khabat Khosravi; Edris Marufinia; Ebrahim Nohani; Kamran Chapy
Abstract
In order to prevent any damages which can be caused by flood at Haraz watershed in the Mazandaran province, it is essential to prepare a flood susceptibility map using logistic regression. About 211 flood locations and 211 non-flood locations were first recognized. Ten flood conditioning factors such ...
Read More
In order to prevent any damages which can be caused by flood at Haraz watershed in the Mazandaran province, it is essential to prepare a flood susceptibility map using logistic regression. About 211 flood locations and 211 non-flood locations were first recognized. Ten flood conditioning factors such as Slope, plan curvature, altitude, distance from river, topographic wetness index (TWI), stream power index (SPI), rainfall, landuse and normalized differences vegetative index (NDVI) were then identified. The maps of all affecting factors were prepared using ArcGIS10.1, ENVI 5.1 and SAGA GIS2 software and they were exported to raster formats. Flood locations were randomly divided into two groups: 70% (151 flood locations) and 30% (60 flood locations) for modeling and validation, respectively. Enter method was selected for weighing the 10 factors in SPSS.18. The factors with their corresponding weights were used in the ArcGIS software for generation of flood susceptibility map. The map was divided into 5 classes. ROC curve and area under curve (AUC) are used for the validation of derived map. The results indicated that for prediction rate, the AUC is 78.3%; thus, the logistic regression has a reasonable accuracy for flood susceptibility mapping. The findings of this research are useful and necessary for scholars, the Mazandaran Regional Water Authority (MRWA), Ministry of Energy, and other agriculture and natural resources-related organizations in order for mitigating losses and damages during flooding events.
Hamidreza Moradi; Alireza Sepahvand; Parviz Abdolmaleki
Abstract
More than 30% of Iran's land is formed from mountainous areas. So each year, landslides cause damages to structures, residential areas and forests, creating sedimentation, muddy floods and finally deposit the sediments in reservoir dams. Therefore, for preventing of this damages and expressing the sensitivity ...
Read More
More than 30% of Iran's land is formed from mountainous areas. So each year, landslides cause damages to structures, residential areas and forests, creating sedimentation, muddy floods and finally deposit the sediments in reservoir dams. Therefore, for preventing of this damages and expressing the sensitivity rate of hillslopes, landslide hazard zonation is considered in prone areas. The purpose of this study is to determine the optimal structure of artificial neural network with different numbers of input factors for the landslide hazard zonation in the Haraz Watershed. First, the number of optimal epochs was determined to prevent network overlearning with trial and error method. Then, 14 neurons were determined in the hidden layer. Finally, the number of neurons was changed from 1 to 9 in the input layer. According to the obtained results, with increasing the number of neurons in the input layer, efficiency of Artificial Neural Network improved for landslide susceptibility mapping. In this research, nine neurons in the input layer, 14 neurons in the hidden layer and one neuron in the output layer were selected as the optimal structure. Root Mean Square Error and Descriptive Coefficient (R2) were equal to 0.051 and 0.962, respectively and the accuracy of landslide hazard zonation map was equal to 92.3%. Meanwhile, the results showed that about 35.14, 26.73, 14.59, 9.88, and 13.63 percent of all studied areas are located in stable, low, moderate, high and extremely hazardous areas, respectively.
