An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network Models
Landslide susceptibility mapping (LSM) characterizes landslide potential, which is essential for assessing landslide risk and developing mitigation strategies. Despite the significant progress in LSM research over the past two decades, several long-standing issues, such as uncertainties related to t...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/12/3200 |
| _version_ | 1797592772289495040 |
|---|---|
| author | Sahand Khabiri Matthew M. Crawford Hudson J. Koch William C. Haneberg Yichuan Zhu |
| author_facet | Sahand Khabiri Matthew M. Crawford Hudson J. Koch William C. Haneberg Yichuan Zhu |
| author_sort | Sahand Khabiri |
| collection | DOAJ |
| description | Landslide susceptibility mapping (LSM) characterizes landslide potential, which is essential for assessing landslide risk and developing mitigation strategies. Despite the significant progress in LSM research over the past two decades, several long-standing issues, such as uncertainties related to training samples and model selection, remain inadequately addressed in the literature. In this study, we employed a physically based susceptibility model, PISA-m, to generate four different non-landslide data scenarios and combine them with mapped landslides from Magoffin County, Kentucky, for model training. We utilized two Bayesian network model structures, Naïve Bayes (NB) and Tree-Augmented Naïve Bayes (TAN), to produce LSMs based on regional geomorphic conditions. After internal validation, we evaluated the robustness and reliability of the models using an independent landslide inventory from Owsley County, Kentucky. The results revealed considerable differences between the most effective model in internal validation (AUC = 0.969), which used non-landslide samples extracted exclusively from low susceptibility areas predicted by PISA-m, and the models’ unsatisfactory performance in external validation, as manifested by the identification of only 79.1% of landslide initiation points as high susceptibility areas. The obtained results from both internal and external validation highlighted the potential overfitting problem, which has largely been overlooked by previous studies. Additionally, our findings also indicate that TAN models consistently outperformed NB models when training datasets were the same due to the ability to account for variables’ dependencies by the former. |
| first_indexed | 2024-03-11T01:57:56Z |
| format | Article |
| id | doaj.art-5cd1206a195d4cf58a3b20fbf91999e8 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T01:57:56Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-5cd1206a195d4cf58a3b20fbf91999e82023-11-18T12:27:52ZengMDPI AGRemote Sensing2072-42922023-06-011512320010.3390/rs15123200An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network ModelsSahand Khabiri0Matthew M. Crawford1Hudson J. Koch2William C. Haneberg3Yichuan Zhu4Department of Civil & Environmental Engineering, Temple University, Philadelphia, PA 19122, USAKentucky Geological Survey, University of Kentucky, Lexington, KY 40506, USAKentucky Geological Survey, University of Kentucky, Lexington, KY 40506, USAKentucky Geological Survey, University of Kentucky, Lexington, KY 40506, USADepartment of Civil & Environmental Engineering, Temple University, Philadelphia, PA 19122, USALandslide susceptibility mapping (LSM) characterizes landslide potential, which is essential for assessing landslide risk and developing mitigation strategies. Despite the significant progress in LSM research over the past two decades, several long-standing issues, such as uncertainties related to training samples and model selection, remain inadequately addressed in the literature. In this study, we employed a physically based susceptibility model, PISA-m, to generate four different non-landslide data scenarios and combine them with mapped landslides from Magoffin County, Kentucky, for model training. We utilized two Bayesian network model structures, Naïve Bayes (NB) and Tree-Augmented Naïve Bayes (TAN), to produce LSMs based on regional geomorphic conditions. After internal validation, we evaluated the robustness and reliability of the models using an independent landslide inventory from Owsley County, Kentucky. The results revealed considerable differences between the most effective model in internal validation (AUC = 0.969), which used non-landslide samples extracted exclusively from low susceptibility areas predicted by PISA-m, and the models’ unsatisfactory performance in external validation, as manifested by the identification of only 79.1% of landslide initiation points as high susceptibility areas. The obtained results from both internal and external validation highlighted the potential overfitting problem, which has largely been overlooked by previous studies. Additionally, our findings also indicate that TAN models consistently outperformed NB models when training datasets were the same due to the ability to account for variables’ dependencies by the former.https://www.mdpi.com/2072-4292/15/12/3200Bayesian networknegative sampleslandslide susceptibility mapping (LSM)uncertaintyPISA-mrobustness |
| spellingShingle | Sahand Khabiri Matthew M. Crawford Hudson J. Koch William C. Haneberg Yichuan Zhu An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network Models Remote Sensing Bayesian network negative samples landslide susceptibility mapping (LSM) uncertainty PISA-m robustness |
| title | An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network Models |
| title_full | An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network Models |
| title_fullStr | An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network Models |
| title_full_unstemmed | An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network Models |
| title_short | An Assessment of Negative Samples and Model Structures in Landslide Susceptibility Characterization Based on Bayesian Network Models |
| title_sort | assessment of negative samples and model structures in landslide susceptibility characterization based on bayesian network models |
| topic | Bayesian network negative samples landslide susceptibility mapping (LSM) uncertainty PISA-m robustness |
| url | https://www.mdpi.com/2072-4292/15/12/3200 |
| work_keys_str_mv | AT sahandkhabiri anassessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT matthewmcrawford anassessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT hudsonjkoch anassessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT williamchaneberg anassessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT yichuanzhu anassessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT sahandkhabiri assessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT matthewmcrawford assessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT hudsonjkoch assessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT williamchaneberg assessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels AT yichuanzhu assessmentofnegativesamplesandmodelstructuresinlandslidesusceptibilitycharacterizationbasedonbayesiannetworkmodels |