In the future, don't go close to mountains during monsoon season - late May to August end; that period is specifically for Southeast Asia. For the rest of the world, you can find it out.
As I've heard tell, this was a glacial collapse into and blocking a river, which then built up a head of pressure behind it, bursting to release a hammer blow of river water and debris that travelled a great distance.
Similar things can happen due to landslides into rivers; it depends on the topology of the land, here long narrow valleys carved out by heavy water flows.
The warning being, don't live in a natural "drainpipe".
In mountainous regions that's often not much of an option, there's a lot of vertical sides and valley bottoms.
From what I've understood, the collapsing glacier either contained enough liquid water or melted enough from the impact that it continued flowing downstream. It covered ~20 horizontal kilometers in about 7 minutes before hitting the border crossing.
It had the look of a large volume of water quickly released.
Massive tonnages breaking off of a glacier is still going to be mostly ice on impact .. there's a chance debris and ice chunks made a temporary dam as the bulk of the now shattered glacier ice released water putting pressure on that weak dam - leading to collapse and release.
We do know that a glacier substantially collapsed.
What contributed to the amount of water in the debris flow isn't clear. Speculation, from trained geologists, is that the friction of the glacier flow itself contributed to melt. I'm speculating that recent rains or higher temperatures might have contributed to melting on the glacier (and have seen some commentary speculating about this).
Speculating on my part: There might also have been accumulated water below the glacier, liberated by the ice flow, or perhaps groundwater which was evacuated along with other material. Events seem to have precipitated too quickly (roughly 7 minutes from galcial collapse to Gyirong Port inundation) for the collapse to have formed, then broken, a dam on tributary waterways upstream of the port.
GLOF is a recognised threat, but this event seems to have been at best only partially explained by that, and may be an entirely different mechanism. The follow-on studies should be fascinating, in the way that horrific disasters often are.
> Speculation, from trained geologists
Probably about on par with speculation from trained geophysicists I'd warrant.
Re "speculation, ...": my point being expertise, but based on background knowledge rather than close event-specific observations or measurements. Yes, geophysicists would be an equivalent level of expertise.
That might take some infrastructure to cover everywhere such things can happen (eg: the beds of rivers in Western Australia's Pilbara and Kimberley regions have a high flash flooding risk in the wet season.
What's the land area you'd like seamlessly and continuously monitored by a mobile object with moving parts ... and how is that a better solution than using static solar powered instrumentation?
Geologists have used the latter for many decades. Monitors upslope of at-risk glaciers, solar/battery powered, and communicating by radio (or Internet / satellite links) could be part of this.
See for example the extensive seismic monitoring of California's San Andreas Fault, particularly at Parkfield and through the Carrizo Plain / CA Hwy 25.
<https://earthquake.usgs.gov/learn/parkfield/safod_pbo.php>
<https://en.wikipedia.org/wiki/Carrizo_Plain>
<https://en.wikipedia.org/wiki/California_State_Route_25>
(Different tech, obviously, but in situ ground-based sensors which share some similarities.)
Multiple sat observations a day is great, obviously, but sudden collapses / changes often won't get flagged for seven or eight hours after the event.