عطاالله شیرزادی

Water infiltration into soil profile is one of the most important components of the water cycle, playing an essential role in determining the hydrological behavior of soil, the amount of runoff generation in a watershed, and the amount of soil moisture for the growth of vegetation and crops. Its variability is influenced by many factors such as precipitation characteristics, soil, vegetation type, land use, land slope and season of the year, which must be determined in each region because these characteristics are highly site-specific. Due to the lack of information on the variability of water infiltration into a soil on different parts of a slope, the aim of this research is to investigate the variability of infiltration in different parts of a slope in different seasons of the year.

In this research, a slope with a length of approximately 60 meters was selected inside the University of Kurdistan campus. Three points on this slope, including the upslope point with 0% slope, the middle slope point with 22% slope, and the downslope point with 28% slope, were considered for the establishment of double rings. The soil texture was clay and clay loam and the type of vegetation was sparse raneglands. Infiltration measurement at these three points was repeated three times in the three seasons of Fall, Winter and Spring with time intervals of one month. During the experiment, the initial soil moisture and the saturated moisture contents of the soil were measured by the gravimetric method. The collected data were analyzed in a randomized complete block design.

The results showed that different slopes had no effect on the intensity of initial infiltration, the average intensity of infiltration and the intensity of final infiltration. The results also demonstrated that the initial ifiltration rate and the average infiltration rate in Fall were significantly different from those of Winter and Spring, while the final infiltration rate was the same in all seasons and its value was not affected by seasonal changes. Further investigations indicated that there is no significant relationship between initial infiltration intensity, average infiltration intensity and final infiltration intensity at three points of the slope and in three seasons of the year with initial soil moisture and soil saturation, which implies that the difference of infiltration intensities in different seasons cannot be justified with changes in initial moisture and soil saturation.

It can be suggested that infiltration at any point is completely dependent on the characteristics of that point and it is not possible to generalize a specific behavior pattern of this component to other points. Therefore, it is necessary to measure the infiltration at each point and even during modeling, the natural characteristics of that point must be fully understood.

Although a flood is an extreme and exceptional flow, every exceptional flow will not turn into a destructive flood, different factors must be changed in nature to cause destruction, damage, and casualties. In general, floods can be divided into four groups flash floods, river floods, urban floods, and coastal floods. Urban floods usually cause fewer casualties and mainly create damage caused by flooding, disruption of traffic, interruption in socio-economic activities, and problems of this kind. The damage caused by non-urban floods is often heavy and sometimes accompanied by high and catastrophic casualties. According to the Mediterranean climate, Iran is the seventh country in the world in terms of flooding. The flood-prone areas of the country are estimated to be around 91 MHA. In other words, 55% of the country's surface has contributed to the production of surface runoff, of which about 42 MHA have moderate to very high flood intensity. The review of sources shows the development of knowledge-based methods, statistical methods, and artificial intelligence algorithms in predicting flood-prone areas in urban and non-urban watersheds in different regions worldwide. However, a hybrid method of the Fuzzy, Delphi, and Analytic Hierarchy Processes (FDAHP) in urban flood susceptibility has not been used. Regarding the questions, what are the most important factors in urban flood occurrence? Is it possible to determine flood-prone areas in urban areas using the FDAHP hybrid model?, this study aims to identify the factors influencing the occurrence of floods and predict flood-prone areas in Sanandaj City.

Materials and Methods  :

In this study, which has a descriptive-analytical-comparative approach, to predict floods in Sanandaj City, the FDAHP was used. First, each of the conditioning factors (14 factors) was scored by flood experts and completed using the scores obtained from other stages of the FDAHP model. After collecting the opinions, the relative weights of the indicators were determined using the hybrid model, and finally, the flood susceptibility map of Sanandaj City was prepared using ArcGIS 10.5 software. The different stages of modeling with the FDAHP method are as follows: 1) experts' opinions, First, with the help of experts (technical and executive experts of the Kurdistan Province and Sanandaj Municipality's Natural Resources and Watershed Administration) the decision-making parameters according to their importance qualitatively or, if possible, quantitatively they rate. (Opinion scales are: very important with a nine score, importance with a seven score, average importance with a five score, low importance with a three score, and no importance with a one score). 2) Calculation of fuzzy numbers. After the preliminary stage, which includes a survey of experts in the form of a qualitative or quantitative questionnaire, fuzzy numbers were calculated based on the results of this survey. 3) Forming the matrix of fuzzy pairwise comparisons. 4) Calculating the fuzzy weight of the parameters. 5) De-fuzzification of model parameters. 6) Evaluation of the accuracy of the output of the spatial flood prediction model.

The findings showed that slopes of less than 10% (flat), areas with an elevation of fewer than 1400 m above sea level, slope aspect and curvature (flat), urban land use with high building density, rainfall of more than 369 mm and rock type Qt2 (Quaternary alluvium) have the highest susceptibility to floods compared to other types in Sanandaj City. Also, the results show that, on the one hand, increasing the distance from the residential areas, the distance from the roads, and the distance from the waterways decreases the susceptibility to urban flooding. On the other hand, with the increase in building density, the density of roads and the density of waterways in Sanandaj increases the susceptibility to flooding. These results are consistent with the other studies that concluded the residential areas with the largest area have the highest risk of vulnerability. Our findings indicated that the density of the waterway, the slope, and the distance from the waterway have the most influence on the occurrence of floods in Sanandaj City. Rainfall, road density, building density, distance from residential areas, distance from roads, flow accumulation, elevation, land use, lithology, and slope curvature are the next priorities in terms of importance in the occurrence of floods in Sanandaj City.

The flood prediction map showed that a large part of Sanandaj City including the City's northern, western, southern, and center, which is crowded, has more potential for urban flooding. For example, the old and dilapidated buildings of the City center are highly susceptible to floods, but the border areas around the City have less exposure to this phenomenon. Therefore, as the distance from the City center and the residential regions increases, the potential for flooding decreases. Based on the value of 80.56% of the area under the curve, the validation results indicated that 80.56% of the areas where urban flooding is visible have been correctly predicted. The FDAHP hybrid model had a high ability to estimate the areas prone to urban flooding and, therefore it can be tested and evaluated as a management tool to identify urban floods in other similar areas. Since our aim concerning the occurrence of floods is more on flood mitigation, according to the mentioned theoretical bases and the existing views in the field of flood prediction maps, it can be stated that the ruined canals and the network of waterways, unauthorized constructions, high density, and elevation of buildings, old and dilapidated buildings and the irregularity of the canals have aggravated the urban flooding. Overall, it can be said that obtaining an accurate and reasonable urban flood prediction map can help City managers and planners identify flood-prone areas to manage the urban flood crisis.

Soil erosion by water is one of the most important processes of land degradation, especially in semi-arid regions. Among the different types of water erosion, gully erosion is one of the most important events affecting soil destruction, changing the landscape and water resources, and land regression [1]. Gully erosion is the most obvious form of soil erosion, which leads to a decrease in soil production capacity and restrictions on land use, and can be a serious danger to roads, fences, and various structures, and also causes significant soil losses and the production of large amounts of sediment [2]. This erosion is also called gully erosion. A gully is a relatively permanent waterway that temporary streams of water pass through during rainfall and carry a large amount of sediment [11]. The formation of gullies is always accompanied by erosion and changes in the appearance of the land and causes the production of a significant amount of sediment, destruction of lands, roads, irrigation networks and filling of dams [9]. Gullies, which are considered major indicators of environmental changes in most cases, are not considered normal forms of erosion due to their rapid growth [8]. In the studies conducted both inside the country and abroad, various methods have been used to evaluate the potential of gully erosion, which are mentioned below. Among the methods used to determine the potential of gully erosion are regression models [17, 10, 4, 15, 23, 22], knowledge-based model of hierarchical analysis [6, 21, 22, 3, 29], fuzzy logic. [12 and 13], Dempster-Shafer model [14], artificial neural network and support vector machine [33], etc. pointed out that, based on this, the watershed of Klocheh Bijar in Kurdistan province has been severely affected by this type of erosion and caused the loss of many agricultural lands in the studied basin have been eroded. Therefore, the gully erosion susceptibility mapping of the studied basin was studied and investigated using Logistic Regression and Fuzzy Logic models, and finally mapping of gully erosion in the studied area and the validity of both used models were verified.

Gully erosion inventory map In this study, 950 points at the top of gullies (head cuts) were recorded as a distribution map of gully erosion using Google Earth images and field survey. Then, they were divided into two parts of training data (70%) and validation (30 percent) were divided. The training data were used in the model learning section with the logistic regression method and the validation data were used to determine the validity/prediction power of the models. Logistic regression model Multiple logistic regression is a multivariate technique that considers several physical parameters that may affect the probability. In this method, the values of the independent variable can be expressed in binary form (0 and 1) and as a numerical quantity. Fuzzy logic model In classical set theory, an element is either a member of the set or not (zero and one). Fuzzy set theory extends this concept and introduces graded membership. So that an element can be a member of a set to some degree and not completely. In other words, a fuzzy set is a set of elements with similar characteristics, where the set has a certain degree from zero to one. Zero means no membership and one mean full membership [24]; Therefore, before implementing the fuzzy model, it is necessary to determine the membership functions for each of the layers mentioned above and set the value of the layers in a range between (zero and one) and then enter the layers into the fuzzy model. To implement the fuzzy technique, a gamma operator is needed, the value of the modulating gamma is between zero and one, zero gamma is equivalent to fuzzy multiplication and one gamma is equivalent to fuzzy addition.

Gully erosion susceptibility map logistic regression Table. 1 shows TOL and VIF values of factors affecting the occurrence of gully erosion. Viewing this table shows that all effective factors have a TOL value greater than 0.1 and a VIF less than 10, which indicates the absence of multiple collinearities between them, and all of them are used as input for the model. They were branched with appropriate logistic regression. Fuzzy Logic Figure 5 shows the prediction map of gully erosion using fuzzy logic. Although the pattern of distribution of different areas from the map obtained with the fuzzy logic model follows the logistic regression model, but it seems that an exponential exaggeration in the area of the area with the probability of gully erosion is very high is seen, which is probably related to choose the numbers for the gamma value during modeling with this method. However, the areas with high and very high probability of occurrence correspond to the top of gullies (head cuts).

Gully erosion is one of the most important natural hazards and the main cause of land degradation in the Klocheh Bijar watershed in Kurdistan province. Recognizing the most important factors affecting the occurrence of this phenomenon as well as predicting areas prone to its occurrence is one of the management and preventive measures to better understand the area before any engineering/structural and biological measures (land management) or a combination of both. Reduction of possible damages. Preparing a gully erosion susceptibility map can be a useful guide for planners, managers, organizations and decision makers regarding the management of these areas. In this research, 950 points at the top of gullies (head cuts) were recorded as a distribution map of gully erosion using Google Earth images and field survey and were divided in the ratio of 70 to 30. Multilinear correlation test was used to check the internal correlation of 20 effective factors, as well as two models of logistic regression and fuzzy logic were used to prepare prediction maps of gully erosion in the study area.

Iran's mostly mountainous topography, tectonic activity, high seismicity, diverse geological and climatic conditions, population growth-induced pressures on natural resources, and land-use changes in recent decades have created natural conditions for a wide range of landslides. Therefore, it is necessary to take appropriate measures to reduce the damage caused by landslides, identify their prone areas, determine the factors affecting them, and prepare their susceptibility maps. Thus, this study sought to identify the most important factors involved in the occurrence of landslides, and investigate the efficiency of the alternating decision tree models for preparing landslide susceptibility maps in the southwestern part of Kurdistan province (the communication route connecting Yozider to Degaga).

First, a distribution map with 175 landslides and 100 non-landslide locations was identified and classified into a ratio of 80% and 20% for training and model validation, respectively. Thirteen factors derived from topographic, land cover and rainfall data were selected for modeling using the Information Gain Ratio (IGR) technique. Then, the ADT algorithm was used to train and prepare the landslide susceptibility maps. Moreover, statistical criteria were used to evaluate the models for both training and validation datasets. Finally, the model's performance was evaluated in terms of the area under the receiver operating curve (AUC).

The highest IGR index values were found to belong to the distance from road, lithology, and road density, respectively. Furthermore, factors such as SPI, curvature, profile curvature, plan curvature, river density, distance from the river, and LS proved to have a negative effect on the modeling results, as zero value was assigned to these indices, which, in turn, led to the creation of noise. On the other hand, final modeling was performed and the 13 remaining factors were removed from the model due to their low impact on shallow landslides in the study area. Then, shallow landslide susceptibility maps were prepared based on the ADT algorithm, using quantile, natural breaks, and geometrical interval methods in the ArcGIS 10.2 environment. Finally, the best method was selected based on the landslides' frequency histogram in each susceptibility class of the maps. The results indicated that the natural breaks method was the best-known method, according to which the landslide susceptibility maps were divided into five classes, including very low susceptibility (VLS), low susceptibility (LS), moderate susceptibility (MS), high susceptibility (HS), and very high susceptibility (VHS).

The results of IGR-based factor analysis revealed that distance from roads, lithology, and road density had the highest effects on the occurrence of landslides, which could be attributed to the existence of landslide-susceptible formations such as marl and shale, and inappropriate human-set policies including road construction and incorrect cutting off of heels, as road construction provides the grounds for more penetration of water into sensitive soil formations and saturation of these soils under force. Moreover, the soil saturation on slopes under the force of gravity can facilitate the occurrence of landslides in the study area, which is consistent with the results found by Pham et al. (2015, 2016, 2019) who reported that the existence of lithological units susceptible to landslides (i.e., marl and shale) in the middle parts of the slopes played an important role in the occurrence of landslides in such areas, considering the fact that marl and shales layers might act as a lubricant for overlying saturated soils, and, therefore,  facilitate the occurrence of landslides. However, it could be argued that the upper floors and higher altitudes of the area are less susceptible to landslides due to the presence of crystalline and basaltic units that are resistant to any mass movement, especially landslides. As for the land vegetation, it was found that landslides mostly occurred in drylands that were formerly semi-dense and grassland forests, indicating the significant role of forest degradation and land-use change in landslide occurrence.
The designed models for both the training and validation data were evaluated using the Kappa, TP, specificity, sensitivity, accuracy, and squared statistical measures. Finally, the models' performance was examined through the AUC. Accordingly, the results of model validation showed that the ADT model had a relatively moderate performance with the sub-curve level of 0.665. Furthermore, the sub-curve level of the ADT model was found to be 0.677 for the training data. Finally, the study area was divided into five classes, including the very high, high and moderate, low, and very low sensitivity. It was also found the severity of landslides increased when moving from low-sensitivity classes to the high-sensitivity ones, indicating higher chances for the occurrence of landslides in areas with high sensitivity. Therefore, considering the high influence of the roads in the model proposed in this study, it is suggested that the priority of taking appropriate measures to prevent and/or control landslides be taken into account so that the effect of road construction in the study area could be reduced. Moreover, in cases where future road development operations are considered along the route, principles of road construction and the stability of the slope must be strictly observed.

Landslides are considered one of the most destructive natural phenomena. Landslides are dangerous natural hazards. Because of their threat, a comprehensive landslide susceptibility map should be produced to reduce the possible damages to people and infrastructure. The quality of landslide susceptibility maps is influenced by many factors, such as the quality of input data and the selection of mathematical models. The main purpose of this study is to presentation, a novel hybrid model namely Rotation Forest based Functional Trees (RFFT), which is a hybrid intelligent approach of two state of the art machine learning techniques of Functional Trees (FT) classifier and Rotation Forest (RF) ensemble, for landslide susceptibility Assessment prediction in Kamyaran city located in Kurdistan Province, Iran. At first, twenty-one factors affecting the occurrence of landslide in the study area including Slope angle, Aspect, Elevation, Curvature, Plan curvature, Profile curvature, Radiation, Valle depth(VD), stream power index (SPI), topographic wetness index (TWI), combination of length-angle of slope (LS), Land use, NDVI (normalized vegetation index), Distance to Faults, Faults density, Distance to Road, Road density, Distance to River, River density Lithology and Rainfall with total of 60 landslide locations have been collected for generating training and testing datasets. Then, based on the Information Gain Ratio Index, eight effective factors were chosen and used for modeling. Performance of the proposed RFFT model was evaluated using some statistical-based measures such as sensitivity, specificity, accuracy, RMSE and area under the ROC curve (AUROC). The results showed that the proposed model performed well in this study (AUC = 0.891), and it improved significantly the performance of the FT base classifier (AUC = 0.819). Therefore, it can be concluded that the proposed RFFT model should be used as a great alternative method for better landslide susceptibility assessment in landslide prone area.Landslides are considered one of the most destructive natural phenomena. Landslides are dangerous natural hazards. Because of their threat, a comprehensive landslide susceptibility map should be produced to reduce the possible damages to people and infrastructure. The quality of landslide susceptibility maps is influenced by many factors, such as the quality of input data and the selection of mathematical models. The main purpose of this study is to presentation, a novel hybrid model namely Rotation Forest based Functional Trees (RFFT), which is a hybrid intelligent approach of two state of the art machine learning techniques of Functional Trees (FT) classifier and Rotation Forest (RF) ensemble, for landslide susceptibility Assessment prediction in Kamyaran city located in Kurdistan Province, Iran. At first, twenty-one factors affecting the occurrence of landslide in the study area including Slope angle, Aspect, Elevation, Curvature, Plan curvature, Profile curvature, Radiation, Valle depth(VD), stream power index (SPI), topographic wetness index (TWI), combination of length-angle of slope (LS), Land use, NDVI (normalized vegetation index), Distance to Faults, Faults density, Distance to Road, Road density, Distance to River, River density Lithology and Rainfall with total of 60 landslide locations have been collected for generating training and testing datasets. Then, based on the Information Gain Ratio Index, eight effective factors were chosen and used for modeling. Performance of the proposed RFFT model was evaluated using some statistical-based measures such as sensitivity, specificity, accuracy, RMSE and area under the ROC curve (AUROC). The results showed that the proposed model performed well in this study (AUC = 0.891), and it improved significantly the performance of the FT base classifier (AUC = 0.819). Therefore, it can be concluded that the proposed RFFT model should be used as a great alternative method for better landslide susceptibility assessment in landslide prone area.Landslides are considered one of the most destructive natural phenomena. Landslides are dangerous natural hazards. Because of their threat, a comprehensive landslide susceptibility map should be produced to reduce the possible damages to people and infrastructure. The quality of landslide susceptibility maps is influenced by many factors, such as the quality of input data and the selection of mathematical models. The main purpose of this study is to presentation, a novel hybrid model namely Rotation Forest based Functional Trees (RFFT), which is a hybrid intelligent approach of two state of the art machine learning techniques of Functional Trees (FT) classifier and Rotation Forest (RF) ensemble, for landslide susceptibility Assessment prediction in Kamyaran city located in Kurdistan Province, Iran. At first, twenty-one factors affecting the occurrence of landslide in the study area including Slope angle, Aspect, Elevation, Curvature, Plan curvature, Profile curvature, Radiation, Valle depth(VD), stream power index (SPI), topographic wetness index (TWI), combination of length-angle of slope (LS), Land use, NDVI (normalized vegetation index), Distance to Faults, Faults density, Distance to Road, Road density, Distance to River, River density Lithology and Rainfall with total of 60 landslide locations have been collected for generating training and testing datasets. Then, based on the Information Gain Ratio Index, eight effective factors were chosen and used for modeling. Performance of the proposed RFFT model was evaluated using some statistical-based measures such as sensitivity, specificity, accuracy, RMSE and area under the ROC curve (AUROC). The results showed that the proposed model performed well in this study (AUC = 0.891), and it improved significantly the performance of the FT base classifier (AUC = 0.819). Therefore, it can be concluded that the proposed RFFT model should be used as a great alternative method for better landslide susceptibility assessment in landslide prone area.

The flood is the second natural disaster in terms of the damage that it causes to human societies every year. At the same time, cities and population centers are exposed to the highest risk and potential for physical damage caused by flood events. The growing trend of floods in Iran in recent years indicates that most parts of the country are at risk. According to the conducted studies, about forty large and small floods occur annually in different parts of the country. Since the flood is the most devastating disaster in the world and a serious threat to life, preparation of flood hazard maps is essential in identification of flood-sensitive areas, and is one of the first steps taken to reduce damage. It should be noted that researchers can easily identify areas with high risk using flood hazard maps to prevent damage. In the modeling of research on natural and environmental hazards such as floods, the complexity of natural systems makes it difficult to use physical models, highlighting the use of hybrid models as a suitable alternative. Therefore, it is necessary to provide appropriate methods and suggestions for estimation of runoff and flood in areas with high risk in order to prevent their occurrence. In recent years, a large number of statistical and probabilistic models have been used for flood hazard mapping, and GIS has been used as a basic analysis tool for spatial management and data manipulation due to its capability of managing large amounts of spatial data. Furthermore, it is possible using GIS prediction models to partition urban areas in terms of flood hazard. The obtained hazard maps can be used for identification of areas with high flood hazard. Today, the accuracy of flood susceptibility maps in urban areas can be increased using hybrid GIS models rather than single ones. Therefore, the purpose of this study was to predict the risk of flood in the city of Sanandaj, Kurdistan Province, Iran using hybrid models in the GIS environment. The factors considered in this study as influencing the occurrence of flood included inclination, slope, elevation, distance from the river, river density, flow accumulation, land use, gradient curvature, lithology, distance from the passage, road density, distance from the building, building density, and rainfall. After collecting the required data and layers, we used two algorithms, including the definitive statistical index (SI) and the evidential belief function (EBF), both in isolation and in combination. The locations of flood susceptibility in Sanandaj were predicted. In order to assess the performance of the model, the AUC curve obtained from the ROC curve was used. According to the adopted assessment criteria (ROC) and the validation data, the EBF model (0.840) exhibited better performance than the SI model (0.827) in flood hazard mapping in the area under investigation. The accuracy rate of the hybrid SI-EBF model was 0.849 based on the ROC results, which demonstrated that the hybrid model performed better in prediction of the spatial hazard of flood than the single models. Finally, the results of the study showed that Sanandaj flood spurts resulted from various environmental and human factors, which can be handled using flood hazard maps and appropriate management measures to reduce flood damage.

  Geomorphological hazards such as mass movements are one of the potentially harmful phenomena. Landslides as one of the most traditional earth movements involved all slope failures in which a mass of materials (soil and rock) moves down the slope due to decreasing of safety factor and overcoming of the destructive forces on the resistance forces on a slope.
Kurdistan province has been frequently exposed to mass movement hazards owing to its characteristics of topography, geomorphology, pedology, geology and climatology. Therefore, the objective of this study is 1) to prepare the spatial prediction map of mass movements for the Kurdistan province to manage the areas prone to the hazards and to use the map in the land use planning projects and development of the rural regions 2) to introduce the Bayesian logistic regression (BLR) ensemble and to compare its efficiency with the logistic regression (LR), and 3) to identify the most significant conditioning factors on mass movements occurrence in the Kurdistan province. Based on the literature review, 18 factors affecting landslide susceptibility were selected for modeling including slope degree, slope aspect, elevation above sea, curvature, profile curvature, plan curvature, stream power index, topographic wetness index, length-angle of slope, lithology, rainfall, land use, distance to fault, fault density, distance to stream, stream density, distance to road, and road density. A total of 895 mass movements in the Kurdistan province were divided into 70 % (626 locations) for modelling using training dataset and 30% (269 locations) to evaluate based on the validation dataset. Additionally, 626 locations were randomly selected as areas where mass movements have not occurred and they were classified into 70% and 30% for training and validation in modelling process. Based on the Information Gain Ratio index to modelling process, 7 factors were then applied to the LR method and 11 factors were used in the BLR method. The evaluation of models were performed using Specificity, Sensitivity, Accuracy, the area under the Receiver Operating Characteristic Curve (AUROC), and the Root Mean Square of Errors (RMSE) indices.
Result and According to the results of spatial prediction of mass movements by quantile approach, about 8.06% of the Kurdistan province area is very sensitive to mass movement occurrence using the LR model; while, the BLR model shows a sensitive area of about 12.46%.
The majority of very sensitive and sensitive areas are located to the west of the province. These regions include Oramanat mountainous areas, Kosalan and Chelchama highlands, the Salvatabad Saddle (East of Sanandaj), the Morvarid Saddle (between Sanandaj and Kamyaran), The Khan Saddle (between Saghez and Baneh), and the Arez Saddle (between Sanandaj and Marivan).
In order to assess the spatial map accuracy of mass movement sensitive areas prediction, both training and validating datasets were used. The results showed that the area under the curve of SRC is 0.714 in LR based on training data, indicating that this approach has a potential of 71.4 percent to predict sensitive areas to mass movement. BLR showed a value of 0.672 using the same data which implies a potential of 67.2% for prediction of sensitive areas.
Based on the validating dataset, the area under the curve of PRC was 0.732 and 0.729 for LR and BLR, respectively. These results confirmed the accuracy of the maps prepared by both models; however, it should be noted that LR was slightly providing better results.
To analyze the probability of mass movement occurrence, the potential of the LR and BLR models were evaluated using Friedman test at 5% significant level. The test revealed that there is no significant difference between these two models and both are reasonably able to evaluate spatial prediction of mass movements. The results showed that there is no significant difference (95% level of significance) between the efficiency and the susceptibility maps prepared by the two models for spatial predicting of landslide in the study area; therefore, the BLR model can be applied as an index model for the study area and other similar areas.
Therefore the hybrid model of Bayesian Logistic Regression has a high capability of identifying sensitive areas to mass movement such that it can be compared with previously successfully testes methods such as artificial neural network, fuzzy logic, decision tree algorithms (Naïve, Random Forest, Baes Forest, …)

One of the slopping processes which created much damage in many locations of Iran and the world is Landslide phenomenon. Identification of susceptible areas to landslide occurrence is one of the basic measures for reduction of the possible risk and management. The main goal of this research is to evaluate Landslide Susceptibility assessment using Support vector machine algorithm. At first, a landslide inventory map with 60 landslide locations for the study area was drawn from various sources. Landslide locations were then spatially randomly split in a ratio of 70/30 for building landslide model and for the model validation.Training and testing of RBF Function the SVM algorithm was evaluated over an assembly of spatial attributes, which included slope angle, elevation, aspect, solar radiation, profile curvature, plan curvature, lithology, land use, distance to fault, distance to road and distance to river with respect of the referent model. Finally the study area was classified into five sensitivity classes’ very high, high, moderate, low and very low. Then Performance of the method has been evaluated using the ROC curve. The results show that area under the ROC curve (AUC) using training dataset is (0/950) and using validation dataset is (0/931). Therefore, analysis and comparison of the results show that RBF Function SVM model performed well for landslide susceptibility assessment in the study area and the results from this the results from this study can be useful for land use planning, mitigate landslide hazards and decision making in landslide prone areas.

Landslides and instability slopes are major dangers for human activities which often cause the waste of economic resources and damage to properties and installations. These occur in the natural slopes or in the changed slopes by human. The main objectives of this study are identifying the effective factors on landslides occurrence in Kurdistan Province, Bijar and evaluating the regions prone to landslide to prepare the susceptibility map using the logistic regression. At first, in this study, by using field visits, questionnaires, geological and topographic maps and reviewing the studies, ten effective factors including elevation from sea level, slope degree, slope aspect, geology, distance from the linear elements (fault, road and river), rainfall and land use were employed. After identifying the factors, they were processed using ARC GIS 10 and ILWIS 33 software. Dependent variable is 144 slopes prone to landslide selected across the region as the landslide data (code 1) and 144 slopes stable against landslide were randomly as land slide free data (code 0). With overlay these data on each of the independent variables, the data necessaries were collected for entry into SPSS 18. The results showed that "slope degree" has the most significant role on landslides. Then, land-use, slope aspect, fault, distance from the drainage network, elevation from sea level, distance from road and litho logy are next effective factors, respectively. The results of the evaluation showed that logistic regression model with PCPT index equal to 83.4; -2LL index equal to 229.226 and ROC index equal to 98.5 percent and landslide susceptibility map based on SCAI index has high verification in the case study. Therefore, 75.489 % of the area has very low susceptibility, 10.037% with low susceptibility, 3.628% with moderate susceptibility, 4.062% with high susceptibility and 6.784% with very high susceptibility. These results can be used in predicting the occurrence of future landslides, decreasing their risks and planning for the land use.

Estimation of flood hydrograph is of necessities in hydrological studies such as flood mitigation projects. This estimation in un-gauged watersheds is usually taken place using geomorphological characteristics of watersheds. The objective of this research is to estimate synthetic unit hydrograph using regional flood frequency analysis and geomorphological parameters of watersheds. 1-hour and 2-hour hydrographs of two watersheds, Kanisavaran and Maranj Watersheds, were generated using maximum discharge data based on regional flood frequency analysis. Estimated hydrographs were compared with observed data and the efficiency of the model was evaluated using Nash-Sutcliffe coefficient, absolute and bias errors. The results showed that multiple regression models give more acceptable results among others for the computation of synthetic unit hydrograph (higher coefficient of determination). The Nash-Sutcliffe coefficient was 0.98 for 1-hour hydrograph while it was 0.93 for the 2-hour hydrograph. The absolute error in 1-hour hydrograph and 2-hour hydrograph was 0.13 and 1.2, respectively. The bias error was close to zero for both hydrographs, indicating that the proposed model is efficient. The model may be used for estimation of synthetic unit hydrograph in similar un-gauged watersheds.

The aim of this research is generating rockfall hazard zonation map by using of overlay index method in Salavatabad saddle that is located in Sanandaj southeastern، Kurdistan province. The first، in this research، we using of field search، local interview and review on carried out studied in the outside and inside of Iran، eight factors effective was recognized on rock falls. These factors include; slope degree، slope aspect، slope curvature، elevation، distance to road، distance to fault، litology and land use. After recognizing، we act to processing with Arc view 3. 2 and ERDAS imagine 8. 5 software. Thirty four Sensitive slopes were used as prone slopes to rock fall data (Dependent variable). For statistical analysis، be used of likelihood ratio statistical analysis. At this research، the value of each various class between 0 and 1 is determined. As for، the chart be obtained from overlay distribution rockfall map with information layers، the each of class of each layer that had more frequent density، 10 value and other classes with aspect of value of rockfall frequent density in class itself، was determined as unit of 100 value. With transmit this coefficient to each class and wights based on the charts as effective factors with rockfalls and expert opinions، the hazard zonation susceptibility of rockfall was generated in Salvatabad saddle regional with five classes. For Verification، rockfall point’s map was overlaid with hazard map. For more confidence، first each of effective variables removed then hazard Zonation map was generated and at last this map was verificated. Result have shown that percent of verification without slope degree is 75. 81 percent، without slope aspect is 78. 50 percent، without elevation of sea is 72. 16 percent، without litology is 74. 64 percent، without distance from road is 76. 23 percent، without distance from fault is 73. 56 percent، without curvature is 64. 00 percent، without landuse is 74. 23 percent and by all variables the percent of verification is 81. 44 so، it is high and acceptable..

Climatic factors and human activities can cause mass movements. Due to the risk of these movements to human communities، one has to control them. In order to control mass movements، different aspects of the movements should be studied in detail. The aim of this research is to evaluate the efficiency of a statistical model - logistic regression - and a probabilistic model - frequency ratio - in rockfall hazard mapping in the Salavatabad saddle in eastern Sanandaj، Iran. The study of 34 sensitive slopes and 34 stable slopes in the study area using field works، local interviews and the literature review showed that 8 factors including slope gradient، slope aspect، slope curvature، elevation above the sea، and distance from road، distance from faults، lithology and landuse are the most effective factors for occurring rockfalls. The relationships of rockfall and these 8 factors were studied by the logistic regression model and the probabilistic frequency ratio model and the generated rockfall susceptibility maps were compared. The results showed that both models work reasonably well for rockfall hazard studies; however، the logistic regression model has an accuracy of 85. 09 percent of the Area under ROC Curve (AUC) in prediction of sensitive areas compared to the accuracy of 76. 74 percent by the probabilistic frequency ratio model. Therefore، the logistic regression model can be used for rockfall hazard studies in the study area.