Machine learning (ML) models are extensively used in spatial predictive modeling, including landslide susceptibility prediction. The performance statistics of these models are vital for assessing their reliability, which is typically obtained using the random cross-validation (R-CV) method. However, R-CV has a major drawback, i.e., it ignores the spatial autocorrelation (SAC) inherent in spatial datasets when partitioning the training and testing sets. We assessed the impact of SAC at three crucial phases of ML modeling: hyperparameter tuning, performance evaluation, and learning curve analysis. As an alternative to R-CV, we used spatial cross-validation (S-CV). This method considers SAC when partitioning the training and testing subsets. This experiment was conducted on regional landslide susceptibility prediction using different ML models: logistic regression (LR), k-nearest neighbor (KNN), linear discriminant analysis (LDA), artificial neural networks (ANN), support vector machine (SVM), random forest (RF), and C5.0. The experimental results showed that R-CV often produces optimistic performance estimates, e.g., 6–18% higher than those obtained using the S-CV. R-CV also occasionally fails to reveal the true importance of the hyperparameters of models such as SVM and ANN. Additionally, R-CV falsely portrays a considerable improvement in model performance as the number of variables increases. However, this was not the case when the models were evaluated using S-CV. The impact of SAC was more noticeable in complex models such as SVM, RF, and C5.0 (except for ANN) than in simple models such as LDA and LR (except for KNN). Overall, we recommend S-CV over R-CV for a reliable assessment of ML model performance in large-scale LSM.

This study evaluates the utility of the ensemble framework of feature selection and machine learning (ML) models for regional landslide susceptibility mapping (LSM) in the arid climatic condition of southern Peru. A historical landslide inventory and 24 different landslide influencing factors (LIFs) were prepared using remotely sensed and auxiliary datasets. The LIFs were evaluated using multi-collinearity statistics and their relative importance was measured to select the most discriminative LIFs using the ensemble feature selection method, which was developed using Chi-square, gain ratio, and relief-F methods. We evaluated the performance of ten different ML algorithms (linear discriminant analysis, mixture discriminant analysis, bagged cart, boosted logistic regression, k-nearest neighbors, artificial neural network, support vector machine, random forest, rotation forest, and C5.0) using different accuracy statistics (sensitivity, specificity, area under curve (AUC), and overall accuracy (OA)). We used suitable combinations of individual ML models to develop different ensemble ML models and evaluated their performance in LSM. We assessed the impact of LIFs on ML performance. Among all individual ML models, the k-nearest neighbors (sensitivity = 0.72, specificity = 0.82, AUC = 0.86, OA = 78%) and artificial neural network (sensitivity = 0.71, specificity = 0.85, AUC = 0.87, OA = 79%) algorithms showed the best performance using the top five LIFs, while random forest, rotation forest, and C5.0 (sensitivity = 0.76–0.81, specificity = 0.87, AUC = 0.90–0.93, OA = 82–84%) outperformed other models when developed using all twenty-four LIFs. Among ensemble models, the ensemble of k-nearest neighbors and rotation forest, k-nearest neighbors and artificial neural network, and artificial neural network and rotation forest outperformed other models (sensitivity = 0.72–0.73, specificity = 0.83–0.84, AUC = 0.86, OA = 79%) using the top five LIFs. The landslide susceptibility maps derived using these models indicate that ~2–3% and ~10–12% of the total study area fall within the “very high” and “high” susceptibility. The obtained susceptibility maps can be efficiently used to prioritize landslide mitigation activities.

There is a fundamental interest in studying travertine rocks, and this is to understand their structure, their geomechanical behavior and other particularities in order to guarantee their proper use in different engineering and architectural applications, and thus, evaluate the sustainability of the travertines, natural resources, the stability of slopes, the preservation of cultural heritage and the mitigation of possible anthropic risks. Travertine has petrological and mechanical properties similar to carbonates from oil fields such as those found in El Presal-Brazil, which currently contain the largest hydrocarbon reserves in the world. Given the impossibility of obtaining rock samples from this deposit to carry out the study, rocks similar to these were used. The present study specifically used samples of Lapis tiburtinus rocks, coming from the west of the city of Tivoli in Italy and these showed resistance to uniaxial and triaxial compression, and showed mechanical resistance due to increased porosity and brittleness. The investigation carried out an analysis of the geomechanical behavior travertine through an experimental program, which includes a petrological, structural, and mechanical characterization. It was determined the travertine is mainly composed of micrite and spastic calcite without the presence of grains or allochemical cements and presents high porosity of the fenetral and vulgar type. Macro and micropores were found to be chaotically distributed in the rock and have low connectivity, which demonstrates the complexity and heterogeneity of the porous structure of Roman travertine. Uniaxial and triaxial compressive strength tests were also carried out, observing a decrease in its mechanical strength due to the increase in porosity, presenting a property of brittleness in its behavior. The results were consistent and valid for this type of rock compared to other studies; determining that there is a correct and adequate operation of the triaxial cell used in the mechanical resistance tests.

Small communities in the Arequipa region of Peru are susceptible to rockfall hazards, which impact their lives and livelihoods. To mitigate rockfall hazards, it is first necessary to understand their locations and characteristics, which can be compiled into an inventory used in the creation of rockfall hazard rating maps. However, the only rockfall inventory available for Arequipa contains limited data of large, discrete events, which is insufficient for characterizing rockfall hazards at the community scale. A more comprehensive inventory would result in a more accurate rockfall hazard rating map—a significant resource for hazard mitigation and development planning. This study addresses this need through a remote method for rockfall hazard characterization at a community scale. Three communities located in geographically diverse areas of Arequipa were chosen for hazard inventory and characterization, with a fourth being used for validation of the method. Rockfall inventories of source zones and rockfall locations were developed using high-resolution aerial imagery, followed by field confirmation, and then predictions of runout distances using empirical models. These models closely matched the actual runout distance distribution, with all three sites having an R² value of 0.98 or above. A semi-automated method using a GIS-based model was developed that characterizes the generation and transport of rockfall. The generation component criteria consisted of source zone height, slope angle, and rockmass structural condition. Transport was characterized by rockfall runout distance, estimated rockfall trajectory paths, and hazard ratings of corresponding source zones. The representative runout distance inventory model of the validation site matched that of a nearby site with an R² of 0.98, despite inventorying less than a third of the number of rockfalls. This methodology improves upon current approaches and could be tested in other regions with similar climatic and geomorphic settings. These maps and methodology could be used by local and regional government agencies to warn residents of rockfall hazards, inform zoning regulations, and prioritize mitigation efforts.