Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Can we blow up a Tornado?

  1. May 21, 2013 #1
    I have an idea for you on defusing or dissipating tornados.

    I thought that a sufficiently large explosion at the right time and place in the interior of a tornado funnel about to touch down would force the tornado to collapse and render it harmless. It could be delivered by ballistic or cruise missile, and I believe we have very powerful non-nuclear bombs at our disposal.

    Even if only those tornados headed towards populated areas could be defused or weakened, a great saving in life and property could be realized.

    My condolences to all the victims in Oklahoma.
  2. jcsd
  3. May 22, 2013 #2

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    Welcome to PF;
    You are thinking of disrupting the wind patterns that make the tornado?

    A big tornado like the one that just hit Oklahoma is driven by atomic-bomb levels energy - the energy system driving the funnel you see has built up over a long time and covers a wide area. You would definitely need nuclear weapon levels of explosives before the tornado would even notice.

    Then there is the problem of delivery - it takes a missile some time to reach it's target and the touchdown point is very hard to predict. You'd be better to try to disrupt the patter as it builds - but even then... see above. If you could figure how to get it to work - there may be an advantage in having the destruction occur someplace we choose.

    However, it is not clear how a bomb would actually help - I'd expect the energy of a nuclear bomb, particularly at the touchdown point, would just make things worse. You could try blowing up lots of them around the tornado hoping to draw the winds away ... but I would expect the tornado to reform since the forces bringing it into existence are still there - only now it's radioactive.

    It's a popular question online.
  4. May 22, 2013 #3


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    and finally the explosion, chemical or nuclear is likely to do just as much damage as the tornado
    so ... pretty pointless

  5. May 22, 2013 #4
    I suppose as our radar tracking gets better, we might have a chance in the future, but I see what you mean about needing an atomic weapon for a monster like the one that just hit Oklahoma. Thanks for the input.
  6. May 22, 2013 #5
    You can't use atomic bombs to destroy a powerful tornado. Even if it worked, the results in terms of heat and blast damage and long-term atmospheric radiation pollution would be equally devastating. An anti-tornado 'barrier' that starves a tornado heading towards a city of inrush energy and so deflects its path would be very expensive but might just work - except for any tornado that touches down within the protected zone. The best and least costly solution I can think of is simply tornado proof buildings - shallow, smooth-surfaced reinforced concrete domes that offer no resistance and so remain unscathed as a tornado passes over it.
  7. May 22, 2013 #6

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    There may be a percentage in controlling where the damage occurs though - if it came to a choice between a tornado ripping through a city and a bomb stopping the tornado outside the city...

    @Banquo: welcome to PF;
    You are right - tornado-proof buildings are probably cheaper, certainly than the nuke option ... though I notice that the main reason many homes in the path of the recent one did not have tornado shelters or proofing was simply that the owners couldn't afford it.
    Sometimes it is a matter of who does the paying.

    But I think the original question has been answered.
  8. May 22, 2013 #7
    Tornado schematics

    German wikipedia has good schematics of a tornado:

    (click on image for larger version)
    Created by Michael Graf / Michael81 in 2006
    Public Domain, according to http://de.wikipedia.org/wiki/Datei:Superzelle_schema.gif

    Translation (mine) ordered following label positions; clockwise starting at 12:00:
    • "Superzelle von Süden her betrachtet" = supercell seen from the south
    • "Konvektives Überschiessen" ~= convective excess
    • "Rotierender Aufwind" = rotating updraft
    • "Eiskristallwolken" = ice crystal clouds
    • "Gleichgewichtsniveau" = level of equilibrium
    • "Mammatuswolken" = mammatus(?) clouds
    • "Niederschlagsfallstreifen" = virga
    • "verdunstender Niederschlag" = evaporating precipitation
    • "Kondensationsniveau" = level of condensation
    • "Warmluft" = warm air
    • "Regenfreie Wolkenbasis" = rain-free cloud base
    • "Zugrichtung" ~= direction of movement / direction of motion
    • "Pseudokaltfront" = pseudo cold front
    • "Kaltluft" = cold air
    • "Fallstreifen" =~ (lit.) fall stripes
    • "Niederschlagsfallstreifen" = virga
    • "Auf- und Abwinde" = up- and downdrafts
    • "Mammatuswolken" = mammatus(?) clouds
    • "Eiskristallwolken" = ice crystal clouds
    Last edited: May 22, 2013
  9. May 27, 2013 #8
    An alternative approach would be to release butterflies. No, I'm not crazy - it's a metaphor.

    Many of of you will be familiar with the poster child of chaos theory, the Brazilian butterfly that, with one flap of its wings so changes the dynamics of the airflow that ultimately a hurricane is triggered in the Caribbean.

    Specifically, at some point in the future our understanding of tornadoes, our modelliing software speeds and our sensors will be of a sufficient calibre to anticipate what initial conditions need to be disrupted, in a small way, to forestall tornado formation. At that time Dennmann57's proposal may become practical.
  10. May 27, 2013 #9
    You have it exactly backwards: that is the opposite conclusion that chaos theory implies. The point of the butterfly effect is that chaotic systems are dependent on such tiny changes to their initial conditions that long time/large scale forecasting is impossible. There is always a degree of experimental uncertainty and, in chaotic systems, this uncertainty (however slight) is enough to severely limit your ability to predict the system's behaviour because of how rapidly the uncertainty grows.
  11. Feb 25, 2014 #10


    User Avatar
    Gold Member

  12. Feb 25, 2014 #11


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    At only $60 billion per 100 miles of wall, why use stone or other material when you can just use gold?

    In any event, the previous idea about using a nuclear detonation to disrupt a tornado is a non-starter for an entirely non-technical reason: such an explosion would violate the Limited Test Ban Treaty signed by the US and other nations in 1963. The treaty prohibits testing of nuclear weapons or any other nuclear explosions in the atmosphere, under water, or in outer space.

  13. Feb 25, 2014 #12

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    Average economic impact of a tornado: $300,000
    Cost of a cruise missile: $1,400,000
  14. Feb 25, 2014 #13


    User Avatar

    Staff: Mentor

    How do you know if a tornado will form? And if one does form, they are usually very short lived, and the blast from a missile if it's over a densely populated area????
  15. Feb 26, 2014 #14

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    Where are you getting those figures?

    iirc. The 1.4mil price tag belongs to the tomahawk cruise missile
    There are other kinds - the cost sans warhead can go below half that (still too high) but what you are really looking for is some kind of UAV capable of not crashing before it gets sucked into the tornado right? The technology in the typical cruise missile is way overkill.

    Tornado season results in damages in the billions:

    But estimating the impact on the economy is tricky

    Certainly you wouldn't want to spend UD$0.5 - 1.4M on every tornado - but maybe it is a good idea for the really big ones like in Oklahoma last year?

    Provided of course that the damages from using an effective tornado buster is not worse than the tornado.
  16. Feb 26, 2014 #15

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    Some of these things remind me of the stories about the town that builds a tall wall around it's borders to keep the birds from migrating or the people who build a building without windows so the light cannot escape at night only to discover they forgot to put any light in to begin with so they try trucking it inside in wheelbarrows.... those sorts of stories.
  17. Feb 26, 2014 #16

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    Cost of the 2011 Alabama tornados: $3,740,000,000
    Cost of the 2011 Joplin Missouri tornado: 2,800,000,000
    Cost of the 2013 Moore Oklahoma tornado: $2,000,000,000 to $3,000,000,000

    We don't have to get every single tornado. Just the ones that cost a whole lot.

    That said, tossing a nuke at a supercell isn't exactly a brilliant idea.
  18. Mar 4, 2014 #17
    A bomb won't work - but I don't thing any of these posts really hits the important reason.

    My experience on this matter is based on the meteorological material that I was required to study to get my pilots license, study beyond that, and certainly from my experience as a pilot watching thunderstorms develop first hand.

    Cumulonimbus clouds form from an unstable air mass. When a parcel of air is moved upward, it expands and experiences adiabatic cooling. Its exact response is also influenced by its moisture content. Once the parcel of air has been displaced upward, its new density can be compared to the density of the air that it just moved into. If it is heavier it will move back downward - thus the air mass is stable. But if it becomes lighter than the surrounding air, then the air mass is referred to as "unstable" and the air parcel will continue to move up.

    As the sun bakes the ground during the day, the lower parts of the atmosphere are heated and begin to rise. This changes the vertical temperature and moisture profile of the air mass and during the summer, and generally make it less stable. With enough direct sun, the effects on the air can be easily seen from a small plane climbing in flight. If you keep a sharp eye out for sky color, you will notice that at certain altitude you will break through the haze layer - and the sky will shift from whitish to a strong blue.

    Of course, the other effect you see is the development of cumulus clouds from infrequent small fluffy things (perfect for sail planes) to more and more massive and taller structures. Each one of those clouds is generated from a rising column of air. If any one of these air columns creates a rain cloud, with perhaps considerable ground resistance, it will partially "spoil" the air in that area - sapping some of its energy - making it more stable. The actual formation of a cumulonimbus cloud occurs when one of these columns is able to build up in a larger and larger area of unstable air.

    So, what would a bomb do. For the most part, it would add energy to the air mass. Not something that would help.

    As for trying to stop these things before they get started... Flying through an area of rising cumulus clouds, you are in a target rich environment. In an area destined for the formation of thunderheads, every couple of thousand feet you will find another 7,000+ foot cumulus tower that could be the next thunderhead. And you wouldn't want to stop the ones that cause a little rain showers, because they're helping you by draining moisture from the air - unless it's one that's in a clear path to more and more unstable air.
    Last edited: Mar 4, 2014
  19. Mar 5, 2014 #18

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    I think the core idea is not so much to randomly nuke tornadoes but create a large pressure difference to disrupt the air mass... hopefully disorganizing it long enough that the conditions for formation pass.

    It's a bit like the idea of using high explosive to put out a fire - it looks like you are just adding more energy to the area the fire is in but, carefully placed, the energy from the blast need not add to the fires energy.
    In detail: https://www.physicsforums.com/showthread.php?t=370808

    However we are talking about disrupting a "long-lived" tornado - typically forming within a supercell thunderstorm - hopefully shortening the life-cycle of an existing one and so minimizing the possible damage. I think that's very "hopeful".

    Big explosions can be pretty disruptive - you don't want to try flying near a nuclear blast either - but whether they can get big enough to disrupt a big tornado? That would be an interesting experiment - I'd like to see anyone get it past an ethics committee though. Other methods that have been considered include large conventional explosions and large blocks of dry ice. The main problem appears to have been deployment.

    NSSL talks about this topic in their tornado FAQ.
    You have to consider that the tornado is part of something bigger: the supercell thunderstorm. Unless you disrupt the supercell thunderstorm itself, you would likely have another tornado, even if you were able to destroy the first. The thunderstorm's energy is much greater than the tornado. No one has tried to disrupt the tornado because the methods to do so could likely cause even more damage than the tornado. Detonating a hydrogen bomb, for example, to disrupt a tornado would be even more deadly and destructive than the tornado itself. Lesser things (like huge piles of dry ice or smaller conventional weaponry) would be too hard to deploy in the right place fast enough, and would likely not have enough impact to affect the tornado much anyway.​
    ... the site seems to be a decent 1st stop if you want to know about tornadoes and tornado research.

    BTW: related thread from 2010 on the question of preventing Tornadoes.
  20. Mar 5, 2014 #19
    That "supercell" is an oversize cumulonimbus cloud, an engine that feeds off unstable air. At it's core is a rising column of air - which can become one or more tornadoes.

    Once the cumulonimbus has formed, trying to kill it with a bomb would be like trying to keep water from flowing down the drain with a bomb. You can make a big splash, but the water will come right back to the drain. In this case, warm moist air is draining into the upper atmosphere - and it can form a drain anywhere.

    All that being said, tornado alley in southern US is a unique area where tornadoes form which much greater occurrence than anywhere else on the planet. Perhaps if cumulonimbus clouds could be made to form earlier in the day in those parts of our country - perhaps with more dark asphalt parking lots - the ones that did form would be less potent.
    When crops are at different stages of growth, they have varying effects on rising thermals - as any sail plane operator will explain. Here's advice from an experienced hang-gliding pilot:
    CO2 would act like a poison - it's colder and as a gas, it's heavier than H2O, N2, or O2. Of course, in the quantities that you would need, it would also be a thermal and drowning hazard to anyone nearby.
  21. Mar 8, 2014 #20


    User Avatar
    Science Advisor

    A tornado is a rotation driven by rising air. A bomb would heat the air and therefore add energy to the tornado. If enough energy could be released continuously on the Earth's surface it might cause a tornadoe's eye to remain in that position, but how do you then stop riding that aggravated tiger.

    There is very little atmospheric coupling between the hemispheres. As far as I know, tornadoes cannot cross the equator because of the coriolis force reversal. The Amazon is in South America, so Amazonian butterfly activity is pretty much irrelevant to tornadoes in North America.
  22. Mar 8, 2014 #21

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    This point has been raised and answered several times in this thread already.
    Note: The same argument would mean that a bomb cannot be used to put out a fire.
    Please see the links in post #18.

    With careful placement, the heat from the bomb can change the heat distribution driving the tornado so the winds do something else.

    Explosives don't just "add heat": there is also a shockwave. A very large bomb detonated close to a tornado as it forms could disperse the thunderhead. Of course, such a bomb would also do far more damage than the tornado could.

    In principle - a sharp increase in pressure close to the center of a tornado could disrupt it - though another one may form someplace else.

    It is technically possible to use smaller explosives carefully targeted to disrupt the heat and pressure patterns ... something you can test out in the small scale in a vortex generator. (Hmmm... I wonder if anyone has formally done it?)

    The main problems are practical: the sufficient and timely placement of the charges is a lot trickier that it looks.

    There are a LOT of discussions of how to disrupt tornadoes and they all seem to involve oversimplifications of one kind or another. Right now, the best practice for minimizing the damage of a tornado is to build strong structures and be able to run fast.
  23. Mar 8, 2014 #22


    User Avatar
    Science Advisor

    True. That is the most economic solution.

    The problem I see is that energy is stored in the atmosphere and is being released by the tornado. It becomes a problem of removing stored energy from the system in a non-destructive way.

    There is a way of thinking in which, if you throw a big enough bomb at a problem it will go away. That is not true with tornadoes, it just feeds and excites them.

    A fire fighters bomb puts out the fire by depriving it of oxygen. It is not a good analogy. The equivalent in tornado is the removal of the thermal gradient. Tornadoes are a problem because, once they have started they are hard to stop.

    To use many small bombs to disrupt a tornado would require a very accurate placement. It is hard to see how that could be done at low altitude without it adding energy and ferocity to the cyclone.

    At high altitude, one or many small bombs would have little effect because the rotating vortex has such a high inertia and is coupled so weakly to the upper atmosphere that it will not just stop because you want it to.

    The last option is to puncture the thermal boundary in the atmosphere at so many points that the energy is released in many small cyclones rather than one big tornado. That requires a technology akin to the gas guns, ground mounted propellers or helicopters, as are now used to break the thermal inversion over fruit growing areas.

    Imagine a fleet of 1000 heavy helicopters, say 1 km apart flying in a line of 1000 km along the thermal inversion boundary. What is the chance that it will destroy the profile before it triggers a cyclone that will then become a destructive tornado. They would need to be heavy inefficient helicopters or have blades that are inefficient with high drag.

    The direction of helicopter rotor rotation has an influence on the probability of initiating a tornado. The aim in the northern hemisphere must be to spin the air in a clockwise direction when viewed from above. That is safest because it cannot be reinforced by the coriolis force and so become self perpetuating. Which way do helicopter blades normally rotate. Is there a standard?

    When two trucks pass on a US highway they generate an anti-clockwise, vertical axis vortex. Maybe driving on the LHS of the road in the northern hemisphere tornado breeding grounds could reduce the problem.
  24. Mar 8, 2014 #23

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    Do you have data for that?

    Images of large nuclear tests show quite a strong shockwave blowing clouds away and vaporizing ground structures. You think it is not possible to build a bomb that would blow away a thunderhead or at least disrupt a forming funnel.

    That is a common misconception.

    In, say, an oil well fire, the pressure wave from the explosion blows away the vapors that are burning - it's like blowing out a candle.

    Please read the links.

    Note: seems that hobbyists have been using firecrackers to disrupt small vortexes in vortex generators. Seems to show you can, in principle, place explosives without adding to the ferocity of the vortex ... and quite easy in miniature.
    i.e. http://www.challenge.nm.org/archive/99-00/FinalReports/033/1stpg.html
    ... looks like backyard messing about to me.

    There are no end of dodgy patents surrounding this subject!

    Academic papers concentrate on approaches to predict tornadoes - though there are some odd ones about space-based approaches to affecting them.

    Iowa State has the kind of tornado lab needed to investigate this.
    You'd think they'd have a FAQ.
  25. Mar 9, 2014 #24


    User Avatar
    Gold Member

    http://wallisty.com/wp-content/uploads/2013/12/tumblr_m8g680zfgS1ru0y9ko1_1280.jpg [Broken]

    If not exploded, then simply unplug it.:bugeye:
    Last edited by a moderator: May 6, 2017
  26. Mar 11, 2014 #25
    If tornadoes don't cross the equator, it's only because they are uncommon except in tornado alley in the US. There are such things as "anticyclonic tornadoes", tornadoes that spin the "wrong way". You can Google "anticyclonic tornado" or check this source:

    Last edited by a moderator: May 6, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook