Unnatural Catastrophes

In the last post we looked at counts of weather-related natural catstrophes according to Munich Re, the giant re-insurance corporation. It was mentioned that part of the increasing trend could be due to changes in the way catastrophes are counted, and in the number of people and value of property which is susceptible to such catastrophes. Although Munich Re adjusts their classification each decade to account for this, any such attempt is bound to be imperfect.

But it was also pointed out that Munich Re tracks catastrophes which are not weather-related (earthquake, tsunami, volcanic eruption), classifying them as “geophysical.” Do these geophysical events also show a trend? If so, are they increasing as fast as weather-related catastrophes?

Munich Re provides a nice graph of the number of all catastrophes, including geophysical:

Geophysical catastrophes are shown in red, the bottom part of each bar. There’s no visually obvious trend, but that’s a question which shouldn’t be decided on the basis of visual inspection so I digitized this graph to get some data. It turns out that there is a trend in geophysical catastrophes, and it’s statistically significant:

The overall rate of increase is only 0.8 per year per year, but it’s not zero. In fact, the overall trend is not linear, a quadratic trend giving a much better fit:

It’s not impossible that entirely natural changes have caused significant decadal variation in geophysical catastrophes. It’s not impossible that global warming has done so! The disappearance of landfast ice caused by man-made climate change affects tectonic stresses, which could impact the likelihood of earthquakes, tsunamis, even volcanic activity.

But it’s also possible that the observed changes are due to demographic/financial factors, the number of people and value of property vulnerable to catastrophes, or due to the way catastrophes are classified and counted. It would be nice to be able to “normalize” catastrophe counts to remove such influences, but properly accounting for changes in population, property value, building codes, advances in early-warning and in recovery-relief technology, even political changes which can impact disaster preparedness and response, is a very thorny issue indeed. When such normalizations are done by those with an agenda, they are too easily skewed one way or the other by the preferred outcome.

But there is a way to normalize weather-related catastrophe counts which does not depend on assumptions about the impact of all those complex factors. It is, to compare the changes in weather-related catastrophes to those which are not weather related. Just this approach was taken by Munich Re itself, as shown in this oustanding video; at the 5:35 mark you can hear the voice of Peter Hoppe from Munich Re, and at 6:00 he directly compares weather-related to geophysical catastrophes, noting that “… they have gone up by far more, and we only can explain this by a change also on the hazard side, which means a change in the frequency and intensity and as you have seen in the slide before we definitely have this.”

Essentially, we could use the counts of geophysical catastrophes to normalize the other types of catastrophe. Just as an experiment, let’s do so — and see what happens.

I decided to “normalize” the weather-related catastrophes by a smoothed estimate of the number of geophysical catstrophes. How to smooth? I’m very fond of my “modified lowess smooth” and also of the plain-vanilla lowess smooth. But there are other ways too, for instance something as simple as a Gaussian smooth (a moving average with Gaussian weights). Here are smoothed values using each of those three choices:

The Gaussian smooth shows less variation at the endpoints because instead of continuing the trend, it tends to do the opposite. Since it’s more “conservative,” showing less variation, I adopted the Gaussian smooth for normalization. I divided the smoothed values by their average so that the normalization function would have a mean value of 1:

Then I divided the value of each of the three types of weather-related catastrophes by this function to define normalized counts for weather-related catastrophes per year. And here they are:

All three classes of weather-related catastrophes show an increasing trend. All three trends are statistically significant. Of course the rates of increase are slightly less than when using non-normalized data. Catastrophic storms have increased by about 5 per year. Catastrophic floods/mass movements by about 6.6 per year. Heat/drought/fire catastrophes about 2.5 per year. For total weather-related catastrophes we’re seeing an extra 14 catastrophes per year:

Even when normalized by geophysical counts, the number of weather-related catastrophes has more than doubled in a mere 30 years.

I’m certain that some people will find an excuse to “explain away” this rather obvious trend. That’s what those in denial do.

I’ll agree with Peter Hoppe, that “… we only can explain this by a change also on the hazard side, which means a change in the frequency and intensity …

59 responses to “Unnatural Catastrophes

  1. Nicely done Tamino.

    Another way to avoid the imperfect and caveat-riddled economic corrections/adjustments (funny, I thought contrarians and “skeptics” were opposed to adjusting the data?) is to just look at the area affected by severe weather in the USA (the so-called Climate Extreme Index hosted on NOAA’s site). Guess what those data show? Yup, a marked increase in the Climate Extremes Index since circa 1970. Don’t take my word for it, go an have a look (Roger Pielke Jnr. included):


  2. David B. Benson

    Well done.

  3. The CEI data are all the more intriguing given that Roger Pielke Jnr says on his blog,

    “Of course, as we have written many times, if your goal is to detect trends in climate phenomena, then it is best to look directly at actual climate data.”

    OK, now will Roger accept the upward trend in the CEI data? Meh!

    • PIelke ignores that flooded/burned/wind-flattened buildings are a handy tool for integrating a very complicated signal; a kind of summing proxy instrumentation.

  4. “The disappearance of landfast ice caused by man-made climate change affects tectonic stresses, which could impact the likelihood of earthquakes, tsunamis, even volcanic activity.”

    I think this is highly unlikely. The scale of tectonic forces compared to the scale of forces from recent glacial melting makes this seem highly improbable. Also, I think the 32 years of data shown for the trend of geophysical events is far too short a time period to establish a trend. Much longer geological time periods would be required (100’s or 1000’s of years?). Thirty two years is noise on a geological time scale.


    • As a geologist, I’ll second Ken’s opinion on this. Loss of glacial ice could have some local isostatic effects but most of the larger isostatic effects are related to the loss of large scale ice sheets. The isostatic rebound from the loss of glacial ice sheets is unrelated to the larger (much) and deeper tectonic forces responsible for major earthquakes etc.

      • Is the “disconnect” due in considerable part to thinking more qualitatively than quantitatively? We use ‘earthquake’ to denote movements the energies of which span many orders of magnitude, right? (I’m sure tokodave could put some numbers on that.) But as a suggestive analogy, there are muscular tics and then there are grand mal seizures!

        Quakes due to tracking are presumably very different phenomena, qualitatively speaking, than the ‘major quakes’ spoken of in the last couple of comments.

      • There appears to be a bit of a consensus forming in Iceland that Vatnajokull may pretty much cease to exist as an coherent icecap within the next century (and by default the same could be suggested for Mydalsjokull, i.e. the one atop Katla). Given what we know about the Laki (Scafta Fires) eruption, as well as the 2010 ‘problems’, I think that isostatic effects on local SL are not the only thing we need to be worrying about in relation to glacial unloading. Hugh Tuffen’s work on this is quite illustrative

      • “tracking” = “fracking”, of course. Blast that auto-correct…

      • There are other things to take account of, for example withdrawing water from underground acquifers, deep mining, heavy building (Dubai, Manhattan, etc), eg. anythings which loads and unloads the system.

    • Is there a disconnect here? I would have been blindly accepting of these comments by KenH & tokodave a year ago. But since then I think I’ve read that fracking causes little Earthquakes all the time. That damming rivers has changed the tilt of the Earth and speed of rotation. That sea level and its distribution have changed (and ocean currents too have been affected by climate change). As someone without much geology or physics training, it’s hard for me to imagine that all of these changes (even if minor) can’t add to or change the tectonic stresses at all and shift the probabilities somewhat. Even small weights can affect the outcome of a horse race.

      • Fracking and earthquakes are a relatively near surface phenomenon, as is isostatice rebound. Bear in mind most major earthquakes are on the order of tens of km in depth and are caused by the shifting of the earth’s tectonic plates; for example, you’re moving one piece of North America past another. I’ve seen some of the articles you note and I’m a bit skeptical of some of the claims, but have not studied the specifics. And of course, as with everything related to fracking, whether the earthquakes are actually related to fracking is sometimes a matter of opinion. The US Geological Survey thinks they can, and that’s good enough for me.

    • Humans have speculated on atmospheric to geologic linkages for quite some time; recent science has found evidence of rain triggered earthquakes:


      But the effect is… well, not Discovery Channel disaster material, anyways. A random Mount Tambora style event piling on top of the ongoing anthropogenic stresses could be very challenging, but that would just be bad timing.

    • Fracking and earthquakes are a relatively near surface phenomenon, as is isostatice rebound.

      Isostatic rebound depends on the viscosity of the mantle to function. Crustal deformation necessarily reaches to the mantle or isostasy couldn’t exist. The top of the mantle below continents is in range of 30-70km deep.

      • We need to distinguish between visco-elastic isostatic effects and elastic effects. As Doug says, visco-elastic effects depend on the viscosity of the mantle and unfold regionally on timescales of tens of thousands of years. Elastic responses, although smaller, are instant and local.

        This means that we may well see a climate-related increase in volcanic and earthquake activity in areas close to (within kilometres of) retreating glaciers, such as Iceland, the Pacific Northwest and southern Alaska.

      • Maybe what I should have said for a more detailed response…Following deglaciation several thousand years ago, recent surface manifestations of isostatic rebound have generally been limited to near surface observations; changes in various shorelines, river courses/steepness and shallow earthquakes as the crust continues to adjusts to the change in stress following deglaciation. It’s possible the recent earthquake in New England would be an example of relatively shallow minor quakes related to this but a quick look didn’t find that. Ice sheets constitute a static load on the crust as opposed to the dynamic stress from tectonic forces. It is certainly possible in some cases for there to be a link between the dynamic and static stressing of the crust/pre-existing structures etc. but I’m not aware of any large scale earthquakes related to loss of glacial ice and neither was the state seismologist here (Montana).

    • There is some evidence of impacts on volcanic behaviour in Iceland in the last few years:


      also worth checking out is:


      For a general perspective on earthquakes and climate driven processes, see:


  5. Horatio Algeranon

    Oxfam published an interesting report (Time’s Bitter Flood: Trends in the number of reported natural disasters) in May 2011 that summarizes attempts to “subtract out” effects like increased population and possible increased reporting of disasters.

    The report uses EM_DAT (a different database than that used by Munich RE) which uses the following definition for “disaster”

    Disasters are entered into EM-DAT when at least one of the following criteria has been fulfilled: ten or more people reported killed; 100 people reported affected (i.e. requiring immediate assistance in a period of emergency); declaration of a state of emergency; or a call for international assistance.

    and concludes

    There is an upwards trend in the number of reported disasters. This is chiefly driven by a steep rise in reported floods in all regions and, to a lesser extent, storms in Africa and the Americas. When weather-related disasters are analysed separately, the average rate of increase is 4.1 per cent per year for the sample of countries which have a first disaster reported from 1980 (a rise of 233 per cent over 30 years), and 4.9 per cent per year for countries whose first report was from 1990 or before (159 per cent over 20 years). The rise in the number of reported disasters, and of floods in particular, is broadly supported by independent data.
    An increase in the number of people exposed to disasters (approximated by population growth) partly explains the trend, but not fully. It is unlikely that reporting bias fully explains the trend either: the methodology used was designed to minimise reporting bias, and both of the methods used to evaluate any remaining reporting bias reduced, but critically did not eliminate, the rise in reported disasters. Although it was not possible to estimate directly, it is therefore possible that an increase in the number of hazards is responsible for some of the increase in reported disasters, even if only a small part. This is consistent with the reported increase in extreme weather events across many parts of the world. The implications of a continuing and steep rise in the numbers of disasters for the millions of vulnerable people living in developing countries are stark.

    The report further states that

    It was not possible to directly analyse the effect of climate change on disaster trends; however, there is insufficient evidence to exclude the possibility that climate change is increasing hazards and hence trends in reported disasters. This effect is unlikely to be very large, because the magnitude of climate change over the past 20-30 years is relatively small when compared with (for example) the growth in the world’s population over that time.

    If Horatio is not mistaken, the recent work of Hansen et al indicates that relatively small shifts in the climate distributions (temperature, humidity, etc) can lead to relatively large changes in the frequency of extreme events (heat waves, deluges), which would seem to be relevant to the latter claim: “This effect is unlikely to be very large…”

    • Horatio Algeranon

      Analysis of the number of reported disasters adjusted for population change (‘normalised’11) shows a statistically significant upwards trend (Figure 7). Weather-related disasters increase by 2.1 per year, compared with 3.4 per year in the non-normalised data (Figure 3), suggesting that increased exposure makes a considerable contribution to the increased trend in reported disasters. Nonetheless, the upwards trend is still statistically significant, indicating that hazards and/or reporting are increasing over time.

      So, when specifically adjusted for population, the trend in total weather related disasters was scaled downward by about 2.1 / 3.4 = 0.62.

      The normalization method used in this post is quite different (and performed on a different database, with different criteria for what is included as a disaster/catastrophe), but scaled the trend in total weather related catastrophes by about 14/18 = 0.78.

  6. Thanks, Tamino.

    It’s not just ice. In one of the AGU meets, last year I think it was, a researcher was showing how the change in monsoons could affect earthquakes – and did IIRC.

    This isn’t the video I was thinking of, but it looks as if it’s on a similar topic.

  7. OK, I’m thinking that the “skeptics” will now want to ban the teaching of statistics, because your consistently clever use of real world data to counter their, at first sight, plausible arguments is bringing them undone.

    Well done.

    • Good point, John Brookes: ESS (Ever Since Sandy) I have noticed a distinct increase of evidence of a *disturbance in the Farce,* i.e., the denialosphere is showing signs of self-ingestion; IOW, they’re beginning to eat their own. This is readily apparent at WTFiUWT, and somewhat at RPJr’s blog. I would be, as a working scientist, lying to say it doesn’t provide with a certain sense of schadenfreude to se them at each others’ throats; however, the time we have lost, and are continuing to lose, due to the Big Oil-funded tactics, overwhelms that with more than a bit of sadness, for the time and opportunity lost, battling these idjits.

  8. Pielke’s argument is found in his WSJ piece. “http://online.wsj.com/article/SB10001424052970204840504578089413659452702.html?mod=googlenews_wsj
    “In studying hurricanes, we can make rough comparisons over time by adjusting past losses to account for inflation and the growth of coastal communities. If Sandy causes $20 billion in damage (in 2012 dollars), it would rank as the 17th most damaging hurricane or tropical storm (out of 242) to hit the U.S. since 1900—a significant event, but not close to the top 10. The Great Miami Hurricane of 1926 tops the list (according to estimates by the catastrophe-insurance provider ICAT), as it would cause $180 billion in damage if it were to strike today. Hurricane Katrina ranks fourth at $85 billion.

    “To put things into even starker perspective, consider that from August 1954 through August 1955, the East Coast saw three different storms make landfall—Carol, Hazel and Diane—that in 2012 each would have caused about twice as much damage as Sandy.

    “While it’s hardly mentioned in the media, the U.S. is currently in an extended and intense hurricane “drought.” The last Category 3 or stronger storm to make landfall was Wilma in 2005. The more than seven years since then is the longest such span in over a century.”

    On his blog he has references to ICAT and to various articles. I would be a bit cautious about referring to this as denial. Its properly reasoned and documented. You can perfectly well accept the reality of global warming, also that probable implications include an increase in extreme weather events, but still be skeptical that we are actually seeing such events right now, still less that they can be tied to the degree of warming we have so far experienced.

    [Response: Using the lack of demonstrable increase in losses to suggest that the large, nay huge, in fact absolutely undeniable increase in the number of disasters is just natural variation, or even nonexistent, is a vile form of denial.]

    • Pielke’s comparison is simply misleading, bogus and wrong.

      To compare previous damages with current ones and then account only for inflation is stupid. We usually have now several days in advance to prepare for the arrival of a hurricane (evacuate people, bring objects of value to a safe place (e.g. expensive things in a house upstairs to reduce flood damage) in a very precisely known area. That obviously leads to less damage and less casualties. Also our houses are now built with all the knowledge that we built up over the decades in how to reduce the damaging effects e.g. of strong winds etc. The people in the great 1926 Miami hurricane didn’t have satellite assisted accurate hurricane path prediction, sophisticated evacuation planes and real-time monitoring of the situation via internet/television.
      To suggest, that the only difference between the damage of a hurricane in the 1920s and a current one is only inflation is simply bullshit…

  9. So, no Roger yet.

    I wonder why?

    [Response: Probably because I informed him in no uncertain terms that he is not welcome here.]

    • His presence would not be useful either. His responses on skepticalscience on one of his misdirections was “come to my blog where all the issues have already been discussed”; a very clear non-desire to respond to the actual issues.

      And as Stefan Rahmstorf can tell you, whatever you say to Roger, he *will* twist it such that you concede that his interpretation is right.

      • guthriestewart

        Reading what others have helpfully posted, it appears Roger is guilty of acting as if the numbers and causes of disaster damage are settled (oddly enough in a way which agrees with his opinion), when in fact they aren’t.

  10. interesting analysis! a question and a comment. question: is there a significant difference in the trend in individual weather-related catastrophes using the lowess or modified-lowess smoothing to normalize? i don’t think there would be, but while the gaussian is the most conservative in terms of the overall variability (minmax ~ 68-100), the lowess and m-lowess (minmax ~ 58-100, via eyeballing method) look like they would weight the 1980-1983 period more heavily.

    comment: i’d suggest the last plot needs a legend to explain what the red curve is (mod-lowess?).

  11. After each big quake, assessment of what did and didn’t work informs not just building codes but also what individuals do.

    Shearwalling — putting plywood on the ‘cripple wall’ studs between foundation and floor joists — is an example.

    I talked to one guy who went down a line of fallen houses after a big quake and looked at the one house on the block that hadn’t gone sideways — the owner had gotten nervous a few days before their big quake and tacked a few sheets of plywood up on his cripple walls with a few dozen nails.

    That was far, far less plywood and far fewer nails than the code spec.

    But it was enough for that particular event. HIs house didn’t go sideways off its foundation, when the neighbors did.

    Point is — for geophysical disasters — information about what prevents or reduces damage now propagates _much_ faster than it did decades ago, and perhaps people get better at paying attention.

    For some kinds of events — like earthquakes — the individual can make a big difference in what happens to one individual building with hammer and nails.

    For climate change events —

    — don’t forget to vote.

  12. You can add Federal Crop Insurance indemnity to the list.
    Thanks to MapleLeaf for the CEI reminder.

  13. I think using the linear fit of the geophysical catastrophes would be better. You are attempting to detect and remove unrelated factors which we have some evidence would be increasing with time (population, wealth). Ideally, the geophysical catastrophes themselves would have no underlying physical time trend, so any trend that does show in this series would be the bias you are trying to remove from the other series. Instead, your quadratic and lowess fits given their shape are capturing fluctuations in the geophysical catastrophes themselves (in addition to any methodological bias trend which is what you are really after) and you are then removing this signal from the weather series. That doesn’t make any sense.

    [Response: Perhaps. But perhaps the nonlinear trend reveals the nonlinear pattern of how geophysical catastrophes are counted. Bottom line: either way, the increase in weather-related catastrophes is undeniable.]

  14. Poul-Henning Kamp

    It is quite interesting that your quadratic fit peaks during the abnormally long interval between the dec98 and dec05 leap seconds.
    We don’t know why there was such a long interval, leap seconds “normally” occur at a intervals of 18 months, but unlike leap-days, they are too unpredictable to be scheduled centuries in advance.
    In practice they are announced 6 months in advance, based on observation relative to a stable inter-galactical coordinate system.
    On teory is that friction between the fluid between core and the mantle, or possible particulate (ie: mountains) in this fluid is responsible.
    If that is true, it is not a far stretch to think that the stress and energy has to go somewhere else.

  15. On the subject of graphing catastrophic events, I’m wondering if anyone has seen the data for Sandy’s impacts put to graph.

    For example, days since landfall versus:

    1) number of homes without power, or
    2) number of people displaced from their homes, or
    3) cumulative estimate of damage costs, or
    4) number of businesses/enterprises shut for business, or
    5) number of people without access to fuel or basic services, or
    6) (sadly) number of deaths directly or indirectly attributable to the storm.

    I’d like to see the “it wasn’t even a hurricane” crowd rationalise such depictions…

    • The ado over whether or not it was a hurricane is pretty pointless, isn’t it? It tempts me to regard it as another example of magical thinking by elements of the WUWT crowd: they really appear, some of them, to think that verbally ‘proving’ something has the power to change the physical reality.

      As your comment suggests, it matters little in any immediate sense to a victim just which meteorological category best fits the storm that tore their life apart.

      • Metaphysical speculation—wondering whether one has been stabbed with a spear, or a lance—is a distraction at best, and certainly provides no support for the thus stabbed, nor insight into the root causes and how to avoid or minimize them moving forward.

    • Bernard, it occurs to me that a quick and dirty way to do that would be to use my Sandy article. I did daily updates ’til now, invariably including the reported death toll and usually some reported measure of power loss.

      So you’d be able to pull numbers for your #6 from one source, and maybe for #1, too.

      It’s here, if you want to give it a shot:


  16. Heads up Tamino. Some choice quotes of assertions made by Pielke Jnr.:

    “Unfortunately (and tellingly), some of the activist blogs on climate purportedly interested in science have banned me from commenting or edit my comments to delete substantive content.”

    “However, on climate time scales there has not been detection (much less attribution) of increasing disasters (intensity or frequency) to human-caused climate change”

    “The peer reviewed literature and the IPCC SREX are consistent on this point ”


    • Horatio Algeranon

      From a Munich RE press release (Oct 17, 2012)

      “A new study by Munich Re shows that North America has been most affected by weather-related extreme events in recent decades. The publication “Severe weather in North America” analyzes all kinds of weather perils and their trends. It reports and shows that the continent has experienced the largest increases in weather-related loss events.

      Nowhere in the world is the rising number of natural catastrophes more evident than in North America. The study shows a nearly quintupled number of weather-related loss events in North America for the past three decades, compared with an increase factor of 4 in Asia, 2.5 in Africa, 2 in Europe and 1.5 in South America. Anthropogenic climate change is believed to contribute to this trend, though it influences various perils in different ways. Climate change particularly affects formation of heat-waves, droughts, intense precipitation events, and in the long run most probably also tropical cyclone intensity. The view that weather extremes are becoming more frequent and intense in various regions due to global warming is in keeping with current scientific findings, as set out in the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) as well as in the special report on weather extremes and disasters (SREX). Up to now, however, the increasing losses caused by weather related natural catastrophes have been primarily driven by socio-economic factors, such as population growth, urban sprawl and increasing wealth.

      Among many other risk insights the study now provides new evidence for the emerging impact of climate change. For thunderstorm-related losses the analysis reveals increasing volatility and a significant long-term upward trend in the normalized figures over the last 40 years. These figures have been adjusted to account for factors such as increasing values, population growth and inflation. A detailed analysis of the time series indicates that the observed changes closely match the pattern of change in meteorological conditions necessary for the formation of large thunderstorm cells. Thus it is quite probable that changing climate conditions are the drivers. The climatic changes detected are in line with the modelled changes due to human-made climate change.

      The Head of Munich Re’s Geo Risks Research unit, Prof. Peter Höppe, commented: “In all likelihood, we have to regard this finding as an initial climate-change footprint in our US loss data from the last four decades. Previously, there had not been such a strong chain of evidence.

  17. > “skeptics” will now want to ban the teaching of statistics

    They label the results to suggest they are not real.

    Look for references to “virtual risks” — meaning risks that someone tells you are there but you can’t see with your own eyes so aren’t believable. Nothing to worry about here, move along …..

    RISKY BUSINESS – John Adams
    “… routinely exploited by the media whenever they discover a new virtual risk. With such risks the balancing act still involves judgements about rewards and …”

    Life’s Adventure: Virtual Risk in a Real World
    Roger Bate – 2012 – Business & Economics
    ” …The ETS [environmental tobacco smoke, second hand smoke] statistic, for example, is provided by the US Environmental … to enter smoky places, like pubs, or requires more restrictive legislation, is open for debate….”

  18. Horatio Algeranon

    There is a fatal problem with this approach, Tamino.

    The remaining upward trend in weather-related disasters after normalization is due to the fact that you normalized by the wrong thing.

    In other words, you abnormalized when you should really be normalizing by sunspot (or cosmic ray) counts.

    Then your graph would undoubtedly show a downward trend in weather related catastrophes.

    Wait for the WUWT post.

    • It was funny until you said to wait for the WUWT post, and then it was hilarious.

    • I’m still waiting for the earthquaking science of a WUWT post using multiple regression to correlate global temperatures with forcings from periodical cicada cycles.

      Coauthored by Willis Eschenbach and Nicola Scafetta…

      • “I’m still waiting for…a WUWT post using multiple regression to correlate global temperatures with forcings from periodical cicada cycles.”…

        You forgot this part, ” …which are closely related to the increase in beri beri complicated by increasing occurrences of athlete’s foot.”


  19. Kerry Emmanuel makes an interesting point about the relative sparseness of data employed by Pielke compared to that used in physical evaluations of hurricane energy. Well worth a read:

    Click to access Emanuel%20Nature%20Dec%2005.pdf

  20. “Risk” is a function of impact times the probability of the event in the stated period.

    Geologic events have high impact, but a rather low probability of occurring at a particular place in a particular 30 year period, The normalization curve (above) is dominated by moderate events that occur rarely (a few times in a 30 year period). There happened to be a cluster of these events in the middle of the sampled time period. However, if a great mega-volcano had gone off at the beginning or end of the 30 year time period, the shape of the normalization curve would have been very different. And, these big volcanos only go off every few hundred thousand years so 20 years is nothing in their schedule. A 30 year sample of geologic events is only going to pick up random noise.

    Risk from AGW is dominated by the absolute certainty of AGW. Everyone on Earth is going to see some impact, every day. These impacts range from higher food prices to higher prices for shelter as storm events knockdown housing and disrupt other supply lines. AGW imposes a significant cost on planning, engineering, and construction all types, but in particular, public safety critical infrastructure — such as electrical, sanitary, and transportation systems,

    Real businessmen plan for the future. Real businessmen know what risks they face.

  21. Tamino is my hero

    [Response: Who are you really? Why are your email and IP addresses different from MapleLeaf?]

  22. Tamino, that was definitely not me @9:51. With all due respect, while I have much respect for you and all that you do, and while I greatly admire your statistical expertise, you are not my “hero”.

    Anyway, as you suspected, someone is trying to pretend that they are me.

    Desperate times for deniers.

    [Response: I suspect it was the same individual who left another comment with an extreme insult directed at you. Both the insult and the impersonation reflect dishonor.]

    • Horatio Algeranon

      — Horatio’s perversion of Enrique Iglesias

      Would you add, if I asked you to add?
      Would you “R”, and never Excel?
      Would you trend, if you saw me trending?
      And would you save my stats, tonight?

      Would you tremble, if I touched your Lisp?
      Would you graph? Oh please tell me this
      Now would you regress, for the one you love?
      Hold me in your ARMA’s, tonight

      I can be your hero, baby
      I can GISS away the pain
      I will stand by you forever
      You can take my math away

  23. Damage estimates are rather useless if they only consider the US where landfall is rare. Does anyone have a measure for the Carribean, perhaps not costs, but lives lost? Even there improvements in weather observation must have made a huge difference.

    Taking this all into account, if normalized for population and wealth costs have remained constant, there must have been a huge increase in hurricane energy.

    • Susan Anderson

      I note the blackout of news about a big typhoon in South China that ran at the same time as Sandy. That part of the world has been battered by several huge storms for the last few months. Then there was the one near India a couple of days later. The devastation in the US is only “sexy” to westerners because they can see and relate to it.

    • I believe it’s all here Eli

  24. If I can, maybe this normalization is not sufficient. For weather related disasters, there is a permanent improving of the forecast, and ealier warmings. For earthquakes, there is no such improvements. I don’t know if this point makes a real difference, but if weather forecasts were still as reliable as in the 50s, perhaps we would have see an even greater increase in weather related disasters. What I want to say is that perhaps a part of the increase in geological related disasters is not expected to be found in the increase in weather related disasters due to better forecasts and warmings ; and maybe the normalization “over divided” the weather related disasters.

    [Response: Since early warning of weather events has improved but that of geophysical has not, the normalization would have “under divided” weather catastrophes, hence underestimated their increasing trend.]

  25. I think the topic is narrowing excessively here. When you look at global PDI or ACE values for hurricanes, what you mostly see is a very noisy data set. Deniers love to say that ‘global ACE values have been trending down for the last 20 years,’ but that seems to me just another cherry-pick deserving a real analysis.

    But the question isn’t just tropical (or, for that matter, extra-tropical) cyclones. The question is climate-related disasters. Hurricane damage is topical now–and dare I say, ‘sexy?’–but just a part of the picture, and probably one of the murkier parts at that. “Climatological” and “hydrological” disaster (drought and flood) are horses of a different statistical color.

    • Susan Anderson

      My amateur opinion is that there is no question we are seeing more hybrids and out of season events. Hurricanes are difficult because they grab so much attention and the science does not seem to be clear on them. I do wish that the whole boiling of mildly extreme to extremely extreme could be bunched together rather than spending so much energy on one particular extreme and fighting over how to include them. (Also, the “fighting”, being fueled by professional PR, is heavily weighted and prevents real discussion all too often.) I’d like to see more integration of observation and modeling, though I know that snipers are fond of insisting on purity in their opposition, though any mud will do on their “side”.

    • The focus on what’s “sexy” at the moment, be it hurricanes, tornadoes, derechoes, or what have you does miss the point. There are a variety of extreme weather events that happen and do not rise to the level of getting a name or a title. There are a variety of metrics to approximate this; declared disasters, billion dollar weather events, forest fire acreage, drought severity, disruptions to the electrical grid, insured losses from catastrophes etc. Some of these have a socio-economic component which is hard to avoid. The observed increase in the amount of precipitation falling in very heavy precipitation has no socio-economic component, forest fire acreage burned can be complex, but it doesn’t have much of socio-economic component, nor do extreme heat events/drought. If Pielke wants to focus on hurricanes, not only is he cherry picking the data, he’s cherry picking the events.

  26. Loss of the 2010 Russian and Pakistan wheat crops were both related to AGW, Both raised global food costs, and these costs are not fully accounted for in the insurance data.

    The “Texas” drought raised the price of forage for sheep, which raised the price of US wool, (which is exported to Europe) raised the price of yarn in Italy and France, which raises the price of wool garments in the designer sections of your local department store, This cost is not reflected in the insurance data. (Anne Romney is paying more for her clothes as a result of AGW.)

    The “Texas” drought raised the price of hay and other feeds, which means reduced profits for the dairy industry. That means higher milk and beef prices. These costs are not in the insurance data for the costs of AGW.

    Higher intensity precipitation events damage infrastructure designed for lower intensity events. Roads can be designed to withstand more precipitation but our local roads were designed for a more arid environment. A couple of precipitation events damaged. these roads and they need to be redesigned and reconstructed from the base up, but instead we patch them over and over, and they are in a constant state of disrepair, which leads to greater wear on vehicles. These costs are not in the insurance data for the costs of AGW.

    Forest fires change the water delivery rates and times for the watershed. And, erosion from burned areas can fill dams them less useful. In western NA, this can be a huge cost. This changes the cost of water for farmers and the cost of food on the East Coast and around the world. Changes in water delivery increase the cost of cotton grown in the west.

    The costs from AGW have started to permeate our lives.

    • Susan Anderson

      “Permeate our lives”, indeed. Aaron Lewis, I hope you don’t mind if I say that I find your commentary consistently outstanding and informative beyond the norm. Thanks.

  27. Most structure foundations are designed on the basis of a particular soil moisture regime. If the soil moisture changes as a result of of AGW, then the foundation will shift or even fail, resulting in cosmetic or even structural damage.

    Thus, there is a cost to housing and other structures as the moisture regime of the soil / sub-soil changes. Changes to soil and sub-soil moisture is an early impact of AGW.

    • As I discovered to my dismay after installing a dehumidifier in a basement w/posts sitting on exposed clay. :-) Fortunately I discovered the problem before none of the doors in the house would close. :-)

      Boneheaded, truly.