A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Quantitative susceptibility assessment of the breach of moraine-dammed lakes due to glacier avalanches
Abstract Glacial lake outburst floods (GLOFs) have caused tens of millions of dollars of damage to infrastructure and have killed thousands of people worldwide. The objective of this study was to quantify the hazard posed by the occurrence of a GLOF due to an ice avalanche into an ice-free moraine-dammed lake, producing a surge wave and raising the water level. This study was carried out in the Himalayan region and in the western cordillera of North America. Some simplifications and combinations of parameters were made to reduce the number of parameters as much as possible. We selected seven factors for the quantitative susceptibility assessment of the GLOF hazard: the dangerous glacier slope factor S, temperature factor T, aspect factor Y, glacier aspect factor X, runout factor of the dangerous glacier R, ratio of the volume factor of the dangerous glacier to the volume of the lake V, and the downstream slope factor of the dam D (lake parameter). A machine learning model (a support vector machine, SVM) was used to create a susceptibility assessment model. Data for 21 drained and 50 undrained ice-free moraine-dammed lakes in the Himalayan region were used for the training of the GLOF susceptibility assessment model. In addition, data for 19 drained and 43 undrained ice-free moraine-dammed lakes in the western cordillera of North America were used to test the GLOF susceptibility assessment model. The susceptibility assessment factor P1 is a combination of these seven factors. The probability of GLOF occurrence increases as the P1 value increases. Using three critical P1 values, we divided the GLOF occurrence probability into four grades: low, medium, high, and very high probability of occurrence. The threshold values for the western cordillera of North America were 6.1–15.4% lower than the corresponding critical values for the Himalayan region because the annual temperature variation is larger in Canada than in Tibet. Thus, this assessment model can be used to assess the susceptibility of other areas of the world to GLOFs when annual temperature variations are considered.
Highlights The study was carried out in Himalayan and the western cordillera of North America. Quantitative susceptibility assessment of the GLOF hazard by seven factors. A machine learning model (SVM) was used to create a susceptibility assessment model. This assessment model can be used to assess the susceptibility of GLOF in other area.
Quantitative susceptibility assessment of the breach of moraine-dammed lakes due to glacier avalanches
Abstract Glacial lake outburst floods (GLOFs) have caused tens of millions of dollars of damage to infrastructure and have killed thousands of people worldwide. The objective of this study was to quantify the hazard posed by the occurrence of a GLOF due to an ice avalanche into an ice-free moraine-dammed lake, producing a surge wave and raising the water level. This study was carried out in the Himalayan region and in the western cordillera of North America. Some simplifications and combinations of parameters were made to reduce the number of parameters as much as possible. We selected seven factors for the quantitative susceptibility assessment of the GLOF hazard: the dangerous glacier slope factor S, temperature factor T, aspect factor Y, glacier aspect factor X, runout factor of the dangerous glacier R, ratio of the volume factor of the dangerous glacier to the volume of the lake V, and the downstream slope factor of the dam D (lake parameter). A machine learning model (a support vector machine, SVM) was used to create a susceptibility assessment model. Data for 21 drained and 50 undrained ice-free moraine-dammed lakes in the Himalayan region were used for the training of the GLOF susceptibility assessment model. In addition, data for 19 drained and 43 undrained ice-free moraine-dammed lakes in the western cordillera of North America were used to test the GLOF susceptibility assessment model. The susceptibility assessment factor P1 is a combination of these seven factors. The probability of GLOF occurrence increases as the P1 value increases. Using three critical P1 values, we divided the GLOF occurrence probability into four grades: low, medium, high, and very high probability of occurrence. The threshold values for the western cordillera of North America were 6.1–15.4% lower than the corresponding critical values for the Himalayan region because the annual temperature variation is larger in Canada than in Tibet. Thus, this assessment model can be used to assess the susceptibility of other areas of the world to GLOFs when annual temperature variations are considered.
Highlights The study was carried out in Himalayan and the western cordillera of North America. Quantitative susceptibility assessment of the GLOF hazard by seven factors. A machine learning model (SVM) was used to create a susceptibility assessment model. This assessment model can be used to assess the susceptibility of GLOF in other area.
Quantitative susceptibility assessment of the breach of moraine-dammed lakes due to glacier avalanches
Yu, Bin (author) / He, Yuanxun (author) / Ye, Peng (author)
2022-12-10
Article (Journal)
Electronic Resource
English
Glacier Dammed Lakes Impacting Different Alaskan Drainages After 30 Years
| British Library Conference Proceedings | 2009
Moraine Lake Outburst Simulation of Glacier Sliding and Risk Assessment
| British Library Conference Proceedings | 2011
| Elsevier | 2025