LandslidesPub Date : 2024-05-28DOI: 10.1007/s10346-024-02274-0
Xiao Feng, Juan Du, Minghua Wu, Bo Chai, Fasheng Miao, Yang Wang
{"title":"Potential of synthetic images in landslide segmentation in data-poor scenario: a framework combining GAN and transformer models","authors":"Xiao Feng, Juan Du, Minghua Wu, Bo Chai, Fasheng Miao, Yang Wang","doi":"10.1007/s10346-024-02274-0","DOIUrl":"https://doi.org/10.1007/s10346-024-02274-0","url":null,"abstract":"<p>Accurate landslide segmentation from remote sensing data is pivotal for efficient emergency response and risk management. In recent years, data-driven deep learning approaches have emerged as a significant area of focus in this domain. However, the limited availability of landslide data often restricts the effectiveness of these approaches. This study introduces the StyleGAN2-transformer framework for landslide segmentation, utilizing generative adversarial networks (GANs) for the first time to create synthetic, high-quality landslide images to address the data scarcity issue that undermines landslide segmentation model performance. Two datasets were developed: one containing a limited set of real landslide images and the other supplemented with synthetic landslide images generated by StyleGAN2. These datasets facilitated comparative experiments to quantitatively assess the impact of synthetic data on the performance of both convolutional neural network (CNN) and transformer series models, employing a suite of metrics for thorough evaluation. The findings indicate that adding synthetic landslide images from StyleGAN2 improves the overall accuracy of most landslide segmentation models significantly, achieving more than a 10% increase. Moreover, integrating StyleGAN2 with transformer models presents an optimized approach, as transformer models surpass CNN models in accuracy when adequate training data are available. Finally, the results also confirm that the StyleGAN2-transformer framework exhibits strong generalizability in a variety of scenarios.</p>","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141166851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LandslidesPub Date : 2024-05-27DOI: 10.1007/s10346-024-02277-x
Li Qingmiao, Zhao Jianjun, Zuo Jing, Ji Feng, Deng Jie, Liu Shuowei, Lai Qiyi
{"title":"Investigation of the slope-type debris flow disaster chain triggered by a landslide-induced road blockage in Yingpan Township, Shuicheng, Guizhou, on July 27, 2023","authors":"Li Qingmiao, Zhao Jianjun, Zuo Jing, Ji Feng, Deng Jie, Liu Shuowei, Lai Qiyi","doi":"10.1007/s10346-024-02277-x","DOIUrl":"https://doi.org/10.1007/s10346-024-02277-x","url":null,"abstract":"<p>On July 27, 2023, at approximately 04:00, a debris flow, with an estimated volume of 25,000 m<sup>3</sup>, struck Lanhua Village in the southern part of Shuicheng District, Guizhou Province, China. This event led to the burial and damage of 15 houses in Lanhua Village. Remarkably, residents successfully observed this debris flow 30 min before its peak arrival, enabling sufficient time for evacuation prior to the onset of this fatal geological disaster. To elucidate the evolutionary process of the Lanhua Village debris flow, a detailed field survey was conducted utilizing unmanned aerial vehicle (UAV) technology. The findings suggest that extreme heavy rainfall served as the primary trigger for the debris flow. Landslides, comprising residual slope deposits at the top of the slope and induced by rainfall, resulted in the obstruction of Highway X244 and the blockage of culverts beneath the road. These blockages contributed to the formation of a peak cross-basin flood, exceeding 7 m<sup>3</sup>/s, which acted as the driving force initiating the debris flow. The thick layer of colluvial deposits covering the slope surface, with approximately 23,000 m<sup>3</sup> of material mobilized by flood erosion, constituted the primary source contributing to the rapid expansion and movement of the debris flow. The topography, characterized by steep upper slopes and gentler lower slopes, was a contributing factor to the debris flow deposits spreading into residential areas, resulting in disaster. The event in Lanhua Village exemplifies a “landslide-blockage-flood diversion-erosion-debris flow” disaster chain.</p>","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141167004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LandslidesPub Date : 2024-05-20DOI: 10.1007/s10346-024-02267-z
Bingli Hu, Suo Lijun, Chonglei Zhang, Zhao Bo, Xie Qijun
{"title":"Mobility characteristics of rainfall-triggered shallow landslides in a forest area in Mengdong, China","authors":"Bingli Hu, Suo Lijun, Chonglei Zhang, Zhao Bo, Xie Qijun","doi":"10.1007/s10346-024-02267-z","DOIUrl":"https://doi.org/10.1007/s10346-024-02267-z","url":null,"abstract":"","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141119087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LandslidesPub Date : 2024-05-17DOI: 10.1007/s10346-024-02268-y
Tao Xie, Cheng-Cheng Zhang, Bin Shi, Zhuo Chen, Yan Zhang
{"title":"Integrating distributed acoustic sensing and computer vision for real-time seismic location of landslides and rockfalls along linear infrastructure","authors":"Tao Xie, Cheng-Cheng Zhang, Bin Shi, Zhuo Chen, Yan Zhang","doi":"10.1007/s10346-024-02268-y","DOIUrl":"https://doi.org/10.1007/s10346-024-02268-y","url":null,"abstract":"","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140963452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LandslidesPub Date : 2024-05-13DOI: 10.1007/s10346-024-02270-4
Safiyeh Tayebi, Md. Akib Jabed, Ana Lorena Ruano, Gwenyth O Lee, Paula F. da Silva, Saleh Ahmed, Edier V. Aristizábal G., Ranjan Kumar Dahal, Arezoo Soltani, Mohammad Imran Khan, Md. Atiqur Rahman, M Ashraful Islam, Ubydul Haque
{"title":"Stakeholder perspectives on landslide triggers and impacts in five countries","authors":"Safiyeh Tayebi, Md. Akib Jabed, Ana Lorena Ruano, Gwenyth O Lee, Paula F. da Silva, Saleh Ahmed, Edier V. Aristizábal G., Ranjan Kumar Dahal, Arezoo Soltani, Mohammad Imran Khan, Md. Atiqur Rahman, M Ashraful Islam, Ubydul Haque","doi":"10.1007/s10346-024-02270-4","DOIUrl":"https://doi.org/10.1007/s10346-024-02270-4","url":null,"abstract":"<p>Expert perspectives drive landslide mitigation and post-disaster policy planning. This study examines landslide risk perceptions among the stakeholders (government officials, academics, policy experts, local community representatives, and representatives of NGOs/civil society) across Brazil, Colombia, Nepal, Iran, and Pakistan, identifying both shared concerns and local heterogeneity. Key informants revealed a discrepancy in their degree of concern about landslides, with government officials exhibiting greater apprehension compared to local community representatives. Local community representatives incorrectly perceived landslides to be the result of natural phenomena. In contrast, governmental and academic stakeholders felt that human-induced triggers, specifically those related to land use and land cover change, were significant contributors to landslide occurrences, necessitating stringent law enforcement. The comprehensive impacts of landslides included economic losses, infrastructure disruption, agricultural losses, and food security concerns, underscoring the multifaceted nature of this hazard. Our results suggest the need for proactive citizen engagement in landslide monitoring, recognizing the importance of local contexts. We end by proposing a dual-pronged policy approach that emphasizes the socio-economic context of each region.</p>","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140934762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LandslidesPub Date : 2024-05-11DOI: 10.1007/s10346-024-02272-2
Nini Johana Marín-Rodríguez, Johnny Vega, Oscar Betancurt Zanabria, Juan David González-Ruiz, Sergio Botero
{"title":"Towards an understanding of landslide risk assessment and its economic losses: a scientometric analysis","authors":"Nini Johana Marín-Rodríguez, Johnny Vega, Oscar Betancurt Zanabria, Juan David González-Ruiz, Sergio Botero","doi":"10.1007/s10346-024-02272-2","DOIUrl":"https://doi.org/10.1007/s10346-024-02272-2","url":null,"abstract":"<p>This scientometric analysis significantly advances the understanding of landslide risk assessment and economic losses, focusing on scientometric insights. This study aims at analyzing the global trends and structures of landslide risk and economic loss research from 2002 to 2023 using scientometric techniques such as co-authorship, co-word, co-citation, cluster analysis, and trend topics, among others. Thus, analysis of 92 studies gathered from Scopus and Web of Science databases reveals a continuous growth in environmental, social, and quantitative research topics. Predominant contributions hail mainly from China and Italy. The research identifies critical themes, including risk analysis, vulnerability, fragility, and economic losses. The current identified research combines advanced statistical methods, including logistic regression, with climate change scenarios and susceptibility assessments to reveal intricate connections between climatic shifts, hydrogeological hazards, and their economic and environmental impacts. This study provides researchers and practitioners with a comprehensive understanding of the status quo and research trends of ontology research landslide risk and its economic losses. It also promotes further studies in this domain.</p>","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140935208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LandslidesPub Date : 2024-05-02DOI: 10.1007/s10346-024-02266-0
Luke Weidner, Gabriel Walton, Cameron Phillips
{"title":"Investigating the influences of precipitation, snowmelt, and freeze-thaw on rockfall in Glenwood Canyon, Colorado using terrestrial laser scanning","authors":"Luke Weidner, Gabriel Walton, Cameron Phillips","doi":"10.1007/s10346-024-02266-0","DOIUrl":"https://doi.org/10.1007/s10346-024-02266-0","url":null,"abstract":"<p>Understanding the triggering factors leading to rockfall is essential in managing their risk to transportation infrastructure. Precipitation and freeze-thaw (FT) are widely studied rockfall triggers, but developing reliable, quantitative methods to forecast rockfall in response to weather events remains challenging. Terrestrial laser scanning (TLS) is a powerful tool for high-accuracy modeling of rock slopes, but the frequency of scanning is often too low to correlate rockfall behavior with weather events or seasonal trends. We conducted a TLS campaign between 2017 and 2022 in Glenwood Canyon, Colorado, to investigate the seasonal variability in rockfall triggering and conditioning mechanisms. A total of 44 scans were collected over 5 years and were processed to allow for consistent detection of rockfalls larger than 0.0036 m<sup>3</sup> in volume. Meteorological variables relating to precipitation, snowpack, and temperature were modeled using the National Weather Service SNODAS product and were used to complete an exploratory analysis of the correlation of various weather indices with rockfall rate over time. It was found that the short-term sum of liquid precipitation and snowmelt (averaged over the scanning interval or the max single-day total) was a strong predictor of rockfall volume rate between 2018 and 2020, especially in the spring and summer months; a linear model of max daily liquid was able to explain 65% of the variance (<i>R</i><sup>2</sup><sub>adj</sub>) in rockfall volume rate in March through August. This implicates springtime snowmelt and rain-on-snow events as strong predictors of rockfall at the study site. We interpret these observations to indicate that snowmelt and rainfall are acting to trigger blocks that have been conditioned (destabilized) over the preceding winter.</p>","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140838986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LandslidesPub Date : 2024-05-02DOI: 10.1007/s10346-024-02259-z
Olga Mavrouli, M. Amparo Núñez-Andrés, Felipe Buill, Nieves Lantada, Jordi Corominas
{"title":"Correlation between rockfall frequency and overhang geometrical attributes","authors":"Olga Mavrouli, M. Amparo Núñez-Andrés, Felipe Buill, Nieves Lantada, Jordi Corominas","doi":"10.1007/s10346-024-02259-z","DOIUrl":"https://doi.org/10.1007/s10346-024-02259-z","url":null,"abstract":"<p>The estimation of rockfall frequency for the quantitative assessment of rockfall hazard is challenging when there are not records of previous events. The aim of this study is to investigate the correlation between the rockfall frequency and the geometry of the overhanging rock blocks, at rocky slopes. The investigation takes place for seven rocky slopes along a coastal road in Northern Spain, and it is based on 15 years of rockfall data. A field survey of these slopes took place between May and June 2022, using a terrestrial laser scanner as well as ground and aerial photogrammetry from UAVs, in order to build three-dimensional digital models. Then, the geometrical attributes of the existing overhanging rocks were measured on the models. The surface area and the width of the latter were calculated. The correlation between the rockfall frequency and the sum of the overhanging areas at each slope was assessed. A good linear correlation was indicated between the normalized per slope area number of rockfalls and total overhang area (<i>R</i><sup>2</sup> = 0.9013) and between the respective normalized per unit of road length parameters (<i>R</i><sup>2</sup> = 0.9594). The magnitude-frequency relationship for the rockfall events that occurred at the seven slopes follows a power law distribution with exponent −0.65.</p>","PeriodicalId":17938,"journal":{"name":"Landslides","volume":null,"pages":null},"PeriodicalIF":6.7,"publicationDate":"2024-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140839126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}