South Carolina Flooding is NOT a 1 in 1,000 Year Event

October 6th, 2015 by Roy W. Spencer, Ph. D.

There is no question that the flooding in South Carolina is exceptional, even historic.

But a once on 1,000 year event? Sorry, but there is no way to determine that…there are simply not enough rainfall statistics over a long enough period of time to establish such a claim.

But we do have information on previous floods over the last 100 years or so. So, let’s look at how the current event compares.

The greatest multi-day rainfall reported on the CoCoRaHS cooperative rainfall monitoring website was 27 inches in Columbia, SC. The Congaree River crested at 31 ft. there on Sunday:

Congaree River gage height at Columbia, SC.

Congaree River gage height at Columbia, SC.

Here’s a photo taken about the time the river crested:

Photo of Congaree River, Columbia, SC, Oct. 4, 2015, taken about the time of cresting.

Photo of Congaree River, Columbia, SC, Oct. 4, 2015, taken about the time of cresting.

Now, for comparison, take a look at a bridge over the Congaree River during the record flood of 1908, when the river crested at 40 ft…about 9 ft. higher than the current flood event:

Congaree River bridge in Columbia, SC, during the 1908 record flood event.

Congaree River bridge in Columbia, SC, during the 1908 record flood event.

Not to discount the misery and likely billions of dollars of damage caused by the current event, but when someone claims that a weather disaster is 1 in a 1,000 year event, they need to back it up.

Unfortunately, there seems to be an trend toward classifying events as “1 in 1,000 years”, when there is no way of knowing such things. This is especially true for floods, where paving of urban and suburban areas causes increasing runoff, making river flooding worse for the same amount of rainfall. This is a big reason why flood events have gotten worse in the last 100 years…it has nothing to do with “climate change”.

For some areas the current flood is no doubt a 1 in 100 year event, or even worse. But remember, it is perfectly normal to have a 1 in 100 year event every year…as long as they occur in different locations.

That’s how weather records work.

116 Responses to “South Carolina Flooding is NOT a 1 in 1,000 Year Event”

Toggle Trackbacks

  1. Thanks, Dr. Spencer. This is a very good post.
    Yes, there is something alarmist in the air. The increase in alarmism is undeniable, even if I have no model to “prove” it.
    The atmospheric river that flooded South Carolina was terrible, no doubt, but the sense of despair sprouting from the notion that it was caused by human emissions of CO2 is unfounded.

  2. jimc says:

    Looks like whoever designed that 1908 bridge did it just right. Maybe he had some older records.

  3. crosspatch says:

    In addition to the 1908 flood, they also had severe flooding in 1903. This was likely a 100 year flood, not a 1000 year flood. Reference:

  4. Doug Lampert says:

    My parents were moved for drainage control after a “500 year” flood that was not noticeably worse than the “100 year” flood while I was growing up.

    And there’s no question that increased pavement upstream was a major reason for even the flood that actually happened.

    No one blamed it on climate change (that I can recall), but bad weather happens.

  5. mpainter says:

    Dr. Roy says
    ” This is especially true for floods, where paving of urban and suburban areas causes increasing runoff, making river flooding worse for the same amount of rainfall.”

    I can verify that some rural areas have experienced increased runoff since pre-settlement days. Much of Texas and other states is rangeland with the grass well-cropped by grazing. Previously, grassland had a thick mat of grass and cover which very effectively reduced runoff, but there is none of this left today. One can observe the entrenchment of drainage, the formation of ravines where formerly there was a low place, ditches and gullies where once there were none, all of this the result of the settling and cultivation in the last 180 years or so. There is no question that floods are more severe and more frequent, and because more runoff means less absorption, more severe droughts. I believe this is true for much of the eastern US.

    The wackos will have us all believe that CO2 did it. And cow farts.

    • Goldminer says:

      Gee, I thought there were some folks living in SC 1,000 years ago and they were sending in the history! [sarcasm]

      It is amazing how quickly left-wingers will believe bunk like this. Pause for a moment left-wingers before you respond and answer this question:

      What is the scientifically verifiable evidence that this IS a 1,000 year flood?

      All this is going to do is set up some more taxpayer funded “research groups” at universities so PROfessORS can increase their already six figure “incomes”.


      • Keith Noren says:


        Read this article about the definition of xxxx year events that the National Center for Environmental Information uses with the media:

        You see, the tails of distributions are definable not though actual histograms but by the number of standard deviations (or other measures of dispersion) away. In this case it was a rain fall amount in a given period (not flood level). And by that defined standard, the South Carolina flooding was over a 1000 year event in many locations.

        I share a concern about the misunderstanding the public (and apparently has) could get from this definition.

        Read also David Hagen’s post / links.

        The degree sarcasm in these comments is astounding; not surprising given the website. But this shows the misguided fervor in the GW skeptics bandwagon.

        Hi, Doug L – read your post with some degree of personal knowledge.

        • mpainter says:

          Keith Noren. You find the sarcasm astounding but not surprising? Quite a mental feat, worthy of the most ardent AGW types.

          In regard to thousand year events, we lack the three thousand year record to determine that with any confidence.

          As far as your alarmist statistics are concerned, save your breath. ClimateAudit has shown that alarmist statistics are invariably cooked.

          • Keith Noren says:

            So you resort to quick and easy sarcastic attacks instead of engaging the statistics involved.

          • mpainter says:

            Keith Noren, you are the typical muddle headed alarmist who can’t make up his mind whether he wants to do snark or do junk science and so he does both. David Roe has now got you talking out of both corners of your mouth. And Keith, this is not sarcasm, this is an insult.

          • Keith Noren says:

            Still at the level of snark/sarcasm/insult I see.

        • David Roe says:

          I find it funny that you use this article to buttress your argument. Key quote in the article YOU PROVIDED: ‘”There’s a lot of uncertainty about it,” said Kenneth Kunkel, a professor focusing on climate variability at North Carolina State University.’

          That alone should give serious analysts pause.

          The “tails of distribution” are something that is only good for publicity, not serious discussion.

          • Keith Noren says:

            The tails of the distribution are difficult to know that is for sure. But how rare an event is can be described by how close a parameter is to the nearest measured value. Thus describing an event as a 1000 year event or a 500 year event or a 100 year event is a relative description of how rare that event is by how far away form the nearest actually measured value and the tightness of that measured histogram like distribution, It has a mathematical definition perhaps with a Gumbel distribution or the EV1 distribution mentioned in Koutsoyiannis’s paper referenced in David Hagen’s comment below.

            Truth is these SC rains and floods have been a very rare event. Thus we cannot fault the press for repeating what the National Centers for Environmental Information has said.

            Is it due to global warming? My judgment says probably yes, since our CO2 level are at historic highs for the past 120,000 years and flooding is happening more frequently since 1990 at several places around the globe (Pakistan, Texas, Alberta, Afghanistan, China to name a few). Read
            for more info on flooding vs date.

            But I’m open to good science to counter that (not just name calling or sarcasm, however). And David Roe you provided food for thought, reasonably free from just sarcasm.

          • Keith Noren says:

            One more thing – that article I provided said a lot more than the single quote you mentioned.

            It also had Michael Mann’s quote:

            “It could happen twice in 10 years,” said Michael Mann, a professor of meteorology at Penn State University. “And, then it might not happen for another 10,000 years.”

            Both Mann and Kunkel are correct.

            By pure stats, the SC rainfall more than qualifies for a 1000 year event given the definitions used. The fact that tails of the distribution sometimes (but not always) occur more frequently than pure stats does not invalidate this defined relative measure of infrequency.

          • Keith Noren says:

            In recognition of the looseness of this relative measure of infrequency, the NCEI labels events as 1000 year, 500 year, 100 year events and nothing more precise than 1 significant digit. Those that complain about it want to take it more precisely than intended when and iff (if and only if) it does not suit their prior biases.

          • mpainter says:

            Keith Noren says “But I’m open to good science to counter that..”
            You would not know good science if it were in your soup.

          • wert says:

            ‘But how rare an event’

            Before calculating rarity you need to define your event space. If you define it post-hoc, you’ll end up with random / arbitrary rarities.

            Given thousand events, there is one that is 1/1000-rare. That is all we know. Assigning rarity based on an assumed distribution is unintelligent. Death by Gauss. Same as calculating percentage of people with negative IQ.

        • mpainter says:

          Truthfully, you deserve no better.

        • Gerald Marts says:

          In measurements there is a certain amount of bandwidth that is need to determine the outcome with any kind of probability that you might be correct. In electronics this is referred to the Nyquist number. Typically a number that is twice the bandwidth is needed. So for a thousand year flood you would need 2000 years worth of data. 100 years does not do it. Below is the definintion of Nyquist rate.

          Nyquist rate relative to sampling[edit]

          When a continuous function, x(t), is sampled at a constant rate, fs samples/second, there is always an unlimited number of other continuous functions that fit the same set of samples. But only one of them is bandlimited to ½ fs cycles/second (hertz),[note 1] which means that its Fourier transform, X(f), is 0 for all |f| ≥ ½ fs. The mathematical algorithms that are typically used to recreate a continuous function from samples create arbitrarily good approximations to this theoretical, but infinitely long, function. It follows that if the original function, x(t), is bandlimited to ½ fs, which is called the Nyquist criterion, then it is the one unique function the interpolation algorithms are approximating. In terms of a function’s own bandwidth (B), as depicted above, the Nyquist criterion is often stated as fs > 2B. And 2B is called the Nyquist rate for functions with bandwidth B. When the Nyquist criterion is not met (B > ½ fs), a condition called aliasing occurs, which results in some inevitable differences between x(t) and a reconstructed function that has less bandwidth. In most cases, the differences are viewed as distortion

  6. c3033 says:

    Thank you for this post. Really helpful when discussing with warmist loons.

  7. Michal Garbers says:

    Thanks for being a voice of sanity in a insane world.

    It is difficult to gauge (pun intended) weather modern civilization increases the affects rains into River Levels or decreases it.

    1) reduction of wetlands
    2) channelizing rivers and streams (down stream increases)
    3) Increased runoff in volume and rate from urban areas

    1) Dams (as long as the Dam does not fail)
    2) Levees (although this can increase the down stream affect. via River channelizing)
    3) Contour Plowing

    While the increases are valid, if the resources are managed correctly the decreases should be predominant.

    The picture of the 1908 flood confirm this was not a 1000 year event.

    Thanks Again,

  8. David L. Hagen says:

    Demetris Koutsoyiannis finds conventional statistics severely underestimate extreme events. His extensive analysis found better distribution shapes.
    The underestimation of probability of extreme rainfall and flood by prevailing statistical methodologies and how to avoid it, Koutsoyiannis, D., , EU COST Action C22: Urban Flood Management, 2nd meeting, Athens, University of Athens, 2006.>Presentation especially slides 15-19

    “It can be shown that the distribution tail of flood is of the same type as that of rainfall Š The EV1 distribution, which has been the prevailing distribution in rainfall underestimates risk significantly”
    “The shape parameter κ of EV2 is very hard to estimate on the basis of an individual series, even in series with length 100 years or more Š However, the results of the analysis of 169 long series of rainfall maxima allow the hypothesis that κ is constant (κ = 0.15) for all examined zones”

  9. Ted Gilles says:

    Dr. Roy: Another excellent report, this time Charleston. I came across the following which is a 44-page comprehensive discourse for us skeptics.

    Ted G.

  10. rah says:

    Tony Heller has done a little records checking. July 1916 S. Carolina saw some pretty severe rains and flooding also.

  11. JeffT says:

    Unfortunately, the general public will believe this. There is very little critical thinking done and people just gobble up all this hysterical news.

  12. Paul Homewood says:

    You don’t even have to go back as far as 1908, Roy

    Hurricane Floyd dumped more rain in 2 days in 1999 on South Carolina, than in 3 this month.

    Moreover this followed Hurricane Dennis that also unleashed untold gigatonnes of rain a week before, thus leaving the ground saturated.

    And I won’t even mention the Great Flood of North Carolina in 1916, which has probably never been matched since (Sorry, just did!)

  13. Denis Ables says:

    Suppose only one really bad natural event happens in each state once in a 1,000 years…….

    If these are evenly spaced out, that’s one bad event in some state every 20 years. (Obviously it doesn’t happen that way, so could be more than one or none in any particular year)

    But…. there is obviously more than one type of bad event (hurricane, tornado, flood, drought, earthquake, sinkholes, etc.) so rare events ( 1 in a thousand years) is likely happening somewhere in the US practically every year.

    The alarmists who insist on associating every dubious event with “climate change” have to do so because they have nothing else. There is not only no empirical evidence showing that co2 (even over geologic periods (when it was much higher than now) has ever had any impact on global temperature – there’s not even a correlation unless the alarmists restrict their claim to a specific cherry picked duration (such as from the mid 70s to about 1998.

    • geran says:

      “There is not only no empirical evidence showing that co2 (even over geologic periods when it was much higher than now) has ever had any impact on global temperature – there’s not even a correlation unless the alarmists restrict their claim to a specific cherry picked duration (such as from the mid 70s to about 1998).”


      (Well stated, Denis.)

    • David Appell says:

      Denis wrote:
      “There is not only no empirical evidence showing that co2 (even over geologic periods (when it was much higher than now) has ever had any impact on global temperature.”

      Wrong Denis. There have been several studies showing this since Harries et al 2001. Here is the latest:

      “Observational determination of surface radiative forcing by CO2 from 2000 to 2010,” D. R. Feldman et al, Nature 519, 339–343 (19 March 2015)

      Press release: “First Direct Observation of Carbon Dioxide’s Increasing Greenhouse Effect at the Earth’s Surface,” Berkeley Lab, 2/25/15

      • geran says:

        Wrong, Davie. Your links only confirm you own pseudoscience.

        (You paycheck might get held up this month due to the Shukla/George Mason flap.)

        • David Appell says:

          Data would be convincing. Your blunt dismissal of observations is not.

          • Aaron S says:


            I think this paper is a great start. But the method needs extended in time to “test” the measurements robustly. Basically, the seasonal signal is equivalent to the decadal trend which concerns me that the singnal to noise is not robust for the green house effect. Also, just because there was warming from the 2000 to 2010 decade does not conclude CO2 did the forcing. My concern is 2000 was a la nina after 97 98 el nino and 2010 was a significant el nino. It is possible (likely) the interval was unfortunate and cherry picked an optimal time (look at satellite data). Again this is great if it holds another decade (2020)

        • geran says:

          Davie, you fear “facts”.

          • David Appell says:

            Feldman et al gave you facts.

            You gave none. You are, as usual, way out of your league.

          • mpainter says:

            Hi David. Your Feldman “facts” are this: a Keeling curve recorded over a ten year interval at a location where the was a cooling trend during that ten years.

          • David Appell says:

            Wrong on both counts.

            The Feldman et al did not use the Keeling Curve — it didn’t need it.

            And the study measured the IR spectrum at the surface, in CO2’s absorption/emission window, not temperature.

            You really should read a paper before trying to dismiss it.

          • mpainter says:

            Hi David. I read the paper. You apparently do not know what a Keeling Curve is. There was a cooling trend. Of course, the paper does not report that, so you are well fooled.
            We are not.

          • geran says:

            Davie, I’m not in YOUR “league”, for sure!

            I don’t use “science” papers for my facts, I use physics books. You probably are not familiar with those.

          • Joel Shore says:

            Oh, the irony…It burns! geran is claiming he uses physics books when in fact he regularly claims things about the greenhouse effect that are directly contradicted by physics textbooks and for which he can’t find support even from physicists who are otherwise skeptical about the severity of AGW (such as Robert Brown, Fred Singer, …), nor even from skeptics who are climate scientists and have a decent training in physics (like Roy Spencer, Judith Curry, …)

          • mpainter says:

            Hi Joel Shore. The greenhouse effect (so called) is real, but it has been mis-characterized by the AGW types. It does not warm the surface. It does not increase tmax, but lowers it. The overall effect of the GHE is to moderate The diurnal temperature range.

          • geran says:

            Joel says: “…he regularly claims things about the greenhouse effect that are directly contradicted by physics textbooks…”

            Joel, if you have time, please provide just one example.


          • Joel Shore says:

            geran: There’s a lot of fodder back in this thread:

            Particularly egregious example is this comment where you show you have absolutely no understanding whatsoever of the greenhouse effect:

          • Joel Shore says:

            mpainter: On Venus, the greenhouse effect certainly does warm the surface. And, it does for the Earth too, where in the the greenhouse effect is necessary to explain why the average surface temperature is higher than 255 K.

          • geran says:

            Joel, don’t you believe your GHE/”back radiation” can warm up a turkey? Are you now in “denial”?

          • mpainter says:

            Earth, not Venus, Joel. I was talking about Earth.

          • mpainter says:

            The Earth is neither black body nor grey body. S-B principles do not apply to the Earth, except to give spurious resultz. The Earth’s accumulates energy and surface temperatures are due to accumulated energy, not to back radiation. The whole AGW hypothesis is founded on error. The GHE does not warm the surface. Insolation does. It is energy that accumulates. Think about it.

          • Joel Shore says:

            No, geran. Because I believe in the 2nd Law of Thermodynamics PROPERLY APPLIED which says that the radiative energy from the warmer turkey that is absorbed by the colder freezer will be larger than the radiative energy from the freezer that is absorbed by the turkey and hence the net flow of energy, or heat, will be from the warmer turkey to the colder freezer.

            Here’s a little hint for you: In your turkey example, what plays the role analogous to the sun in the greenhouse effect example or do you think the existence of the sun is a minor detail that we can just ignore?

          • Joel Shore says:

            mpainter: The Earth is very close to a blackbody in the wavelengths of interest, i.e., the far infrared. And, in fact, the emission spectrum from the Earth as seen from satellites confirms that in wavelengths where there are not strong absorption lines from greenhouse gases, the radiative temperature is about what you’d expect. However, there are “bites” out of the spectrum that match very well the full-fledged line-by-line radiative transfer modeling:

          • geran says:

            Joel, you are so entangled in your pseudoscience you do not even realize when you are being spoofed. The “ice baking a turkey” example is a parody of the IPCC GHE/CO2/AGW pseudoscience. Ice cannot bake a turkey. Earth’s atmosphere does NOT warm the surface.

            See, a REAL greenhouse does “trap heat”. In fact, if the greenhouse owner does not open the windows on a warm day, the greenhouse can kill the plants. But, the Earth’s atmosphere is NOT a greenhouse. It does NOT “trap heat”.

            In a REAL greenhouse, there is no need to “adjust” temperatures. Everything is REAL. No “excuses” necessary, such as “the heat is hiding in the oceans”.

            You have been tricked by the IPCC. And, then you have tricked yourself.

            Leave the “dark side”, and come over to REALITY.

          • mpainter says:

            Joel Shore. I think that you know better. Or you should. BB’s radiate their acquired energy immediately, with no accumulation of energy. The earth accumulates energy and thus surface temperature is higher than BB theory . BB theory does not apply the Earth. Your posit of 255 K as Earth’s temperature is spuriously derived. Earth’s surface temperature is due to the accumulation of heat, not to BB principles.

          • Joel Shore says:

            The “ice baking a turkey” example is a parody of the IPCC GHE/CO2/AGW pseudoscience. Ice cannot bake a turkey. Earth’s atmosphere does NOT warm the surface.

            Oh…So, it is a parody, which means apparently that it doesn’t have to bear any relationship to what you are parodying? Because you can make a humorous statement about turkeys, that somehow shows us that the greenhouse effect cannot exist or illustrates how silly it is? To me it just illustrates your utter ignorance of both physics and what the greenhouse effect is. (Not surprising that in another thread , our host here admonished you “geran, you need to learn some physics,” an admonishment that has apparently fallen on deaf ears.)

            You have been tricked by the IPCC.

            I don’t need the IPCC to tell me about matters of basic thermodynamics. I have a PhD in physics and have written papers in top physics journals in the field of statistical physics, which is the underpinning of thermodynamics. I have taught the 2nd Law to introductory physics students. And, I have come to the same conclusion that the climate-skeptic physicists and climate scientists who I have mentioned have come to regarding your silly statements about the greenhouse effect and your implication that your claims have any support whatsoever from physics.

            And, I’d be curious to have you enlighten us as to what physics textbooks support your view. Here are a few (two introductory texts and one upper level undergraduate text) that explicitly contradict it:

            * Knight, Jones, & Field, “College Physics”

            * Young & Freedman, “University Physics”, 13th edition

            * Kittel & Kroemer, “Thermal Physics”, 2nd edition (which notes its discussion of the greenhouse effect right on the front cover: )

          • geran says:


            (My last comment did not make it through. Testing to see if this one makes it.)

          • David Appell says:

            mpainter says:
            “The greenhouse effect (so called) is real, but it has been mis-characterized by the AGW types. It does not warm the surface. It does not increase tmax, but lowers it.”

            Also, the Easter bunny doens’t deliver eggs, he actually steals them!

            Explains why my mom always made waffles on Easter Sunday….

          • David Appell says:

            geran says:
            “See, a REAL greenhouse does “trap heat”. In fact, if the greenhouse owner does not open the windows on a warm day, the greenhouse can kill the plants. But, the Earth’s atmosphere is NOT a greenhouse.”

            Everyone AND their mother knows this.

            They understand the concept of an a-n-a-l-o-g-y.

            Look it up — I dare you.

          • mpainter says:

            David Appell says: “Also, the Easter bunny doens’t deliver eggs, he actually steals them!”

            Sounds like you grew up in a rough neighborhood: Easter bunnies that mug the kids and steal the eggs.

  14. David Appell says:

    Roy wrote:
    “But a once on 1,000 year event? Sorry, but there is no way to determine that…there are simply not enough rainfall statistics over a long enough period of time to establish such a claim.”

    Come on, Roy, you know what how they’re doing this calculation: they’re measuring water level at some point over time, and calculating its mean and standard deviation, and, assuming the distribution is Gaussian, determining how many standard deviations out the present flood was relative to the mean. Then the probability of such an event is determined from the Gaussian distribution.

    How much historical data would you like? It’s really a question of statistical uncertainty, for the mean and for the standard deviation, which of course depend on the number of data points. The 1/1000 probability also has a statistical uncertainty, but I’m not surprised most media articles don’t give that — it’s pretty wonky — or, sadly, that most journalists don’t even know enough to ask.

    • Grant Ceffalo says:

      Here is a quick and dirty Gaussian-statistics based analysis:

      The probability of having a “1000 year” event in any given century would be 0.1. Having two in subsequent centuries would be 0.1*.01 = .01 = 1%; which is less than the 95% confidence interval.

      The maximum time interval for two such significant floods, and achieve a 5% probability would be a “500-year flood” (447 years) the probability in any century would be 0.2236, giving the chance of having two in subsequent centuries as 0.2236*0.2236 = .05 = 5%

      Using a common coin toss event, the coin would have to come up “heads” seven times in a row in order for the 2015 flooding to be a “1000 year flood.” If it is a 500 year flood, it is four or five coin tosses.

      However, given the other flooding events pointed out, the frequency for this event seems to be less than 100 years. As Roy correctly states, with less than 1000 years of history in this location, all we are doing is ESTIMATING the frequency, while we are constantly changing the flood-contributing conditions constantly (as other posts illustrated).

      I would be willing to guess the individual first person calling this a “1000-year event” was not trained in statistics, but did like big round numbers. Everyone after that was using a sound-bite.

      • David Appell says:

        I want to see the statistics from the meterologists who have followed this event in detail and made the calculation, not what any anonymous Joe has written in a blog comment.

        • mpainter says:

          Translated: I want to know what Jagadish Shunkla sez.

          Hi David. What’s wrong with Roy Spencer’s evaluation? Is he not qualified by your lights?

          • David Appell says:

            Roy presented no statistics that are relevant to the question.

            (A photo doesn’t constitute statistical evidence, but thanks.)

          • mpainter says:

            Statistics “revelant to the question” would be based on 2,000 years of data, at the minimum. Statistics based on 150 years of data is simply more fabrication by the AGW zealots, who have no other kind of science.

        • geran says:

          Davie says: “I want to see the statistics…”

          No, you want to see “adjusted” statistics.

      • Keith Noren says:

        I like that you get down to numbers, but I’ll point out:

        – that the statistical distribution used for this parameter – rainfall amount per unit of time and over a defined area- cannot be Gaussian. For instance, a histogram of rainfall amount over say 1 hour in a given area is certainly bottom weighted (i.e.. their wlll be more hours where 0 rainfall occurs than any other amount). Thus we are stuck with dealing with the upper end distributional shape and what is termed extreme value theory. Read:

        – your guess that the fist person calling this a 1000 year event just “did like big round numbers” is cynical. Truth is the NCEI who made this announcement only uses big single significant digit numbers in recognition of the uncertainty involved but still is desirous of a telling the public some relative measure of “how rare”. To understand that this was not just pulled out of their behind, read:

        Previously posted.

        • mpainter says:

          Can’t be hard science w/o data. It falls into the category of alarmist propaganda.

        • Grant Ceffalo says:

          You accurately point out that it is not a Gaussian event; I only used Gaussian statistics because Appell called for a Gaussian analysis. This kind of event is more accurately represented by a lognormal, Poisson or even binomial distribution; the frequency of events renders the event type ineligible for Gaussian analysis. I fully appreciate that a few events can be used to estimate the probability of extremely improbable events, including multi-billion year extinction events.

          However, this CNBC article does not actually cite any numbers whatsoever, and just explains that “there is a process.”

          I am well familiar with measuring and estimating improbable events, which is why I agree with Appell in stating that I would like to see the actual statistics used to generate this “1,000 year flood” statement. The photos and other evidence do seem to be trout in the milk.

          • Keith Noren says:

            I am familiar with statistical estimation of improbable events as well from my work experience (missile defense optical discrimination – Doug Lambert can attest to that). Also familiar with the earthshine spectrum with it’s 12-20 micron CO2 absorption band (not yet saturated) and how increased CO2 concentrations affects earth’s thermal equilibrium (gradually sinking into the oceans where the pH has increased and 40% of phytoplankton – bottom of food chain- have been killed off)

            I too would like to see the distribution assumed by the NCIE and actual histogram involved. But given the 1000 year rainfall was quoted as being at 17”/hr level and the actual rainfall rate was 20”/hr, it is clear that it was more than a 1000 year event that has real numbers behind them. I’m not so cynical as to believe it was all picked out of thin air.

            This sort of posturing by GW skeptics (which is what all this website is about) shows that it is sort of pseudo-science whining with a heavy political motivation.

            mpainter can call me what he wants, but I’m really the realist here.

        • David Appell says:

          Keith, the question is about deviations from the mean rainfall, not the amount of rainfall itself….

  15. Bob Burban says:

    No amount of slick sophistry can match the two pics provided by Roy Spencer in this posting.

  16. Slipstick says:

    “Sorry, but there is no way to determine that…”. Really? Somebody should tell the Hydrometeorological Design Studies Center they’re wasting their time. Regarding the 1908 flood vs. the current event, one should consider the changes in flood control, including the Saluda Dam, built in the late 1920’s, that have taken place since 1908.

  17. Vincent says:

    This immediate reaction of a ‘once-in-a-hundred-year flood’ or even a ‘once-in-a-thousand-year flood’, every time a major flood occurs, is something that is typical in Australia, the land of droughts and floods.

    We had a major flood in Brisbane, Queensland, early 2011, and a bit earlier in Rockhampton, north of Brisbane, Dec 2010-Jan 2011, as a result of the same weather event.

    I remember vividly the news reports at the time. Worst flood on record, and a confirmation of the effects of AGW.

    However, the Australian Bureau of Meteorology has records stretching back to 1859. Check out the following link.

    In terms of flood height, here one can see that the 2010/2011 flood in Rockhamton was not the worst flood on record. It wasn’t even the second worst, or the third worst. It was the fourth worst, although fourth by a narrow margin of 10 centimeters.

    The worst occurred in 1918 with flood levels of 10.11 meters. By comparison, the ‘initially claimed’ worst flood on record in 2011, rose to 9.20 meters, almost 1 meter lower.

    What is alarming for me, is the fact that the local council in Rockhammpton was still considering proposals resulting from an inquiry into the previous major flood in 1991, when the 2011 flood struck.

    As I’ve written in a previous topic on Fred Singer, on this site, we have a problem in sacrificing our long-term future for the sake of immediate economic gain.

    Let’s suppose that AGW, in addition to natural climate change, is in fact a reality. Why would anyone imagine that we could be capable of tackling the problem if we can’t even tackle the effects of natural, extreme weather events, the occurrence of which are definitely more certain than the effects of AGW?

  18. mpainter says:

    Ha! Wanna see a flood? See Pecos River flood, 1954. A flash flood that crested at 94.6 feet. Took out a steel truss bridge at 65 feet and the concrete piers to boot.

  19. John Moore says:

    I don’t know if they used it, but don’t hydrologists and geologists have data about flood extent and frequency going way back in time? They don’t need to measure precipitation – presumably they can use coring to find the history.


  20. dp says:

    I wonder how far below the surface the river bottom was then compared to today, and what the capacity of the basin is today compared to then. I suspect dredging has taken place over time but don’t know for a fact. Flood levels are a combination of water volume and basin capacity. They are also a function of rate of flow which can have several causes including rate of rainfall and the number of affected confluences.

    I’m not convinced we know everything needed to make this comparison based on evidence at hand.

  21. MarkB says:

    . . . , but when someone claims that a weather disaster is 1 in a 1,000 year event, they need to back it up.

    The methods are as described in this document:

    Handy online precipitation frequency calculator:

    • mpainter says:

      No, Mark B. Not with a record that goes back for only a century or two. Guesswork does not count as reliable data.

  22. Stevek says:

    I’m sure somewhere in the universe, some planet is having a 1 in a million year rainfall.

  23. Laurie says:

    Here ya go . . . Glad you said it . . . So I didn’t have too.
    Has some really pretty and chatty pictures!
    A few pictures . . . [alright charts and graphs]

    Ok not as good as the first link in my opinion but, it will work . . .

  24. Jim Dean says:

    The Churches in South Carolina have been busy the last couple of days delivering water, food, towels, blankets and toiletries to the areas of need as soon as the roads are clear enough to get through. Where the roads are not open, we’ve shown that where there is a will, there is a way. Fishing boats, pontoon boats, rafts, canoes, kayaks have also been used to get much needed supplies to those in trouble. My church has plans to get 35K more bottles of water, 10k cans of food, several thousands of other supplies to our troubled surrounding areas by Saturday night, no matter what it takes. All water, food and other supplies have been brought to the church by patrons and local businesses. South Carolina is showing the country how it is supposed to be done the same way we showed the country what unity looks like after the Charleston massacre. It is encouraging to see so many willing to help others in times of need. May God Bless the giving people of South Carolina!

  25. dave says:

    !…1 in 1000 Year Event…”

    This does not seem to have much to do with statistics.

    Instead, it is more the standard “signs and portents” meme.

    To be be followed by the standard “repent warning”:

    “At that time, there will be wailing and knashing of teeth!”

    “But I don’t have any teeth.”


  26. dave says:


    Or even gnashing.

  27. WizGeek says:

    Seriously? I find it amusing that most everyone seems obsessed with needing to appear erudite by painting all other opponents as inane. Let’s all take to heart civility in discussions. Address the topic, not the person. Present your logic with supporting detail rather than levy allegation and opinion. Take time to consider opposing assertions, do some investigative research, and strive to increase your knowledge and understanding. Give people time to grow in their understanding–it may take years. Stay focused. Be courteous. Not one of us has all the answers or has a complete understanding of any subject matter. We must all learn from each other in the most collaborative way, or we most certainly will be ambushed by the onslaught of ignorance while we’re foolishly distracted.

  28. Alan McIntire says:

    I checked to see that South Carolina makes up about 1/1600 the land area of the world, so an area that size SOMEWHERE in the world would show 1 in 1000 record flooding or drought twice that- or about

    2/1000 time 1600= 3.2 times per year we should be reading about record events like that SOMEWHERE in the world.

    • mpainter says:

      If you try to explain that to the alarmists, they do not listen. If it’s a flood/drought/hurricane/blizzard, it’s CO2. That is why they have lost credibility (rather, one of the many reasons).

      • Norman says:

        Alan McIntire & mpainter,

        Great posts and I agree completely. That is why Climate Science is no longer a science but a propaganda tool for a Green agenda.

        If it were to be a real science anytime one invoked a slight global warming as the cause of catastrophic weather one would have to demonstrate clear and logical mechanisms as to how a slight temperature increase could then explain a 2 foot rain fall in a small local region. The meteorological explanation given on WUWT was that the hurricane blocked and turned the normal flow of water vapor toward South Carolina which had a low pressure system above it. This could easily explain the high rainfall experienced rather than say it was proof that global warming is leading to catastrophic weather. Thanks again.

        • mpainter says:

          Probably most tropical depressions moving ashore carry with them the potential for the sort of extreme precipitation that deluged poor South Carolina, depending on whether meteorological circumstances combine in similar adverse fashion.

  29. Am currently reading Bill Bryson’s Book “One Summer: America, 1927”
    He recounts about the Great Mississippi Flood of that year, which was easily the worst river flood in the history of the United States. What was that about climate becoming more extreme?

    • Eli Rabett says:

      The course of the river (and many others such as the Tennessee) have been changed to limit flooding. TVA was a response to really bad floods which had the side benefit of many hydroelectric dams. The electricity was useful for isotope separation at Oak Ridge (why do you think it was put there) as well as industrializing the south.

  30. Ben Sturgis says:

    You need to understand what is meant by “1000 year event”

    1-in-1,000 chance that a rainfall of such intensity will happen in a given year.

  31. Bart Thompson says:

    Dr. Spencer,
    While all sorts of claims may have been made by the media, the original media claim on a local station in Greenville, SC (where I reside) was different.

    The original language was “1 in a 1000” event. That meant (according to the news anchor) that “there is a 1 in 1000 chance of such a flood in any one year.” This is different from the “once in 1000 years” subsequently used by other media.

    The probability calculation would be (I think) like this:

    The chance of the flood not occurring in one year is 0.999. The probability of it not occurring in a period of “n” years is (0.999)**n. So, in a 100 year period, the chance of it not occurring is 90.4%. So, there is a 10% chance that it would occur over that time frame.

    There is another fact to consider. Here is South Carolina, water management is much better than it was 100 years ago. We have better control over water release via a network of controllable dams with large man-made lakes. Thus, for a given amount of rainfall in a 48 hour period, the actual flooding would be mitigated now, compared to an earlier event of identical rainfall. Hence, earlier floods could have been more severe in terms of water crest, yet the total rainfall could have been less than that seen in SC last week.

    Certainly, I’m not supporting alarmist news reports … just saying that, in this case, the original statistic does not seem completely far fetched.

  32. mpainter says:

    Such a concept is useful as an excuse when dams, bridges, etc. get swept away or when levee systems are overwhelmed. Then the parties responsible can say “only once in a thousand years”. But the concept is meaningless when there is only a century or two of data. Meaningless, but very useful to the alarmists.

  33. Eli Rabett says:

    Ask not whether the storm was a one in a thousand year event caused by climate change. Ask about the cause of the high sea surface temperatures that fed the storm

  34. Eli Rabett says:

    Oh yeah, who the hell let that truck move over the bridge when the footings were covered with water. Seriously

  35. mpainter says:

    Ask wherefore art the hurricanes that the alarmists screeched about.

  36. jimc says:

    Beyond my comprehension level, but interesting.

    “He has fixed two errors and the new corrected model finds the climate’s sensitivity to carbon dioxide (CO2) is much lower than was thought.”

  37. Jim Dean says:


    Thank you for your kind words. I’m starting to feel my age after helping 5 other people move 50K plus bottles of water (several times) over the last 4 days. I’m grateful that we had the support of so many people that donated time and supplies so our church could be as effective as it has been this past week. We are still chugging away serving the coastal areas, the coastal areas are the last areas to reach there peak water levels. I’m relieved to see the midlands and the low country water levels steadily declining. It’s going to take months to get things back to some semblance of normalcy. Thanks for all prayers and support. Jim

  38. Billagain says:

    I was surprised when I heard the “1000-year flood.” The first thing I thought of was where the hell was the airport in 1015?

    I’ve been in a couple of those rains in Florida. Happens with stalled fronts or static nor-Easters. I was in St. Petersburg for an 18-inches in 18 hours event in the ’70s. About twenty years ago, downtown Jacksonville flooded with over 6 inches of rain but barely wet the beaches nineteen miles away.

  39. boris says:

    did anybody notice the 1908 flood was in August?

  40. Gunga Din says:

    I really think that the “1 in a 1000 year” or “1 in 1 year” flood nomenclature should be changed.
    It seems to really refer to a 1 in a (plug in the number)chance that such a flood will occur in any given year.
    The caGW hype aside, that nomenclature can give a false sense of security. If a “50 year flood” occurred last year and someone is in in their 50’s, they might think that building a home in a “50 year flood plain” is perfectly safe for the rest of their lifetime.

    • lewis says:

      Gunga Din:

      Agreed. Those who plan for flood mitigation are prone to use words such as 1000 year flood or 100 year or 500 year in order to generalize how often one should occur, not really making any type of prediction. They are well aware of the fact that a historic flood this fall does not mean there won’t be one next year. Or the year after.

      More to the point is the planning of flood plains and plans based on these 1000 year floods. Should flood plains be restricted from building? Perhaps flood plains should be for agriculture only? Wouldn’t NYC and Miami and New Orleans enjoy those new rules?!

      • Bart Thompson says:

        Gunga Din, Lewis:

        Per my earlier post, the original newscast in SC claimed “1 in 1000 chance of flood occurring in one year.” Whether or not it’s right, this is a number amenable to statistical treatment.

        In 100 years, you have about a 10% chance.

        In 1000 years, you have about a 63% chance.

        In any one year, you only have a 1/1000 chance of it occurring, regardless of history.

        • dave says:

          “In 1000 years you have about a 63% chance [of exactly one such flood].”


          an 18% chance of exactly two such floods in 1000 years,

          a 6% chance of exactly three such floods in 1000 years,


          a Poisson process,

          no change in the rate.

        • dave says:

          “…a 63% chance…”

          Actually, a 63% chance of one or more floods in 1000 years.

  41. dave says:

    There is something about the number “a thousand” that brings out foaming-at-the mouth excitement. Thus Neil Kinnock, the sometime Leader of the Labour Party in England, boasting about reforms in access to education supposedly brought about by socialism, famously said,

    “Why am I the first Kinnock in a thousand generations to be able to get to University?”

Leave a Reply