Parameters Derived from and/or Used with Digital Elevation Models (DEMs) for Landslide Susceptibility Mapping and Landslide Risk Assessment: A Review

Saleem, Nayyer; Huq, Md. Enamul; Twumasi, Nana Yaw Danquah; Javed, Akib; Sajjad, Asif

doi:10.3390/ijgi8120545

Cited by 77 publications

(23 citation statements)

References 99 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Too low resolution cannot guarantee the rationality of the obtained LSP, while too high resolution will greatly increase the model computation complexity [62]. Although some studies focusing on the issue of LSP considering the different spatial resolutions of grid units, show that the LSP performance decreases when the resolution of grid units rises from 10 m to 100 m [63], a lot of literature shows that a 30 m grid resolution is suitable for LSP and can obtain satisfactory LSP results [7,11,21,30,35,37,50,59,[64][65][66][67][68][69]. Meanwhile, the original grid resolutions of the DEM and remote sensing images used in this study are both 30 m, which can not only effectively represent the topographic characteristics, but also avoid excessive computation.…”

Section: Sensitivity Analysis On Resolution Of Grid Unitsmentioning

confidence: 99%

Landslide Susceptibility Prediction Based on Remote Sensing Images and GIS: Comparisons of Supervised and Unsupervised Machine Learning Models

et al. 2020

View full text Add to dashboard Cite

Landslide susceptibility prediction (LSP) has been widely and effectively implemented by machine learning (ML) models based on remote sensing (RS) images and Geographic Information System (GIS). However, comparisons of the applications of ML models for LSP from the perspectives of supervised machine learning (SML) and unsupervised machine learning (USML) have not been explored. Hence, this study aims to compare the LSP performance of these SML and USML models, thus further to explore the advantages and disadvantages of these ML models and to realize a more accurate and reliable LSP result. Two representative SML models (support vector machine (SVM) and CHi-squared Automatic Interaction Detection (CHAID)) and two representative USML models (K-means and Kohonen models) are respectively used to scientifically predict the landslide susceptibility indexes, and then these prediction results are discussed. Ningdu County with 446 recorded landslides obtained through field investigations is introduced as case study. A total of 12 conditioning factors are obtained through procession of Landsat TM 8 images and high-resolution aerial images, topographical and hydrological spatial analysis of Digital Elevation Modeling in GIS software, and government reports. The area value under the curve of receiver operating features (AUC) is applied for evaluating the prediction accuracy of SML models, and the frequency ratio (FR) accuracy is then introduced to compare the remarkable prediction performance differences between SML and USML models. Overall, the receiver operation curve (ROC) results show that the AUC of the SVM is 0.892 and is slightly greater than the AUC of the CHAID model (0.872). The FR accuracy results show that the SVM model has the highest accuracy for LSP (77.80%), followed by the CHAID model (74.50%), the Kohonen model (72.8%) and the K-means model (69.7%), which indicates that the SML models can reach considerably better prediction capability than the USML models. It can be concluded that selecting recorded landslides as prior knowledge to train and test the LSP models is the key reason for the higher prediction accuracy of the SML models, while the lack of a priori knowledge and target guidance is an important reason for the low LSP accuracy of the USML models. Nevertheless, the USML models can also be used to implement LSP due to their advantages of efficient modeling processes, dimensionality reduction and strong scalability.

show abstract

Section: Sensitivity Analysis On Resolution Of Grid Unitsmentioning

confidence: 99%

Landslide Susceptibility Prediction Based on Remote Sensing Images and GIS: Comparisons of Supervised and Unsupervised Machine Learning Models

et al. 2020

View full text Add to dashboard Cite

show abstract

“…In general, DEM accuracy is related to the quality of data collection procedures [58], which consist of map scale and interpolation methods. In fact, to obtain the submeter accuracy of DEM, ground survey methods, LiDAR, and aerial photography techniques should be used, but these techniques are costly [59]. This research focuses on finding the most effective estimation procedures that are widely known for constructing a largescale DEM using a comparative approach.…”

Section: Resultsmentioning

confidence: 99%

Impact of interpolation techniques on the accuracy of large-scale digital elevation model

et al. 2020

View full text Add to dashboard Cite

AbstractThere is no doubt that the tremendous development of information technology was one of the driving factors behind the great growth of surveying and geodesy science. This has spawned modern geospatial techniques for data capturing, acquisition, and visualization tools. Digital elevation model (DEM) is the 3D depiction of continuous elevation data over the Earth’s surface that is produced through many procedures such as remote sensing, photogrammetry, and land surveying. DEMs are essential for various surveying and civil engineering applications to generate topographic maps for construction projects at a scale that varies from 1:500 to 1:2,000. GIS offers a powerful tool to create a DEM with high resolution from accurate land survey measurements using interpolation methods. The aim of this research is to investigate the impact of estimation techniques on generating a reliable and accurate DEM suitable for large-scale mapping. As a part of this study, the deterministic interpolation algorithms such as ANUDEM (Topo to Raster), inverse distance weighted (IDW), and triangulated irregular network (TIN) were tested using the ArcGIS desktop for elevation data obtained from real total station readings, with different landforms to show the effect of terrain roughness, data density, and interpolation process on DEM accuracy. Furthermore, comparison and validation of each interpolator were carried out through the cross-validation method and numerous graphical representations of the DEM. Finally, the results of the investigations showed that ANUDEM and TIN models are similar and significantly better than those attained from IDW.

show abstract

“…a prohibition on logging, restrictions on development density, regular checks, and maintenance of drainage systems). To facilitate the use of susceptibility maps, a landslide protocol is necessary to link the terrain domains outlined in the terrain zonations to specific actions (Rossi et al 1982;Van Den Eeckhaut et al 2009). Design of the appropriate landslide protocol for Gdynia should be prepared involving relevant stakeholders and considering the negative consequences of landslides.…”

Section: Discussionmentioning

confidence: 99%

Landslide susceptibility mapping of Gdynia using geographic information system-based statistical models

Małka

2021

Nat Hazards

View full text Add to dashboard Cite

This work aims to prepare a reliable landslide susceptibility model and to analyse the factors contributing to landslides in a dynamic environment by considering the city of Gdynia, Poland as a case study. Geological, geomorphological, hydrological, hydrogeological, and anthropogenic predisposing factors are considered using geographic information systems. Ground types at different depths (1 m and 4 m b.g.l.) are used in the statistical susceptibility assessment for the first time. Landslide susceptibility maps are developed using two techniques in presenting landslides, 13 conditioning factors, and three statistical methods: landslide index, weight of evidence, and logistic regression. The considered factors have an influence on mass movement formation, but their roles are different. Many of these passive factors are interrelated and some of them are also related to active factors, i.e. triggers. Consideration of many thematic layers in the statistical approach allows for the selection of the most appropriate geo-environmental variables. The most significant conditioning factors that affect the likelihood of landsliding include land use and land cover as well as topography. The susceptibility maps generated by the index model and many interrelated passive factors appear to be over-predicted. The logistic regression model and only independent controlling factors (slope angle, slope aspect, and lithology) are sufficient to compile a reliable susceptibility map of Gdynia. Prediction rate curve plots show that the susceptibility map produced using logistic regression exhibits the highest prediction accuracy. The results emphasize the need to check independence in the selection of instability factors and the use of an independent subset of landslides for validation.

show abstract

Parameters Derived from and/or Used with Digital Elevation Models (DEMs) for Landslide Susceptibility Mapping and Landslide Risk Assessment: A Review

Cited by 77 publications

References 99 publications

Landslide Susceptibility Prediction Based on Remote Sensing Images and GIS: Comparisons of Supervised and Unsupervised Machine Learning Models

Landslide Susceptibility Prediction Based on Remote Sensing Images and GIS: Comparisons of Supervised and Unsupervised Machine Learning Models

Impact of interpolation techniques on the accuracy of large-scale digital elevation model

Landslide susceptibility mapping of Gdynia using geographic information system-based statistical models

Contact Info

Product

Resources

About