the home of online investigations

You can support the work of Bellingcat by donating through the following link:

Tactical Nuclear Ambiguity (Part II): Mushroom clouds, Flashes, and Bangs

June 17, 2015

By Dan Kaszeta

If we are going to use video and still images to make pronouncements about whether or not certain explosions are nuclear or not, then we must make full advantage of the audio and visual cues available to us on the image or recording.  The purpose of this post is to provide some basic information on the effects of smaller nuclear weapons, so that we have some basis for comparison when confronted with an image or a video.

Sources for Data and Formulas: I principally use Samuel Glasstone’s book “Effects of Nuclear Weapons.” I make significant use of both sides of the circular slide rule that came with many editions of this book. Rather a lot of the distances I cite are based on using the slide rule, so there is some ± error, so the calculated distances should be considered approximate.

Can a conventional explosion be as big as a nuclear weapon?

First, we must dispense with the notion that nuclear weapons have a monopoly on large explosions. The smallest of nuclear weapons were things like 155mm artillery shells and “atomic demolitions munitions”, with very small yields, often in the high tens of tons of explosive equivalent, although the exact figures remain classified. In principle, the brute force effect of a very small nuclear explosion can overlap significantly with the effects of a very large conventional explosion.

Location of the Explosion

Try to determine the location of the explosion. For large incidents there will be numerous videos and stills available, and the geolocation skills of others in the Belligcat community can help. I am not much of a geolocation specialist, but I have easily found others willing to help. By fixing the location of the explosion on a map, we can do other useful calculations.

Location of Vantage Point / Distance to the Explosion

For a particular video of an explosion, it is useful to determine the distance between the camera and the explosion.   While knowing the location of the vantage point is interesting and useful, the distance between the two is also of great use if the exact vantage point is not known.   We can use the time differential between the flash of the explosion and the arrival of the loud bang to estimate the distance.   This is because the flash travels at the speed of light (basically instantaneous as far as the sort of distances we are talking about) and the bang travels at the speed of sound.

The speed of sound varies somewhat with air temperature and altitude but not enough to really affect our rough calculation as we are looking for a basic estimate.   At 15° C at seal level, the speed of sound is 340 meters per second. In cold temperatures or at higher altitudes, it will be somewhat less e.g. 325 m/s at -10° C at sea level.

Try to use the sound of the bang, not visible blast effects for this estimation. If there are a lot of explosions going on out of frame, you can’t necessarily state that blast effects on camera are from the explosion being observed.  Blast effects may not be moving at the speed of sound, for various reasons.

Was the explosion in the air or at the surface?

If you can observe the actual location of the burst, was it at the surface or in the air? This can be significant in analyzing the tactics of the attack. With very small nuclear weapons the greatest radius of casualty production is prompt radiation, not blast or heat. In other words, the radius of immediate radiation effects is higher than the radius of significant destruction from the explosion or heat for the smallest nuclear weapons. As both ionizing radiation and heat are direct line of sight phenomena, this means that the tactical effects (if you are trying to kill people or stop tanks or such) – the radius of lethal effects – are optimized with a burst at some altitude above ground. (The ideal altitude is eminently calculable.)   A blast at ground level may be intended for use against a hardened target. If the nuclear fireball (see below) touches the ground, the amount of radioactive fallout will be much larger than the fallout caused by a burst at significant altitude.   As fallout occurs at some period of time after a nuclear explosion, I will not be discussing it in detail in this post.

A detonation at some significant altitude is probably not tactically sound for defeating a heavily buried / hardened target, but is optimized for a wider radius of effects against less-hardened targets on the ground.

Was it one explosion or two?

A common statement in alleged incidents is that “no conventional bomb could be that big” or “the (plane/shell/warhead) that caused the explosion couldn’t make an explosion that big”. This disregards the many scenarios where a smaller explosion causes a larger explosion. A bomb or artillery shell could cause larger explosions in an ammunition storage site, fuel depot, or an industrial facility. Indeed, when bombing such targets, creating a large secondary explosion is often the desired effect of the attack.

If you have videos, slow them down and look closely. Was there smoke, fire, or a smaller explosion in the same position in the seconds before the large, allegedly nuclear explosion? That is a strong clue that the large explosion was a secondary explosion.


How bright was the flash? The direct flash of a nuclear weapon is extremely bright and is likely to temporarily or permanently blind people. It may overload a CCD camera temporarily or even permanently damage it.  Was the flash directly observed? Or did you see the flash on other items in the picture. It should be noted that in overcast conditions the flash may behave somewhat differently as what appears to be the flash may actually be a reflection.

How long was the flash?: The illumination time of a nuclear explosion can be used to provide a rough indication of the flash. There is a defined period, which can be calculated mathematically, during which time the fireball will cause a brilliant flash. The flash of a nuclear explosion is actually quite quick.

You cannot really use flash time and flash time alone to make a nuclear/non-nuclear decision. A conventional explosion can often have a flash/illumination time similar or even much longer than a much larger nuclear explosion, particularly in situations where there is a lot of flammable material involved. I have personally witnessed a large conventional fuel explosion of approximately 500 gallons of thickened gasoline. The explosion caused an illumination time well in excess of 5 seconds.

It is possible to use the flash time to make a few educated guesses. The following chart is from US Army Field Manual 3-3-1 (1994 ed) and can be used to do a rough order of magnitude (± 0.1 x to 10x) estimate of the explosive yield of an explosion. Unlike Veterans Today’s dubious claims, you cannot use this table to make a nuclear versus non-nuclear explosions.

illum time table

Using this table, the thickened gasoline explosion I helped cause in 1993 would be between 6 to 600 Kilotons of nuclear yield, due to its approximate 5 seconds of illumination time. This is, of course absurd. The flash times of small nuclear weapons, the ones historically developed for battlefield use, are 2 seconds or less.

How big was the fireball?

You can use well-understood mathematic formulas from the Glasstone textbook on the effects of nuclear weapons to derive the radius/diameter of a fireball from a nuclear weapon. I have derived the approximate sizes of fireballs for various yields of weapons using Glasstone’s formulas and they are in this table below:


Can you use an object in the video or still image to estimate the size of the fireball?   Compare the fireball to the chart above.

Cloud characteristics:

First of all, it is important to understand that a mushroom shaped cloud, while an icon of the nuclear era, is not indicative of a nuclear explosion. Conventional explosions can easily cause mushroom shaped clouds. So, do not fall into the erroneous “It was a mushroom cloud, it must be nuclear” – there are numerous examples of conventional explosions causing mushroom clouds, including hundreds if not thousands of examples documented from the pre-nuclear era. Volcanoes can cause mushroom clouds. The Wikipedia page on mushroom clouds is not a bad overview of the basic principles of how and why they form.

The following painting is of the Siege of Gibraltar. It is called “Vue du siège de Gibraltar et explosion des batteries flottantes” and dates from about 1782. The artist is unknown.


The following photograph is from the eruption of Mt. Redoubt, a volcano in Alaska, in 1990. The photo is from the US Geological Survey.


A nuclear explosion will, of course cause a mushroom cloud. Because the size and shape of a nuclear mushroom cloud are well-understood phenomenon, there are mathematic formulas to calculate their size. The characteristics of a mushroom cloud can be used to estimate the explosive yield of a nuclear detonation.  As with flash time, It should be noted that we cannot use this method, directly, to assess whether an explosion is nuclear or non-nuclear, particularly at the very low end of the scale of possible nuclear weapons. However, the following table will give you an idea of how big a mushroom cloud, once it has stabilized, should be from a nuclear explosion of a particular explosive yield. As you can see, even a very small yield will give quite a dramatic cloud.

Cloud Parameters

How tall and wide is the cloud in the alleged incident? I think we are still evolving when it comes to things like estimating the height of something in the YouTube universe. However, we sometimes have things like mountains of known height and buildings of known height in the video footage that we can use an a gauge to give us an approximate height of the suspect cloud

Immediate blast, thermal, and radiation effects:

The question the skeptical observer should ask is – at the distance at which we believe the video/still vantage point was from the explosion, what should have happened to the photographer and his camera?

Blast effects

A principal measurement of blast from an explosion is the pressure of the overpressure caused by the movement of air from the explosion, often measure in pounds per square inch (psi). For purposes of the discussion here, the blast wave, once it exceeds the perimeter of the fireball, moves at about the speed of sound. I have no special data table to tell me at what distance a digital camera gets ruined by blast effects. Window glass shatters with approximately 1 psi of blast overpressure. Certainly, some cameras and smart phones will sustain damage at this level of overpressure. However, I conservatively use figure of blast overpressure of 2 pounds per square inch (psi) as likely to crack a camera lens or seriously damage a smart phone or camera. The table below, calculated from the circular slide rule calculator provided with the Glasstone book (I have one – example pictured, and no purer example of my membership in the CBRN geek fraternity exists) the 2 psi blast overpressure radius (for a surface detonation) for a variety of small-ish nuclear detonations. I apologize for not including anything less than 1 KT in the following tables as the slide rule bottoms out at 1 KT.

blast distance

Thermal Effects

A nuclear explosion causes a lot of heat and light. In general terms, the amount of heat generated is greatly in excess of what would come from a conventional explosion. Again, the amount of heat from the nuclear explosion can be calculated using equations and tools from the Glasstone textbook. It is also important to note that with small nuclear weapons a very large percentage of the thermal energy is emitted in a fraction of a second. For a 1 KT explosion, about 70% of the thermal energy is transmitted in 250 milliseconds or less. Conventional explosions often have a more spread-out transmission of thermal energy, particularly when a lot of materials such a propellants or fuels are in a secondary explosion.

This chart shows some of the various thermal effects that can be expected from various sizes of explosion:

thermal distance

Evaluate the images and video for thermal effect. Did anything in the image near the vantage point catch fire? How much was melted or ignited at distances between the observer and the explosion.

Radiation Effects

I used the calculator to come up with approximate distances for various immediate radiation doses. The 1 Gray (Gy) (i.e. 100 rem) radius would cause most of the people exposed without shielding (i.e. standing in the open) to eventually get acute radiation sickness after a latent period. Very few would die from it. The 10 Gy dose, however, would make everyone sick quite quickly, and nearly everyone exposed to that level would die.  The 1 Gy and 10 Gy distances are shown in this table.

Radiation Dose distance

We can’t really use the biological effects to judge what’s on the video, as it takes time for signs and symptoms to develop. However, immediate radiation affects cameras. In the old film days, the plastic or cellulose film acted as a scintillator plastic, i.e. gamma photons would react with the plastic and cause flashes of light that would expose the film, even though it was inside the camera. In this age, we are mostly using digital charged coupled device (CCD) cameras. The CCD chip of a camera, and indeed plastic in and around the lens, can scintillate, causing strange effects on the recorded image. This is not an area where I am a specialist (my nuclear weapons effects training occurred in the film era), but there are several interesting studies (here and here) on the effects of ionizing radiation on digital cameras.  At a distance from a nuclear explosion, there should be a general graying effect on the video / still image, and a number of hot pixels, more or less randomly distributed. (The lens is focusing visible light photons, but is not focusing gamma photos or neutrons.) There are many practical demonstrations of radiation effects on CCDs.   A YouTube video shows random pixellation from a Radium 226 source on a CCD.  Another one shows effects from Americium.

Putting it All Together:

Using these various data tables, is any of the data consistent? In particular, consider the following questions:

  • Is the size of the visible fireball big enough to be a nuclear weapon?
  • How is the observer / photographer / camera affected?
  • Are the flash and thermal effects dramatic enough, particularly at the distance of the observer / vantage point?
  • Has anyone reported any burns or flash blindness?
  • Is the flash time long? If it is, it probably isn’t a small nuclear explosion.
  • How big is the cloud?
  • Are the cloud size, fireball size, and flash time all consistent with each other?


Dan Kaszeta

Dan is the managing director of Strongpoint Security Ltd, and lives and works in London, UK. He has 27 years experience in CBRN response, security, and antiterrorism.

Join the Bellingcat Mailing List:

Enter your email address to receive a weekly digest of Bellingcat posts, links to open source research articles, and more.


  1. Frank Phillips

    Good write up on the identification.

    The only real way to identify an unexpected nuclear event is radiation and thermals. It is an interesting concept that a nuclear strike would be unexpected, especially in the military space. Other than dosimetry, I do not think I have every read anything on validation of a strike in any manual, which would be a rather new concept strategically speaking.

  2. Bob

    A very interesting article, but are we expecting to see potential nuclear explosion any time soon?

    • Dan Kaszeta

      No… I hope not. If you read part one of this post, there has been a non-trivial number of people suggesting that “Explosion X was a nuclear weapon” and this post is part of an effort to combat some of the silliness out there

      • Bob

        Yes I did read that. It was also interesting. Strange how the world media didn’t pick up on what that group were purporting to be a mini nuke. But yes there is a lot of silliness out there.

          • smorgasbroed

            What I learn from what you say is that … sorry … but people are stupid, illiterate or both or a combination of all. Or willing to pour gasoline on burning fires.

            Dan, perhaps you should list a few videos of real nuclear explosions and the damage they do. People seem to have absolutely no clue how devastating nuclear explosions are. Is it even possible to have such small nuclear explosions like those conventional ones referred to in some vids?

          • Dan Kaszeta

            Good idea.

            I haven’t spent huge amounts of hours delving through all the archive film footage, but to date the smallest above ground detonation I have found video of is the Ivy Flats operation in 1962. This operation tested the very small W54 warhead, which I believe had the smallest yield of any device in the US arsenal. This video is actually quite interesting. The detonation itself is around 9:00 in the video, but it is worth watching the whole thing for the context. The yield of this detonation was extremely small for a nuclear weapon, somewhere around 10 to 20 tons of HE equivalent.

          • PavewayIV

            I’m curious why anyone would assume any of these Yemeni vids (or the Damascus one, for that matter) are air bursts? The one on the side of the pretty steeply-sloped hill had no apparent surface targets. Less clear on the three others, but Damascus was certainly a penetrator of some type. The targets seem to be underground bunkers, not surface structures or military units in the open (like and armor group). The explosions are all well-beyond the 2000 lb. class JDAMs or SBUs that the U.S. sold Saudi Arabia – it’s reasonable to assume their secondaries from ammo or rocket storage. Those are not something a Middle East country keeps in open surface pens that an air-burst could set off. If any of these were earth-penetrating weapons, then you never saw the blast.

            There’s no evidence they were nuclear of course, but comparing the secondaries with historical footage of air- or surface-blast nuclear weapons isn’t going to tell much.

            There is no footage (that I’ve ever seen) of an EW weapon aside from sketchy stills of a purported Chinese one and a very-high-altitude anti-ballistic missile shot.

      • Harald K

        What would be more interesting was if Bellingcat could get into forensic seismology… it can be used for a lot more than nuclear explosions, and surely there’s a lot of public seismology data.

        • Dan Kaszeta

          I freely admit that seismology isn’t my speciality. From a nuclear monitoring standpoint there’s a brief introduction here and I will gladly post more links. Indeed, if someone wanted to do a “forensic seismology for laymen” intro article, it could easily get published here on Bellingcat.

  3. MBobrik

    A very fascinating read. Hope I will never have to use what I learned from this article ( but I am afraid, given the direction this world is heading, I will… )

  4. smorgasbroed

    Thanks for providing footage of the Davy Crocket. The detonation was quiet small which surprised me a lot. But the footage is very bad so it is really hard to tell. What is very good visible though is the short and very bright flash.

    So what type of scenario is possible today and possible to be carried out by some terrorist group. What would be doable, ignitable, carryable. Dirty suitcase bombs … is it realistic today that these exist and are of interest to someone or is this all to hard to get, disguise, keep secret etc.?

    • Dan Kaszeta

      Making a very small nuclear warhead, like the Davy Crockett, is actually very very hard. But if you want to read about small weapons, I wrote an article about this in 2012.

      • smorgasbroed

        Thanks Dan … I scrolled all the way down to the TL;DR section, which did not exist back in 2012 😉 …

        I wondered that you did not relate to the Clooney Movie “Project Peacemaker” which exploits exactly this topic as a very realistic thread.

        I ask myself why Hollywood is pressed to censor every nipple but must not show a disclaimer for make-believe and unrealistic horror scenarios which people might grow some nightmares from …

    • Dan Kaszeta

      Peter, thanks for that video. To be honest, the image quality is poor on that one. But very few thermal effects appear to be present outside the direct blast. For a very small nuclear explosion the thermal effects would be in a greater radius than the blast radius. The illumination time seems really rather too long for a sub kt nuclear, several seconds. Whereas in a sub-kt nuclear explosion, the vast % of thermal and visual light will be over quickly, <0.2 secs.

      Interesting video, nonetheless.

  5. kevloral

    I think you should also mention the easiest way to distinguish between a conventional explosion and a nuclear one: the double flash the latter characteristically show.

    • kevloral

      Forgot to mention that, depending on yield, the frame rate of the typical youtube video may not be high enough to distinguish that double flash though.

    • Steve Cook

      The double flash was used as part of the detection and measurement using the AWDREY (Atomic Weapons Detection Recognition and Estimation of Yield) equipment used by the United Kingdom Warning and Monitoring Organisation (UKWMO) during the Cold War.


Leave a Reply

  • (will not be published)

You can support the work of Bellingcat by donating through the following link: