بهناز بیگدلی

The purpose of this research is site selection of temporary accommodation centers in District 22 of Tehran with the help of effective criteria. These criteria are including Red Crescent, distance to bridge, distance to schools, distance to the main road, distance to the police station, distance to the gas source, distance to mosques, distance to food distribution centers, distance to metro lines, distance to Metro station, distance to health centers, distance to park and green space, distance to hospital, distance to fire stations, population density, earthquake risk, and slope. The novelty of this study is the use of an appropriate and sufficient combination of criteria for locating temporary accommodation sites. In this regard, the geographically weighted regression (GWR) method with two Gaussian and tri-cube kernels was used. This method is compatible with two unique properties of spatial data, namely spatial autocorrelation, and spatial non-stationarity. The values of R2 and RMSE obtained by the GWR method with tri-cube kernel were 0.9413 and 0.3470, respectively, which indicates the high compatibility of the tri-cube kernel with respect to the Gaussian kernel. The results also show that Chitgar Park and Azadi Sports Complex are some of the largest and most suitable areas for the construction of temporary shelters after the earthquake crisis in the case study.

Accurate monitoring of surface water is one of the important and necessary applications in the use of remote sensing systems. Meeting the needs raised in the use of remote sensing data collected from the earth's surface in many applications, using only one product and classification algorithm is not sufficient and possible, and for a more accurate understanding, data fusion can be a better option. In this system, various approaches such as water extraction indices or classification algorithms are used to identify water areas. In this research, an fusion approach of Landsat-8 and Sentinel-2 optical sensor images was used. Firstly, the spatial resolution of these sensors was enhanced from 30 to 10 meters by Pansharpening them and preserving spectral information. Then, water extraction indices such as NDWI, MNDWI, AWEI_sh, AWEI_nsh, and WI were applied to the integrated images. Subsequently, using classification algorithms such as SVM, Maximum Likelihood, Minimum Distance, Neural Network, and Random Forest, the study area was classified into two categories of water and non-water areas. Finally, the results obtained from all classification algorithms for pre and post-flood images of Mazandaran province in the 2019 flood event were merged using the majority voting method, which is considered an integration approach at the decision-making level. Random forest classification algorithm with overall accuracy of 97.76 and 94.12 and Kappa coefficient 94.49 and 91.41 for images before and after flood had the best classification performance among the algorithms used in this research. The fusion of classification algorithms showed an improvement in the separation performance of water and non-water areas with an increase in the overall accuracy of separation to 98.41 and 95.24 and Kappa coefficient 96.12 and 92.81 for the images before and after the flood.

Housing prices are one of the indicators that identify the factors affecting it and could help to increase the efficiency of plans and to present housing planning strategies and policies. Despite many exchange rate fluctuations in recent years, there is a need to create a model that pays attention to the economic factors affecting housing prices in addition to the ordinary housing features. Since the housing price modeling is one of the issues that has a spatial component, therefore, in presenting the model related to housing prices, its location should also be considered. Therefore, in this study, the analysis of the spatial distribution of housing prices in district 5 of Tehran municipality and the factors affecting that have been investigated. In this regard, housing sales data in this region in 2018, 2019, and 2020 have been used to model housing prices. The research results have been obtained by the Multiscale Geographically Weighted Regression (MGWR) method, which provided better results compared to those by both the Geographically Weighted Regression (GWR) and Ordinary Least Squares (OLS) methods. The adjusted coefficient of determination in OLS, GWR, and MGWR algorithms was obtained equal to 0.762, 0.821, and 0.853, respectively. The MGWR method is one of the methods that can model the spatial heterogeneity of housing price data. According to the results, the exchange rate variable (dollar price) has the greatest impact on housing price modeling.

Streamflow forecasting models play a crucial role in hydrological issues, such as the determination of reservoir inflows and flood forecasting. In this research, artificial intelligence hybrid models including ANFIS and GA-ANN have been used for short-term (daily) streamflow forecasting. This research aims to predict the outlet of the Latiyan basin, Tehran province, from 2017 to 2018. For this purpose, a snow-covered area (SCA) is obtained from the processing of Sentinel-2 optical satellite images. Then, in order to extract the effective snow, the fusion algorithm is applied for Sentinel-1 and 2 integrations. Finally, the artificial intelligence model with the help of the effective snow parameter along with other daily hydrometric and meteorological data including daily precipitation, temperature, and discharge is applied to forecast the daily outlet of the basin. Also, to improve the model performance, the seasonal index has been used to identify streamflow trends and better model training. The results showed that the prediction model using satellite data has improved its performance by 37%, which shows the direct effect of the snow parameter on the basin runoff. In addition, the trend of changes in the effective snow parameter has a favorable agreement with the flow trend of the basin, especially in the peak flows. Also, using seasonal information as an input parameter can improve the results of the prediction models by approximately 22%. In addition, the AI method based on fuzzy inference (ANFIS) showed better performance than the developed neural network method (GA-ANN) based on statistical indices.

Making a balance between the demand for healthcare services and the response to it requires the operation of new hospitals. But the important problem is finding the optimal for the construction of a hospital. The present research used a mixed method for determining the optimal location for a hospital in District 2 of Tehran. The mixed method is based on the Geospatial Information System (GIS), Best-Worst Method (BWM), and Multi-Criteria Decision-Making Methods, WASPAS, and TOPSIS. Fewer pairwise comparisons in the weighting process of the proposed method have increased the accuracy and reliability of decision-making results. The combination of spatial modeling and ranking has also reduced the search space for suitable places to build a hospital. For this purpose, after determining the appropriate criteria, the weighting process was carried out with BWM, and the spatial layer of each criterion was prepared using GIS. Based on the weighting results, the criteria of distance from existing hospitals and distance from healthcare centers have had the highest and lowest weight, respectively. In the next step, the land suitability map was prepared by combining the spatial layers. Almost 88% of the spatial modeling results corresponded with the realities of the region, and the western half, especially the southwestern part, for the construction of a new hospital, had a higher proportion than other parts. Finally, sites number 2, 5, and 8 among 11 candidate sites were determined as the most optimal places for the construction of a new hospital in the studied area. It is suggested that one or more hospitals should be built in the designated optimal sites in accordance with the population living in areas without optimal access to hospital facilities so that in addition to improving health spatial equity, the cost of citizens’ access to hospitals could be reduced.

Determining the optimal location for the construction of the hospital will increase citizens' satisfaction with health services and increase their quality of life. The aim of the present study is to determine the optimal location for the construction of a hospital in the second district of Tehran’s metropolis. For this purpose, an integrated methodology including GIS, SWARA weighting method, and COPRAS ranking method has been presented. SWARA and COPRAS multi-criteria decision-making methods are among the best methods and have received less attention in the field of determining the optimal location of the hospital in combination with the GIS. The results of the weighting method showed that the criteria of distance to existing hospitals and the geospatial distribution of PM2.5 particles are the most and least important criteria in the decision-making process of the optimal location of the hospital, respectively. The results of geospatial modeling showed that the southwestern parts of the district have a very high potential for hospital construction and almost 48% of the region has the potential to build a hospital. In other words, there is no spatial justice for the citizens of half of the region in using the medical services of the hospitals. The results of ranking candidate sites showed that sites 5 and 6 were the most suitable places to build a new hospital in the second district. According to the research results, the weighting method used has the same accuracy as the fashionable weighting method of the analytic hierarchical process, despite the executive and computational complexity.

The flood is the most important natural disaster that causes human and financial losses every year, then its management is very important. The most basic step in flood disaster management is the preparation of a flood susceptibility map, which integrating the geospatial information system and multi-criteria decision making is an efficient approach to handle it. In order to spatially model flood susceptibility, the present study has presented an approach of fusion the weights of the effective criteria on flood susceptibility using the Dempster-Shafer information fusion theory (DST). The purpose of the fusion of the weights of the criteria is to increase reliability, reduce the uncertainty of the weighting process, and increase the accuracy of flood susceptibility modeling. The present research has presented a hybrid weighting method by integrating the results of two weighting methods (Analytical Hierarchy Process (AHP) and Best-Worst Method (BWM)). The hybrid weighting methods of previous researches are mainly based on simple mathematical operators, and complex operators such as DST used in the present research are less used. The results of the fusion of the weights obtained from the two weighting methods of AHP and BWM indicate the very high weight of the flow accumulation criterion (0.437) and the very low weight of the vegetation index criterion (0.004). Comparing the results of the research with the facts of the studied area showed that the presented hybrid weighting approach with 96% accuracy Compared to each of the basic weighting methods used, is of higher performance. Also, the susceptibility modeling resulting from the new BWM weighting method has been more accurate compared to the common method of the AHP. According to the results of the research, more than 92% of the studied area has moderate to high flood susceptibility and less than 8% of the area has less than moderate susceptibility, which indicates that the studied area is prone to flooding.

The purpose of this research is to select temporary accommodation centers after the earthquake crisis to meet the needs of the victims. Therefore, this research has tried to identify temporary accommodation sites in District 22 of Tehran with the help of effective criteria. The required data has been obtained from the Tehran Disaster Mitigation and Management Organization and the results of the general population and housing census of 2017. The novelty of this research is to present a new combination approach to determine the effective criteria for locating temporary shelter sites. In this regard, the combination of geographically weighted regression (Gaussian and tri-cube kernels) with a binary particle swarm optimization algorithm was used. The recommended combination method is suitable for spatial regression problems because it is compatible with two unique properties of spatial data, i.e. spatial autocorrelation and spatial non-stationarity. The best value of the fitness function (1-R2) for Gaussian and tri-cube kernels was obtained at 0.04616 and 0.0097, respectively, which indicates the high compatibility of the tri-cube kernel with effective criteria. According to the obtained maps, Chitgar Park and Azadi Sports Complex are some of the widest and most suitable areas for the construction of temporary shelter sites after the earthquake crisis in the case study. By identifying temporary shelters, relevant organizations can provide the appropriate infrastructure for these selected centers so that there is no need to spend time to provide these services in the event of an earthquake.

In this study, the spatial-temporal distribution analysis of housing prices in District 5 of Tehran Municipality and the factors effective on it were investigated. To this end, the data related to housing buying and selling in this district in the years 2018, 2019, and 2020 were used to model the housing price. The results were obtained using GTWR method, which gave in a better measure compared to GWR and OLS methods. The adjusted coefficient of determination in OLS, GWR, and GTWR algorithms were found to be 0.759, 0.798, and 0.835, respectively. GTWR is a method that can model the spatial-temporal heterogeneities that exist in the housing price data. Based on the obtained results, the currency exchange rate (dollar to rial) has the highest effect on modeling housing prices. After that, the physical characteristics of housing – such as its footage and age – are important in modeling housing prices. Finally, the access rate to urban services – such as distance to hospitals, sports centers, educational centers, religious sites, green space, highways, and urban public transportation stations – can improve the modeling of housing prices. The findings of this study show that using dollar-to-rial exchange rate as the independent variable, we can model the housing price with a proper precision.

Identification and monitoring of surface water using remote sensing have become very important in recent decades due to its importance in human needs and political decisions. Therefore, surface water has been studied using remote sensing systems and Sentinel-1 and Sentinel-2 sensors in this study. In this paper, two data fusion approaches and decision fusion improve the accuracy of surface water identification for the two study areas in Mazandaran province, Iran, due to dense forests, turbid rivers, and land Agricultural products have been proposed. Data fusion approach by increasing the spatial resolution of Sentinel-2 optical sensor from 20 meters to 10 meters and using radar dataset and water indices in one layer, monitoring surface water with much higher accuracy by classification using supervised classifiers such as Support Vector Machine (SVM), Neural Network (NN), and Random Forest (RF) were performed. Then, by combining the results using the Maximum Voting (MV) method, the accuracy of the results compared to the data fusion approach increased by 4 percent to 5 percent. Since narrow and confused rivers are not easily identifiable and extractable, the proposed approach was able to extract an accurate map of surface water.

Hospitals are among the most important service centers, and the selection of the optimal site for them is a very important (yet complex) undertaking, as it can bring about optimal spatial distribution of hospitals and can make them optimally accessible for citizens. In the present study, in order to optimally locate hospitals in District 5 of Tehran metropolis, a combination of the geographical information system, objective weighting methods, and multi-criteria decision making method was used. The geographical information system was used to analyze and manage the optimal hospital locating criteria, the CRITIC weighting method was implemented to account for the correlation between the criteria, and Shannon's entropy method was used to model the existing uncertainty in the criteria. CODAS multi-criteria decision making method was used due to its novelty and the evaluation of alternatives based on two criteria. Based on the results obtained from CRITIC weighting method, distance from health centers, and based on the results of Shannon's entropy method, distance from industrial areas were the most important optimal hospital locating criteria. The results of ranking the candidate sites using CRITIC-CODAS and Shannon’s entropy-CODAS were almost the same, and both methods identified the sites on the western side of the District (that did not have any hospital) as the appropriate sites. The results of the study indicated the high accuracy of combined objective weighting and multi-criteria decision making methods in optimal locating of the hospitals. It might be asserted that these methods can replace thematic weighting methods such as analytical hierarchy process.

Due to the increase in the occurrence of floods, especially in the cities, and the emergence of human, financial, and environmental risks due to its increase, the flood zoning areas are of great importance. Therefore, in this study, it was tried zoning the areas of floods with the help of determining effective criteria. The criteria used in this research include Modified Fournier Index, Topographic Position Index, Curve Number, Flow Accumulation, Slope, Digital elevation model, Topographic Wetness Index, Vertical Overland Flow Distance, Horizontal Overland Flow Distance, and Normalized difference vegetation index. The novelty of this study is to present a new combination approach to determine the effective criteria in flood hazard zoning (Maneh and Samalqan County). In this regard, the combination of geographically weighted regression (Gaussian and tri-cube kernels) and binary particle swarm optimization algorithm was used. The recommended combination method is suitable for spatial regression problems because it is compatible with two unique properties of spatial data, i.e. spatial autocorrelation and spatial non-stationarity. The best value of the fitness function (1-R2) for Gaussian and tri-cube kernels were obtained 0.0745 and 0.0022, respectively, which indicates higher compatibility of the tri-cube kernel than the Gaussian kernel. It was also found that the criteria used have a significant effect on the rate of flooding in the study area.

Remote sensing has wide applications in many scientific and research fields, including road engineering and transportation, the most important of which is extraction of road network and preparation of a schematic map of road network. Extraction of road network from satellite images is a complementary technology for accessing information, which simplifies the interpretation and analysis of image and improves the quality, which is one of the most important goals of planners. The main purpose of this study was to automatically extract road network on Shahroud-Miami route so that the resulting road network map can be used as the input of pavement management system (PMS). The proposed method in this study is based on the technique of merging and combining images of Sentinel-1 and Sentinel-2 satellites with the majority voting method in order to make maximum use of spectral and spatial information of multiple images (detail increase) instead of single image using texture features. Then, for the monitored classification, three classifications of artificial neural network (ANN), support vector machine (SVM) and maximum likelihood of similarity (ML) were used in two general classes of road and non-road. Random and homogeneous experimental samples and evaluation samples from the existing images and maps of the region were used to assess the accuracy of classification. Results of this study showed that integration of the results of classifications with the majority voting method improved the accuracy by 4% for Sentinel-1 satellite and 6% for Sentinel-2 satellite in identifying the route and the road network. Also, the kappa coefficient in the majority voting method has increased by about 0.11 for Sentinel-1 satellite and by about 0.06, compared to ANN (the best effective classification performance).

Healthcare is one of the most important urban services and proper access to healthcare centers is critical. Determining the hospital’s prospect location is a difficult problem of multi-criteria decision-making (MCDM) and the use of Geographic Information Systems (GIS) improves this decision. In this study, integration of GIS and MCDM was used in order to determine the prospective location of the hospital in the fifth district of Tehran. For this purpose, firstly, the effective criteria in the process of determining the prospect location of the hospital was selected by reviewing the literature and expert's opinions, and the spatial layer of each criterion was prepared. Then, ten sites in this district were selected and the criteria values were calculated for each. Then, the criteria weights were determined by the Criteria Importance Through Inter-criteria Correlation (CRITIC) weighting method. Weighting results showed that the seismic vulnerability was the most effective criterion in the decision-making process. In the next step, the sites were evaluated and ranked by two MCDM methods, that is, the Combinative Distance-based Assessment (CODAS) and the Vlse Kriterijumska Optimizacija Kompromisno Resenje (VIKOR), and the results showed the similarity of the performance of the two methods. The ranking results of the two methods showed that sites No. 7 and 1 were the most suitable and the most unsuitable place to build a new hospital, respectively. In order to evaluate the accuracy of the ranking, the ranking results of the two methods were compared with the ranking performed by experts. Evaluation of the accuracy of the methods showed that the CODAS method for selecting the prospect location of the hospital has higher accuracy than the VIKOR method.

Landslide as a terrifying disaster can cause human and economic losses and the destruction of cultural and natural heritage. While the need for a method to directly predict the location of landslides has vital importance but currently, the prediction is not possible. The zoning of landslide hazard can be an efficient indirect approach. This paper proposes a method for landslide hazard zoning based on the decision fusion and Analytical Hierarchy Process (AHP) in the Semirom of Isfahan province.

In the first step of the proposed methodology of this research, GIS information layers of the study area are collected. Then by using of fuzzy and non-fuzzy hierarchical analysis method and based on expert knowledge, the layers and sub-layers were weighted. In addition, two different overlay methods including weighted overlay and fuzzy overlay are applied for zoning of the AHP and fuzzy AHP results. Combination of both AHP and fuzzy AHP methods with two overlay methods create four zoning maps for the area. The Fuzzy Overlay tool makes it possible for the analysis of the possibility of a phenomenon belonging to multiple sets in a multi-criteria overlay analysis. Not only the fuzzy overlay determines the influential members in the occurrence of a phenomenon but also analyzes the relationships between the memberships of several sets. Weight overlapping is one of the most effective methods used to overlay analysis to address multiple-criteria questions such as location selection and appropriate models. This method will adopt the values in the input raster to a common evaluation criterion for suitability or priority, risk, or appropriate scale. The cell values of each row of inputs increase with the increase of importance of the raster. It also combines the resultant cells to produce the output raster. After obtaining four zoning maps, a decision fusion strategy is applied for the fusion of these maps. Decision fusion systems or in general data fusion or combination strategies combines various decisions made from different methods or data to ultimately make decisions that are more precise and reliable than the result obtained from a single decision. One of the most important and effective methods for integrating decisions is based on the concept of voting. In this method, one vote is assigned to each decision. The simplest form of this method is known as the majority voting. In this method, if all decision-making methods have the same weight and accuracy, the decision of all strategies for an input sample is considered to be the same weight, and the decision with the highest score will be introduced as the winning class for the input sample.

The study area is located approximately 60 kilometers from Semirom city. Also, this area is located in Marbur River watershed. Generally, different factors can be effective in slope instability and landslide, which in this research, slope, aspect, distance to fault, distance to roads, distance to drainages, distance to residential areas, lithology and rainfall were selected for assessing the landslide phenomenon. These effective layers are obtained from information data such as Digital Elevation Model (DEM), fault lines, rivers location, streams location, residential areas, roads location, lithology and synoptic stations. The digital elevation model (DEM) of the region is prepared with 30 meters pixel size from the USGS website. By using DEM in GIS, slope and aspect maps in five classes are created. Faults map of the studied area is obtained from 1:100000 geology map of the Geology organization center of the country. Also, by using Euclidean distance in GIS, distance to faults layer is created in five classes. For preparation of rainfall map, the rainfall content of the studied area has been used from the average rainfall data of the Iran Meteorological Organization in the last 10 years of 19 meteoroidal stations. Based on the rainfall information, the area is divided into five classes. Roads map of the area is obtained from 1:25000 map of National Cartographic Center.  The distance to road layer is created from roads map of the area and divided into five classes. For drainage and residential area maps, a 1:25000 map from NCC is applied. Also, distance to residential area layer is created by this map in five classes. For assessment of the lithology in this area, a 1:100000 geology map is applied.

Results showed that the zoning methods provide satisfactory results, but eventually the results were improved with the decision fusion strategy. For validation our finding the results were compared with historical landslides. Based on the results, it was concluded that zoning by four different combinations: hierarchical analysis and overweight analysis, hierarchical analysis and fuzzy overlay, fuzzy hierarchical analysis and weighted overlay, and fuzzy hierarchical analysis and fuzzy overlaying, have a precision of 80%, 86%, 75% and 88% respectively. After integrating the results of these four methods, the accuracy of the zoning increased to 90%.

In recent years, rapid population growth has led to increase per capita water use in various sectors including agriculture and industry and a growing gap between water demand and water supply has emerged. Therefore, identifying and tracking changes in groundwater resources as an alternative and reliable source of surface water resources are so important to region located in the Middle East with dry weather and large volumes of drought and climate changes. In the current study, the potential and evaluation of water resources are investigated in the west and northwest formations of Shahroud in an area of about 480 square kilometres using remote sensing and spatial information techniques. In this region, due to the presence of carbonate rocks as well as the function of erosion and tectonic forces in parts of the region, relatively suitable aquifers have been formed. For this purpose, information layers including lineament, lithology, slope, aspect, waterways, precipitation and type of precipitation were evaluated by Analytical Hierarchy Process (AHP) method. In this study, information layers were prepared using the ability of RS and GIS in three main stages: extraction of geological map, extraction of area lineaments and extraction of other information layers. The proposed methodology applied two different types of remote sensing sensors including Landsat 8 as optical data and Sentinel-1 as radar data. In preparing lithology map with emphasis on calcareous formations, the proposed method applied four techniques including independent component analysis (ICA), minimum noise fraction (MNF), band ratio (BR) and color composition on Landsat 8. Using the pixel purity index, the endmemberes were extracted and the obtained maps were classified by Support Vector Machine (SVM) and maximum likelihood (ML). Among the classification methods, SVM has a higher ability to classify than ML and identified formations with higher accuracy in the region. Finally, tree decision making system was used to improve the classification of images. The geology of the area was extracted in four classes focusing on calcareous formations. The map was compared with the geological map prepared by the surveying organization and the following results were obtained: Kappa coefficient 0.83, Accuracy of Ku and Jd Formation (calcareous formations) with 99.2% accuracy, Shemshak Formation (Js) with 80.2% accuracy, Lalun Formation (Cl) with 81.2% accuracy and Alloviume (Q) with 78% accuracy were identified. Subsequently, lineaments area were extracted using integration of Landsat 8 and Sentinel 1 radar remote sensing data based on semiautomatic methods. Based on the results, the lineaments were in good agreement with the faults in terms of orientation and numbers at different lengths. Also, the densities of the lineaments extracted in different formations of the geological map of the region had 99% compliance with the densities of faults in the formations. Therefore, by combining band6 of Landsat8, VV and VH polarization Sentinel1, the area lineaments can be extracted with high accuracy. Other thematic layers were extracted by using remote sensing and GIS Techniques using 30 m SRTM. It was also observed that the groundwater potential map is mainly controlled by precipitation, lithology, and lineament density factors.

In the present study, firstly, the effective factors of Golestan national park forests fire have been identified. Then, by using these factors, forest fire development modeling is performed based on cellular automata. Multiple linear regression and genetic algorithm are used to determine the effective factors on Golestan national park forests fire. In order to investigate the effect of spatial resolution of the maps are used on the results of modeling, effective factors have been generated in different spatial resolution and these data are used as the input of the proposed algorithm. Also, the neighboring filters 3×3, 5×5, and 7×7 are used to investigate the effect of the neighboring filter in the forest fire development process. Cellular automata is used for modeling Golestan national park forests fire development, and the artificial bee colony is proposed to calibrate it. The results of this study show that using the proposed algorithm with 3×3 neighboring filter is more accurate than the other neighboring filters. In the best case, the Kappa index, the overall accuracy, and the relative operating characteristic are 0.924, 0.960, and 0.494, respectively that these results are for spatial resolution of 30 meters on November 17, 2010.

Nowadays, Determining the effective factors on fire is so important, because the plenty areas of forests around the world are destroyed annually by fire and recurrence of that in the long term can irreparably damage to the earth and its inhabitants. It helps us to identify most dangerous locations and times in forest fire. Hence, we can prevent many of driving factors of forest fire by law enforcement, efficient forest management policies and more supervision. In the current study, we identified the effective factors on the fire in Golestan forest through integration of three different methods including multiple linear regression, logistic regression and multivariate adaptive regression spline with Genetic Algorithm.

Golestan Province is in the North of Iran and 18% of it is covered by forests. Golestan Province is a touristic province and several roads pass through its forests and according to statistical records, most of the occurred fires were in proximity of these roads. Our study area is located in 36°53′-37°25′N and 55°5′- 55°50′E and its area is about 3719.5 km2. We selected this area, because includes the most of fires have been occurred in Golestan Province in recent years.

A big fire was occurred on 12 December, 2010 in our study area and we used it as the dependent variable. The actual burnt area and some other data, such as Digital Elevation Model (DEM), the roads network, the rivers, the land uses, and soil types in the area were provided from Golestan Province Department of Natural Resources. Also, geographic coordination of the synoptic weather stations near the area and their data, including maximum, minimum, and mean temperature; total rainfall, as well as maximum wind speed and azimuth in December 2010 were obtained from National Meteorological Organization of Iran.
The land use and soil layers were in scale of 1:100000 and the roads and the rivers layers were in 1:5000 and all of them were provided in 2006. The region DEM is generated from topographic maps of Iran National Cartographic Center in scale of 1:25000 with positional resolution of 30m and we produced the slope and the aspect layers from it in ArcGIS software with the same resolution. The roads and the rivers were in vector format, hence, we used the Euclidean Distance analysis to generate rasters that each cell of them shows the distance from the nearest road or river.
At first we had 5 weather stations, which is very few for GWR. In this regard, we generated 1000 random points in the area and interpolated data to these points using Ordinary Kriging method with exponential semivariogram model in 30m resolution in ArcGIS software.
The multiple linear regression (MLR) model is the generalization of simple linear regression that is modeling the linear relation between one dependent variable and some independent variables. The general formula of MLR is seen below: (1)The unknown coefficients are obtained using least squares adjustment as follows:  (2)The logistic regression (LR) model is a nonlinear model for determination of the relation between a binary dependent variable and some independent variables. If we use the values of 0 and 1 for non-fire and fire points respectively, then the probability that a point be a fire point is obtained by Eq. (3):   (3)If the number of parameters is insignificant compared to the observations, then we use the unconditional maximum likelihood estimation shown by Eq. (4) to compute the unknown coefficients of this model. (4)The multivariate adaptive regression spline (MARS) model is a flexible non-parametric model that requires no assumption about the relation between the dependent andindependent variables. Hence it has a high ability in determination of complex nonlinear relations among the variables. The general formula of MARS is seen below:   (5) is the m’th basic function that is obtained by Eq. (6):  (6)These basic functions are chosen in such a way that leads to minimum RMSE of model.
We use the genetic algorithem (GA) with the fitness function of the normalized RMSE to select the optimum combination of effective factors on forest fire.

In this paper we study the dependence of the forest fire to 14 factors shown in table 1, in the study area. Our results are shown in figures 1 to 3.
 

This research shows that both of the biophysical and anthropogenic factors have significant effects on forest fire in our study area. Just two factors were identified as impressive factors in all three cases including the minimum temperature and the maximum speed of wind. This study concluded to the NRMSE=0.4291 and R2=0.9862 for the multiple linear regression, NRMSE=0.9416 and R2=0.9912 for the logistic regression and NRMSE=0.1757 and R2=0.9886 for the multivariate adaptive regression spline and totally the multivariate adaptive regression spline method showed a better performance in comparison to the other two methods.

Forest areas are among the most important natural and ecological resources on the Earth and are considered as one of the main pillars of sustainable development in any country. Fires ruins almost 5500 hectares of Iran’s forests yearly. In this research, firstly, the fire points were identified using the fire data of Forest Organization in combination with MODIS sensor data between 2012 and 2017. Due to the fact that more than 75% of fires were happened in the hot season of the year (June, July, and August), the data of the three months was used for modeling. Then, the effective parameters in fire occurring were evaluated and the dependent parameters were removed. Accordingly, two methods, including multiple linear regression and multivariate adaptive regression spline were studied to predict the fire risk. Some important parameters including the root-mean-square error (RMSE), R2, the correct estimation percentage of fire and non-fire points, and error distribution were used to evaluate. After modeling, it was found that the multivariate adaptive regression spline has better performance—where its RMSE of test data was 0.1628, its R2 of test data was 0.893, and its correct estimation percentage of test fire points and test non-fire points was near 94% and 88% respectively, as well as its error distribution was better than the other method. This actually shows that modeling with a local method is very better than modeling with a global method. Therefore, the risk map resulted by multivariate adaptive regression spline has better reliability compared to those of the other method. Finally, the high-risk areas were recognized using the risk map of this method. The traits of these areas were a short distance to residential areas and roads, having rich soil with organic materials, relatively high temperature, and low height.
 

In 2000, a convention was established in the United Nations to improve the quality of human life in which the principles of the Millennium Development Goals were adopted. One of these goals was to ensure the stability of the environment and natural resources. In the contemporary world, the value of forests is about 120 billion dollars and the livelihood of almost 9.1 people is dependent on forest (in)directly.
According to the opinion of global experts including FAO, if the forest cover of a country is less than 25% of that country’s area, that country is in critical condition in terms of the human environment. Almost 190000 hectares of Iranian forests have been ruined by fire in a 28-year period. Forest fire not only changes the natural ecosystem and ruins many plant and animal species of a region, but also makes other destructive effects like air pollution, respiratory problems, soil erosion, increased flowing surface waters, increased acidity of soil, decreased fertility, tourism industry losses, manufacturing industry and economy losses, and even climate change.
Immediate and accurate detection of the fire location and the ability to determine the effective parameters on it, as well as the detection of the areas with high-risk of fire is among the main concerns of environmental protection and disaster management. We can prevent the fire by training people, making effective regulations and management policies, and increased monitoring to deal with fire triggers. Moreover, in the case of fire occurrence, we must take necessary actions like deploying fire-fighting equipment near hazardous areas and making easy access to these areas. In fact, nowadays, the increasing importance of protecting the forests and natural resources has led to change the focus from crisis management to risk management.

The modeling was not possible without non-fire points. Accordingly, at the beginning, some points are randomly selected in the whole area with a certain distance from the fire points and are identified as non-fire points. To implement the methods in MATLAB programming environment, firstly, the parameters used in the modeling are extracted using the maps of these parameters for fire and non-fire points. These parameters are used as inputs in each of these methods. 
Constantly, 70% of the selected data were used as the training data and 30% of them were used as the test data. Initially, the multivariate linear regression and then the multivariate adaptive regression spline were used for modeling. The steps of the research implementation are shown in Figure (1).
After implementation of the modeling, the evaluation parameters of each method were provided to compare. Then, the risk map of the area was provided using trial points and Inverse Distance Weighting (IDW) and by employing 12 lateral points for each method (Figures 2 and 3). The points with a high risk were extracted from the resulted map. Then, the main traits of these points are considered as the traits of high-risk points.
 
Fig. 1. The steps of the research implementation

Fig. 2. Fire risk map provided using the MLR method on test data

Fig. 3. Fire risk map provided using the MARS method on test data

 
After removing the dependent parameters from the effective parameters on the fire, the optimal effective parameters are presented in Table (1). These parameters are divided into three groups including climate, ground physical, and human parameters.
The modeling of fire risk was done by two methods. In the training and testing data section, the RMSE and R2 are presented in Table (2) for multivariate adaptive regression spline and multivariate linear regression methods, respectively. The results achieved by the training data section indicate that the training procedure is more accurate (R2 closer to 1) and with less error (less RMSE) in the multivariate adaptive regression spline than those achieved by the multivariate linear regression method. The appropriate amount of evaluation parameters for test data shows that the model does not experience over-fitting in these methods.
Table 1. Effective parameters on fire occurrence in the case-study area

In the linear regression method, the two parameters of the correct estimation percentage of fire points and non-fire points have a low value, hence, the worst possible scenario has happened and the risk map has the least amount of reliability. In the multivariate adaptive regression spline, the fire and non-fire points are simultaneously estimated with a high accuracy. This makes the risk map provided by the multivariate adaptive regression method becomes to be more reliable.
As seen in the results, the risk map provided by the multivariate adaptive regression spline method has a very higher reliability compared to the risk map provided by multivariate linear regression method. Hence, the risk map resulted by the first method was used to determine the features of the areas with a high risk of fire (Figure 4).
Since the fire risk has a normal distribution, the areas which satisfy Equation (1) are among the 2.5% of the areas that have the most fire risk.
(1)
where  is the average,  is the standard deviation, and R is the fire risk. The main features of the mentioned areas can be used as the important tools for decision making. The extraction of high-risk areas is done in ArcGIS environment. Statistical analysis of effective parameters’ features in these areas shows some key points. These features include low distance from the residential regions (less than 2 km), low distance from the road (less than 2 km), having mollisol, relatively high average temperature (more than , and low height (less than 50 m).

Fig. 4. High risk map provided using the MARS method on test data

This research attempted to identify the optimal method for modeling of fire points risk using climate, ground physical, and human parameters. Therefore, an accurate local method (MARS) was used along with a non-local method (MLR).
In the test data and the training data sections, the MARS method had the lowest RMSE and a value closer to 1. The outputs showed that the MARS method had a more accurate performance in the estimation of the fire and non-fire points compared to the MLR method. This indicated the high reliability of the MARS method. After determining the optimal method for the modeling of the area’s fire occurrence, the points of the area with high risk of fire were detected. After doing a statistical analysis it was found that these points have some fundamental features including low distance from the residential regions (less than 2 km), low distance from the road (less than 2 km), having mollisol, relatively high average temperature (more than  and low height (less than 50 m).

Recently, Synthetic Aperture Radar (SAR) data are of high interest for different applications in remote sensing specially land cover classification. SAR imaging is independent of solar illumination and weather conditions; it is not affected by rain, fog, hail, smoke, or most importantly, clouds. It can even penetrate some of the Earth’s surface materials to return information about subsurface features. However, SAR images are difficult to interpret due to their special characteristics, i.e., the geometry and spectral range of SAR are different from optical imagery. In addition, the exhibition of the property of speckle caused SAR image is visually difficult to interpret. Consequently, optical data can be applied in fusion of SAR data to improve land cover classification. On the other hand, Light Detection and Ranging (LiDAR) data provide accurate height information for objects on the Earth, which makes LiDAR become more and more popular in terrain and land surveying. Regarding to the limitations and benefits of these three remote-sensing sensors, fusion of them improved land-cover classification. For this purpose, it is necessary to apply data fusion techniques. In recent years, significant attention has focused on multisensory data fusion for remote sensing applications and, more specifically, for land cover mapping. Data fusion techniques combine information from multiple sources, providing potential advantages over a single sensor in terms of classification accuracy. In most cases, data fusion provided higher accuracy than single sensors. Furthermore, fusion of sensors with inherent differences such as SAR, optical and LiDAR data need higher level of fusion strategies. Ability to fuse different types of data from different sensors, independence to errors in data registration step and accurate fusion methods could be mentioned as the benefits of decision-level fusion methods rather than other level fusion (pixel and feature level fusion methods).
This paper presents a method based on the simultaneously using of RADAR, multispectral and LiDAR data for classification of urban areas. First, different feature extraction strategies are utilized on all three data then a feature selection method based on Ant Colony Optimization (ACO) is applied to select optimized features. Maximum Likelihood (ML), Support Vector Machine (SVM) and K-nearest neighbor (KNN) are applied to classify optimized feature space as three classification methods. Finally a decision fusion method based on Weighted Majority Voting (WMV) is applied to provide final decision. A co-registered TerrraSAR-X, WorldView-2 and LiDAR data set form San Francisco of USA was available to examine the effectiveness of the proposed method. The results show that the proposed method based on the simultaneous use of three radar, optical and LiDAR data can increase the accuracy of some classes more and some other classes less. Also, the results of the data fusion can provide a different improvement compared to the results of the classification of each data individually. Generally, the results show that the use of multisensor imagery is worthwhile and the classification accuracy is significantly increased by such data sets. There are also several practical issues to be considered in future studies. Note that only a decision ensemble system was explored in this study. Additional research is needed in areas such as further study on more powerful feature spaces on each data, further processing of DSM of LiDAR, novel fusion strategies such as rule-based and object based classification strategies.

Regarding to the limitations and benefits of remote sensing sensors, fusion of remote sensing data from multiple sensors is effective at land cover classification. All these data have different characteristics, e.g., different spatial and spectral resolutions, different angle of view, and different abilities and disabilities. For many applications, the information provided by individual sensors is incomplete, inconsistent, or imprecise. Fusion of information from different sensors can produce a better understanding of the observed site, which is not possible with single sensor. Particularly, Light Detection And Ranging (LiDAR) provides accurate height information for objects on the earth, which makes LiDAR become more and more popular in terrain and land surveying. On the other hand, hyperspectral imaging is a relatively new technique in remote sensing that acquires hundreds of images corresponding to different spectral channels. The rich spectral information of HS data increases the capability to distinguish different physical materials, leading to the potential of a more accurate image classification. As hyperspectral and LIDAR data provide complementary information (spectral reflectance, and vertical structure, respectively), a promising and challenging approach is to fuse these data in the information extraction procedure.
This paper presents a multiple fuzzy classifier system (Multiple Classifier System or MCS) for fusions of hyperspectral and LiDAR data based on Decision Template (DT). After feature extraction on each data, the classification was performed by fuzzy K-Nearest Neighbor (KNN) on hyperspectral and LiDAR data separately. In a multiple fuzzy decision system, a set of decisions is first produced and then combined by a specific fusion method. The output of the fuzzy classifiers that provide the class belongingness of an input pattern to different classes is arranged in a matrix form defined as decision profile (DP) matrix. Then, a fuzzy decision fusion method (Decision Tempate) is utilized to fuse the results of fuzzy KNNs on hyperspectral and LiDAR data. In order to assess the fuzzy MCS proposed method, a crisp MCS based on (Support Vector Machine) SVM as crisp classifier and Naive Bayes (NB) as crisp classifier fusion method is applied on hyperspectral and LiDAR data. The experiments were executed on a hyperspectral image and a LiDAR derived Digital Surface Model (DSM); both with spatial resolution of 2.5 m. The dataset have captured over the University of Houston campus and the neighbouring urban area by the NSF-funded Centre for Airborne Laser Mapping (NCALM). Also hyperspectral image has 144 spectral bands in 380 nm to 1050 nm region. Training and testing samples were selected from different areas of the images. They are spatially disjointed.
Fuzzy MCS on hyperspectral and LiDAR data provide interesting conclusions on the effectiveness and potentialities of the joint use of these two data. Overall accuracies of fuzzy classifiers on LiDAR and hyperspectral data are %75 and %88 respectively. Fusion of these two fuzzy classifiers produced %96 as overall accuracy. Second scenario for joint use of hyperspectral and LiDAR data is fusion of these two data through a crisp decision fusion system. The results show that fuzzy classifier provided higher accuracies than crisp classification based on SVM for both data. In the presence of mixed coverage pixels in remote sensing data, crisp classifiers may produce errors while fuzzy classifiers are not affected by such errors and in principle can produce a classification that is more accurate than any crisp classifier. Also, fusion of ensemble of fuzzy classifiers based on Decision Template method produced more accuracy than fusion of crisp SVMs based on Bayesian Theory.

With recent technological advances in remote sensing sensors and systems, very high-dimensional hyper spectral data are available for a better discrimination among different complex land-cover classes. However, the large number of spectral bands, but limited availability of training samples makes the problem of Hughes phenomenon or ‘curse of dimensionality’ in these data. Moreover, these high numbers of bands are usually highly correlated and the information provided can contain several data redundancies. Because of these complexities of hyperspectral data, traditional classification strategies have often limited performance in classification of hyperspectral imagery. Referring to the limitation of single classifier in these situations, classifier ensemble systems may have better performance than single classifiers especially on hyperspectral data with this high level of complexities. This paper presents a new method for classification of hyperspectral data based on a band grouping strategy through a SVM ensemble system. Proposed method used a band grouping process based on a mutual information (MI) strategy to split data into few band groups. After band grouping step, the proposed algorithm aims at benefiting from the capabilities of SVM as classification method. So, proposed method applied SVM on each band groups that produced in previous step. Finally, this paper applied Naive Bayes (NB) as a novel and robust classifier fusion method for combining classifiers in classifier ensemble system. NB is a precise classifier fusion based on the concepts of Bayesian theory. Experiments are applied on two common hyperspectral data. Obtained results show that the classification accuracy is significantly improved by the proposed method in comparison with standard SVM on all bands of hyperspectral data. Also, these results confirm the high performance of band grouping strategy in contrast to using of standard SVM on all feature space.

Support Vector Machine (SVM) has emerged in recent years as a popular approach for classification of remote sensing data. SVMs don’t require huge training samples and have little possibility of over fitting however; the accuracy of SVM mainly depends on the parameters selection of it. So, one of the significant research problems in SVM is the selection of optimal parameters that can increase the accuracy of this classifier. Regularization constant C and kernel function parameters exert a considerable influence on the accuracy of SVM. In recent years, the development of parameter optimization for SVM is supported by evolutionary algorithms and bio-inspired metaheuristic algorithms such as swarm-based methods. This paper evaluates the potential of one of the swarm-based bio-inspired optimization methods called Firefly Algorithm (FA) for SVM optimization. FA is a metaheuristic algorithm, inspired by the flashing behavior of fireflies. The primary purpose for a firefly's flash is to act as a signal system to attract other fireflies. Firefly algorithm is