Looks like a warning siren could have saved some lives. But apparently this was a black swan.
What is your definition of black swan? This paper is from 2022 and scientists have been warning about climate change for decades. It's not like this possibility wasn't anticipated. It seems likely we'll see more glacial events as the atmosphere warms.
It usually takes entire icebergs calving off to produce such seismic events and intense monsoons to create such flooding.
It seems that past history (especially recent) will not be a good predictor of future events.
While I'd wait for confirmation about the cause of this specific incident, the point you seem to be ignoring is that, due to climate change, events that would have been considered "black swans" historically will become more and more common, to the point that they are expected to occur.
It’s the inflection point when our priors change.
And I have no idea where you got the "rationalized as inevitable in hindsight" thing. I've never heard of that in relation to black swan events before.
This flood is like The Big One in California. We know it is going to happen, it is virtually a certainty. The only question is when and where.
If glacial floods come as a surprise to you, it is only because you have been deliberately ignoring all the climate change warnings. So, a short recap: we'll see massive droughts, giant wildfires, huge floods due to rainfall, significantly more intense hurricanes, mass starvation due to large-scale crop failure, and low-lying countries drowning due to sea level rise. Oh, and add a high likelihood of the AMOC collapsing triggering a cooldown of the East Coast and Europe to the list.
I've been reviewing what coverage I can find of the present event in Nepal/China, in context with Section 5.1, "Comprehensive approach to disaster risk management", of TFA.
It seems that what would be required would be:
- Active surveillance of glacial lake formation.
- Active mitigation of glacial lakes, reducing risk.
- Warning and response infrastructure, procedures, and protocols along the likly flood path.
From video of what I presume are among the worst-hit areas (see: <https://twitterwebviewer.com/?tweet=2092542957060841592>), there appear to have been at best a few seconds of warning. Very robust structures are demolished virtually instantly. Even had people been sheltering within those structures, or headed up-slope above the water flow, either structural integrity would prove insufficient, or response time inadequate to reach a safe elevation above the debris flow. The latter problem is compounded by what appears to be an extraordinarily strong air blast (likely hurricane-strength or greater) as well as massive amounts of airborne debris, much of it rock.
Terrain is severely constrained in the region, and the only buildable regions are in river valleys themselves. The Gyirong Port border crossing is located deep within a river channel, and despite robust construction simply could not survive the event.
See:
<https://en.wikipedia.org/wiki/Gyirong_Port> (Wikipedia article on the port.)
Map / satellite view: <https://www.google.com/maps/@28.2792579,85.3752907,1341m/dat...>
(Take a few minutes to examine the region around the port itself ... there are several communities and businesses all well within the disaster zone, and I suspect that obliteration to those matches or exceeds what was captured on video at the crossing itself.)
From the first video at this Twitter post, a camera situated roughly 20m above the base terrain did not survive the event (though it transmitted for roughly a minute).
A second camera (I believe near the same location) shows similarly catastrophic damage, and no chance for anyone at base level to climb to a safe elevation before being overtaken by the debris flow.
Wikipedia notes "Floodwaters reached settlements as high as 270 feet (80 m) above the river level", citing <https://english.onlinekhabar.com/rasuwa-flood-in-pictures.ht...>.
Climbing that far would require 4--8 minutes for even a highly-fit person. Available response time in the videos I've seen is on the order of 1--2 minutes. Anyone beginning within the debris zone is already walking dead.
As the Cowboy Junkies said:
But if there's on thing in my life
That these years have taught
It's that you can always see it coming
But you can never stop it
Anticipation isn't sufficient. Other than entirely abandoning such locations during all periods of high risk, loss of life is likely, and catastrophic property and infrastructure damage inevitable.- The crossing was substantially renovated in January of this year following an earlier flood which washed out the bridge connecting the station (in China) to Nepal.
- The crossing, part of China's Belt and Road Initiative (<https://en.wikipedia.org/wiki/Belt_and_Road_Initiative>), by needs must sit on the principal highway through the region. Which ... follows the river valley. Now, it's possible for that highway to be raised substantially off the valley floor, but consider that "inundations reached heights of 800m (2,600 ft) in a valley some 7km (4.3 mi) from the origin" (<https://www3.nhk.or.jp/nhkworld/en/news/20260827_22/>, via Wikipedia).
- The inundation reached the Gyirong port seven minutes after the initial avalanche. Even with absolutely timely warnings, that is only just barely enough time for a fit person to reach a safe elevation above the port itself. Wikipedia now mentions two people (of apparently hundreds present) being rescued from Gyirong port.
- Early reconnaissance footage above the site of the former Gyirong port shows nothing but bare land. All of the apparently quite substantial structures on the site were simply scrubbed from existence. The forces involved are formidable. <https://www.globaltimes.cn/page/202608/1369215.shtml>
I believe that the end of this video shows post-incident video of the same site, though frankly it's unrecognisable. I'm sure we'll have more definitive evidence soon. <https://www.msn.com/en-ca/video/other/china-video-shows-muds...>.
- The event is, as has been noted, foreseable, but also unprecedented. There are no events of comparable magnitude in the area, despite previous flooding. The likelihood that this week's incident represents a potential new baseline must be considered.
- Debris flows were moving at roughly 50m/s, or 180 kph (about 110 mph). That's not only an absolutely formidable speed, but a formidable force, where kinetic energy = mass * velocity squared.
The only possible safe refuge I could see would be some sort of underground bunker, into which all staff and travelers would have to locate to in a matter of seconds, and then sealed against water intake, this in a busy border crossing with multiple nationalities and languages in a region with a < 70% literacy rate.
And that's for the Gyirong border station itself. Note that that was a Chinese facility, where China is the vastly wealthier and better-resourced partner to Nepal. Nepal's capabilities to effect similar measures at an equivalent facility or in numerous communities throughout the region is quite limited.
Also affected were numerous other facilities, including a shrine (being visited by numerous Hindu pilgrims), apparently utterly destroyed and all present killed, numerous villages, etc. Roads are heavily damaged if present at all, bridges are washed away, and in a region with tens to hundreds of thousands of inhabitants (based on various county/district Wikipedia pages I've examined), a dozen or so rescue helicopters have so far been summoned.
Oh, and a new glacial/debris dam and lake have formed which is feared may pose an additional outburst flood event within days (<https://www.theguardian.com/world/2026/aug/27/nepal-tibet-fl...>).
Ridgelines and slopes are terrible to build on (and often use) because there is little to no actual flat usable area without extensive construction. They also tend to be unstable and slide/erode.
That said, if I was in that area, of course i'd never build/live on anything but the highest ridgeline or slope around.
One of those 'long term/outlier risks cause daily costs' type of things. :s
<https://en.wikipedia.org/wiki/Lake_Agassiz>
<https://www.nature.com/articles/nature08954>
<https://www.sciencedirect.com/science/article/abs/pii/S00335...>
- The site captured in the video, the Gyirong Port border crossing between China and Nepal, was struck by the flood roughly seven minutes after the presumed avalanche thought to have triggered the event.
- Flood inundation is seen up to 800 m (2,600 ft) above river level, and the camera position in the video at approximately 20--25m (roughly 6--8 stories) above ground level was engaged in the debris flow.
- It would take a fit person 4--8 minutes to climb even 250m or so above ground level, presuming instant notification and a clear evacuation path. Even with an instant warning of upstream conditions survival was overwhelmingly unlikely.
- Subsequent footage appears to show complete obliteration off all structures, with no trace remaining. This is based on preliminary imagery, without precise location information, and in a landscape changed beyond all recognition.
- Astoundingly, two survivors from the Port are mentioned in Wikipedia presently.
Though yes, one of the peculiarities of flood risk is that elevation equals safety. That's modulo a number of other factors though.
In particular, with an event of this magnitude, wind blast is a major effect. Toward the end of one video a line of buses in a car park is seen shifting backward, I suspect due to wind. These can reach well above hurricane force, and of course carries its own debris and threats, where even grass and straw can penetrate deeply. There's also the risk of secondary landslides and the like in steep terrain. One of the regions affected by this event has an etymology from "mud wall", which may be a reference to the geology of the canyon walls themselves. Those could easily be unstable.
Best option is to be lucky and not be there when things go horribly wrong.
Rescue operations in Nepal are in good hands, but this is sadly because those same response orgs had to respond to a similarly brutal cloud burst floods back in 2013 [0], 2021 [1] and 2025 [2].
It also shows the very real and brutal effect climate change is already having, and will only get worse.
[0] - https://en.wikipedia.org/wiki/2013_North_India_floods
[1] - https://en.wikipedia.org/wiki/2021_Uttarakhand_flood
[2] - https://en.wikipedia.org/wiki/2025_Kishtwar_district_flash_f...
That’s just pure horror, nightmare material
You could probably count 100+ people in the death zone of the border crossing video alone. The final death toll will be terrible, with most never found.
For indians, you also have the religious aspect because the immigration center you go to to visit Mt Kailash is there.
That immigration building was the white one that got completely shredded in the popular video going around.