Serveh Darvand; Hassan Khosravi; Hamidreza Keshtkar; Gholamreza Zehtabian; Omid Rahmati
Abstract
The purpose of this study was to compare machine learning models including Support Vector Machine, Classification and Regression Tree, Random Forest, and Multivariate Discriminate Analysis to prioritize susceptible areas to dust production. To determine the dust days, hourly meteorological data of Alborz ...
Read More
The purpose of this study was to compare machine learning models including Support Vector Machine, Classification and Regression Tree, Random Forest, and Multivariate Discriminate Analysis to prioritize susceptible areas to dust production. To determine the dust days, hourly meteorological data of Alborz and Qazvin provinces and satellite images of the same days for the period 2000 to 2019 were used. 420 dust collection points were identified and the map of their distribution was prepared. The maps of factors affecting the occurrence of dust, including landuse map, soil orders map, slope map, slope aspect map, elevation map, vegetation map, topographic surface moisture, topographic surface ratio, and geology mam were prepared. Using the mentioned models, the impact of each of the effective factors of dust was determined and prioritization maps of dust harvesting areas were prepared. Models were evaluated using the ROC curve. According to the results, the elevation factor is more important in all models than the other parameters used in the model. The modeling results also showed that the Random Forest )RF( and Multivariate Discriminate Analysis (MDA) models had the highest values of accuracy (0.96), precision (0.94), Probability of Detection (POD) (0.98), and False Alarm Ratio (FAR) (0.051) compared to the others. The performance of the RF and MDA models is better than the other models, followed by the Support Vector Machine (SVM) and Classification and Regression Tree (CART) models, respectively. Also, in evaluating the models using Receiver Operating Characteristic (ROC), the RF model was selected as the best model.
Somayeh Movahedi; aboalhasan fathabadi; null null; Ali Heshmatpour
Abstract
In this study using Frequency Ratio (FR), Statistical Index (SI), Weights Of Evidence(WOF), Logistic Regression (LR), Random Forest (RF) models the probability of gully formation was calculated in Aytamar watershed and susceptibility maps was prepared. First the thematic maps of 13 gully conditioning ...
Read More
In this study using Frequency Ratio (FR), Statistical Index (SI), Weights Of Evidence(WOF), Logistic Regression (LR), Random Forest (RF) models the probability of gully formation was calculated in Aytamar watershed and susceptibility maps was prepared. First the thematic maps of 13 gully conditioning factors including lithological formations, distance to faults, faults density, altitude, slope-length, slope angle, slope aspect, plan curvature, profile curvature, distance to roads, land use, distance to rivers, stream power index and topographic wetness index was prepared. Then landslide inventory map was combined with each gully conditioning factor and all models weights and parameters were calculated. Area under curve for test data was calculated as 0.74, 0.78, 0.75, 0.86 and 0.96 for Frequency Ratio (FR), Statistical Index (SI), Weights Of Evidence(WOF), Logistic Regression (LR), Random Forest (RF) models, respectively. Random forest, Frequency Ratio and Logistic Regression have the least the area of high susceptibility zone, respectively. With respect three validation criteria multivariate methods including Random Forest and Logistic Regression had the best performance among all models.
ALIREZA Arabameri; khalil rezaei; mojtaba yamani
Abstract
Gully erosion is one of the erosive processes that mostly change the shape of the earth surface and has severe environmental and economic damages. The aim of this research is modeling between geo-environmental parameters effective in gully erosion and gully occurrence in the study area and gully erosion ...
Read More
Gully erosion is one of the erosive processes that mostly change the shape of the earth surface and has severe environmental and economic damages. The aim of this research is modeling between geo-environmental parameters effective in gully erosion and gully occurrence in the study area and gully erosion susceptibility mapping using evidential belief function (EBF) data driven model in toroud watershed that has high susceptibility to gully erosion. At first, a gully erosion inventory map is prepared, using extensive field surveys and 80 gullies which have been identified, 70 percentage (56 gully location) randomly selected to modeling, while the remaining 30 percentage (24 gully location) are used to validation. In modeling, if there was high correlation among parameters, reduce accuracy of model, thus has done multi-collinearity test among independent variables. Tolerance and the variance inflation factor (VIF) are two important indexes for multi-collinearity diagnosis. Finally 15 parameters including geomorphological, geological, environmental and hydrological are selected for modeling. In evidential belief function model four relationships were calculated: belief (Bel), disbelief (Dis), uncertainty (Unc), and plausibility (Pls) and belief function are used for gully erosion susceptibility mapping. Area under the curve are used for model validation. According to results, EBF model with prediction rate (1) and success rate (0.959) had excellent accuracy and capability in identification of prone areas to gully erosion in study area. The results indicates that 21.79 percentage (90.84 km2) in study area located in high and very high susceptibility class.
ALIREZA Arabameri; kourosh shirani; khalil rezai; mojtaba yamani
Abstract
landslides situation recognized using interpreting the aerial photos and extensive field measurements. Among total number of 200 identified landslides, %70 (140 landslides) of them have been utilized for model executing and %30 (60 landslides) of them for verification randomly. This research criteria ...
Read More
landslides situation recognized using interpreting the aerial photos and extensive field measurements. Among total number of 200 identified landslides, %70 (140 landslides) of them have been utilized for model executing and %30 (60 landslides) of them for verification randomly. This research criteria including geomorphological parameters, hydrological parameters , geological parameters and environmental parameters . The Shannon’s entropy model have been used for defining the criteria weight and Area density model for defining classes weight, then the regionalization map obtained by combining the criteria and classes weight in ArcGIS 10.2 software environment and classified to 5 classes very little, little , moderate, high and very high according to natural fractures. The Roc curve have been used for model verification. The clerical accuracy results indicated that the compound model have the high accuracy 0.877 (87.7%) for identifying the regions susceptible to landslide. According to the results, slope length, slope and topography wetness index have had the most effect in occurring the landslide. Among total area of region (168547 hectar), 27.39% (46165.02 hectar) have been placed in high and very high sensitive. The prepared regionalization map can be useful for planning land use and building the infrastructure installations such as road.