Article
Engineering, Geological
Ugur Ozturk, Massimiliano Pittore, Robert Behling, Sigrid Roessner, Louis Andreani, Oliver Korup
Summary: Contemporary landslide research often focuses on predicting and mapping susceptibility to slope failure using generalized linear models, however the impact of sample size, location, or time on model performance remains largely untested. Findings suggest that considering only lower parts of landslides can significantly improve model performance, particularly for medium-sized landslides. Additionally, model performance shows marginal variation when progressively updating and adding more landslide data through time.
Article
Chemistry, Multidisciplinary
Zhu Liang, Weiping Peng, Wei Liu, Houzan Huang, Jiaming Huang, Kangming Lou, Guochao Liu, Kaihua Jiang
Summary: Shallow landslides in the Himalayan areas pose serious threats to humans and economic development. Landslide susceptibility mapping (LSM) is an effective way to minimize the risk of landslides. This study compared the performance of conventional algorithms (information value and logistic regression) and advanced algorithms (categorical boosting and conventional neural networks) in LSM in Yadong county. The CNN model exhibited the best performance, while the LR model performed the worst. The results suggest that LSM accuracy can be further improved by utilizing advanced algorithms and identifying more representative features.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Electrical & Electronic
Pingheng Li
Summary: This study explores the pollution risk and resource utilization potential of vegetable waste on land resources through the analysis of remote sensing data and literature collection. By establishing a land use reference database and using geographical information system, key areas for controlling vegetable waste pollution are determined, and the possibility of waste utilization is compared. In addition, the study conducts a preliminary analysis of the causes and cumulative characteristics of polluted soil.
JOURNAL OF SENSORS
(2022)
Article
Geosciences, Multidisciplinary
Halil Akinci, Mustafa Zeybek
Summary: This study compared the performance of machine learning models and traditional statistical methods in producing landslide susceptibility maps, with machine learning models showing higher success and prediction rates, especially with the RF model performing the best.
Article
Geosciences, Multidisciplinary
S. Boussouf, T. Fernandez, A. B. Hart
Summary: A landslide susceptibility analysis was conducted in the Rio Aguas catchment in southeast Spain using the Maximum Entropy (MaxEnt) and Geographically Weighted Logistic Regression (GWLR) models. The analysis utilized a previous landslide inventory and 12 predictors related to morphometry, hydrography, geology, and land cover. The results showed excellent prediction with both models, but GWLR had better results for AUC and MaxEnt for the degree of fit. A consensus model combining both methods achieved an AUC value of 0.99 and a degree of fit of 90%.
Article
Chemistry, Multidisciplinary
Gaetano Pecoraro, Gianfranco Nicodemo, Rosa Menichini, Davide Luongo, Dario Peduto, Michele Calvello
Summary: This paper presents a procedure to assess the risk level of stretches of roads exposed to slow-moving landslides at the municipal scale. It proposes an analysis method that combines landslide susceptibility maps, a road-damage database developed using Google Street View images, and ground-displacement measurements from satellite SAR images. The results demonstrate the importance of integrating these different approaches and data to understand the behavior of slow-moving landslides affecting road networks.
APPLIED SCIENCES-BASEL
(2023)
Article
Environmental Sciences
Svetlana Gantimurova, Alexander Parshin, Vladimir Erofeev
Summary: This article introduces the methodology and techniques of using UAV data for landslide research, and presents the results of landslide susceptibility assessment conducted in practical applications using an indirect heuristic approach in a GIS environment.
Article
Environmental Sciences
Gonzalo Maragano-Carmona, Ivo J. Fustos Toribio, Pierre-Yves Descote, Luis F. Robledo, Diego Villalobos, Gustavo Gatica
Summary: This study employs machine learning methods to assess the changes in landslide susceptibility in the Central Andes and proposes an implementation for a rainfall-induced landslides early warning system. The results demonstrate that the method accurately predicts landslide changes based on geomorphological features and precipitation conditions. The study also suggests strengthening monitoring of unstable slopes and incorporating landslide early warning into risk management strategies to mitigate the impact of landslides on densely populated areas.
Article
Geosciences, Multidisciplinary
Xinfu Xing, Chenglong Wu, Jinhui Li, Xueyou Li, Limin Zhang, Rongjie He
Summary: The study introduces a revised logistic regression method for dynamic landslide susceptibility prediction under cumulative daily rainfall, achieving a high accuracy of 91.9% in the assessment of landslide susceptibility in Shenzhen. This method utilizes five kinds of cumulative daily rainfall and updates the susceptibility model with the latest landslide events.
Article
Geosciences, Multidisciplinary
Yunhao Wang, Luqi Wang, Songlin Liu, Pengfei Liu, Zhengwei Zhu, Wengang Zhang
Summary: This study utilizes three machine learning models-random forest, logistic regression, and extreme gradient boosting-to assess the landslide susceptibility of Wushan County and compare the predictive performance of each algorithm. The results show that the random forest model has higher accuracy, F1 score, and area under the receiver operating characteristic curve values. Soil thickness is found to play a significant role in the occurrence of a landslide. Therefore, the random forest algorithm is recommended for evaluating landslide susceptibility in Wushan County.
GEOLOGICAL JOURNAL
(2023)
Article
Multidisciplinary Sciences
Alireza Arabameri, Nitheshnirmal Sadhasivam, Hamza Turabieh, Majdi Mafarja, Fatemeh Rezaie, Subodh Chandra Pal, M. Santosh
Summary: The study introduced novel hybrid ensemble models for gully erosion susceptibility mapping in Northern Iran, evaluating the relative importance of predictor factors and identifying the most influential variables for mapping GES. The CDT-RF model was found to be the most robust and accurate in predicting gully erosion susceptibility.
SCIENTIFIC REPORTS
(2021)
Article
Environmental Sciences
Husam A. H. Al-Najjar, Biswajeet Pradhan, Bahareh Kalantar, Maher Ibrahim Sameen, M. Santosh, Abdullah Alamri
Summary: This study examined the effectiveness of six feature transformations in landslide susceptibility modeling, with the use of Ohe-X transformation significantly improving model performance. A case study in the landslide-prone area in Cameron Highlands, Malaysia was conducted to test this novel approach.
Article
Environmental Sciences
Dil Kumar Rai, Xiong Donghong, Zhao Wei, Zhao Dongmei, Zhang Baojun, Nirmal Mani Dahal, Wu Yanhong, Muhammad Aslam Baig
Summary: Landslide distribution and susceptibility mapping are crucial for hazard and disaster risk management. This study investigates the landslide condition in the Dailekh district, Western Nepal, using inventory data and various contributing factors. The results show that both topographic and non-topographic factors significantly affect landslide occurrence and susceptibility in the Nepal Himalaya region. The reliability of the methods used for landslide susceptibility mapping is also verified.
CHINESE GEOGRAPHICAL SCIENCE
(2022)
Article
Engineering, Multidisciplinary
Ahmad Hammad Khaliq, Muhammad Basharat, Malik Talha Riaz, Muhammad Tayyib Riaz, Saad Wani, Nadhir Al-Ansari, Long Ba Le, Nguyen Thi Thuy Linh
Summary: This study utilized machine learning techniques to analyze and assess the landslide susceptibility in the Hattian Bala district of NW Himalayas, Pakistan. Historical satellite imageries were used to generate spatiotemporal landslide inventories, and a spatial database was created for various factors. The results showed that the Random Forest model outperformed the Logistic Regression model. The study aims to minimize losses and effectively manage landslide hazards in the region.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Geosciences, Multidisciplinary
Congwei Yu, Kan Liu, Bin Yu, Jie Yin
Summary: Fujian Province in China is prone to frequent landslide disasters, making the study of landslide susceptibility in this region significant. The study analyzed the landslides that occurred in Sanming, Fujian in 2019 using DEM data and aerial imagery, and established logistic regression models for landslide susceptibility. The study found that the model based on a 500 m * 500 m grid was the most suitable for evaluating landslide susceptibility.
Article
Geosciences, Multidisciplinary
S. C. Rai, A. K. Saha
ARABIAN JOURNAL OF GEOSCIENCES
(2015)
Article
Geosciences, Multidisciplinary
Nairwita Bandyopadhyay, Ashis Kumar Saha
ARABIAN JOURNAL OF GEOSCIENCES
(2016)
Article
Geosciences, Multidisciplinary
N. Bandyopadhyay, C. Bhuiyan, A. K. Saha
Article
Geography, Physical
C. Bhuiyan, A. K. Saha, N. Bandyopadhyay, F. N. Kogan
GISCIENCE & REMOTE SENSING
(2017)
Article
Remote Sensing
Biswajit Mondal, Ashis Kumar Saha, Anirban Roy
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2019)
Article
Multidisciplinary Sciences
Biswajit Mondal, Ashis Kumar Saha, Anirban Roy
Summary: Shoreline identification in West Bengal, India is a complicated process due to its size and inaccessibility, but geospatial technologies have helped. However, issues like low spatial resolution and mixed pixels still remain. Analysis using NDVI-Tasseled-Cap transformation technique showed that the coastal region is eroding at a rate of -3.36 to -4.70 meters per year, with islands particularly vulnerable.
PROCEEDINGS OF THE INDIAN NATIONAL SCIENCE ACADEMY
(2021)
Article
Environmental Sciences
Biswajit Mondal, Anirban Roy, Ashis Kumar Saha
Summary: In the era of climate change, vulnerability analysis in coastal regions is crucial in mitigating the impact of various hazards and protecting coastal resources. This study analyzes vulnerability along the coastal region of West Bengal, using geospatial techniques to assess the vulnerability of key parameters such as geomorphology, shoreline change, sea level rise, and mangrove density. The results show that a significant percentage of coastal and mangrove grids are highly vulnerable, with several mangrove-dominated islands under severe threat. The vulnerability maps created in this study provide valuable guidance for sustainable coastal mangrove ecosystem management.
REMOTE SENSING APPLICATIONS-SOCIETY AND ENVIRONMENT
(2022)
Article
Environmental Studies
Sakshi Naithani, Ashis Kumar Saha
Summary: This paper demonstrates the role of livelihood assets, strategies, and local social networks in disaster response and recovery in the post-disaster setting of the 2013 Kedarnath disaster in India. By identifying post-disaster macro-spaces and assessing micro spaces of livelihood and social capital in two villages, the study highlights the importance of geographical location and isolation in influencing the recovery trajectory of villages. Recovery strategies were seen to shift from pilgrimage-based to skill-based opportunities for families suffering loss of livelihoods.
DISASTER PREVENTION AND MANAGEMENT
(2021)
Article
Hospitality, Leisure, Sport & Tourism
Sakshi Naithani, Ashis Kumar Saha
ASIA PACIFIC JOURNAL OF TOURISM RESEARCH
(2019)
Article
Gastroenterology & Hepatology
Ashis Saha, Somnath Maitra, Subhas Hazra
INDIAN JOURNAL OF GASTROENTEROLOGY
(2013)
Article
Computer Science, Information Systems
AK Saha, MK Arora, RP Gupta, ML Virdi, E Csaplovics
INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE
(2005)
Article
Engineering, Geological
AK Saha, RP Gupta, I Sarkar, MK Arora, E Csaplovics
Article
Remote Sensing
AK Saha, RP Gupta, MK Arora
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2002)