Chinanews.com, Chengdu, January 17th (Reporter He Shaoqing) Landslide early warning and prediction is recognized as a worldwide problem.

In the more than ten years after the "5.12" Wenchuan earthquake, how did Chinese geological researchers establish a prediction model for earthquake-induced landslides from scratch?

How to pry open the "blind box" of post-earthquake disasters?

How does China's disaster prevention and control come to the forefront of the world?

  Focusing on the above issues, Fan Xuanmei, the winner of the "Scientific Exploration Award" in 2022 and deputy director of the State Key Laboratory of Geohazard Prevention and Geoenvironmental Protection of Chengdu University of Technology, accepted an exclusive interview with Chinanews.com to interpret this.

Fan Xuanmei accepted an exclusive interview with a reporter from Chinanews.com.

Photo by Tang Qihao

  What factors may induce post-earthquake geological disasters?

  Fan Xuanmei introduced that aftershocks and rainfall are the main factors that induce post-earthquake geological disasters.

Immediately after a strong earthquake occurs, the earthquake area is prone to strong aftershocks.

Under the influence of aftershocks, secondary landslides may occur in some mountain bodies shaken loose and cracked in the main earthquake and in places where landslides have already occurred.

Similarly, heavy rainfall after the earthquake can easily lead to secondary landslides or debris flows in the earthquake area.

  In order to predict these possible geological disasters, the State Key Laboratory of Geological Hazard Prevention and Geoenvironmental Protection of Chengdu University of Technology has established a "three-check" system integrating air, space and ground.

  "In addition to collecting satellite remote sensing data on a large scale, we will also strengthen monitoring of some key areas before and after the rainy season." Fan Xuanmei said that if heavy rainfall occurs in the "9.5" Luding earthquake area in 2023, then Moxigou, There will be a high probability of debris flow disasters in areas such as Wandong River and Hailuogou.

The Chengli National Geological Disaster Laboratory team is preparing to deploy monitoring instruments in several key watersheds to observe rainfall, mud levels in ditches, water levels, and the stability of earthquake-induced landslide accumulations on slopes.

Fan Xuanmei at work.

Photo provided by the interviewee

  Why establish an earthquake-induced landslide prediction model?

  For more than ten years after the Wenchuan Earthquake, Fan Xuanmei's team went to the "4.14" Yushu Earthquake, "4.20" Lushan Earthquake, "8.3" Ludian Earthquake, and "8.8" Jiuzhaigou Earthquake and other earthquake relief teams. On the front line, collect valuable images and data, and based on more than 400,000 disaster data induced by more than 50 earthquakes around the world, combined with the latest artificial intelligence algorithm, a near real-time prediction model for earthquake-induced landslides was established.

  "After the Wenchuan Earthquake, the main rescue force went to Wenchuan as soon as possible, instead of Yingxiu and Beichuan, which were the hardest hit at that time. This was because we did not have timely and comprehensive satellite data to obtain disaster information immediately after the earthquake. damage information.” Fan Xuanmei pointed out that the biggest use of the earthquake-induced landslide prediction model is to fill the information gap in the 72-hour golden rescue time after the earthquake, and provide first-hand support and decision-making information for emergency rescue after the earthquake.

An intelligent prediction model for earthquake-induced landslides.

Photo provided by the interviewee

  Fan Xuanmei introduced that the satellite will not be fixed at a certain position to take pictures of a fixed point on the earth, but will continue to revolve around the earth.

If a satellite happened to be above the earthquake area when the Luding earthquake occurred, then this satellite may have photographed the disaster.

If it's not a coincidence, then we need to wait for the satellite to turn over the Luding earthquake next time before we can take images of the disaster area.

Even sometimes, when a good satellite image is obtained, a month has passed since the earthquake occurred.

  "If you rely entirely on satellite data to assess the disaster situation after the earthquake, you will most likely miss the best time for rescue." Fan Xuanmei said that the earthquake-induced landslide prediction model can be based on big data and artificial intelligence, and according to the earthquake information, quickly judge which places are geologically hazardous The most concentrated, where the houses and roads are most damaged, so that the rescue force can go to the location that needs rescue most at the first time.

Fan Xuanmei at work.

Photo provided by the interviewee

  How do Chinese researchers pry open the "blind box" of post-earthquake disasters?

  Fan Xuanmei introduced that after the Wenchuan earthquake, Chinese researchers combined satellite technology, artificial intelligence, big data and other technologies with disaster prevention and mitigation, and finally opened the "blind box" of post-earthquake disasters.

  Fan Xuanmei revealed that Chengli National Laboratory for Earthquake Disasters is currently conducting scientific research on earthquake disaster chains.

If the landslides and mudslides after the earthquake formed a barrier lake-burst flood, it may affect hundreds or even thousands of kilometers downstream.

At present, researchers are studying how to better predict the occurrence of disaster chains and avoid large-scale casualties that may be caused by disaster chains.

  Fan Xuanmei said that in recent years, both the experience and scientific research achievements of Chinese scientific researchers in the field of disasters have taken a leading position in the world.

In the future, Chinese knowledge and Chinese wisdom in the field of disaster prevention and reduction should be sent abroad to help more people.

(Finish)