Search

A comprehensive integrated framework was designed to categorize individual streams and watersheds based on their temporal variation, and to determine the key drivers controlling watershed hydrology in the Qorveh-Dehgolan Basin. Four stream networks were extracted from multi-temporal Pan-Sharpened Landsat 7 and Landsat 8 imageries. Nineteen geomorphometric indices are measured and classified into the linear aspect, aerial aspect, and relief aspect. The results revealed a significant decline in stream networks from 2000 to 2017, reducing basin order. The factor analysis identified three principal factors including 'factor of basin magnitude' (affecting low-altitude watersheds), 'factor of steepness' (influencing high-altitude watersheds), and 'factor of texture'. The focal mechanism highlighted the complex fault mechanisms in the study area. Finally, RadViz clustering showed that for the lower order streams in the high-altitudes watersheds, stream width showed higher temporal variability in comparison to other parameters. In contrast, flow accumulation showed high variance in higher-order streams. [ABSTRACT FROM AUTHOR]

The application of multivariate statistical analysis (MSA) in analyzing drainage basin morphometry, in conjunction with seismic events (focal mechanism) and associated tectonic influences, has been discussed less in the framework of existing geomorphometric studies. Hence, in this study, an integrated approach was designed combining the geomorphometric analysis with MSA, focal mechanism and associated structural lineaments (using a remote sensing and GIS framework) to address the role of active tectonism on the drainage network characteristics in Qorveh-Dehgolan basin located in western part of Iran (Kurdistan province). Eight watersheds were delineated from burnt ASTER DEM using SWAT model and categorized into high-altitude watersheds and low-altitude watersheds. Four-sets of the drainage network were extracted from satellite imageries (Landsat 7 and 8) over a period of 17 years (2000–2017) to monitor the temporal stream network behavior. Overall, nineteen morphometric parameters were calculated (and categorized into linear, aerial, and relief aspects) exhibiting significant changes in the drainage network from 2000 to 2017 (the stream order declined from 5th to 4th order). The higher-altitude watersheds showed various anomalies in the calculated indices (indicating the potential influence of tectonic activity), while the lower altitude watersheds did not exhibit any specific anomalies. In addition, higher concentration of longer lineaments was observed in areas where the relatively resistant, fractured bedrock lithology (meta-volcanic and meta-sedimentary rocks) provided favorable conditions for origin and propagation of the lineaments. Further analysis of focal mechanism solutions (FMS, beachball diagrams) showed that the study area is primarily influenced by moderate earthquake events (3 ≤ Mw ≤ 6) and is influenced by a complex fault mechanism (primarily normal faults and strike slip faults). The morphometric parameters were further subjected to the factor analysis (principal axis factoring using Varimax rotation transformation) resulting in derivation of "factor of basin magnitude", "factor of steepness" and "factor of texture" as three principal factors impacting the drainage network. Finally, it was concluded that "factor of steepness" played a decisive role in defining hydrology of higher-altitude watersheds (an indication of tectonic influences in these watersheds) and thereby defining the overall hydrology of the study area as well. Overall, this study highlighted the significant role of the tectonic activity associated with structural lineaments and the seismic events in defining the nature and behavior of the drainage network and certain anomalies caused in the basin hydrology under these structural controls. [ABSTRACT FROM AUTHOR]

Landslide is a natural hazard which causes huge loss of properties and human life in many places of the world. Mapping of landslide susceptibility is an important task for preventing and combating the landslides problems. Main objective of this study is to use different artificial intelligence methods namely support vector machines (SVM), artificial neural networks (ANN), logistic regression (LR), and reduced error-pruning tree (REPT) in the development of models for landslide susceptibility mapping of Muong Lay district of Vietnam. In total data of 217 landslide locations of the study area was used for the development and evaluation of the models. Nine landslide-conditioning factors were used for generating the datasets for training and validating the models. Results show that the SVM outperformed all other methods namely ANN, LR and REPT. Thus, it can be suggested that the SVM method is more useful in developing accurate and robust landslide prediction model. [ABSTRACT FROM AUTHOR]