Insights into geospatial heterogeneity of landslide susceptibility based on the SHAP-XGBoost model

Zhang, Junyi; Ma, Xianglong; Zhang, Jialan; Sun, Dongmei; Zhou, Xinzhi; Mi, Changlin; Wen, Haijia

doi:10.1016/j.jenvman.2023.117357

Cited by 132 publications

(39 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So far, this level of model explainability was already presented in three recent articles (Collini et al, 2022;Zhang et al, 2023;Dahal and Lombardo, 2022). However, what they all missed is translating the information offered by the SHAP values across the geographic space, which is what we will present in the next section.…”

Section: Local Interpretationmentioning

confidence: 77%

On the use of explainable AI for susceptibility modeling: examining the spatial pattern of SHAP values

Wang¹,

Zhang²,

Dahal³

et al. 2023

Preprint

View full text Add to dashboard Cite

Hydro-morphological processes (HMP, any natural phenomenon contained within the spectrum defined between debris flows and flash floods) are globally occurring natural hazards which pose great threats to our society, leading to fatalities and economical losses. For this reason, understanding the dynamics behind HMPs is needed to aid in hazard and risk assessment. In this work, we take advantage of an explainable deep learning model to extract global and local interpretations of the HMP occurrences across the whole Chinese territory. We use a neural network architecture and interpret the model results through the spatial pattern of SHAP values.In doing so, we can understand the model prediction on a hierarchical basis, looking at how the predictor set controls the overall susceptibility as well as doing the same at the level of the single mapping unit.Traditional statistical tools usually lead to a clear interpretation at the expense of large performance, which is otherwise reached via machine/deep learning solutions, though at the expense of interpretation. Explainable AI is the key to combine both strengths and in this work, we explore this combination in the context of HMP susceptibility modeling. Specifically, we demonstrate the extent to which one can enter a new level of data-driven interpretation, supporting the decision-making process behind disaster risk mitigation and prevention actions.

show abstract

Section: Local Interpretationmentioning

confidence: 77%

On the use of explainable AI for susceptibility modeling: examining the spatial pattern of SHAP values

Wang¹,

Zhang²,

Dahal³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Elevation directly influences the number and intensity of landslides and is considered one of the most important factors in landslide occurrence. It has a significant influence on features such as hydrological condition, slope, precipitation and vegetation in an area (Zhang et al, 2023).…”

Section: Determining Topographical Factorsmentioning

confidence: 99%

Landslide susceptibility prediction using machine learning and remote sensing: Case study in Thua Thien Hue province, Vietnam

Nguyen,

et al. 2023

Geological Journal

View full text Add to dashboard Cite

Landslides lead to widespread devastation and significant loss of life in mountainous regions around the world. Susceptibility assessments can provide critical data to help decision‐makers, for example, local authorities and other organizations, mitigating the landslide risk, although the accuracy of existing studies needs to be improved. This study aims to assess landslide susceptibility in the Thua Thien Hue province of Vietnam using deep neural networks (DNNs) and swarm‐based optimization algorithms, namely Adam, stochastic gradient descent (SGD), Artificial Rabbits Optimization (ARO), Tuna Swarm Optimization (TSO), Sand Cat Swarm Optimization (SCSO), Honey Badger Algorithm (HBA), Marine Predators Algorithm (MPA) and Particle Swarm Optimization (PSO). The locations of 945 landslides occurring between 2012 and 2022, along with 14 conditioning factors, were used as input data to build the DNN and DNN‐hybrid models. The performance of the proposed models was evaluated using the statistical indices receiver operating characteristic curve, area under the curve (AUC), root mean square error, mean absolute error (MAE), R2 and accuracy. All proposed models had a high accuracy of prediction. The DNN‐MPA model had the highest AUC value (0.95), followed by DNN‐HBA (0.95), DNN‐ARO (0.95), DNN‐Adam (0.95), DNN‐SGD (0.95), DNN‐TSO (0.93), DNN‐PSO (0.9) and finally DNN‐SCSO (0.83). High‐precision models have identified that the majority of the western region of Thua Thien Hue province is very highly susceptible to landslides. Models like the aforementioned ones can support decision‐makers in updating large‐scale sustainable land‐use strategies.

show abstract

“…Yang et al (2022) considered Anhua, Xinhua, Taojiang, and Taoyuan in Hunan Province as the study area, and they adopted four different landslide susceptibility evaluation models in which a Bayesian algorithm was used to improve the hyper-parameters and to obtain a better result. Zhang et al (2023) selected different areas of typical mountainous and hilly areas to construct landslide database. Then, a landslide susceptibility evaluation model is constructed based on XGBoost algorithm and landslide database, and the prediction results of the landslide susceptibility evaluation model are interpreted by SHAP algorithm.The previous studies on landslide hazard assessment have achieved good results that can be used as a reference for subsequent researches.…”

Section: Introductionmentioning

confidence: 99%

Distribution of ancient landslides and landslide hazard assessment in the Western Himalayan Syntaxis area

Cui

Yang

Xu³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

The Western Himalayan Syntaxis area is located near the Pamir Plateau. The geological structure is active and geological disasters occur frequently in this area. In this study, we employed the Google Earth platform and visual interpretation to identify ancient landslides in the region and to establish a regional ancient landslide database. Then, nine landslide-influencing factors (elevation, slope, aspect, curvature, distance to the river, distance to a glacier, lithology, distance to fault and distance to the epicenter of earthquakes above magnitude 5) were examined using ArcGIS software. The spatial distribution of landslides were analyzed statistically. Finally, an IV model and WoE model were used to evaluate the regional landslide hazard and the evaluation results were verified via a confusion matrix and a receiver operating characteristic (ROC) curve. The landslide database contained 7,947 landslides in this area with a total area of 3747.27 km2. Landslides were mostly developed at an elevation of 4,000–5,000 m, a slope of 15–25°, a north aspect, curvature of −0.33 to 0.33, distance to the water system of 1,000–2000 m, distance to a glacier of 2000–5,000 m, Permian sandstone, siltstone, argillaceous sandstone and Triassic siltstone, conglomerate and fine conglomerate, and distance to a fault of 20,000–50,000 m. The accuracy of the IV and WoE models was relatively high. The comprehensive accuracy of the confusion matrix of the two models was above 70% and the AUC value of the ROC curve was above 75%. The landslide database of the Western Himalayan Syntaxis was established and the landslide distribution and hazard assessment results can be used as a reference for landslide disaster prevention and mitigation and engineering construction planning in this area.

show abstract

Insights into geospatial heterogeneity of landslide susceptibility based on the SHAP-XGBoost model

Cited by 132 publications

References 29 publications

On the use of explainable AI for susceptibility modeling: examining the spatial pattern of SHAP values

On the use of explainable AI for susceptibility modeling: examining the spatial pattern of SHAP values

Landslide susceptibility prediction using machine learning and remote sensing: Case study in Thua Thien Hue province, Vietnam

Distribution of ancient landslides and landslide hazard assessment in the Western Himalayan Syntaxis area

Contact Info

Product

Resources

About