… in the USA. In the long run … over, say, a hundred years.
My aunt, who has spent her life living either in the path of tornado or hurricanes has always preferred tornado alley. She can’t stand the anticipation. With a hurricane you have weeks to prepare for it, you watch it on the news getting closer for days, you know it’s coming. With a tornado, you often find out there’s a chance of it about an hour before it happens, but the time between seeing it and it getting close to you is minutes.
Based on that, I’d guess tornado kill more people.
Well, comparing these lists of the deadliest tornadoes and hurricanes on record the immediate answer would have to be hurricanes.
Also, keep in mind that hurricanes can only happen along the coastline.
For directly attributable deaths, tornados seems to claim more victims than hurricanes but I’m not sure how many of the flood statistics are cause by hurricanes…
See figure 6: http://www.csccc.info/reports/report_23.pdf
Numbers for the 20th Century:
**Hurricane Deaths 1900-1999
Total**: 15,360
**Average: **153.60
**Median: **10.5
Max: 8,000
**Min: **0
**Tornado Deaths 1900-1999
Total: **14,943
**Average: **149.43
**Median: **85.5
**Max: **794
**Min: **12
What this suggests (and what is in fact the case) is hurricanes have far greater potential per incident but massive variance in how many deaths they can cause. There are many years in the 20th century in which no person died in America to Atlantic Hurricanes.
Further, this particular 100 year window is dramatically impacted by the first year in the series. In 1900 8,000 people were killed by the Galveston hurricane. If we changed the 100 year window to start in any year between 1901-1911 the hurricane numbers would be far different. While it would add Katrina to the calculations if you went from say, 1906-2005, Katrina was 1500 versus Galveston’s 8,000.
If you were to look at the second 50 year period in the series 1900-1999, the numbers change dramatically:
**Hurricane Deaths 1950-1999
Total: **2,102
**Average: **42.04
**Median: **18.50
**Max: **400
**Min: **0
**
Tornado Deaths 1950-1999
Total: **4,460
**Average: **89.2
**Median: **62
**Max: **519
**Min: **15
Without breaking down the numbers also for the first half of the series, I can tell you that deaths from both hurricanes and tornadoes have decreased in the last 50 years of the 20th century compared to the first 50 years of the 20th century.
Hurricanes primarily kill via flooding (and flood deaths are officially attributed to the hurricane, at least by NOAA statistics, and the NOAA reports the overwhelming majority of hurricane deaths are from flooding.) We have come a long way in preventing flooding deaths, even factoring in the deaths during Katrina.
We have far more sophisticated flood protection systems, we also have far better warning systems, people who live in areas frequently hit by major Atlantic hurricanes have probably evacuated a few times in their lifetimes and it is an understood and accepted part of living in the region.
Over a long period of time what I’m seeing suggests that sans-technology, hurricanes have far greater potential to cause death. There are several hurricanes in history that have claimed over 20,000 lives (although not in the United States.) The largest ever was in the late 1700s during the American Revolutionary War, and that hurricane actually almost depopulated several Caribbean islands.
Unfortunately and sadly, some Caribbean islands, especially the poorest ones, remain very susceptible to hurricanes and Hurricane Mitch attests to how deadly Atlantic hurricanes can be for the Caribbean.
The trend in the United States (even if you picked a sliding dataset that includes Katrina) is that we are getting better and better and surviving hurricanes. We’re getting better at surviving tornadoes, too, but from what I’m seeing there are a few things unique about tornadoes that make them significantly harder to lessen purely through technology, primarily:
-
Tornadoes in the grand scheme of things affect very small areas when they hit. For this reason many people who live in tornado alley their entire lives may have lived through many tornadoes passing through their areas, without suffering any ill effects. This creates a mindset where these people have gotten in the habit of disregarding tornado sirens and warnings.
-
Tornadoes develop far faster than Atlantic hurricanes, and are far more unpredictable. Even with improvements in weather tracking and forecasting there are real limits to how many minutes in advance we can tell a community a tornado is coming.
-
Tornadoes hit a far more variable area than hurricanes. Most tornadoes in the world happen in tornado alley, but tornadoes can actually happen anywhere. Where they are very uncommon, people will be ever less prepared for them. Many homes in tornado alley have a basement precisely because people want a place to go during tornadoes. But in some parts of America and the world, basements are uncommon, meaning residents have to take shelter entirely in the part of the house that is directly being hit by the wind.
The fact that most tornadoes happen in tornado alley and tornado alley is flat, also has lead to grave and dangerous misunderstandings about tornadoes. For example people who live in mountainous states are under the impression that mountains have the ability to stop tornadoes and that mountains are protection from tornadoes. One of the tornadoes that has killed > 100 people in the last 100 years was in Shinnston, West Virginia, in the Appalachians. It touched down, followed the mountains and killed over 100 people.
Because of the essentially global target area of tornadoes, and the fact that any specific spot is very unlikely to be hit by one, it basically means incidents like the Shinnston tornado will probably be very difficult to ever plan for, and un-economical to build structures for, such incidents are so uncommon in Shinnston that it would be very unlikely residents there build houses with tornadoes in mind.
Even in tornado alley, hardening all homes for tornado damage is not really economical because most homes will never be hit by a tornado in the lifetime of the homeowner. However if you live in the highly frequent impact areas for an Atlantic hurricane your house almost certainly will be subject to hurricane winds a few times during your life, so it is worthwhile to design and build houses that can survive those winds. Protection from flooding obviously has to happen more at the large scale, and even then you can’t totally eliminate floods (meaning some houses will always be destroyed by flooding) with proper education and evacuation techniques you can however intensely reduce the number of people who die in flood waters of Atlantic hurricanes.
Not sure if it would be enough to cause an anomaly, but a hurricane that comes into the continental US tends to “downgrade” to a strong storm system that often happens to spawn tornadoes. Would deadly tornadoes from such events be in the hurricane column?
Hurricanes spawn tornadoes at full power too. Hurricane spawned tornadoes don’t tend to get as big and deadly as those that mid-western supercells are capable of, though they could still throw a fully loaded Greyhound bus into the lake.
Hurricane Andrew released a number of tornadoes, so where does that fit in to the OPs criteria?
As noted in general on a year to year basis tornados kill more folks than hurricanes. And this year, the tornados have been plentiful, about as big and strong as they come, and have hit some prime targets rather than the typical sparsely populated areas.
But, IMO, one day sooner or later we are going to have a “perfect storm” hurricane strike and the death toll is gonna be a whopper. Like Japanese earthquake/tsunami bad or worse. I can easily envision “evacuating” Miami turning in SNAFU and tens of thousands of folks riding out a Cat 5 hurricane stuck on the highway thats 5 feet above sea level.