虎嗅

The deadly mudslides in Nepal seemed to have precursors. Can such events be predicted in the future?

原文:尼泊尔夺命泥石流原来有预兆,那未来可以预警吗?

Summary of the Core Content in Plain Language

This is a professional analysis report published by the journal Nature in September 2026 regarding the recent major high-altitude ice and rock collapse disaster on the China-Nepal border. The deadly flash floods caused by this disaster have resulted in over 1,000 deaths and more than 4,000 people being missing. Retrospective analysis of satellite data revealed that several weeks before the incident, and even just 7 days before it occurred, there were clear signs of accelerated movement and instability in the glacier and the underlying rock mass in the affected area. However, no effective warnings were issued at that time. Experts have pointed out that the issue is not with the monitoring technology’s ability to detect such signals, but rather with the fact that the global monitoring and early warning systems for disasters caused by the combination of ice and rock collapses are virtually non-existent. This incident serves as a stark reminder for the world to enhance its preparedness for high-altitude disasters.

---

Detailed Interpretation of Various Aspects

1. This disaster was not an ordinary flash flood; it was a series of events that no one could predict in advance

Many initially thought it was just a result of heavy rain or the collapse of a glacier lake, but that’s not the case at all. The disaster originated thousands of meters above sea level, where a large mass of glacier and the underlying rock suddenly collapsed. Experts are still unsure which part of the sequence triggered the collapse first—perhaps the rocks on the slope broke down first, dragging the glacier down with them, or the glacier broke at its base, causing the already unstable rock mass to collapse. As the ice and rocks rushed down, they either blocked the river channels, causing temporary floods, or they swept up any ice and soil along the way, turning into mudslides that traveled dozens of kilometers downstream. This type of disaster is not part of the conventional risk assessment and monitoring frameworks.

2. Satellite data detected changes 7 days before the disaster; why was no action taken?

Reviewing the radar data from the European Space Agency’s Sentinel-1 satellite, it was evident that the ice and rock mass in the affected area was moving down the mountain at a rate of 10 millimeters per month (1 centimeter per day), and the speed increased as the disaster approached. However, this was dismissed as normal. Firstly, it is common knowledge that glaciers can move 10 to 200 meters per year, so a 1-centimeter movement per month seems harmless. It’s like noticing a car going from 120 miles per hour to 10 miles per hour on the highway; you might not realize it’s moving faster just because it had been moving slowly before. Secondly, there were practical barriers: the area was inaccessible, and ground monitoring equipment was not installed. It’s impossible to monitor every glacier worldwide. Additionally, the collapse occurred on the China-Nepal border, with the affected area split between the two countries, preventing the exchange of monitoring information. Even if one side noticed any abnormalities, they couldn’t alert the other side in time for evacuation.

3. Why were current high-altitude disaster warning systems ineffective?

Most existing high-altitude disaster warning systems focus on well-known risk areas. For example, Tibet has installed monitoring devices for the Subrenma Co Lake, and Nepal has equipped several glacier lakes in the Khumbu region with automatic alarms. However, the site of this disaster was not on any of these lists. It was not a glacier lake that had been monitored for a long time, nor a slope that had previously experienced landslides. The current systems do not regularly scan the entire high-altitude area for minor changes in ice and rock structures. It’s like only monitoring the water level in a puddle in front of your house without checking the thousands of square meters of surrounding terrain for potential hazards.

4. A new warning system to prevent such disasters will be a multi-layered, automated approach

Experts suggest that we need to move away from relying on a single camera or sensor to detect risks. Instead, a three-tiered warning system should be established:

  • The first layer would conduct a comprehensive scan of the entire high-altitude area using radar satellites like Sentinel-1, which can operate regardless of clouds or darkness, to identify areas with accelerated movement of ice and rock.
  • The second layer would focus on the identified suspect areas, using higher-resolution commercial satellites for closer observation. If it’s possible to deploy personnel, ground-based equipment such as seismic sensors, cameras, and water level gauges should be installed to monitor changes 24/7.
  • The third layer would assess the potential danger: not all movements of ice and rock require immediate evacuation of downstream communities. It’s necessary to consider the terrain to determine whether the collapse would result in a small, harmless incident or a major flood that could affect many people. This requires integrating information about the population and villages downstream to distinguish between minor issues and urgent threats.

After this disaster, this concept, which has so far only existed in theoretical papers, is likely to become a necessity for both China and Nepal, as well as the entire Himalayan region.