Hurricane Fatalities, 1900–2010: Update

by Indur M. Goklany

Much to my surprise, I heard from NHC’s Dr. Eric Blake today on the fatalities data for hurricanes. I was sure he would take a break to catch his breath after Irene (and before Katia), and since I’m sure the NHC gets no respite during hurricane season. Regardless, I really want to thank him for his prompt response.

He confirmed that the discrepancy between the fatality numbers for 2005 from the 2007 and 2011 versions of the NHC fatalities data is due to a reevaluation of the older data. [I had feared it might be due to a typo.]

Based on this information, I have revised the earlier figure that Anthony published on WUWT. The updated figure follows.

Figure 1: U.S. hurricane deaths and death rates per year, 1900–2010. Death rates are estimated per 100 million population. Sources: Updated from Goklany (2009), using USBC (2011) and Blake et al. (2011).

References

Blake ES, Rappaport EN, Landsea CW. 2007. The Deadliest, Costliest, and Most Intense United States Hurricanes from 1851 to 2006 (and other Frequently Requested Hurricane Facts), Apr 15, 2007. Available at http://www.nhc.noaa.gov/Deadliest_Costliest.shtml. Accessed Sep 26, 2009.

Blake ES, Landsea CW, Gibney, E.J. 2011. The Deadliest, Costliest, and Most Intense United States Hurricanes from 1851 to 2010 (and other Frequently Requested Hurricane Facts), August, 2011. Available at http://www.nhc.noaa.gov/pdf/nws-nhc-6.pdf. Accessed 22 August, 2011.

Goklany, IM. 2009. Deaths and Death Rates from Extreme Weather Events: 1900-2008. Journal of American Physicians and Surgeons 14: 102-09.

U.S. Bureau of the Census. 2011. Statistical Abstract of the United States: 2011, visited 14 August 2011.

About these ads
This entry was posted in hurricanes. Bookmark the permalink.

54 Responses to Hurricane Fatalities, 1900–2010: Update

  1. In 1900,the population of the US was just over 76,000,000 or about a quarter of today’s estimate, and the graph shows a significantly decreasing death toll despite the population growth figure. That would lead me to think that Humanity has shown the resilience to cope with the worst Mother Nature can throw at us.

  2. Alan the Brit says:

    Don’t bore us with the trivial details & reality, this is a sensational story, lets double the number & add a nought, it will look great in the headlines!!!!! :-)) “OVER 800 DIE IN HURRICANE IRENE DISASTER!” Much better than “Thankfully, only just over 40 deaths caused by floods from tropical storm Irene!”

    Give it a few days & we’ll be hearing the usual drone of “whilst no particular weather event can be specifically attributed to Climate Change, this is what computer models have predicted will occurr as a result!” A’la Met Office “some areas will have more floods, some areas will have less floods. Some areas will have more drought, some area will have less drought. Some areas will have more rainfall, some areas will have less rainfall!” Keep covering ALL the bases fellas, then you know you’ll be right. Sarc off! This prediction lark is childs play, literally!

  3. Minus 100? Is it raining babies?

  4. MikeA says:

    Well it’s obvious that, in the modern era, death rates have gone up, even with modern warning systems. But what does it mean and what is the cause?

  5. It seems to me that the values for 1900 – 1909 should be a lot higher. The estimated number of deathes from the Great Galveston Hurricane on September 8, 1900 range from 6,000 – 8,000. Shouldn’t both the number of deaths and the death rate be increased?

  6. jim karlock says:

    Just to mention something else the green shirts hate: automobiles.

    Automobiles provide a way to evacuate large areas quickly and save lives.
    Mass transit can’t do that! (Of course automobiles also use less energy that transit buses per passenger-mile, therefore transit DOES NOT SAVE ENERGY OR REDUCE CO2 – but that is a whole other area of greenshirt deception and propaganda.)

    Thanks
    JK

  7. Gilles says:

    I think that the cost of climate catastrophe must be corrected for the increase of both population, and wealth. A possible way of doing that is to compare with the cost of earthquakes, that are not supposed to be dependent on climate change (although.. who knows !). But the 2007 AR4 evaluation was dominated by the high cost of 2005 hurricanes. The recent occurrence of powerful and deadly earthquakes (Haiti, Chile, Japan) will for sure considerably impact the “correction factor”. Let’s see if the IPCC will keep the same way of counting the costs of meteorological events ….

  8. Jer0me says:

    I love the Y axis. It starts at -100! So what would constitute a negative death rate? People staying home with no Radio or TV, perhaps?

    Aside from that, without the obvious outlier of 1900 – 1910, there would be no trend. If you start in 1940, the trend is up slightly.

  9. Mark says:

    How does the total number of deaths also compare to the number of tropicals (unless this graph. Is solely bases on Cat 1 – Cat 5 hurricanes) that have made landfall in the US?

  10. H.R. says:

    More people, and more people along the coasts, yet fewer deaths. Obviously we’re doing something right. Just think what those numbers would be like if there was a 50-mile “no build” zone along the coasts.
    (“Shhh… H.R.! Don’t even think that! Some do-gooder in Washington might think it’s a great idea.”)

    Also, I think perhaps later numbers may be inflated as the earlier numbers probably counted only drowned and/or mangled bodies while newer numbers include some such as a 104-years old lady dying because the oxygen pump went out when hurricane interrupted electricity.

    It’s probably “better than we thought.”

  11. Philip Finck says:

    A great graph — I’ll use it in a talk

  12. tallbloke says:

    I anticipate we are in for a stormy decade. Not because of global warming, but because of imbalances in the system as the ocean cools due to the sleepy sun.

  13. Patrick Davis says:

    I wonder how many people were killed through jaywalking over the same period?

  14. Alan the Brit says:

    H.R. says:
    September 1, 2011 at 3:06 amAlso, I think perhaps later numbers may be inflated as the earlier numbers probably counted only drowned and/or mangled bodies while newer numbers include some such as a 104-years old lady dying because the oxygen pump went out when hurricane interrupted electricity.

    Don’t forget, these “good guys” will use every available statistic they can to boost disaster. remember that here in the UK, as it may well have been in the USA, the medical profession et al boosted “premature deaths” from active & passive smoking by including people dying in their 70s & 80s!!!!!!!

  15. Bruce Cobb says:

    It seems, despite their best efforts to inflate the numbers the “death toll” angle just isn’t working for the CAGW cult and MSM. Diane Sawyer’s recent prattling about “the body bags” was sheer idiocy, as well as a blatant attempt at sensationalism.
    So now, their main focus will apparently be the number of “billion-dollar” disasters. For that scheme, they play a little shell game, purporting to show that those billion dollars have been inflation-adjusted. To do that, they add all the figures together for a previous period, the 70′s, say, then do the same for the most recent decade, and inflation-adjust the earlier decade. Fine and dandy. Voila, the numbers go up because population centers have gotten bigger, and people are building more in flood-prone areas. Who’d have thought? Now for the shell game. They count the number of billion-dollar+ disasters from both periods, but do they inflation-adjust those billion dollars? No. The goal is to show that the number of billion-dollar disasters has increased, and surprise, surprise, they have.

  16. HaroldW says:

    Why is the death rate normalized by US population? Surely it should be calculated in terms of the population in the affected areas only, namely Gulf Coast & East Coast.

  17. Brian D says:

    It should be noted that the deaths in the chart are in thousands. Confused me at first until I looked at the paper.

  18. JJwright says:

    Brilliant
    A linear curve fit to something that obviosly is not linear.
    Well done!

    Tallbloke September 1, 2011 at 4:01 am
    Well done in getting that in before AGW provides the stormy decade!

  19. The Y axis should be logarithmic. We are dealing with extreme values that can never be negative. The distribution should not be expected to be linear and the linear trend is meaningless.

  20. Bernie says:

    Does anybody have any data on whether hurricanes might actually “save” lives in the sense that there is less driving, less in public drinking, less crime, etc.? Katrina obviously is an outlier.

  21. novareason says:

    H.R. I have a sneaking suspicion that your guess about inflated numbers is dead on. The other thing I can think of is… increasing American stupidity (people not getting the French out) aka Those Stupid Surfers Effect… all the people who didn’t evacuate during Katrina… in other words, back when people actually listened to things like “Mandatory Evacuation”.

  22. A. C. Osborn says:

    Why is the last column for 2000-2010 and not 2000- 2009 as all the other results are?

  23. Richard111 says:

    I understand there were deaths on the roads of people following evacuation orders.
    Are those deaths blamed on tropical storm Irene?

  24. Donny Cook says:

    @Mikea The problems with the deaths isn’t because of “global warming” or “climate change”- it’s up to the people to take appropriate actions- take for example the stage collapse a few weeks back- a severe t-storm watch was issued for the region hours before this happened- severe t-storm warnings were issued west of the region and as this moved east there was lightning in the area, severe t-storm warnings were issued the idiots in charge didn’t shutdown the program, the people refused to leave even when there was lightning in the area and an omminous sky loomed over the region- look what happened-the stage collapsed- deaths that SHOULD have been prevented

  25. A plot of the logs of death rate gives an R^2 of 0.618.

  26. PaulM says:

    Trying to draw a straight line through this sort of data is not very sensible, as Maurizio’s comment shows.

  27. Louis Hooffstetter says:

    “…the discrepancy between the fatality numbers for 2005 from the 2007 and 2011 versions of the NHC …data is due to a reevaluation of older data.”

    SAY WHAT !?!? Fatality numbers for 2005 were “reevaluated” and changed? WHY ??
    Were lost bodies of people killed by hurricanes in 2005 somehow discovered between 2007 and 2011? Did the cause of death of some people somehow change from ‘cancer’ to ‘hurricane’? My BS meter is pegged and its siren is screaming at full volume.

    This appears to be the same kind of revisionist science that climatologists use to show it’s worse than we thought.

  28. bob says:

    ” MikeA says:
    September 1, 2011 at 1:57 am
    Well it’s obvious that, in the modern era, death rates have gone up, even with modern warning systems. But what does it mean and what is the cause?”

    The states with fewer deaths are those who don’t usually get hurricane strength storms. People in the Gulf States are weather wise, and know that when they build a house next to a river, there is a chance of flooding. Folks from Texas to Arizona know that flash flooding does not mean for everybody to run to the mall at the same time.

    I remember one tropical storm in Texas that dumped 36 inches of rainfall in the hill country, and 100 foot deep canyons were filled with rushing water in Bandera, Texas.

    The mid-Atlantic and New England states were not prepared for two to three foot rainfalls.

  29. bob says:

    OOPS! I meant to say, The states with more deaths are those who don’t usually get hurricane strength storms.

  30. Ed Caryl says:

    According to the census bureau, the death rate from traffic accidents is about 11 per 100,000 people annually (2009). In the area affected by the hurricane, the population is on the order of 60 million. If the storm prevents that population from driving, or the drivers slow down and become more cautious, 18 traffic deaths will be prevented every 24 hours. Keeping that in mind, did the hurricane increase the death rate? Probably not.

  31. Pamela Gray says:

    I beg to differ tallbloke (how tall are you?). I anticipate we are in for a stormy decade, not because of global warming, but because of imbalances in the 7th house and because Jupiter aligns with Mars. By the way, my CB handle was Strawberry Shortcake.

  32. Gary says:

    Jer0me said: “I love the Y axis. It starts at -100! So what would constitute a negative death rate? People staying home with no Radio or TV, perhaps?”

    Most likely the -100 is an artifact of the graphic program (Excel does this). The zero line just doesn’t show a number on the scale. None of the data bars go negative so your complaint is trivial.

  33. richard verney says:

    As one might expect from events that are randon and chaotic in nature, there is no real trend. With better forecasting and the ability to evacuate quicker, one may expect to see a downward trend over that which existed 100 years or so ago.

    The 1900/09 decade appears an outlier. If that is removed, the decade 2000/10 does not appear remarkable, and one would probably consider 1940/9 and the period 1970 t0 1999 to be rather quiet years.

  34. geo says:

    I’d think population density has quite a bit to do with it as well. There are 4x as many people in the US now as there was in 1900, and the coastal population density would have increased by even more than that.

  35. geo says:

    For example, the population of Florida in 1900 was 528K. Now it’s nearly 19M.

  36. cromagnum says:

    OT, but related

    Can you make a chart for the Bastardi hurricane power scale? He premises that the real power is a combination of Wind Speed and Low Pressure. He uses a numeric scale 1-10, with one decimal place.

    Then post that chart over at the Tropical-Cyclone page for future reference.

  37. Theo Goodwin says:

    MikeA says:
    September 1, 2011 at 1:57 am
    “Well it’s obvious that, in the modern era, death rates have gone up, even with modern warning systems. But what does it mean and what is the cause?”

    It is the modern attitude that any death occurring during a tornado is caused by the tornado, any death during a flood is caused by the flood, and so on.

  38. Bob Diaz says:

    I’d say that it’s the improved forecasting, better warning systems (like radio & TV), and good transportation (cars), that can account for the drop. However, the joker in me wants to say, “it’s proof that AGW saves lives!!!”

  39. Rick Lynch says:

    To me, the interesting thing is that there have only been 3 hurricanes of category 4 or 5 in the last 40 years. The previous 40 years had 8. The 40 years before that also had 8. This should put the lie to the AGW refrain of more and stronger hurricanes as the earth warms.

  40. John says:

    The deaths and death rates for 2000 -2010 are to a great extent due to Katrina. If the levee boards had done their jobs instead of spending their time on levee board casinos and levee board small airports, levees wouldn’t have failed and the numbers would have been as low as earlier decades. That is based upon an account by a person who served for two years on a levee board and wrote that never once in their meetings was the issue of levee safety or adequacy brought up. And/or if the Mayor of New Orleans had taken the advice of the President and others, and evacuated the city earlier, again the death rate would have been lower. Not to minimize the force of Katrina, but the death toll could have been far, far less.

    Just as in the case of the recent flooding in Australia, if human responses had been thoughtful, the disasters wouldn’t have occurred. (In Australia, the operators of a dam ignored forecasts of heavy rain, when the dam was already filled with water. Instead of releasing water several days early, so that there would have been a constant rate of high but not flood waters, they released water at the last minute, when the deluges had begun to fall. They did this because they were managing the dam’s water for drought, but were unable for some reason to take serious account that in La Nina years, that part of Australia gets lots of rain.)

  41. jet pack says:

    I thought there were 1800 deaths from Katrina?

  42. jet pack says:

    Wow, massive fail on the chart with out the correct units labeled.

  43. Eric Anderson says:

    John, good point. Katrina blew through New Orleans without massive damage. I remember watching a live report from the area after the hurricane had essentially gone through, with the reporter noting that other than some heavy rain and a bit of minor damage New Orleans had largely been spared.

    Just a few hours later the levees broke . . .

  44. Ray says:

    The rate and deaths were going down… until FEMA was form and you can see that it has continuously risen from 1979.

  45. DJ says:

    What about puppy deaths?

    You do love puppies, don’t you??

  46. nonegatives says:

    Looks like Reagan was successful in stopping the Soviets from using their secret weather controlling device. Would a /sarc tag help those critical of the linear fit?

  47. MikeA says: September 1, 2011 at 1:57 am

    Well it’s obvious that, in the modern era, death rates have gone up, even with modern warning systems. But what does it mean and what is the cause? I might suggest that more people are dying!

    In the UK we are seeing more and more building on flood planes. We are also seeing more and more “incapable” people as the government takes on everything from handing out bandages for scraped knees to putting out small chip pans fires. As an example, in the past organisations like the Scouts would actively help out in disasters. These days they need a training course to tie a shoe lace and the idea of going out in a boat without to a CRB (criminal records check) on each and everyone they rescued, and then getting to sign disclaimers in case they bump their knees getting in the boat. In other words, people don’t help out in disasters any longer because we are being told to leave everything to the “professionals” … who then arrive at iced up ponds and can’t help someone drowning (who then drowns in front of them) — or arrive at mineshafts and won’t go down … because they don’t have the right training … who then drowns as they all stand around doing nothing at the top.

  48. Ed Caryl says: September 1, 2011 at 6:16 am

    … death rate from traffic accidents is about 11 per 100,000 people annually (2009)…affected… population ~ 60 million. If the storm prevents that population from driving, 18 traffic deaths will be prevented every 24 hours.

    And if it doesn’t stop them driving and there is a fatality it gets blamed on the storm?

  49. Hoser says:

    Just do a linear fit to the log (or ln) of the deaths. Then you get exponential decay and no negative projected deaths.

    Pamela, if talkbloke is tall, and shortcake is short, how short are you? I find short female humans are particularly dangerous. ;->

    I was Jabberwocky. 10-10

  50. Interesting graph. Considering all the possible consequences and circumstance that could create a death caused 100% by the hurricane makes the graph inconsequential. These people who lives in areas affected by hurricanes who have died from hurricanes, individually have their own list of circumstance that cannot be identified and quantified.

    Per hour, a student pilot is the safest among pilots. More accidents occur with two professional pilots at the controls. It is a function of how one observes the data.

  51. tadchem says:

    Linear regressions to produce ‘trend lines’ are simple and easy – pocket calculators could do it in the 70′s.
    Judiciously chosen regressions are marvelous tools for making interpolations – guesses at numbers that fall *within* the range of the data.
    They are useless for extrapolations, as I tried frequently to persuade the engineer-types in my classes, and as this chart demonstrates.
    The world is not linear.

  52. tadchem and others — thanks for your comments. The linear trend line is merely to show that it is not increasing over time (as it should have if warming is the major determinant of casualties). I don’t use it to model future deaths or death rates. As someone with training in engineering and physics, I am generally leery of extrapolating.

  53. Patrick Davis says:

    “John says:
    September 1, 2011 at 7:11 am”

    And those such as Flannery forecasting permanent drought won’t be held accountable.

Comments are closed.