In July 2020, Facebook page "Meteorologist Greg Pollak" posted a video (Facebook link):

WHOA! Check out how fast this river of lava was moving in Hawaii back in 2018 👀🌋 Note: This is not a time-lapse video

🎥: Ken Boyer

The video is a river of lava that is hooning; I would estimate 60-80 km/hr.*

I first saw the video on my phone, and I was like "No way! Fake", and then I watched on on a bigger screen and I was like "Err... way?"

The comments include a lot of people calling it fake.

Was there a river of lava flowing that fast?

* I see from early responses that people are focussing on my estimate more than I expected. To clarify; the only reason I gave an estimate is to protect against link rot; there is a good chance the video will not be available to some viewers at some point in time, and I wanted the question to be able to stand on its own.

My estimation technique was to imagine it was a cross road, and the lava was a car, and guess what the speed limit on the road would be. Hardly precise...

  • 8
    Well there is info about lava from that area flowing around that speed volcano.oregonstate.edu/faq/how-fast-does-lava-flow
    – Joe W
    Jul 18, 2022 at 0:41
  • 8
    So, from an image processing perspective, the only way to fake this is to have all that lava simulated in super realistic CGI. Note the droplets of lava that jump up and down. They behave "naturally". If you'd alter the speed of it, you'd feel that it was fake, you would notice it going too fast, physics would be accelerating them at double the force of gravity. Yet these look very natural, so definitely you can scratch speedup of the lava part of the video. Either Hollywood level CGI, or real. Jul 18, 2022 at 10:50
  • 2
    Just FYI this is "normal" so to speak. You can find any number of "videos of lava rivers".
    – Fattie
    Jul 18, 2022 at 13:54
  • 4
    Without information on the field of view, or a reference of scale, it's impossible to judge the exact speed. I estimate 2 to 3 seconds for a spot on the visible surface to cross from right to left of the frame, so for it to be 60kph, we need to be seeing a 30 to 50 metre stretch (100 to 160 feet); if it's 30kph (as suggested by some of the answers) we're seeing 15 to 25 metres (50 to 80 feet). 30kph is still fast enough to overtake Usain Bolt before he reaches 20 metres.
    – IMSoP
    Jul 18, 2022 at 14:23
  • 2
    Having spent a great deal of my life intently studying rapids in rivers (of water ;-), I thought I would share a little perspective. Flows always seem much faster than they really are--people usually guess at least twice the true speed. The fastest laminar flow I ever studied and measured (this excludes waterfalls) was the initial drop at Lava Falls in the Grand Canyon, which (at a flow of 35K cfs) ran about 7-8 meters per second. To my eye, this lava flow is running around 10 mps, perhaps a tad faster depending on just how far away from the camera it is. That works out to 36 kph.
    – whuber
    Jul 20, 2022 at 13:16

4 Answers 4


This seems possible. The U.S. Geological survey states the following:

Fluid basalt flows can extend tens of kilometers from an erupting vent. The leading edges of basalt flows can travel as fast as 10 km/h (6 mph) on steep slopes but they typically advance less than 1 km/h (0.27 m/s or about 1 ft/s) on gentle slopes. But when basalt lava flows are confined within a channel or lava tube on a steep slope, the main body of the flow can reach velocities >30 km/h (19 mph).

Lava flows destroy everything in their path


It's Ken Boyer's video from 5pm 15 June 2018.

A more complete version is here: https://www.facebook.com/watch/?v=1469403073205921

Ken Boyer Photography says:

Nothing is even slightly altered. It's straight from his camera. Really wish people would stop assuming this is fake when they obviously don't have a clue what they are looking at.

  • 30
    There are two clues in the video which support that it isn't speeded up. a) the two people walking and b) the occasional rock thrown up out the lava stream does not have exaggerated vertical motion. Jul 18, 2022 at 7:42
  • 61
    This is useful background, but doesn't really feel like a strong answer: plenty of faked videos are claimed to be real by their creators. An independent corroboration that lava was flowing at that speed on Hawaii in 2018 would be more valuable; or at least evidence like some of the links posted that it is plausible.
    – IMSoP
    Jul 18, 2022 at 8:45
  • 29
    It looks like a telephoto lens was used. Which makes sense if you want to be far away from a fast river of lava. It also compresses the apparent distance between people and lava, and, since it shows a smaller portion of a river than a wide-angle would, it makes it look like the lava crosses the image faster. It's basically the same effect which is used for large moons above the horizon : youtube.com/watch?v=afHfMMC-MJE Jul 18, 2022 at 8:51
  • 6
    @WeatherVane Captain Disillusion would be quick to point out that the power lines provide points where two separate shots could be spliced together - one sped up, and one not. Jul 18, 2022 at 10:55
  • 8
    @user253751 a third clue is the smoke rising from the lava, sometimes to left, sometimes to right. That too looks natural not speeded up. The speed-up faking would have to be very clever, not to speed up the lava splashes and smoke too. Jul 18, 2022 at 12:41

According to the USGS report:

Fissure 8 lava fountains 52-70 m tall showered spatter onto the cone overnight into 19 June (figure 411). [...] Standing waves were visible within the channel and cascades/rapids were visible near the base of the 50-m-high cone. The maximum flow velocity in the channel was measured at 28 km/hour.

Now this report is really long and that statement is particular to one day's observation, so it's possible it was higher on some other day, but for most days only the height of the lava fountain is reported.

The eruption was filmed from a drone which referred to in the caption of an earlier figure (399) as the source of the lava flow measurement.

There's brief ground-level video posted by USGS in a montage (on Twitter) which bears some similarity with that one (but it's not exactly the same), with fairly similar commentary.

The lava flows out of fissure 8 and into a main channel, where the lava is moving up to 15 miles per hour, about the speed an average human can sprint over good ground. The flow slows down significantly before it reaches the ocean. [...]

“In this video taken from the Leilani Estates subdivision, lava at fissure 8 pulses above the cinder cone adding fragments of lava (spatter) that build the cone higher,” the USGS explains. “From fissure 8, lava flows freely over small cascades (rapids) into a well-established channel. Near the vent, lava is traveling about 24 km per hour (15 mi per hour). Lava slows to about 2 km per hour (1.5 mi per hour) near the ocean entry at Kapoho.” [...]

Late yesterday, the eruption vigor increased, and the rate of lava flow from fissure 8 ticked up significantly.

Standing waves visible in channel that moves 7.7 m/s (17 mph) near vent.

7.7 m/s is 27.72 km/h, so it's probably referring to the same event/measurement as the previous quote.

Now for "back of the napkin" verification... In the video from the question, there's a flying bit of lava that takes about one second (from 0:04 to 0:05) to cross the width of the (two-lane) road. According to Wikipedia:

In the United States, the Interstate Highway standards for the Interstate Highway System use a 12 ft (3.7 m) standard lane width, while narrower lanes are used on lower classification roads.

So, ignoring camera FoV/perspective issues, that's about 7.4 m/s for the flying bit.

Just so this doesn't get drawn out in the comments, there are numerous sources of error for this kind of estimate, including:

  • actual width of the road
  • actual time the blob is in the air (no millisecond timestamp on FB)
  • FoV-related, i.e. distance from camera to end of the road vs distance from end of the road to the center of the flow, which the fast moving part. (The proportion of these two times the width of the road determines actual distance travelled by the lava blob.) You can see on the USGS drone video that this latter distance, from the end of the road to the center of the lava flow looks appreciable, e.g. relative to the size of houses. On the other hand, the road hardly appears to widen in the video in question, despite the camera being positioned on its axis, which suggests a pretty long telephoto lens being used, so the camera could be located pretty far off from the end of the road.

So, for these reasons, this my calculation can easily be off by a factor of 2 or even 4 in either direction.

Fast lava flows aren't too common but not unheard of. One in Africa (Mount Nyiragongo in 1977), was supposedly clocked at 40mph, which according to Guinness is [still] the world record on that matter.

Addendum. I cannot find anything too conclusive on this, but it's possible that the swirly motion of the lava from the OP's video amplifies the subjective impression of linear speed, based on some experiments, which however found a limited subjective increase, around 10%. The hot points of the lava are quite bright relative to the background, which may also increase the subjective impression of speed; the latter paper also notes "substantial intersubject variation" in this regard.

  • 1
    There aren't any roads of Interstate Highway standards anywhere on that island. That is a local residential neighborhood road with lanes much narrower than an Interstate. It's hard to get a precise measurement from aerial photography, but, judging from the Google Maps measurement tool, most of the roads in that particular subdivision have lanes of around 7ft or less with Leilani Ave itself being maybe 8ft lanes. I haven't driven on those particular roads myself, but most of the other residential roads in that area that I have driven on have indeed quite narrow lanes.
    – reirab
    Jul 18, 2022 at 16:56
  • Also, this video appears to be shot through a lens that is making the background look much larger than it actually is relative to what's in the foreground. Look at the people, for example. They're larger than the truck that is actually closer to the camera.
    – reirab
    Jul 18, 2022 at 17:02
  • 2
    @reirab: I freely admit that my calculation can be off by a factor or 2 or 3 in either direction for numerous reasons (incl. actual size of the road, not having ms timestamps, so the "one second" could be nearly two etc.) I don't see anything unusual about the size of the people relative to the small, SUV-size truck though. Almost certainly this is a long telephoto lens. Jul 18, 2022 at 17:09
  • 3
    @reirab there isn't a lens that does that more than a telephoto lens. The people don't look larger than the truck to me, but measure slightly less than its height, which is credible. The apparent distance travelled by flying lava isn't going to be less than the road width that Fizz is comparing width, so the estimate would be the minimum speed. Jul 19, 2022 at 8:28
  • @WeatherVane: true when it comes to lenses and sharp images, but I suspect that differential heating of the atmosphere (hotter farther away) may cause such an effect, albeit the father away object will also appear blurry; see e.g. en.wikipedia.org/wiki/Astronomical_seeing Jul 19, 2022 at 9:03

You don't mention your method for estimation so it's hard to respond to the specifics of your question. In general, I don't see any reason why a 35 to 50 MPH flow isn't possible once the initial flow has paved the way by heating up the ground and laying down a thick layer of insulating hardened lava to smooth the way.

Here's a link to an old VHS video that I have a personal copy of (just wish it were available in DVD quality now). It's an overview of the film work by Katia and Maurice Krafft, a French volcanologist couple. They are directly responsible for educating government leaders about the serious nature of volcanic eruptions in time to get them to order the needed evacuations for pending eruptions.

Regardless, the video is fascinating to watch and well worth your time despite it being almost an hour long. There's some flows around the 52 minute mark that look like they could be as fast as the ones you pointed to. Again, they are in well established channels, likely well paved by solidified but still melting point hot lava.

As an aside, volcanologists have to be a little nuts to walk in that close to danger. They wistfully talk in the video about wanting to ride a flow in a titanium canoe someday.

They never managed that. They died while doing some filming of an eruption at Mount Unzen:

In June 1991, while filming eruptions at Mount Unzen in Japan, they were caught in a pyroclastic flow, which unexpectedly swept out of the channel that previous smaller flows had been following and onto the ridge they were standing on. They were killed instantly along with 41 other people, including fellow volcanologist Harry Glicken, several firefighters and journalists also covering the eruptions.

  • 1
    That was my first thought when watching the OP video. Why are those guys unconcernedly dawdling? With a flow of that speed, a slight diversion upstream would vaporize them in seconds. Jul 18, 2022 at 15:20
  • 2
    FWIW, the fastest lava flows that have been recorded were around 40 mph according to Guinness World Records.
    – reirab
    Jul 18, 2022 at 17:16
  • 1
    I've updated the question to explain my "estimation method". I didn't expect it to be the focus of the question; sorry for being misleading.
    – Oddthinking
    Jul 18, 2022 at 17:54
  • 2
    @Oddthinking - Oh, not at all. I was just remarking on it because it's tough to answer specifically without that info. Given the graininess of the video I posted a link to, it's even harder to tell if it's been sped up at all. I don't think it has been as they weren't so much trying to sensationalize as to do accurate research. They are generally well respected for their contributions.
    – user63491
    Jul 18, 2022 at 18:57
  • 3
    @RayButterworth, a "slight diversion upstream" would push the lava out of the channel, and lava moves much slower when it's not confined to a channel.
    – Mark
    Jul 18, 2022 at 21:51

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .