logistic regression model

This study aimed to systematically explain the potential of the landslide occurrence to provide a prediction model of the possibility of this phenomenon in the Yamchi catchment in Ardebil province. In this regard, both approaches of discrete and continuous variables were used by means of overlay and logistic regression, respectively. Independent variables included elevation, slope, aspect, lithology, annual rainfall, roughness, general curvature, topographic wetness index, vegetation index, distance to fault, distance to stream and distance to road. The results, firstly, revealed the areas with high landslide potential by the matching layers of independent variables with the landslide layer in the geographical information system (GIS). These areas were in the middle elevation, high slopes, northern slope, high roughness, erodible formations, high rainfall, medium vegetation, surroundings of faults and rivers. Secondly, the results of the logistics regression model by providing a prediction equation of probability of landslide occurrence showed that the resulting model with pseudo r2 and ROC 0.22 and 0.86, respectively, had good power and efficiency to predict landslide through the catchment. In addition, the resulting beta coefficients for independent variables indicated that the importance of the variables was as follows: vegetation index distance to road, rain, lithology, distance to fault, elevation, topographic wetness index, roughness index, aspect, slope, and distance to river. In the end, the need to pay serious attention to the supporting and protection of vegetation cover of the mid -range and upstream of the catchment was determined because of unstable geomorphic conditions of these areas.

Slope instabilities are one of the major natural hazards in mountainous areas that cause significant damage to human activities each year. The purpose of this study was to compare the artificial neural network model with the logistic regression model to evaluate the risk of mass movements and amplitude instability and to identify the effective factors in this phenomenon in Ojan Chay basin. The purpose of the results of the statistical models is to determine the regions with potential of occurrence of instability and finally to prepare a hazard zoning map for the study area. And then the layers are prepared. Distribution map of slope instabilities that occurred in the basin was prepared and integrated with the map of factors affecting the movements and slope distribution map of the slope. Distance from fault, land use, distance from village and road, distance from drainage network were calculated in ArcGIS software environment. ROC, Pseudo R square and Chi Square coefficients were used to evaluate the outputs of the models used. The results showed that the percentages of high risk zones in neural network model and logistic regression were 10.32% and 5.06%, respectively, which mainly include the lithologically restricted zones of these areas. . Also the neural network model with ROC value is 0.89 more efficient than logistic regression for zoning the occurrence of domain instabilities; based on zoning using neural network model, respectively, 40.32, 22.15, 18.32, 8.89,10.32 of the area is classified as very low, low, medium, high and very high risk classes

Several factors have contributed to the occurrence of the landslide that could increase the risk of landslide in any area. Identifying these factors and their value can help to appropriate landslide zonation. The aim of the study is to find ways to reduce the damages caused by them, which makes it necessary to zoning the susceptible areas that play an undeniable role in watershed management. Therefore, by using statistical models and assessing them, the sensitive areas to the occurrence of landslide are identified. In this research, the landslide hazard zonation was performed based on the data-driven method. Based on this method, the zoning was done based on the use of slope data, aspect, elevation, precipitation, vegetation, geology, land use, distance to fault, distance to river, and distance to road. To validate the model, the ROC curve has been used which is a new and efficient method for verification. The purpose of this research is to investigate various influencing factors that affect the landslide occurrence in the Nazlochai basin.

In the methodology section, the satellite imagery processing (to identify and extract landslides, vegetation extraction, and land use) and logistic regression model have been discussed for landslide hazard zonation. In this study, by reviewing the previous sources (Mir Nazari, et al., 1393, Abedini, et al., 1393, Ayalew, et al., 2004, Ebadinejad, et al., 2007) and by investigating various factors (morphometric, climatic, and human) in Nazlochai basin, ten effective factors (elevation, slope, aspect, distance to river, distance to road, distance to fault, lithology, landuse, precipitation, and vegetation) on the landslide occurrence in the area were considered. The ArcGIS software was used to digitize and provide information layers for landslide hazard zonation, and the ENVI software was used for image processing, vegetation extracting, and land use mapping. Existing landslides were identified and characterized using various tools including aerial photos, satellite imagery (Google Earth), existing information, Global Position System (GPS), and field surveys.

The obtained coefficients indicated that the occurrence of landslide in the studied area had a direct relation with lithology, slope, and aspect factors, and weak relation with landuse, distance to fault, precipitation and distance to river. Lithology investigation of the region indicated that the more landslides have occurred on calcareous and conglomerate stones, which could be due to the development of the slopes and the accumulation of destructive materials on them. Slope is one of the slippery factors due to gravity and decreasing shear strength of soil in slopes of more than 10% to 45% leads to instability which in most researches is considered as an effective factor, too. Also, north slopes are more susceptible to landslide than the southern slopes due to the reduction of normal pressure and shear strength of the soil. By considering the Pseudo R-square index (equal to 0.34), which is greater than the threshold (0.2), this model shows acceptable fit. The area under the ROC curve was equal to 0.958, which shows a strong correlation with predicted landslides by the logistic regression model. Finally, the study area was classified into 5 landslide hazard classes include very low, low, medium, high, and very high.

In this research, landslide hazard zonation has been done using the logistic regression model in the Nazlochai basin. The coefficients of variables indicated that the occurrence of landslide in the study area had a direct relationship with the lithology, slope, and aspect factors; and weak relationship with landuse and distance to fault. Thus this indicates the probability of landslide occurrence increases by changing in lithology, slope, and aspect

Landslide susceptibility mapping is  essential for  land use  planning and decision-making especially in  the mountainous areas. The main objective of this  study is to produce landslide susceptibility maps (LSM) at Dalahoo basin, Iran  using two statistical models such as an  index of entropy and Logistic Regression and to compare the  obtained results. At the  first stage, landslide locations identified by Natural Resources Department of Kermanshah Province is used to prepare of LSM map. Of the 29 lanslides identified, 21 (≈ 70%) locations were used for the landslide susceptibility maps, while the remaining 8 (≈ 30%) cases were used for the model validation. The landslide conditioning factors such as slope degree, slope aspect, altitude, lithology, distance to faults, distance to rivers, distance to roads, land use, and  lithology  were extracted from the spatial database. Using these factors,  landslide susceptibility and weights of each factor were analyzed by index of entropy and Logistic Regression models. Finally, the ROC (receiver operating characteristic) curves for landslide susceptibility maps were drawn and  the areas under the curve (AUC) were calculated. The verification results showed that the index of entropy model (AUC = 86.08%) performed slightly better than conditional probability (AUC = 80. 13%) model. The produced susceptibility maps can be useful for general land use  planning in the Dalahoo basin, Iran.

Today, a wide range of spatial analysis models are used in environmental risk zoning. Some models, such as hierarchical and fuzzy analyzes, despite the inclusion of uncertainty in the input variables, are unable to explain quantitatively the output uncertainty. In this study, the aim of evaluating the capabilities of the Dempster-Schaeffer algorithm is to explain the uncertainty in the outcomes for landslide hazard zonation in the south of Chalus. Therefore, after field studies and review of similar studies, a map of 10 factors was provided in the GIS environment and was introduced as input data along with a map of the distribution of landslides to the model. Landslide hazard zonation was performed by integrating different weights in the Dempster-Sheffer model and in order to evaluate the output of the model, a logistic regression model was used; the performance of the two models was based on the output results of the models and using two indicators of the density ratio (Dr) And the sum of utility (Qs) was evaluated and verified. The results of Dr showed that both models had good performance in identifying high-risk classes compared to low risk classes. Based on the Qs index, the Dempster-Schafer model with QS = 98/2 was good compared to Logistic regression model with QS = 91/66 has a better relative utility. Therefore, the D-S model is more successful in identifying risk classes (finiteness) and consequently hazard classes (uncertainty) in the region.

The investigation of land cover and land use changes has been important since ancient eras and these changes mainly occur in two ways: the first types of these changes take place by the means of natural factors such as erosion, tectonic forces and floods and the second types take place by human factors. In recent years in many parts of the world, human oriented activities were the most effective factors in changing land use, land cover and urban development. Nowadays population growth,on one hand and limites resources,on the other hand, result in changing of the application of lands. Land use changes in agricultural areas neighboring the cities is an important factor which has a vital role in supplying food and maintaining food security as well as earning money for urban population which have rapid growth. Vicinity of agricultural and horticultural areas with cities of Iran as ecological reserves, has increased the necessity of conducting this study. The present study aims to minimize the extent of use changes and lessen challenge of ecological destruction via development scenarios.
In this study logistic regression model has been used, which is a special kind of multiple regressions and in which the dependent variable is discrete. Contrary to other linear models, the least square method was used in logistic regression. Maximum probability method was used for estimation of characteristics and factors of each independentvariable. Therefore,the logistic regression equation is as follows: { }{ }{ }….{ }
Different data and softwares were used toward implementing the regression model. The soft wares used are:Ecognition¡Idrisiselva and Arc GIS. Images were received through satellite Landsat 5 in 1986 and Landsat 8 in 2014 from USGS site (United State Geological Site). Regarding the location of Mahabad which is situated in pass 168 and rows 34 and 35, we received two frames each year. Pre-process operations including image mosaic, subset and radiometric correction were implemented on images before classification operation by Flaash method in ENVI environment. For increasing accuracy of multi-spectral images of Landsat 8 bands which have low local resolution and high spectral resolution fusion operations were implemented combined with 15 meters panchromatic band. Classification was conducted through the Fuzzy method in the Ecognition environment. Then, classified images were transformed into ASCII format in Arc GIS environment for entering the software environment of Idrisiselva, and their format were changed into RST format in IDRISI environment software for preparing variables. Logistic regression command was used for modeling in the software Idrisi environment. To do this, we entered the amount of change in urban development from 1986 to 2014 into the model environment as a dependent variable and other involved factors in urban development as independent variables, respectively. We used Fuzzy method for scenario building and zoning of developed areas, in which developed areas were categorized into four classes: very proper, proper, improper and very improper.
The investigation of land use changes from 1986 to 2014 determined that land uses decreased with urban development every year. Land use changes are as follow: building lands increased from 989.86 ha to 2000.02 ha. Water resources (dikes and rivers) had decreasing trend from 841.85 ha to 564.16ha. Irrigated arable lands decreased from 2889.17 ha to 1195.378 ha in 1986, and also rocky lands decreased from 10045.42 ha to 7572.78 ha. In this study, investigation of urban development from 1986 to 2041 shows that land use change will be increased in proportion with population growth.
Results of the study showed that in the city of Mahabad, land use changes doubled with the twofold growth of the population in the mentioned period. Development in the third period was lower than twofold. Generally, changing land use could be predicted for future years through using different models of forecasting and through using geographical information system techniques and remote sensing as well, and allowing controlling and management of urban development trends, considering ecological and risk factors, and prohibit from entering residential areas into risky areas. These bestow help and information to urban managers to solve the problems. By introducing solutions urban development can be managed and controlled in a desired way. Considering urban development and reducing arable lands by making urban belt prohibits urban development in agricultural lands and prohibits approaching to natural risks areas, stabilizes land use in a programmed form and prevents hasty urban development. On the other hand, for identifying areas prone to urban development, and for preventing unlawful building by making barrier in the way of rushed urban development,we should establish an exact plan from an economic point of view.

Slope instabilities are considered as one of the major dangers to human activities which occurred in natural slopes and the slopes made by humans. This study aimed to identify the factors affecting the occurrence of slope instability using logistic regression models and artificial neural network in Ahar Sattar khan dam basin. The results of statistical models to determine the potential areas of instability and ultimately create a hazard zonation map for the study area. In this regard, the most important factors in landslides such as slope, aspect, elevation, rainfall, distance from the road, fault and drainage, land use and exploration and peculiarities of each of them were identified. Standardization base of histogram model is by cutting the classes of each layer with occurred landslides. The models show that the very high-risk area in the neural network and logistic regression are 724 and 5/56 per cent respectively which cover the areas close to Sattar khan dam, mainly including the lithology of these areas located in the regions with lower resistance. Besides, Statistical methods Logistic prove that faults and lithology have an immense impact on the occurrence of landslide in this area. The ROC index value for neural network and logistic regression models are 0/85 and 0/81. So it can be said neural network model for zoning of landslides is more efficient, so any planning and construction must be compatible with the conditions of geomorphology and geology of the area leading to as least human and financial losses as possible.

Mass movements are among the most destructive natural disasters in mountainous areas which cause great damages of over millions of dollars to human being's life and infrastructures across the world. Landslide, as a type of mass movement, is a complex process that occurs under the influence of internal and external parameters. The main effective parameters in hillside instability are earthquakes, precipitation and human activities. Landslide risk assessments estimate the probability of their occurrence in a place with a return period. Most of the methods based on educational samples and assignment of a weight to each of the parameters and sub-parameters establish the relationship between effective factors in landslides and spatial analyses. Among these methods are logistic regression, multivariate statistical analysis, artificial neural network, fuzzy model, and neural fuzzy model. In this study, for optimal management in Goijabel basin, we used the logistic regression method, as a statistical method which creates a mutual relationship between the dependent parameter and independent parameters, as well as artificial neural networks with perceptron multilayer algorithm, which allocates a weight to each of the factors in landslide using educational-testing and training data based on non-linear functions.
Study area: The study area is located 10 km from the south west of Ahar, in East Azerbaijan Province. The application site lies between the latitudes 38° 21′ 42″ N to 38° 27′ 39″ N, and the longitudes 46° 47′ 21″ E to 46° 56′ 53″ E, and covers an area of 74.62km2.
Material and To implement the methods used in this study, firstly, the maps of effective factors in landslide occurrence were provided and extracted, including lithology static factors, distance from fault, elevation, slope and aspect and dynamic factors of distance from road, distance from river, land use and land cover.
Fuzzification of effective factors, logistic regression, parameters such as Chi Square, Pseudo R Squared and ROC are used for validation of the logistic regression model. The other method used in this research is perceptron neural network. Accordingly, effective factors in landslide occurrence in the area under study were prepared as the model's input. In the next step, each extracted layer was sectioned with the landslides layers, and based on the histogram and the area of landslides in each of the layers, a reclassification was done in all of the layers.
Validation of the models used was by ROC method. The results of this model are based on statistical classification analyses. This method indicates higher area under the curve in the neural network model. In other words, the neural network model with the numerical average of 0.91% is introduced as a more efficient model compared to the logistic regression model with the numerical average of 0.89%. Using educational data and hidden nodes and other indexes such as coefficient of acceleration, this model can provide better modeling of the areas sensitive to landslides occurrence.
In this study, for an optimized management in the region and avoidance of potential damages caused by mass movements, especially landslides, the zoning of the areas sensitive to these movements was performed using logistic regression models and artificial neural network with perceptron multilayer algorithm. For the modeling, 9 independent parameters were used including precipitation, Lithology, land use and land cover, elevation, slope, aspect, distance from drainage network, distance from fault and distance from road. After standardization of each of the parameters, 9 factors were introduced as the independent variable, the landslides as binary layer for logic model, and standardized factors as input neurons and landslide layer as training of artificial neural network model with perceptron multilayer algorithm. The result of ROC validation shows that the area under the curve in the artificial neural network model has been higher than that of the logistic regression model, indicating a precision of 0.91% compared to logistic regression with a precision of 0.89%. Therefore, an area of over 9% is in high and very high -risk zones and 9/5% of the area is in medium-risk zones. Medium-risk zones can be affected by mismanagement and human constructions such as roads and be changed into high and very high risk zones, considering the fact that in this modeling, the areas close to man-made effects in both models are zoned as areas sensitive to landslide risk.

Population growth and urban expansion in the last decade in Maraghe city, changed the land cover / land use in the surrounding area. As an area of city has been reached of 350 hectares in 1353 to about 1,799 hectares in 1390. Understanding the factors affecting urban growth modeling and their impact on urban management and planning are essential to professionalism. Factors influencing can be divided into two categories of the biophysical and socio-economic variables. According to the logistic regression model capabilities to identify the impact and amount of effects of socioeconomic and biophysical factors in the growth of the city, from the model has been used for analyzing of driving forces and predicting of future urban growth at 22 year. The Pseudo R-square and the ROC for this model were obtained 0.2764 and 0.8537 due to are in the acceptable ranges. For the calibration of modeling have been used from ROC method. Namely, to the number of independent variable, modeling performed without of a variable in every of modeling and the ROC rates exploited. The result indicates that variable of distance from urban lands has the most effects in urban growth and subsequent variables are the elevation, distance from industrial land-uses, distance from agricultural lands and distance from roads.

The population growth, industrial development, bio-climate changes and scarcity of land resources are the main reasons and causes of forest degradation in developing countries. To control and decrease forest degradation, the governments need to know where, when, how fast, and why (with what causes) such degradations happen. On the basis of such knowledge, a general and sustainable management of these resources will be possible. The science and technologies of GIS and remote sensing could be a perfect tool for answering the above questions. Remote sensing can be the basis of fast and inexpensive data collection and the analytical capabilities of a GIS can be used for analyzing the types, location and rates of changes. In this research, the Landsat TM and ETM+ images of years 1987 and 2001 are used for land use classification and analysis of changes at the forest area of Arasbaran in north-west of Iran. The classification is mainly aimed at the separation of forest from non-forest areas. A few methods have been studied to calculate and show the occurred changes. These include methods that only describe the change areas (such as subtraction and division methods) and those that describe the area, amount and type of the changes (such as comparison after classification). By classifying the forest and non-forest areas of years 1987 and 2001 and overlaying them, a map was extracted representing the stable forest area and deforested area. From the topographic data of the study area, some other raster maps were created showing elevation, slope, aspect and distance from population areas. Information of these maps were entered into a statistical model (a logistic regression model) having the above-mentioned classified map as the dependent parameter and all other maps as the independent parameters. It was resulted that the parameters of distance from populated areas, elevation and aspect have a meaningful relation with the deforestation phenomenon. From such an analysis, the importance of each factor in the phenomenon was defined and the areas that are in higher risk of deforestation and need an urgent protection were defined.