Comparison of Different Machine Learning Methods for Debris Flow Susceptibility Mapping: A Case Study in the Sichuan Province, China

Abstract

Debris flow susceptibility mapping is considered to be useful for hazard prevention and mitigation. As a frequent debris flow area, many hazardous events have occurred annually and caused a lot of damage in the Sichuan Province, China. Therefore, this study attempted to evaluate and compare the performance of four state-of-the-art machine-learning methods, namely Logistic Regression (LR), Support Vector Machines (SVM), Random Forest (RF), and Boosted Regression Trees (BRT), for debris flow susceptibility mapping in this region. Four models were constructed based on the debris flow inventory and a range of causal factors. A variety of datasets was obtained through the combined application of remote sensing (RS) and geographic information system (GIS). The mean altitude, altitude difference, aridity index, and groove gradient played the most important role in the assessment. The performance of these modes was evaluated using predictive accuracy (ACC) and the area under the receiver operating characteristic curve (AUC). The results of this study showed that all four models were capable of producing accurate and robust debris flow susceptibility maps (ACC and AUC values were well above 0.75 and 0.80 separately). With an excellent spatial prediction capability and strong robustness, the BRT model (ACC = 0.781, AUC = 0.852) outperformed other models and was the ideal choice. Our results also exhibited the importance of selecting suitable mapping units and optimal predictors. Furthermore, the debris flow susceptibility maps of the Sichuan Province were produced, which can provide helpful data for assessing and mitigating debris flow hazards.

Landslide Geo-Hazard Risk Mapping Using Logistic Regression Modeling in Guixi, Jiangxi, China

“…It is thus a challenge to objectively, quantitatively, and accurately conduct hazard risk assessment at a regional scale [25,26]. Machine learning algorithms have also been utilized for landslide research [8,10], and a series of models are available, such as support vector machines (SVMs) [27], artificial neural networks (ANNs) [28], random forests (RFs) [29], boosted regression trees (BRTs) [30,31], and limit learning machine (LLM) [32]. These algorithms have been reported to have high accuracy for landslide risk assessment, though they have clear disadvantages such as a complex modeling process, unstable model performance, and weak explanatory capability [33,34].…”

Debris Flow Susceptibility Assessment Using the Integrated Random Forest Based Steady-State Infinite Slope Method: A Case Study in Changbai Mountain, China

“…This is due to the proposed RF-SSIS method inheriting the excellent diagnostic performance of the RF model (i.e., heuristic/probabilistic model) for the region where a debris flow disaster already existed [63]; meanwhile, this method further refined the debris-flow-prone area from the suitable area terrain condition based on the physico-mechanical properties. This is the reason why the proposed RF-SSIS method had better predicting performance than the RF model; however, the prediction accuracy did not improve very well, because under the support of historical data, the RF model exhibited very high prediction accuracy for debris-flow exist areas [14]; so the space for improvement was limited and difficult to further refine. Therefore, even though the determination of FP was improved significantly, the RF-SSIS method classified just 758 more TP pixels than did the RF model; the determination on TP was less improved, causing an insignificant improvement in the prediction accuracy.…”

“…Cama and Lombardo [11] adopted a binary logistic regression model to estimate the susceptibility of debris flow in Messina, Italy. Furthermore, a considerable number of studies have been performed using heuristic or probabilistic models to assess the susceptibility of debris flow [12][13][14]. Despite the differences in the approach used, each model requires an accurate, reasonable, and complete catalog of historical debris flow events [15], and the support of detailed and sufficient environmental data to make the results relatively objective and reproducible.…”

Adapting the Sudden Landslide Identification Product (SLIP) and Detecting Real-Time Increased Precipitation (DRIP) algorithms to map rainfall-triggered landslides in the West-Cameroons’ highlands (Central-Africa)

“…Nowadays, the quantitative approach is supported by several machine learning algorithms for better accuracy. They can be single or hybrids, and amongst them are processing such as the support vector machine (SVM), Random Forest (RF), Fisher's Linear Discriminant Analysis (FLDA), Bayesian Network (BN), Logistic Regression (LR), and Naïve Bayes (NB), or more recently the AdaBoost, MultiBoost, Bagging, and Rotation Forest (Marjanovic et al, 2011;Goetz et al, 2015;Pham et al, 2016a&b;Ada and San, 2018;Pham et al, 2018;Shirzadi et al, 2018;Cavanesi et al, 2020;Xiao, et al, 2020;Xiong et al, 2020).…”