfall-mean-more-flooding-Answer-Not-always.html

    Myles Allen also discussed a study that indicated reduced risk of flooding from trends towards lower mountain snowpack.  His study covered a particular type of flooding event in autumn – one where increased precipitation plays a large role.  So he’s not “guilty” of anything you’re accusing him of.  Of course, reduced snowpack and depleting glaciers in many places means depletion of water resources many depend on – not a good thing really.  Warming has also lead to increased wildfire activity in some areas, as Gore correctly noted.

    http://www.sciencemag.org/content/313/5789/940.full

    Regardless, since the increase in atmospheric greenhouse gases is primarily responsible for recent warming (CO2 being the top forcing), and the direct and indirect effects of warming have lead to the factors above, your claim above that CO2 has had no impact on these events is also false.

    Lastly, a season of local weather is not very good evidence of anything, especially snow events (remember high snowfall totals are caused in part by higher precipitation).  In your last paragraph you appear to be making the mistake Gore is being attributed to.

  • EdG

    #8. Keith says “EdG,

    “Your analogy is absurd. Gore hasn’t defrauded anyone…”

    What Madoff and Gore actually did is beside the point… which is their credibility. Madoff is a handy spectacular example of somebody who would definitely not help as a spokesman for a cause, and that also applies to Gore.

    Details aside, I see no essential difference in the two. Bottom line, look at how much money Gore has made flogging the AGW story. 

    “You also mirror Gore in all your repetitive talking points (every comment seems to mention “crying wolf” and in your overstating (Morano-style) of the supposed “melting credibility of the AGW team.””

    Yes. Well, “crying wolf” is a precise description and a parable I assume everyone knows. Seems far more accurate than your/Allen’s softball that “he “overplays” the climate links,”

    The boy in the Crying Wolf parable overplayed the threat of wolves.

    But, if it helps you to put people in simplistic little boxes and pretend that I must be a ‘Morano clone’ by all means do so. You can even imagine that I’m working for the Big Oil Denial Machine, drive a lumbering SUV to the mailbox, and strangle kittens to pass the time. Whatever.

    But I guess that means I can imagine you are a Gore clone too, right? If I was simple, I might.

  • Fred

    NewYorkJ, #44

    Neither hurricanes nor tornadoes nor floods have increased.  No increase in any of these extreme weather events has occurred.

    Yes, rainfall has increased.  Such a pattern may have nothing to do with CO2 levels.  Even solar influences have been implicated in changing rainfall.  See:

    http://bourabai.narod.ru/landscheidt/relationship.htm

    You talk about “recent warming.”  The pattern of temperatures over the last several decades more closely matches changes in solar activity than in CO2 levels.  See:

    http://www.sciencebits.com/NothingNewUnderTheSun

    You state “higher snowfall totals are caused by higher precipitation.”  But snowfall can only occur with cold temperatures.  If there was warming it would have rained, not snowed.  Anyway, last winter in the UK temperatures were significantly below normal.

    True, one season’s temperatures in one location are not great evidence of anything.  But around 2000 a prominent climate scientist in the UK went public with a widely circulated prediction that snow would be a thing of the past in the UK by 2010.  Its getting tiresome listening to warmists predicting climate changes that never come.   

  • hunter

    Fred,
    Thanks for the heads up. I, too, appreciate KK- he is willing to at least let each of the sides speak. The whole AGW movement is winding down. Unless leaders get really silly and try to prevent the pushback on the deeply flawed EPA decision on CO2 which violated their own policies and procedures. If someone seeks to completely circumvent the rules and science, and pretend the EPA is justified in its war on America, we will come out OK.
    Australia is chomping at the bit to toss out their CO2 obsessed government, Germany has already politely ignored the CO2 issue, China and India see AGW, justifiably, as a form imperialism, and the rest of the world is laughing at us, not with us.

     

  • Dean

    The US Climate Extremes Index is a good example of the complexity of detecting extreme events. In allowing you to construct a graphic, you can choose between 10 event type measurements, 10 different geographical bases, and 9 different time periods during the year. I suppose that is 900 options. And still you never know for sure if the choice you make might result in different extremes canceling each other out. Unlike other kinds of statistics where sample size is almost always an improvement in statistics, difference in geography and season can result in canceling out, or at least hiding valid results in the noise.

    None of us can really know what exact metric is most appropriate for detecting such changes. That is what scientists and statisticians will do. Try lot’s of options and get the response from the scientific fraternity. Over time, certain metric will become more widely used.

  • Fred

    hunter,

    Your analysis of the worldwide status of the AGW movement is insightful and appreciated. 

  • Tom C

    So many words spilled on this one without the simple fact being acknowledged:

    What Gore said is a lie.

    “…directly responsible…” No scientist has said such a thing.  Even the execrable Trenberth.

  • NewYorkJ

    Fred (#46),

    Floods have increased in many areas certain seasons, generally when the heavy precipitation events are primary drivers.

    Higher intensity Atlantic basin hurricanes have increased, which is consistent also with future projections for that basin, but it will take more data to confirm with very high confidence any projections.

    Research on tornadoes and climate change is fairly thin and nothing representing a consensus.  Jeff Masters put it best:

    “At this time, it does not appear that there has been an increase in violent EF4 and EF5 U.S. tornadoes in recent decades. Preliminary research using climate models suggests that we may see an increase in the number of severe storms capable of producing tornadoes over the U.S. late this century. However, this research is just beginning, and much more study is needed to confirm these findings.”

    as did the IPCC (with regards to global records):

     “There is insufficient evidence to determine whether trends exist in small scale phenomena such as tornadoes, hail, lighting, and dust storms.”

    This was before 2011, which was a record year in the U.S….too early to tell if this is part of a major shift.

    Heavy precipitation has increased globally due to global warming.

    Solar activity can’t explain recent warming.  Trends have been in the opposite direction.

    http://www.skepticalscience.com/solar-activity-sunspots-global-warming-intermediate.htm

    Fred: You state “higher snowfall totals are caused by higher precipitation.” 

    Don’t put quotes around something I didn’t say.

    Fred: But around 2000 a prominent climate scientist in the UK went public with a widely circulated prediction that snow would be a thing of the past in the UK by 2010.

    It was “widely circulated” among political spheres for a bit of point-scoring, but your characterization here is not correct.  Find the direct quotes.  Check and revise.  Take some time also to understand what the science is generally indicating before attacking it.  You won’t find many accurate characterizations on wuwt, usually just simplistic strawmen and little nuance aimed at cheap and often dishonest point-scoring – very anti-science.

    Tom C: “”¦directly responsible”¦” No scientist has said such a thing

    It’s not clear if Gore said that phrase.  It’s not a direct quote from the Guardian article.

  • Fred

    NewYorkJ @46

    Your contentions of extreme weather are decisively refuted by the facts.  No increases n floods, hurricanes, or tornadoes.  Lets look at tornadoes.  Yes, 2011 saw an uptick in tornadoes.  Why?  Because, according to Joe Bastardi, of a “tremendous drop in mid-level temperatures.”  Regarding the uptick in tornadoes in 2011 he remarked “its despicable to try to use this as an argument for global warming….it could snow cheese in New York tomorrow and that would be from global warming.”

    This is from:
    http://wattsupwiththat.com/2011/06/03/bastardi-on-the-non-existent-climate-tornado-linkage/
    Regarding my statement that a prominent AGW climate scientist predicted snow would be gone from UK by 2010 you state “It was ‘widely circulated’ among political spheres for a bit of point-scoring, but your characterization here is not correct.  Find the direct quotes.”

    Here it is from March 20, 2000:
    “However, the warming is so far manifesting itself more in winters which are less cold than in much hotter summers. According to Dr David Viner, a senior research scientist at the climatic research unit (CRU) of the University of East Anglia,within a few years winter snowfall will become ‘a very rare and exciting event’.
    ‘Children just aren’t going to know what snow is,’ he said.”

    Available at:

    http://www.independent.co.uk/environment/snowfalls-are-now-just-a-thing-of-the-past-724017.html

    Points scored.
     

  • kdk33

    #48:  The US Climate Extremes Index is a good example of the complexity of detecting extreme events.

    With all due respect, I see this quite the other way ’round.  The time window and the geography window just plot subsets of a larger data set, Some of the variable choices are just combinations of other variables.  These don’t qualify as “different”, IMHO. 

    AFAICT, ignoring subsets and combinations, there are only 6 plots: min temp, max temp, PDSI (Palmer drought severity index), days w/o rain, days w/ rain, 1 day precipitation.

    I think this shows how easy it is.

    I’ve not looked hard at how these “extreme variables” were defined, but I’m flexible.  As soon as I can access software I’ll regress these against temperature (admittedly, these are jut for NA, but It’s a start).

  • NewYorkJ

    Fred: Your contentions of extreme weather are decisively refuted by the facts.

    My statements above are all supported by the evidence.  Yours are a mix of false statements, red herrings, and misrepresentations.

    Fred:  according to Joe Bastardi

    Anyone relying on Joe Bastardi for anything is not focused on “facts”.

    http://www.skepticalscience.com/lessons-from-past-climate-predictions-arctic-sea-ice-extent.html

    As for your David Viner quotes, let’s compare what you attributed to him vs what he said.

    Fred: But around 2000 a prominent climate scientist in the UK went public with a widely circulated prediction that snow would be a thing of the past in the UK by 2010.

    DV: Heavy snow will return occasionally, says Dr Viner, but when it does we will be unprepared. “We’re really going to get caught out. Snow will probably cause chaos in 20 years time,” he said.

    Nowhere in the article is Viner quoted as saying snow will be a “thing of the past” by 2010.  Looks like it’s the usual telephone game.  The author of the article claimed the former and you threw in “2010” for good measure.

    Points deducted.

  • Fred

    New YorkJ @ 54,

    Spinning your way out of this will not work.  In 2000 Viner says in a “few years” snowfall will become a “rare and exciting event.”  2010 qualifies as more than a few years out and snow certainly has not become any rarer in the UK than it was in 2000.  Maybe the article didn’t say snow in the UK would be gone by 2010 but then I hadn’t seen it in several years.  And Viner’s prediction is most certainly very wide of the mark.

    The fact are clear and uncontrovertible that there has been no changes in extreme weather events such as hurricanes, floods, and tornadoes over recent years.  The recent uptick in tornadoes in the US recently that you mentioned was due to cooler conditions. 

    Just for good measure, note this 2011 presentation showing a recent downward trend in flooding events:

    http://itia.ntua.gr/getfile/1128/2/documents/2011EGU_DailyDischargeMaxima_Pres.pdf

    Computer models that are worthless, no indication of predicted extreme weather events, a better correlation of temperature with solar conditions than with CO2 – AGW theory is on its way out.

     

     

  • NewYorkJ

    Fred:  In 2000 Viner says in a “few years” snowfall will become a “rare and exciting event.” 

    That is not the equivalent of your claim:

    But around 2000 a prominent climate scientist in the UK went public with a widely circulated prediction that snow would be a thing of the past in the UK by 2010. 

    Maybe the article didn’t say snow in the UK would be gone by 2010 but then I hadn’t seen it in several years. 

    Since you made no specific charge regarding the article itself, that is irrelevant.  You’re still obligated to correct your misrepresentation of Viner. 

    And Viner’s prediction is most certainly very wide of the mark.

    Not exactly.  It’s not a specific detailed prediction for starters.  And the U.K. did have mostly very mild winters with little snow since then, only recently had that changed.  And it certainly has returned to hit them off guard.

    The fact are clear and uncontrovertible that there has been no changes in extreme weather events such as hurricanes, floods, and tornadoes over recent years.

    This is clearly incorrect, as #44 indicates.  Note again that many factors influence flooding, extreme precipitation increases being one of them, snowfall, heating leading to drops in soil moisture, and human influences on rivers being some of them.

    The recent uptick in tornadoes in the US recently that you mentioned was due to cooler conditions. 

    Bastardi is not a reliable source for that claim.  In April, 2011, which saw record activity, monthly anomalies for the hard-hit regions were well above average

    http://data.giss.nasa.gov/cgi-bin/gistemp/do_nmap.py?year_last=2011&month_last=08&sat=4&sst=1&type=anoms&mean_gen=04&year1=2011&year2=2011&base1=1951&base2=1980&radius=1200&pol=reg 

    But I wouldn’t worry about splitting hairs on that one.  The jury’s still out on what affect, if any, global warming will have on tornadoes.  I do think 2011’s record year is notable, but it’s too early to tell if that’s the start of any trend.

    http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=910

  • Fred

    New YorkJ:

    The graph of CO2 levels bears no relationship to those of the occurrences of the categories of extreme weather events.  Below is a link to a graph of tornadoes:

    http://lwf.ncdc.noaa.gov/img/climate/research/tornado/tornadotrend.jpg

    Here is a graph of hurricane activity:

    http://policlimate.com/tropical/index.html

    And this link shows a steady to falling flooding activity showing a slight trend towards less intense flooding in recent years:

    http://itia.ntua.gr/getfile/1128/2/documents/2011EGU_DailyDischargeMaxima_Pres.pdf

    No relationship exists between any of these extreme weather events and CO2 levels.
     

  • kdk33

    Well, reviewing the NOAA data has been interesting”¦

    Per earlier comments, I contend that, to establish causality, meteorological variables should first be plotted/regressed against temperature.  After all, it is via warming that CO2 affects weather.

    To do this I need a suitable independent variable (temperature) and suitable dependent (meteorological) variables.  So temperature must be reliably measurable and changes plainly detectable and statistically significant ““ to this point Steven Mosher violently cedes, and to claim otherwise would surely be heresy, so I assume no objections.  I next need meteorological variables that can be measured reliably enough to detect significant changes.  Such variables must exist, else the climate sciences would instantly self-destruct; the enterprise would devolve to one word and one word only: weather.

    Some contend that Extreme events are rare, hence un-measurable.  This, of course, opens the door to the magic kingdom, from whence science has been banished and through which I will not pass.  I contend this is not true for 2 reasons: 1) while extreme events are rare at a particular location, it is much less rare to have extreme events occurring somewhere, 2) as long as the event distribution is smooth without multiple maxima then I can use less-rare events as indicators for more-rare events. (In a previous comment I said “continuous”, which I now think is wrong.  I should say smooth, or perhaps continuously differentiable, or something like that”¦  Clearly, piece-wise continuous won’t work.  Any math major is free to help me out here”¦.) If weather event distributions did not have these properties, traditional statistics would cease to have meaning, and the climate sciences are rife with statistics, so again I assume no objections.

    For temperature I considered GISS and Hadcrut-3 land-sea annual anomalies.  The numbers vary slightly, but the conclusions are exactly the same.  Since this is a US-centric exercise, I settled on GISS temperatures from 1910 to 2010.  The range is about 1 C.  A warmist-reasonable expectation for sensitivity is 2.5 C / CO2 doubling.  So the 1C range (40% of 2.5) is ample.

    The data:  I used US is annual meteorological data from here:
    http://www.ncdc.noaa.gov/extremes/cei/
    A simple graphing utility can be found here:
    http://www.ncdc.noaa.gov/extremes/cei/
    The data files here:
    http://www1.ncdc.noaa.gov/pub/data/cei/
    Definitions here:
    http://www.ncdc.noaa.gov/extremes/cei/

    There are two data types: binned temperature data and meteorological data.  Initially, I thought the temperature data would be uninteresting; I was wrong.  The temperature data is binned into four categories (really high/low means 10th percentile ““ high or low, accordingly )

    1) percentage of the US with really high maximum temps
    2) percentage of the US with really low maximum temps
    3) percentage of the US with really high minimum temps
    4) percentage of the US with really low minimum temps

    These all have, at 95% confidence, statistically significant trends with temperature.  A linear regression yields the following slopes: (units == percentage points / deg. C
    1)      slope = 15.7
    2)      slope = -17.6
    3)      slope =  33.0
    4)      slope = -26.2

    Warming acts more on minimum temperatures than maximum temperatures.

    For fun, I regressed the same four variables, but divided them into winter (Dec.-Feb.) and summer (Jul.-Aug.).  The trends for 1, 3, and 4 were statistically significant and they were:

    1)      summer slope = 10.3, winter slope = 11.6
    2)      summer slope = 30.9, winter slope = 16.1
    3)      summer slope = -19.1, winter slope = -21.6

    I did a similar regression splitting the data into warm season (Ap. ““ Sep.) and cold season (Oct. ““ Mar.).   All trends were significant except for really high maximum temps in the winter.  The slopes were:

    1)      Warm season slope = 13.3, cold season slope = n/a
    2)      Warm season slope = -12.9, cold season slope = -12.9
    3)      Warm season slope = 31.8, cold season slope = 21.9
    4)      Warm season slope = -22.6, cold season slope = -21.9

    Surprisingly (to me) then, warming acts about the same on winter and summer temperatures, except for really high minimum temps where the effect is much more pronounced in the summer.  (I did not do ANCOVA, but I suspect the other slopes are statistically the same.) My understanding has always been that CO2 induced warming acted more on winter time than summer time temperatures – make of it what you will; perhaps I’ll have to change my understanding.

    Now, to the weather data, the variables were:
     1)     percentage of US recieving greater than normal amount of rain in 1-day events (a heavy precipitation index)
    2)       percentage of US with greater than normal consecutive days with rain
    3)       percentage of US with greater than normal consecutive days without rain
    4)       percentage of US with much higher than normal PDSI (too wet)5)       percentage of US with much lower than normal PDSI (too dry)6)       CEI ““ the government’s climate extreme index
    7)       Ad-hoc extreme index #1 ““ the sum of 2-5
    8)       Ad-hoc extreme index #2 ““ the sum of 1-5
    9)       Consecutive rain extremes ““ the sum of 2-3
    10)   PDSI extremes ““ the sum of 4-5

    (Again, much higher/lower than normal means 10th percentile ““ high or low, accordingly)

    There are, at 95% confidence, no statistically significant trends for the extreme variables 6-10. So, the observational data does not support the claim that warming leads to weather extremes.

    There were statistically significant trends for 1, 2, 3, and 4.  The slopes were as follows:

    1)      slope = 8.2
    2)      slope = 12.0
    3)      slope = -14.7
    4)      slope = 7.2

    So, warming leads to more rain (which is consistent with my understanding).  The extra rain comes as more frequent rain events and as an increase in heavy precipitation events.  This might suggest an increased flood risk – it’s hard to imagine any more-rain scenario that wouldn’t – but, as NYJ so kindly and repeatedly points out: floods are a function of many other variables, both natural and anthropogenic ““ land use, flood control, water management; meteorological history (it’s a path function), etc).

    Warming is not affecting drought frequency.

    The demonstration of the proper first step to establishing causality being complete (;-)), this particular exercise using NOAA data suggests that:

    1)      Warming acts more on minimum temps than maximum temps.
    2)      Warming Acts about the same on summer time and winter time temperatures
    3)      Warming causes more rain ““ both as more frequent rain events and as heavier precipitation events
    4)      Warming isn’t affecting drought frequency
    5)      Warming does not drive weather to extremes

     Feel free to show me the error(s) of my ways.
     

  • Fred

    kdk33,
    Awesome work.  Very credible in that it fits in with other results, i.e. warmer temperatures are associated with more rainfall.

    Keep in mind that during the 1910 – 2010 period the earth was coming out of the Dalton (solar) Minimum, which lasted from around 1790 – 1830.  This was quite a cold period.  Thus, as you already know, natural forces are a very good candidate for causing the warming observed to occur between 1910 and 2010.

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collide-a-Scape

Collide-a-Scape is a wide-ranging blog forum that explores issues at the nexus of science, culture and society.

About Keith Kloor

Keith Kloor is a NYC-based journalist, and an adjunct professor of journalism at New York University. His work has appeared in Slate, Science, Discover, and the Washington Post magazine, among other outlets. From 2000 to 2008, he was a senior editor at Audubon Magazine. In 2008-2009, he was a Fellow at the University of Colorado’s Center for Environmental Journalism, in Boulder, where he studied how a changing environment (including climate change) influenced prehistoric societies in the U.S. Southwest. He covers a wide range of topics, from conservation biology and biotechnology to urban planning and archaeology.

ADVERTISEMENT

See More

ADVERTISEMENT
Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »