Showing 6 results for Logistic Regression
S. Z. Mosavi Khatir, A. Kavian, A. K. Soleimani,
Volume 14, Issue 53 (10-2010)
Abstract
In this research, logistic regression analysis was used to create a landslide hazard map for Sajaroud basin. At first, an inventory map of 95 landslides was used to preduce a dependent variable, which takes a value of 0 for absence and 1 for presence of landslides. Ten factors affecting landslide occurence such as elevation , slope gradient, slope aspect, slope curvature, rainfall, distance from fault, distance from drainage, distance from road , land use and geology were taken as independent parameters. The effect of each parameter on landslide occurrence was determined from the corresponding coefficient that appears in the logistic regression function. The interpretation of the coefficients showed that road network plays the most important role in determining landslide occurrence. Elevation, curvature, rainfall and distance from fault were excluded from the final analysis because these variables did not significantly add to the predictive power of the logistic regression. After transferring final probability function into Arc/view 3.2 software, landslide susceptibility map was prepared. The results of accuracy assessment showed that overall accuracy of produced map is 85.3 percent. Therefore, 53% of the area was located in very low hazard, 18.3% in low hazard, 21% in moderate hazard and 7.7 % residual area is located in high hazard regions. Model and then susceptibility map verity was assessed using -2LL, Cox and Snell R2, Nagelkerk R2, and was validated.
A. Jafari, H. Khademi, Sh. Ayoubi,
Volume 16, Issue 62 (3-2013)
Abstract
Digital soil mapping includes soils, spatial prediction and their properties based on the relationship with covariates. This study was designed for digital soil mapping using binary logistic regression and boosted regression tree in Zarand region of Kerman. A stratified sampling scheme was adopted for the 90,000 ha area based on which, 123 soil profiles were described. In both approaches, the occurrence of relevant diagnostic horizons was first mapped, and subsequently, various maps were combined for a pixel-wise classification by combining the presence or absence of diagnostic horizons. Covariates included a geomorphology map, terrain attributes and remote sensing indices. Among the predictors, geomorphology map was identified as an important tool for digital soil mapping approaches as it helped increase the prediction accuracy. After geomorphic surfaces, the terrain attributes were identified as the most effective auxiliary parameters in predicting the diagnostic horizons. The methods predicted high probability of salic horizon in playa landform, gypsic horizon in gypsiferous hills and calcic horizon in alluvial fans. Both models predicted Calcigypsids with very low reliability and accuracy, while prediction of Haplosalids and Haplogypsids was carried out with high accuracy.
K. Shirani, A. R. Arabameri,
Volume 19, Issue 72 (8-2015)
Abstract
This research was conducted to prepare landslide susceptibility zonation (LSZ) map for the Dez-e-Ouliabasin using logistic regression model. For this purpose, at first, the most important factors affecting land sliding including slope, aspect, elevation, precipitation, the distance from road, the distance from fault, the distance from drainage, land use, and lithology were determined. Then, thelandslide inventory mapwas preparedby using field digital checks, GPS and satellite images. In the next step, the landslide susceptibility zonation mapwas preparedby usinglogistic regression method. According to the obtained coefficients for LSZ maps, the most important factor in the study area was elevation layer. The Receiver Operating Curve (ROC) index value was calculated (0.92), which indicates a very high level and suggests thatthe observed mass movements have a strong relationship with the logistic regression model.
M. Erfanian, H. Farajollahi, M. Souri, A. Shirzadi,
Volume 20, Issue 75 (5-2016)
Abstract
The aim of this study is to prepare the groundwater spring potential map using Weight of Evidence, logistic regression, and frequency ratio methods and comparing their efficiency in Chehlgazi watershed, province of Kurdistan. At first, 17 effective factors in springs occurrence including geology, distance to fault, fault density, elevation, relative permeability of lithological units, slope steepness, slope aspect, plan curvature, profile curvature, precipitation, distance to Stream, drainage Stream density, Sediment Transport Capacity Index (STCI), Stream Power Index, topographic wetness index (TWI) and land use/land cover (LU/LC) were selected. The validation processes of methods were conducted by relative performance characteristic curve (ROC). The area under an ROC curve (AUC) for the weight of evidence, logistic regression and frequency ratio was 85/8%, 79% and 89%, respectively. The results showed that all methods are suitable estimator for mapping the groundwater spring potential in the study area. But the frequency ratio method with the most amounts is the best method to produce and map the groundwater spring potential. Also, validation of the mappings based on the percentage of pilot springs, training springs and all springs showed that the logistic regression, WoE and frequency ratio, with 45, 56 and 45 percent of spring occurrence on the high potential classes respectively, had the highest validation.
F. Abbaszadeh Afshar, ِ S. Ayoubi, A. Jafari,
Volume 21, Issue 1 (6-2017)
Abstract
Mapping the spatial distribution of soil taxonomic classes is important for useful and effective use of soil and management decisions. Digital soil mapping (DSM) may have advantages over conventional soil mapping approaches as it may better capture observed spatial variability and reduce the need to aggregate soil types. A key component of any DSM activity is the method used to define the relationship between soil observations and environmental covariates. This study aims to compare multiple logistic regression models and covariate sets for predicting soil taxonomic classes in Bam district, Kerman province. The environmental covariates derived from digital elevation models, Landsat imagery, geomorphology map and soil unit map that were divided into two different sets: (1) variables derived from digital elevation models, remote sensing and geomorphology map, (2) variables derived from digital elevation model, remote sensing, geomorphology map and the soil map. Stratified sampling schemes were defined in 100000 hectares, and 126 soil profiles were excavated and described. The results of accuracy model showed that data set 2 increased accuracy of model including overall accuracy, kappa index, user accuracy and reliability of the producer. The results showed that the multiple logistic regression model can promote traditional soil mapping and it can be used to large group of other scientific fields.
K. Nosrati, M. Heydari, M. Hoseinzadeh, S. Emadoddin,
Volume 22, Issue 3 (11-2018)
Abstract
Ziarat drainage basin, in the southern part of Gorgan city, is exposed to mass movement, especially landslide occurrence, due to geologic, geomorphologic, and anthropogenic reasons. The objectives of this study were to predict landslide susceptibility and to analyze the effective factors using rare events logistic regression. In view of this, the map layers of the variables including geology, land use, slope, slope aspect, distance of road, distance of fault and distance of river were prepared using topographic and geologic maps and aerial photo interpretation. In addition, the map layers of the soil variables including the percent of clay, silt, sand, and saturation water as well as plasticity limit index were determined based on the laboratory analysis of 32 soil samples collected from landslide sites and 32 soil samples obtained from non-occurrence landslide sites. The controlling factors of landslide were determined using rare events logistic regression analysis; then based on their coefficients, the landslide risk zoning map was prepared and validated. The landslide risk zoning map was classified in five different hazard classes ranging from very low risk to very high risk; the very high risk class with 16.8 km2 was assigned as the having the highest percent of the catchment area. The results of the model validation showed that the rare events logistic regression model with the receiver operating characteristic (ROC) of 0.69 could be a suitable prediction model for the study area. The results of this study could be, therefore, useful for corrective actions and watershed management landslide high-risk zones.
