Ensemble of fuzzy-analytical hierarchy process in landslide susceptibility modeling from a humid tropical region of Western Ghats, Southern India.

The western flanks of the Western Ghats are one of the major landslide hotspots in India. Recent rainfall triggered landslide incidents in this humid tropical region necessitating the accurate and reliable landslide susceptibility mapping (LSM) of selected parts of Western Ghats for hazard mitigation. In this study, a GIS-coupled fuzzy Multi-Criteria Decision Making (MCDM) technique is used to evaluate the landslide-susceptible zones in a highland segment of the Southern Western Ghats. Fuzzy numbers specified the relative weights of nine landslide influencing factors that were established and delineated using the ArcGIS, and the pairwise comparison of these fuzzy numbers in the Analytical hierarchy process (AHP) system resulted in standardized causative factor weights. Thereafter, the normalized weights are assigned to corresponding thematic layers, and finally, a landslide susceptibility map is generated. The model is validated using the area under the curve values (AUC) and F1 scores. The result reveals that about 27% of the study area is classified as highly susceptible zones followed by 24% area in moderately susceptible zone, 33% in low susceptible, and 16% in a very low susceptible area. Also, the study shows that the plateau scarps in the Western Ghats are highly susceptible to the occurrence of landslides. Moreover, the predictive accuracy estimated by the AUC scores (79%) and F1 scores (85%) shows that the LSM map is trustworthy for future hazard mitigation and land use planning in the study area.

Development of new computational machine learning models for longitudinal dispersion coefficient determination: case study of natural streams, United States.

Natural streams longitudinal dispersion coefficient (Kx) is an essential indicator for pollutants transport and its determination is very important. Kx is influenced by several parameters, including river hydraulic geometry, sediment properties, and other morphological characteristics, and thus its calculation is a highly complex engineering problem. In this research, three relatively explored machine learning (ML) models, including Random Forest (RF), Gradient Boosting Decision Tree (GTB), and XGboost-Grid, were proposed for the Kx determination. The modeling scheme on building the prediction matrix was adopted from the well-established literature. Several input combinations were tested for better predictability performance for the Kx. The modeling performance was tested based on the data division for the training and testing (70-30% and 80-20%). Based on the attained modeling results, XGboost-Grid reported the best prediction results over the training and testing phase compared to RF and GTB models. The development of the newly established machine learning model revealed an excellent computed-aided technology for the Kx simulation.