{"title":"ML-CASCADE: A machine learning and cloud computing-based tool for rapid and automated mapping of landslides using earth observation data","authors":"Nirdesh Sharma, Manabendra Saharia","doi":"10.1007/s10346-024-02360-3","DOIUrl":null,"url":null,"abstract":"<p>Landslides pose a significant threat to humans as well as the environment. Rapid and precise mapping of landslide extent is necessary for understanding their spatial distribution, assessing susceptibility, and developing early warning systems. Traditional landslide mapping methods rely on labor-intensive field studies and manual mapping using high-resolution imagery, which are both costly and time-consuming. While existing machine learning-based automated mapping methods exist, they have limited transferability due to low availability of training data and the inability to handle out-of-distribution scenarios. This study introduces ML-CASCADE, a user-friendly open-source tool designed for real-time landslide mapping. It is a semi-automated tool that requires the user to create landslide and non-landslide samples using pre- and post-landslide Sentinel-2 imagery to train a machine learning model. The model training features include Sentinel-2 data, terrain data, vegetation indices, and bare soil index. ML-CASCADE is developed as an easy-to-use application on top of Google Earth Engine and supports both pixel and object-based classification methods. We validate the landslide extent developed using ML-CASCADE with independent expert-developed inventories. ML-CASCADE is not only able to identify the landslide extent accurately but can also map a complex cluster of landslides within 5 min and a simple landslide within 2 min. Due to its ease of use, speed, and accuracy, ML-CASCADE will serve as a critical operational asset for landslide risk management.</p>","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":5.8000,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Landslides","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s10346-024-02360-3","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, GEOLOGICAL","Score":null,"Total":0}
引用次数: 0
Abstract
Landslides pose a significant threat to humans as well as the environment. Rapid and precise mapping of landslide extent is necessary for understanding their spatial distribution, assessing susceptibility, and developing early warning systems. Traditional landslide mapping methods rely on labor-intensive field studies and manual mapping using high-resolution imagery, which are both costly and time-consuming. While existing machine learning-based automated mapping methods exist, they have limited transferability due to low availability of training data and the inability to handle out-of-distribution scenarios. This study introduces ML-CASCADE, a user-friendly open-source tool designed for real-time landslide mapping. It is a semi-automated tool that requires the user to create landslide and non-landslide samples using pre- and post-landslide Sentinel-2 imagery to train a machine learning model. The model training features include Sentinel-2 data, terrain data, vegetation indices, and bare soil index. ML-CASCADE is developed as an easy-to-use application on top of Google Earth Engine and supports both pixel and object-based classification methods. We validate the landslide extent developed using ML-CASCADE with independent expert-developed inventories. ML-CASCADE is not only able to identify the landslide extent accurately but can also map a complex cluster of landslides within 5 min and a simple landslide within 2 min. Due to its ease of use, speed, and accuracy, ML-CASCADE will serve as a critical operational asset for landslide risk management.
期刊介绍:
Landslides are gravitational mass movements of rock, debris or earth. They may occur in conjunction with other major natural disasters such as floods, earthquakes and volcanic eruptions. Expanding urbanization and changing land-use practices have increased the incidence of landslide disasters. Landslides as catastrophic events include human injury, loss of life and economic devastation and are studied as part of the fields of earth, water and engineering sciences. The aim of the journal Landslides is to be the common platform for the publication of integrated research on landslide processes, hazards, risk analysis, mitigation, and the protection of our cultural heritage and the environment. The journal publishes research papers, news of recent landslide events and information on the activities of the International Consortium on Landslides.
- Landslide dynamics, mechanisms and processes
- Landslide risk evaluation: hazard assessment, hazard mapping, and vulnerability assessment
- Geological, Geotechnical, Hydrological and Geophysical modeling
- Effects of meteorological, hydrological and global climatic change factors
- Monitoring including remote sensing and other non-invasive systems
- New technology, expert and intelligent systems
- Application of GIS techniques
- Rock slides, rock falls, debris flows, earth flows, and lateral spreads
- Large-scale landslides, lahars and pyroclastic flows in volcanic zones
- Marine and reservoir related landslides
- Landslide related tsunamis and seiches
- Landslide disasters in urban areas and along critical infrastructure
- Landslides and natural resources
- Land development and land-use practices
- Landslide remedial measures / prevention works
- Temporal and spatial prediction of landslides
- Early warning and evacuation
- Global landslide database