Statistics

On The Attribution Of A Single Event To Climate Change

typhoon

Rained yesterday here in the city of cities. Must be because of climate change, right? Hey. The climate did change and it did rain. What more evidence do you want?

Bonus trivia question: name a period in which the climate on this island earth never changed.

That’s right, guppies. It’s a trick question. There is no such period! The climate has always changed; therefore, it is rational to suppose that it always will.

“Briggs, you fool. When people say ‘climate change’, they don’t mean the climate changed. They mean the nature of the climate has remained stationary up until some point, after which is changed—and even fools like you know this means a change in the statistical nature of the climate.”

So that if you say, as I’ve heard you say, that temperature is “normally distributed” and yesterday the high was 62F and today the high is 38F, you’d say the temperature didn’t change?

“Well, you know what I mean.”

Funny definition, that. Besides, even if the model—your normal distribution—remained the same from day to day, it’s still true that some thing or things caused the high to plunge (i.e. change), right?

“The climate didn’t change. The temperature did.”

That either makes no sense or it’s a circular argument. Your model—your normal distribution—doesn’t say diddly about what caused the observed change. Do you agree?

“It doesn’t have to! If the model doesn’t change, the climate didn’t change!”

You avoided the question while committing the Deadly Sin of Reification. You’re defining “climate” as your model, and not as the real-life observations. At least you’re in good company. Like that of Gerrit Hansen, Maximilian Auffhammer (cool name), and Andrew R. Solow who wrote a peer-reviewed paper of the same name as today’s post in the Journal of Climate, which makes the same mistakes you make (vol 27, 15 Nov 2014, pp 8297-8301).

These authors define a “stochastic” “point process”, which is to say, a probability model, which describes the uncertainty in event occurrences. Like, say, blog posts. Once a day here—a good and highly accurate model! Which is not meant as a joke. Why?

Their probability model, like any probability model, announces, conditional on specified premises, which usually include past observations, the probability some proposition is true. Thus, given our blog point process model and the history of posts at the venerable WMBriggs.com, you might say the probability “A post shows tomorrow, 11 December 2014” is high. Tune in tomorrow to see how useful this model is.

Again, this isn’t a joke. I started this blog seven, eight years ago. Back then the climate was different, and so was my posting frequency. It was on the order of twice to thrice a week.

That means were I to fit a point process model to this history of posts, it would show a correlation with climate change. There would be parameters inside this model which would measure this association, parameters which I could use to quantify the correlation, I mean.

Now the real question: is the changing climate causing me to write now daily posts?

Of course not!

And it’s silly to suggest that it is, even though the parameters in our model show a “significant” correlation (with time or climate change). The two things—climate and my fevered imagination—have nothing to do with one another.

Just to be perfectly crystalline transparently forcefully clear, a parameter in a point process model, which might be pegged to time or some other external thing, cannot say why the observed series changes or doesn’t change.

Here’s our authors (pp. 8297-8298):

Given that an event has occurred after the climate has changed, was it or was it not caused by climate change? This question implies that, once climate has changed, the point process of events represents the superposition of a point process of events that would have occurred in the absence of climate change and a point process of events that would not have occurred in the absence of climate change and are, therefore, attributable to climate change. Moreover, these point processes must be independent; otherwise, the former would inherit a climate change effect through the latter.

They assume a model which has a rate indexed by a parameter, and after “climate change” this parameter increases, and the increase is “attributed” to climate change.

As you can see, this whole thing, start to finish, confuses the nature of causality. What does “once climate has changed” even mean? If we knew the climate changed at some point, and how this new “climate” (and the old) caused, say, hurricanes or lightning strikes, then we don’t need a probability model. We’d just say, “There will be this many hurricanes and lightning strikes”—and we could not be wrong.

We certainly don’t need a model to tell us if a hurricane struck. We can just look. And if we don’t know the precise causes of the hurricane, it’s silly to claim it was “caused” by climate change. Some thing or things caused the hurricane before the climate changed and some thing or things will cause hurricanes after the climate changes. The probability model just can’t say anything definitive about causes.

The authors:

Over the 30-yr period 1950–79, there were a total of 39 intense North Atlantic hurricanes while over the following 33-yr period, 1980–2012, there were 53 such hurricanes. If we assume that the effect of climate change over the entire 63-yr period was to increase the rate of these hurricanes, then the ML estimate of the estimated probability that a hurricane in the later period is attributable to climate change is 0.19 and an approximate 0.95 confidence interval for this probability is (-0.17, 0.44).

Again, this makes no causal sense. Add emphasis: “If we assume that the effect of climate change….was to [cause an] increase” then the estimated probability that the hurricane is “attributable”, i.e. caused by, “climate change” ought to be 100% or nothing.

Scientists spend far too much time on these vaporous models when they’d be better off searching for causes and in understanding physics.

Categories: Statistics

58 replies »

  1. Do we suppose that the network of weather satellites in the 1950s was as sensitive as those spanning the entire Atlantic in the 2000s? Oh, wait a minute. There was no such network, and hurricanes often took folks by surprise. The number of “severe” storms [whatever that means] counted in the 1950s will not include many of those that are tallied in this wondrous modern era of ours, when even mere tropical storms are graced with names. I wonder if this might have an impact on the count?

    It was quite common to use the expected distribution to determine whether a new tool design produced different results in dimensions and performances after it was put into the press. But you will notice that we already knew that the tool had been changed. It is quite another matter to determine whether a tool has been changed given only a shift in the pattern of results. It might be a new tool installed — but it might be the set-up, or a change in material properties of the feedstock, or a new operator took charge of the machine, or…. In one fascinating case, the shift was caused by QC using a different ruler to measure the registration of the printing.

  2. Technically, climate IS a model. It’s a statistical construct. MIT defined it as “the statistics of weather”. I have started explaining that climate is not “real”, but rather a “model” and changes in it may or may not reflect anything in reality. It is useful for selling real estate—you can point to the “climate in an area” as a selling point. By the time the buyer figures out “climate” has no real meaning and is buried in 10 feet of snow in September in the “warm climate” the agent sold him, it’s too late! (This can, unfortunately, be fatal in some cases. Cold kills.)

    The last quote you have in a shaded box really is telling. They assume what they set out to “prove”. And they say peer-review catches scientific errors. Really–either the reviewers were asleep or they need serious retro-training in logic (or maybe just a first run course, since they may not have ever heard of the idea.)

    Here’s an interesting study–note that Australia had to add another color to their heat map so it’s REALLY serious. (Peter Stott must not be familiar with the Dust Bowl in the US or he would not have such difficulty explaining this as part of nature.)
    http://www.weather.com/science/environment/news/report-human-caused-climate-change-2013-weather-extremes-20140929

  3. Sheri, in my forthcoming post “Integrity in Science–the Lessons of Climategate” I will give specific instances of the perversion of the peer-review process by the warmists, and other instances where they have failed the scientific method.
    I would disagree that “Climate is a model”. If you use historical records and LEGITIMATE proxies for temperature you can give a historical record of temperature that validate, for example, the Medieval Warm Period (Greenland was green not so long ago) and the Little Ice Age.

  4. Bob: I look forward to reading your post!

    Maybe I am not understanding what a model is. From what Briggs writes, I get the feeling that anything that is averaged or otherwise statistically manipulated into a single value (such as an anomaly from the mean, the mean temperature, etc.) is a model. To avoid being a model, you just report the data. I’m thinking that your definition is different? Also, how do you define “climate”? I’m asking for clarification, not be ornery!

  5. The problem is that the useful word climate has morphed from coarse distinctions of tropic, temperate, arctic etc to whatever it means today. Maybe the term climate control as applied to the interior of buildings should now be called weather control. I feel the need to get in a quick comment before the hoards descend.

  6. Sheri, I guess what I mean by “model” is what a physicist would think of–e.g. a representation of a physical situation. What Briggs (and you?) are thinking of I guess are statistical or mathematical representations of data, possibly useful for prediction.
    Scotian, my definition of “climate” is that corresponding to the Oxford dictionary:”The weather conditions prevailing in an area in general or over a long period” . I guess to make a statement about that you do have to do some sort of averaging or weighting process.

  7. Bob: I guess I’m going with anything that is not just pure data is a model. In physics, I can see where models are essential, especially at the quantum level. With temperature, averages seem to be models, except they don’t really describe a physical situation. At least not for me.

    The dictionary definition for climate is pretty loose. Perhaps that’s why the term “global warming” was much more accurate than climate change to describe the theory. Just politically unpalatable.

  8. Guppies? I thought we were groupies.

    “Climate” has acquired too many meanings to convey real understanding in most cases. We assume we know, but do we? To compound the confusion, few bother to specify the “long period” of time over which it integrates “weather.” Without context it’s a guessing game.

  9. “Bonus trivia question: name a period in which the climate on this island earth never changed.”

    Trivially true. But the climate has been quite stable over the Holocene, which has allowed civilization to develop and flourish:

    “A Reconstruction of Regional and Global Temperature for the Past 11,300 Years,” Shaun A. Marcott et al, Science (2013)
    http://www.sciencemag.org/content/339/6124/1198.abstract
    graphs:
    http://www.sciencemag.org/content/339/6124/1198/F1.large.jpg

    “The climate has always changed; therefore, it is rational to suppose that it always will.”

    Which completely avoids the issue. Climate rarely changes willy nilly, especially globally where conservation of energy is required. It changes when it’s forced to change. And right now there is no known forcing causing our rapid warming except the increase in GHGs.

    The Sun isn’t doing it — we know this from monitoring its irradiance with satellites, or, before that, estimating it via proxies. And we know it from the nature of this change — a warmer Sun would, besides increasing surface temperatures, cause the stratosphere to warm. Instead the stratosphere is cooling — a fingerprint of an enhanced greenhouse effect.

    Then there is the rate of change. Temperatures today are increasing very fast by historical and geological standards — much faster than, say, when the Earth left its most recent glacial period. (Same for ice melt and sea level rise.) That requires an explanation beyond the superficial “climate always changes,” because climate rarely changes this fast.

    So why is it changing? Radiative transfer and planetary climate science, using well established physical laws, say the world should be warming due to our large emissions (again, by historical and geologic standards) of greenhouse gases. The climate is too complex to say exactly how much and exactly how fast, but it certainly predicts warming. And we’re observing warming — a lot of it.

  10. “MATT RIDLEY: BEWARE THE CORRUPTION OF SCIENCE
    http://www.thegwpf.com/23657/

    Of course charging corruption is about the only tactic left to Ridley, a pseudo-skeptic whose grasp of the science isn’t very impressive:

    https://twitter.com/richardabetts/status/542474397278539776

    “Sep ’14 @mattwridley accepts climate obs http://www.rationaloptimist.com/blog/whatever-happened-to-global-warming.aspx … Dec ’14: @mattwridley now doesn’t believe climate obs http://www.mattridley.co.uk/blog/policy-based-evidence-making.aspx

  11. It wasn’t an insult, just an accurate description. Ridley, like many others of his ilk, is not a real skeptic, who abound in science. As those links to his own writings show, he accepts the temperature data when he wants to make a certain point, and rejects it when he wants to make a different point. Totally inconsistent and unscientific — that’s what makes him a pretend skeptic — a pseudo-skeptic.

  12. You’re right Sheri.
    Briggs or some other expert in information theory. Suppose I count the number of words in meaningful comments and the number of words in non-meaningful comments. How do I calculate the amount of “information” in the total comments for a thread? Or should I use the number of characters?
    My object is to compare the information content for threads where evangelical warmists post and those where they do not.
    Advice will be gratefully acknowledged.

  13. Bob: While you’re calculating, would you also compare the information content for threads where evangelical cynics post and those where they do not? Thanks.

  14. Bob: You’ve dived into a vaste area that many on the net would love to be able solve. There are true evangelists who post useful information (on my blog, at first, there were two or three people having excellent scientific discussions that were very lengthy) while there are those that merely fling insults (if allowed). So number of characters won’t really work. One must also take into account how quickly the discussion ends when the evangelist shows up. That one is easier to calculate. I’m not a certified expert on information theory, but perhaps my experience can shed some light on this problem. Will get back to you after my research is finished. 🙂

  15. Appell, like many others of his ilk, is not a real scientist, who abound in science. As his own comments show, he accepts the temperature data when he wants to make a certain point, and rejects it when he wants to make a different point. Totally inconsistent and unscientific — that’s what makes him a pretend sceintist — a pseudo-scientist.

  16. Insults are tiresome, Briggs.

    PS: I’m still waiting for ANYONE here to discuss the data and the science. Anything beyond passive-aggression and glib dismissals would work in a pinch….

  17. MATT RIDLEY: BEWARE THE CORRUPTION OF SCIENCE

    It’s sad but Eisenhower was right. No one wants to kill off the cash cow or the goose providing golden eggs. We are getting what we paid for — a wind-up doll that says whatever is needed to support the message currently in vogue.

  18. DAV: Here’s what I never get about your argument.

    1) Young scientists often make their name (and all would like to) by going AGAINST the status quo — it’s the quickest way to move up the ladder. They all know this and aspire to it.

    2) If climate scientists are only in it for the cash, why, over the last 25 years, have they been writing and saying they are increasingly confident of the basics of AGW? (See each subsequence IPCC AR.) How would saying you’re more and more sure of your conclusion get you more grant money? Why wouldn’t you instead say you’re more uncertain then ever?

    3) In what way is your claim falsifiable?

  19. So is the reason for the presence of the pseudoscientist to give us a bad example or what? Virtually everything he posts is not news to readers here and we all know how to insult each other. Oh well, back to researching the information ratio of comments on blogs. So much more fun.

  20. I gather from the comments that applesauce is being tossed around again. I made a filter using a Firefox add-on. Wasn’t sure it worked but now I know.

    I once had to dig through a pile of cow manure for something that was accidentally buried. Not an experience I look forward to repeating. I’m not normally in favor censorship but I long ago became convinced there aren’t any diamonds under most piles and some in particular so why dig?

    If nothing else scrolling has become easier on my Droid with less content to skip.

  21. It’s made using Adbock. It was tricky to modify because it was mostly designed to key off URLs. I’m not sure what I have is general enough to distribute. I made changes to the code.

  22. “Scientists spend far too much time on these vaporous models when they’d be better off searching for causes and in understanding physics.”

    Good luck with that Briggs!

    There is no “climate science” at the present time. All we have is politics, scams, talk of a magic molecule (CO2), and grant money seeking. At some point in time, perhaps when NYC is under a 4 mile deep glacier, people will stop funding these fools and stop listening to them. At that point, we might go back to doing real science in regards to this planet’s climate. (maybe not even then)

    “The whole aim of practical politics is to keep the populace alarmed (and hence clamorous to be led to safety) by menacing it with an endless series of hobgoblins, all of them imaginary.” – H. L. Mencken

  23. For those interested, here’s the original source
    https://adblockplus.org/en/source

    Looking a bit more it may be possible to build a filter using the add-on as-is. There is a way to build a custom filter and it is capable of removing some content. E.g., I blocked all Twitter references which I could care less about and seemed a bottleneck when loading. That filter was already available online.

    I’ll look into how I might use the customizer to make post filters. Who knows? Maybe it’ll make me rich someday!

  24. DAV: I just added Adblock and I love it! No more annoying, poorly photoshopped ads with Oprah, Ellen, etc! As for posts, I just use my delete key based on the name at the top of the post when going through email (I had a couple of very determined individuals who tried modifying their posts so I wouldn’t recognize them, but all in all, it’s effective.) I may try playing with the Adblock.

  25. Young scientists often make their name (and all would like to) by going AGAINST the status quo — it’s the quickest way to move up the ladder. They all know this and aspire to it.

    This is part of the mythos, that is to say the “theory” of science, but historians of science take a more nuanced view. Often, the way up the ladder is to agree with the senior scientists on whose team you are working. As Planck once remarked, a new theory advances “funeral by funeral” as the older scientists die off.

    It has already been remarked that we have for the first time gone through an entire generation of physicists without a major upheaval in the science.

    ++++
    Models
    http://tofspot.blogspot.com/2014/02/americas-next-top-model-part-i.html
    http://tofspot.blogspot.com/2014/03/americas-next-top-model-part-ii.html

  26. Sheri,

    I don’t really like to block ads, though. It could be the site’s bread and butter. OTOH, I never click on them except by accident. They do get in the way however and cost me time and money. Wouldn’t be so bad if all sites placed them in unobtrusive places but more and more are designed to be in-your-face. They are becoming common on phone apps and there they are more easily clicked by accident. Harder to eliminate unfortunately.

  27. YOS,

    I never really counted them but I get the impression a lot of the scientist critics are retired. Why would that be if the way to success is going against the grain?

  28. Ye Olde Statisician commented:
    “Often, the way up the ladder is to agree with the senior scientists on whose team you are working. As Planck once remarked, a new theory advances “funeral by funeral” as the older scientists die off.”

    I’ve always taken Planck’s remark to mean that the older scientists with their older ideas need to be cleared out in order for the new and fresh ideas of younger scientists to flourish, and not that it meant the younger scientists agree with older scientists just because they’re older and have budgets.

    To imagine that every young scientist in the world is just going along with some AGW crackpottery solely in order to have a job and get a grant is beyond ridiculous.

    And, of course, the younger scientists don’t endlessly beat their heads on long-established, consensus science, such as thermodynamics or the greenhouse effect. Their ideas and work at at the edge of the consensus, and very very rarely impact the heart of it.

  29. DAV: I did consider whether blocking ads was depriving anyone of income. My research says, no, I don’t recall ever buying due to any ad I saw on a page. I’m more likely to buy from an email from a retailer I do business with. Also, some ads were so offensive, I left the pages and didn’t bother to read the posts. A lose, lose situation, as far as I could tell. It’s really not much different from fast forwarding on the DVR or muting the annoying commercials on TV. If an ad is entertaining, I watch. If not…….

  30. Appell must be slavering at the mouth that you’ve written a post he thinks he can understand. But let’s not waste time on cranks.

    Regarding any sort of definition of ‘climate change’ – well the climate does change, so by default it’s a confusing phrase to use. Of course it’s vagueness is its strength. If you’re unsure exactly what is being talked about, there is less chance that you can be shown to be talking nonsense. However, I’ve always assumed ‘climate change’ is short for ‘anthropogenic climate change’. That is, that the climate changes in ways that it would otherwise not change into, were it not for human influence. Before you can even begin to debate the issue, you need to look at trends in weather related data sets. If you can’t find trends, you’re not even postulating about what a particular correlation might have as a ’cause.’ You’re therefore firmly in the realm of talking gibberish. At this point you say, “event X is ‘consistent with’ theory Y”. You use “consistent with” when you don’t have evidence, and you don’t have correlations either. “Consistent with” is powerful and useless as it applies to nearly any random event. E.g., the missing garden gnome in my front yard is “consistent with” my alien abduction theory.

  31. YOS, I think there are two problems:
    1) too much money and professional advancement involved in being with a popular movement–when the CRU (East Anglia) has $42 million in grants from various government and private sources, then there’s a lot on the line in keeping on doing what you’ve been doing. Moreover, professional advancement in scientific academia is quite often predicated on the amount of grant support one gets.
    2) Nowadays very few research directors teach their students about the ethics of scientific practice . I can’t prove this by statistical data, but it seems evident from the practices of some of the evangelical warmists. Also, very few students (and their directors) have any knowledge of the history or philosophy of science. Again, an unverifiable statement, so maybe it’s just my prejudice.

  32. Will: love the garden gnome analogy! We really do need a definition of climate if there is to be any science in this. We also need the definition if we are to decide what action needs to be taken. If “climate change” means the change humans are causing”, then the argument is more or less circular. If it means “more changes in the anamoly from the average global temperature” than if humans were not burning fossil fuels, then we need to know what said changes have to do with weather and local climates and much more evidence of causality and not just “consistent” with (although ancient alien theorists can weave some pretty convincing “consistent with”s!) One would think we also needed much more data and a longer timeline (or maybe not—we are just adding more and more beginning and ending points. Scratch that for now.)

    Science needs these definitions before we can attribute any changes in weather or climate to humans, much less one extreme weather event.

  33. @Sheri

    Yes, you’re right. If you look at the modelling predictions made in AR4, you’ll note that the models are starting to predict natural cooling and human produced CO2 is offsetting that, causing some warming. (See the diagrams that show the modeled warming trends with/without CO2.) They do this, I suppose, because the planet was warming naturally before 1950 and seems to have warmed more or less at roughly the same rate after 1950. So there is little room for an anthrogenic influence in the trend, *unless you assume* that nature switched paths after 1950 and started to cool things down. The logic is tortured… Anyway, coming back to your point, the interesting thing is that at least from 1950-2014, it seems that human influence actually caused the climate *not to change*. (Countered the expected natural cooling.) It’s all rather strange, but they are a rather strange research group.

  34. Did they really say that the 95% confidence interval for the probability began at -0.72? I thought I understood probability but I’m lost now.

  35. Rich: If it’s the confidence interval, yes, those are both positive and negative. It means there’s a 95% probability that all values lie between the values in the confidence interval (it’s the error bars—95% or two standard deviations from the mean in a normal distribution). If I am understanding your question correctly. No confidence interval given on that statement. 🙂

    (I didn’t find the .72 value, but am addressing CI’s in general.)

  36. Will Nietzche wrote:
    “You use “consistent with” when you don’t have evidence, and you don’t have correlations either. “”

    You use “consistent with” when systems are complex and models can only give probabilistic results. You look at trends and see how they compare to scientific expectations. AGW influences climate, an simple physics says all climatological and even meteorological events are now influenced by the extra energy in the system. That doesn’t mean AGW created them, though it could have — the climate system is too complex to make black-and-white distinctions. What causes a tornado to form here but not there? What causes one year to have X hurricanes and another Y? Even basic meteorology is unable to answer such questions,

  37. Sheri:
    Here’s the full quotation:
    “then the ML estimate of the estimated probability that a hurricane in the later period is attributable to climate change is 0.19 and an approximate 0.95 confidence interval for this probability is (-0.17, 0.44).”

    So the variable of interest is a probability, estimated at 0.19. The lower limit of the 95% confidence interval for this variable is -0.17. That’s a negative probability. What does that mean. How can a probability be negative?

    I recant my former view. I do understand probability and this paper contains meaningless gibberish produced by mindlessly running stats software.

  38. Actually, the following sentence does not appear in Briggs’s quote:
    “The negative lower bound of this confidence interval indicates that a decline in the
    rate of intense hurricanes between these periods cannot be ruled out.”

    The CI is based on standard deviation and is not necessarily bound by “normal” probability. I suspected that the negative number meant that the rate could be declining and checked the original document to find out. That is the author’s apparent interpretation.

  39. I plead advancing age with too much too remember as the reason for forgetting that little over a month ago, Briggs covered CI’s! 🙂

  40. Sheri commented:
    “We also need the definition if we are to decide what action needs to be taken. If “climate change” means the change humans are causing”, then the argument is more or less circular….”

    see:
    “Climate change,” Annex III – Glossary, IPCC 5AR WG1 pg 1450 (2013).
    http://www.ipcc.ch/report/ar5/wg1/

  41. What has not been hard to do is to count land falling hurricanes in the US, these can’t be missed as “fish storms” can be. There is also a 100 year record here. There is no apparent trend.

    http://rogerpielkejr.blogspot.com/2012/11/us-hurricane-intensity-1900-2012.html
    http://rogerpielkejr.blogspot.com/2012/12/global-tropical-cyclone-landfalls-2012.html

    Using the North Atlantic as a trend determining area is nonsense when global data is available. When you are analyzing a global statistic and you have global data, it makes no sense to use subsets of this data and expect to find a better “truth”. Using less data results in less SNR and higher variability. One could claim the now 8 years and counting Cat3+ hurricane landfall drought in the US is caused by AGW. Want to guess how many activists are peddling this correlation? Compare that to if we were in an 8 year period of a record setting count of Cat3+ US hurricane landfalls. It would be rampant speculation.

    Global hurricane frequency
    http://rogerpielkejr.blogspot.com/2012/12/global-tropical-cyclone-landfalls-2012.html

    Many more measurements can be found here:
    http://models.weatherbell.com/tropical.php

    The opportunistic hurricane climate link media frenzy that occurred after the 2005-2006 hurricane seasons and Katrina have fallen flat with the recent “hiatus” in landfalls.

    Almost every extreme weather climate link is tenuous at best and “science by the power of suggestion” at worst. I can only surmise the AGW activists felt they needed a here and now impact to push the political agenda forward. This always seemed like a bad idea and bound to backfire as it is quite easy to refute.

    In most cases there isn’t even a correlation to temperature or CO2. Once you get the correlation, the really hard part is proving causation. Given the nature of extreme events (sporadic, highly variable, require long trends to determine anything) you can take 10 different event trends and pick any time period and half of them would likely be increasing at different rates and half of them decreasing. Claiming the increasing ones are AGW and the decreasing ones are natural variability is a bit of a stretch.

    If you want to make a claim that these events will get worse in the future, we shall see what happens. If you want to claim they are already worse and caused by AGW, then you are jumping the shark.

  42. I’ll have to recant my recantation it seems. It’s me that’s going mad.

    “I suspected that the negative number meant that the rate could be declining and checked the original document to find out. That is the author’s apparent interpretation.”

    I’m not having any problem with confidence intervals here. I’m having a problem with what the numbers mean. If I analyse some height data and find that the best estimate of a man’s height is 1.8 metres with a confidence interval of 1.5 to 2.0 the ‘1.5’ and the ‘2.0’ mean metres. Not standard deviations, not Christmas elves but metres. That’s the meaning of the variable of interest.

    So if I’m offered an estimate of a probability then the numbers for the limits of its confidence interval mean probabilities not the rise and fall of the rate of hurricanes or the Roman Empire or anything else. And probabilities cannot be negative. They are defined not to be negative. Negative probabilities make no more sense than negative heights.

    (I couldn’t find a free-access version of the paper and given its use of negative probabilities I’m unwilling to buy it).

  43. My only explanation is based on quantum mechanics, so I looked further (QM doesn’t apply to macro, I know). However, I did find a statement from Berkeley
    (http://www.stat.berkeley.edu/~stark/SticiGui/Text/confidenceIntervals.htm) that the negative results and results that exceed 100% can be stated as zero and 100%, rather than exceeding the actual limits of the distribution. That might have been better in this case? It kind of makes sense to me that if the CI is negative, the number of hurricanes would be decreasing, but I may have been drawn into a statistical trap there.

  44. By the way, on the one year anniversary of that failed forecast, the CBC highlighted the new color addition without mentioning that it was a failed forecast and not the actual temperature, thereby misleading their audience.

  45. Briggs,

    On The Attribution Of A Single Event To Climate Change

    When can we expect an article from you entitled, “On the Falsification of a Complex Theory Considering Only Two Observables”?

    Bonus trivia question: name a period in which the climate on this island earth never changed.

    Never mind, that answered my question.

  46. Sheri,

    Technically, climate IS a model. It’s a statistical construct. MIT defined it as “the statistics of weather”.

    I saw a tidbit over at WUWT the other day saying that the 30 year definition was chosen because that was the extent of the observational record

    I have started explaining that climate is not “real”, but rather a “model” and changes in it may or may not reflect anything in reality.

    Why pick on just climate? Next time your local weather bureau says, “Last night it rained 1.3 inches” you really should rip them a new one for it. It didn’t rain 1.3 inches everywhere in town, I guarantee it — it only maybe rained 1.3 inches +/- some observational uncertainty at the instrument site and it’s assumed that a similar amount of rain fell within some who knows what distance of that station … likely based on a statistical model which interpolates those observations with other neighboring stations.

    It gets even worse going to three dimensions and talking about wind shear, cloud layers, cloud coverage, relative and absolute humidity, dew point, pressure …

  47. Brandon: I’m so sorry your California weathermen are soooo incompetent. Our weathermen report the amount of rain “officially at the airport”. They may also report the amount downtown and elsewhere. Same with snow–very clever of them since I live next to a mountain. Better yet, they call the forecast a MODEL!!!! He will even say the models, as in plural, do not all agree.

    Also, you really don’t understand a physical measurement is different from a statistical model. That explains a lot. You’re also making my point–rainfall statistics for a place tell you little or nothing. Our current new anchor said yesterday she was so amazed that it’s in the 50’s in December. Everyone told her Wyoming was really cold and nasty in the winter. That’s what the climate is supposed to be. It was below zero with several feet of snow for all of December when I moved here. Reality is not the same as climate. (I don’t have time to go into interpolation between stations, etc, but again, you are making my point.)

    Again, your last paragraph is making my point. You have been very helpful in pointing out my objections!

Leave a Reply

Your email address will not be published. Required fields are marked *