Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility Mapping
Despite yielding considerable degrees of accuracy in landslide predictions, the outcomes of different landslide susceptibility models are prone to spatial disagreement; and therefore, uncertainties. Uncertainties in the results of various landslide susceptibility models create challenges in selectin...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/20/3347 |
| _version_ | 1797551089817485312 |
|---|---|
| author | Mohammed Sarfaraz Gani Adnan Md Salman Rahman Nahian Ahmed Bayes Ahmed Md. Fazleh Rabbi Rashedur M. Rahman |
| author_facet | Mohammed Sarfaraz Gani Adnan Md Salman Rahman Nahian Ahmed Bayes Ahmed Md. Fazleh Rabbi Rashedur M. Rahman |
| author_sort | Mohammed Sarfaraz Gani Adnan |
| collection | DOAJ |
| description | Despite yielding considerable degrees of accuracy in landslide predictions, the outcomes of different landslide susceptibility models are prone to spatial disagreement; and therefore, uncertainties. Uncertainties in the results of various landslide susceptibility models create challenges in selecting the most suitable method to manage this complex natural phenomenon. This study aimed to propose an approach to reduce uncertainties in landslide prediction, diagnosing spatial agreement in machine learning-based landslide susceptibility maps. It first developed landslide susceptibility maps of Cox’s Bazar district of Bangladesh, applying four machine learning algorithms: K-Nearest Neighbor (KNN), Multi-Layer Perceptron (MLP), Random Forest (RF), and Support Vector Machine (SVM), featuring hyperparameter optimization of 12 landslide conditioning factors. The results of all the four models yielded very high prediction accuracy, with the area under the curve (AUC) values range between 0.93 to 0.96. The assessment of spatial agreement of landslide predictions showed that the pixel-wise correlation coefficients of landslide probability between various models range from 0.69 to 0.85, indicating the uncertainty in predicted landslides by various models, despite their considerable prediction accuracy. The uncertainty was addressed by establishing a Logistic Regression (LR) model, incorporating the binary landslide inventory data as the dependent variable and the results of the four landslide susceptibility models as independent variables. The outcomes indicated that the RF model had the highest influence in predicting the observed landslide locations, followed by the MLP, SVM, and KNN models. Finally, a combined landslide susceptibility map was developed by integrating the results of the four machine learning-based landslide predictions. The combined map resulted in better spatial agreement (correlation coefficients range between 0.88 and 0.92) and greater prediction accuracy (0.97) compared to the individual models. The modelling approach followed in this study would be useful in minimizing uncertainties of various methods and improving landslide predictions. |
| first_indexed | 2024-03-10T15:39:49Z |
| format | Article |
| id | doaj.art-c2a5dcae50444cf5b624e88ac0589a29 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T15:39:49Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-c2a5dcae50444cf5b624e88ac0589a292023-11-20T16:58:47ZengMDPI AGRemote Sensing2072-42922020-10-011220334710.3390/rs12203347Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility MappingMohammed Sarfaraz Gani Adnan0Md Salman Rahman1Nahian Ahmed2Bayes Ahmed3Md. Fazleh Rabbi4Rashedur M. Rahman5Environmental Change Institute, School of Geography and the Environment, University of Oxford, South Parks Road, Oxford OX13QY, UKDepartment of Civil Engineering, Chittagong University of Engineering and Technology (CUET), Chittagong 4349, BangladeshDepartment of Electrical and Computer Engineering, North South University, Bashundhara, Dhaka 1229, BangladeshInstitute for Risk and Disaster Reduction (IRDR), University College London (UCL), Gower Street, London WC1E 6BT, UKDepartment of Geography and Environmental Studies, University of Chittagong, Chittagong 4331, BangladeshDepartment of Electrical and Computer Engineering, North South University, Bashundhara, Dhaka 1229, BangladeshDespite yielding considerable degrees of accuracy in landslide predictions, the outcomes of different landslide susceptibility models are prone to spatial disagreement; and therefore, uncertainties. Uncertainties in the results of various landslide susceptibility models create challenges in selecting the most suitable method to manage this complex natural phenomenon. This study aimed to propose an approach to reduce uncertainties in landslide prediction, diagnosing spatial agreement in machine learning-based landslide susceptibility maps. It first developed landslide susceptibility maps of Cox’s Bazar district of Bangladesh, applying four machine learning algorithms: K-Nearest Neighbor (KNN), Multi-Layer Perceptron (MLP), Random Forest (RF), and Support Vector Machine (SVM), featuring hyperparameter optimization of 12 landslide conditioning factors. The results of all the four models yielded very high prediction accuracy, with the area under the curve (AUC) values range between 0.93 to 0.96. The assessment of spatial agreement of landslide predictions showed that the pixel-wise correlation coefficients of landslide probability between various models range from 0.69 to 0.85, indicating the uncertainty in predicted landslides by various models, despite their considerable prediction accuracy. The uncertainty was addressed by establishing a Logistic Regression (LR) model, incorporating the binary landslide inventory data as the dependent variable and the results of the four landslide susceptibility models as independent variables. The outcomes indicated that the RF model had the highest influence in predicting the observed landslide locations, followed by the MLP, SVM, and KNN models. Finally, a combined landslide susceptibility map was developed by integrating the results of the four machine learning-based landslide predictions. The combined map resulted in better spatial agreement (correlation coefficients range between 0.88 and 0.92) and greater prediction accuracy (0.97) compared to the individual models. The modelling approach followed in this study would be useful in minimizing uncertainties of various methods and improving landslide predictions.https://www.mdpi.com/2072-4292/12/20/3347landslidesremote sensinguncertaintyK-Nearest NeighborMulti-Layer PerceptronRandom Forest |
| spellingShingle | Mohammed Sarfaraz Gani Adnan Md Salman Rahman Nahian Ahmed Bayes Ahmed Md. Fazleh Rabbi Rashedur M. Rahman Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility Mapping Remote Sensing landslides remote sensing uncertainty K-Nearest Neighbor Multi-Layer Perceptron Random Forest |
| title | Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility Mapping |
| title_full | Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility Mapping |
| title_fullStr | Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility Mapping |
| title_full_unstemmed | Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility Mapping |
| title_short | Improving Spatial Agreement in Machine Learning-Based Landslide Susceptibility Mapping |
| title_sort | improving spatial agreement in machine learning based landslide susceptibility mapping |
| topic | landslides remote sensing uncertainty K-Nearest Neighbor Multi-Layer Perceptron Random Forest |
| url | https://www.mdpi.com/2072-4292/12/20/3347 |
| work_keys_str_mv | AT mohammedsarfarazganiadnan improvingspatialagreementinmachinelearningbasedlandslidesusceptibilitymapping AT mdsalmanrahman improvingspatialagreementinmachinelearningbasedlandslidesusceptibilitymapping AT nahianahmed improvingspatialagreementinmachinelearningbasedlandslidesusceptibilitymapping AT bayesahmed improvingspatialagreementinmachinelearningbasedlandslidesusceptibilitymapping AT mdfazlehrabbi improvingspatialagreementinmachinelearningbasedlandslidesusceptibilitymapping AT rashedurmrahman improvingspatialagreementinmachinelearningbasedlandslidesusceptibilitymapping |