support vector machine

Energy, as one of the most critical and essential factors of production, plays a vital role in human life. With fossil fuel resources depleting, researchers are exploring alternatives such as biodiesel, a renewable biofuel with properties similar to diesel. Given the growing importance of liquid biofuels, particularly biodiesel, in global markets, it is crucial to ensure high-quality fuel production to gain consumer trust. Additionally, from a commercial perspective, considering fuel storage duration, it is necessary to determine the type of fuel and the biodiesel-to-diesel ratio using accurate, fast, and cost-effective tools.

This study aims to identify and differentiate various blends of biodiesel and diesel fuel (2%, 5%, 10%, and 20% by volume) derived from different vegetable oil sources (rapeseed, sunflower, and waste cooking oil) over different storage periods (immediately after production, 1 month, 2 months, and 3 months after production). Biodiesel was first produced from rapeseed oil, sunflower oil, and waste cooking oil using methanol and a potassium hydroxide (KOH) catalyst. Each biodiesel blend was mixed with diesel fuel at the specified ratios and analyzed using an electronic nose system equipped with 10 sensors. Data were collected over different periods (monthly) and analyzed using methods such as linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and support vector machine (SVM).

The results demonstrated the effectiveness of classification methods in separating pure fuels over four months. The accuracy rates were 97%, 100%, 82%, and 82% for SVM; 100%, 100%, 100%, and 98% for QDA; and 100%, 96%, 100%, and 100% for LDA, respectively. These methods were also capable of distinguishing pure fuels (D100, K100, WCO100, SUN100) from biodiesel-diesel blends at various storage times with high precision. Among the sensors in the system, four sensors (MQ2, TGS2620, MQ4, and TGS2602) showed the highest sensitivity to biodiesel fuels. The analysis revealed that the separation power of the models decreased during the second month of storage, with the lowest performance observed after two months. This suggests that the most significant structural and physicochemical changes in biodiesel properties occurred during this period. Furthermore, the similarity in performance parameters for fuels derived from sunflower oil and waste cooking oil indicates their shared origin. The QDA model outperformed the LDA and SVM models in separating and classifying fuel blends. Using the SVM technique, all 160 data points (40 for pure fuel and 120 for biodiesel-diesel blends) were evaluated. The SVM model achieved a specificity of over 96% for identifying and classifying pure and blended fuels immediately after production. This parameter increased to 94%, 98%, and 99% after the first, second, and third months of storage, respectively.

The study highlights the effectiveness of electronic nose systems combined with advanced classification methods for analyzing biodiesel-diesel blends. The QDA model demonstrated superior performance in fuel classification, while the SVM model also showed high accuracy in distinguishing pure and blended fuels. The findings underscore the importance of monitoring fuel quality over storage periods, as significant changes occur within the first two months.

The coefficient of water distribution uniformity in sprinkler irrigation systems is one of the important indicators that are effective in evaluating their performance and only high values can justify the implementation of these systems. The purpose of this research is to use support vector machine (SVM) and gene expression programming (GEP) models to simulate the coefficient of water distribution uniformity in the farm-conditions of Malekan plain in the northwest of Iran, which is in the catchment area of the Urmia lake is experiencing severe water stress.Field tests were carried out on seven farms equipped with a classic stationary sprinkler irrigation system with a movable sprinkler (Komet 162, 163) with variables of sprinkler intervals on laterals and manifolds, operating pressure and wind speed, and distribution uniformity coefficient data were obtained. The values of the indicators (RMSE, MAE, R2) were obtained in the training and test steps, respectively (3.5087, 2.6827, 0.8634) and (1.1787, 0.9494, 0.9833) for GEP. The values of the evaluation indices (RMSE, MAE, R2) for the most optimal SVM model in the test and training steps were obtained (4.8917, 4.2704, 0.7884) and (2.6790, 2.4113, 0.9185) respectively. In the training step, the value of CU(DDR(max)) for GEP and SVM model was calculated as 7.0540 and 5.2925 respectively. The value of this index in the test step for these two models was 20.83 and 9.28 respectively. The comparison of the value of this index also showed that the GEP model is more accurate than the SVM model.

Introduction and Goal:

The process of identifying landforms is a subject that has been studied by many researchers. All geomorphological definitions are based on the study and identification of landforms. Understanding landforms and their distribution is a fundamental need of applied geomorphology and other environmental sciences. In this regard, remote sensing technology, due to the production of satellite images with high spatial and spectral resolution, can be a valuable tool for identifying and classifying landforms. Desert pavement is one of the most important landforms in arid and desert regions. Mapping pavements and their types provides a basis for evaluating the region in terms of its structural and geomorphological characteristics, which is useful in many environmental management and planning issues and can be used as a model for similar regions. In this research, using ETM+ sensor data and based on the considered criteria, the characteristics of desert pavements in Semnan township were identified and classified.

The study area, with an area of ​​47645.98 hectares, is located in Semnan township. The geographical coordinates of the region are 28˚53¢ to 43˚53¢ east longitude and 20˚35¢ to 40˚35¢ north latitude. The aim of this study was to investigate, separate, and identify desert pavement classes as a type of desert landform using remote sensing and Landsat ETM+ satellite images in southern Semnan. Therefore, by conducting field surveys and sampling of the study area, the percentage of desert pavement cover density was measured and the location of each sample was recorded with GPS. Support vector machines, neural networks, spectral angle maps, spectral information divergence, and fuzzy artmaps were used to classify desert pavements in Envi 4.5 and IDRISI Selva software environments. Then, the accuracy of each classification of each method was compared with the training samples using the coefficients of complete accuracy, kappa, user accuracy, and producer accuracy. Finally, the spatial zoning map of each method was drawn in the Arc GIS 10.2 software environment.

In this study, the best band combination for detecting and separating desert pavements in southern Semnan was the 6-4-3 band combination with an optimal index factor of 45.71 (OIF), which was in the mid-infrared and visible bands (VNIR + TIR). Based on the kappa coefficient, the support vector machine (85.05), fuzzy artmap (81.44), neural network (55.17), spectral angle map (53.89), and spectral information divergence (50.22) methods had the highest ability in spectral separation of different classes of desert pavements in southern Semnan, respectively. The support vector machine and fuzzy artmap classification methods obtained the highest kappa coefficients for the classes, respectively, and the lowest kappa coefficients and complete accuracy were obtained in the neural network, spectral angle map, and spectral information divergence methods, respectively. Since the classes, bands, and other conditions used were the same for all methods, the difference in accuracy depended only on the calculation instructions of the methods.

Conclusions and suggestions:

Remote sensing technology, due to the production of satellite images with high spatial and spectral resolution, can be a valuable tool for identifying and classifying landforms. Preparing a map of desert pavements and their types is a basis for evaluating the region in terms of structural and geomorphological characteristics, which can be useful in many environmental management and planning issues. Therefore, it is suggested that other classification methods based on shear orientation, combined methods, frequency coverage, images from sensors with better spatial and spectral resolution, and considering characteristics such as particle diameter be used to be effective in preparing maps of desert pavement layers.

Up-to-date information on growing area is necessary for cropping pattern, and remote sensing is a useful tool in achieve this goal (Hunt et al., 2019). Among various image classification algorithms and vegetation indices for crop type identification, machine learning methods and normalized difference vegetation index (NDVI) have had better results (Mousavi et al., 2020). In addition, attention to plant phenological stages and using multi-temporal images have led to more accurate crop distinction. The aim of this study was estimation of wheat growing areas in Shushtar county in Khuzestan province in Iran.

Time series of NDVI index was obtained from eight Sentinel-2 satellite images from November to June, which correspond to the wheat growth stages. Then, maximum likelihood and support vector machine algorithms were used to identify wheat fields. To implement the classification methods, training and test samples were needed, which were obtained by matching the NDVI diagram of the wheat crop growth stages and field surveys.

Error matrix was used to evaluate the classification results. Based on this, the support vector machine with overall accuracy and Kappa coefficient of 98.5 and 96.5 percent, respectively, had higher accuracy than the maximum likelihood with overall accuracy and Kappa coefficient of 97.8 and 95 percent, respectively. In addition, considering the wheat crop phenological stages using the time series of NDVI in training samples, selection increased the classifications accuracy. Based on the support vector machine results, the total wheat growing areas was estimated to be 48233 hectares, which showed a small deviation from the available surveys.

Using the pattern of NDVI time series to obtain training samples showed the efficiency of plant indices in estimation of wheat growing areas according to the crop phenological stages. In addition, the results revealed that the support vector machine classification was more accurate than the maximum likelihood in the study areas, and it was considered as a base method. One reason for the appropriate performance of the support vector machine was appropriate distribution and adeqaute number of training samples based on wheat phenological stages. It can be concluded that the time series of the normalized difference vegetation index in combination with the support vector machine classification provided the possibility of quick and accurate estimation of the wheat growing areas in Shushtar county in Khuzestan in Iran.

Gully erosion is a particularly destructive form of water erosion that can lead to alarming rates of soil loss, especially in the vulnerable landscapes of dry and semi-arid regions. This type of erosion is recognized not only for its immediate impact on land but also as a critical environmental challenge that requires our urgent attention. As a result, there has been a growing emphasis on developing effective predictive models that can elucidate the temporal and spatial dynamics of gully erosion-specifically, how it forms, expands, and evolves over time. This endeavor has captured the interest of soil conservation experts and researchers alike, who understand the profound implications of this issueIn recent years, remote sensing and data mining techniques have emerged as valuable tools for identifying and mapping areas susceptible to gully erosion. These innovative methods provide essential insights for land managers and policymakers, enabling them to make informed decisions. Furthermore, the effectiveness of predictive models hinges on their advanced capabilities, which enhance their learning potential and improve the identification of relationships among various factors. Creating a sensitivity map is an essential strategy for land use planning, as it actively contributes to reducing land degradation and safeguarding our natural resources. Understanding the connection between gully occurrences and influential factors is not only beneficial; it is crucial for sustainable land management and environmental preservation.

This research investigates the sensitivity of the upper basin of the Boustan Dam to gully erosion using object-based techniques and data mining algorithms. To achieve this, field visits were conducted to select 81 gullies for analysis. The study examines several factors, including slope, aspect, slope length index (LS), elevation, plan curvature, distance from the river, drainage density, topographic wetness index (TWI), height above the nearest drainage (HAND), average annual rainfall, distance from roads, distance from faults, land use, geomorphology, soil texture, and satellite bands B7, B5, and B3. Additionally, the normalized difference vegetation index (NDVI), normalized difference built-up index (NDBI), and normalized difference water index (NDWI) are considered, along with geological aspects. QuickBird satellite images from 2021 and Orfeo software were utilized to monitor and identify gullies in the area through image segmentation. Initially, a collinearity analysis of 23 effective erosion occurrence indices was performed, resulting in the removal of distance from the fault, digital elevation model (DEM), NDWI, NDBI, and satellite bands B3, B5, and B7 due to their collinearity exceeding five. Following this linear operation, all remaining indices were integrated with the segmentation map obtained from the Orfeo environment. Finally, three models -Random Forest, Maximum Entropy, and Support Vector Machine- were employed to model the selected indices using Python (Colab).

The results from the object-oriented method in the Orfeo software further demonstrated its effectiveness in accurately identifying gullies. With an impressive accuracy rate of 91.3%, this method has proven to be highly reliable in generating machine learning maps with high precision. Findings indicate that the key factors contributing to gully erosion include the rainfall index, distance from the river, Height Above Nearest Drainage (HAND) index, distance from the road, and valley index. Torrential rain emerged as a significant driver of gully erosion, while the distance from the river was crucial due to the concentration of surface and subsurface flows toward waterways. The HAND index played a prominent role in modeling the sensitivity of the study area compared to other sub-indices derived from DEM, as it exhibited promising applications in assessing natural hazards. Locations close to roads were found to be more vulnerable to water erosion, and valleys were identified as especially susceptible to gully erosion due to their conducive conditions for rapid water flow and erosion. Extensive field studies support this observation. Furthermore, zoning results generated using these indices indicated that, within the random forest model, 544.23 hectares of the area are at high or very high risk of erosion. This model outperformed the Maximum Entropy and Support Vector Machine models in predicting erosion-prone areas. Finally, the ROC curve was utilized to validate the model, yielding AUC values of 0.95 and 0.94 in the random forest model during the training and validation stages, respectively. These results indicate the model's high accuracy in predicting areas highly susceptible to gully erosion.

This study effectively used object-based image analysis algorithms and data mining techniques to create a sensitivity map of the region. The object-based method efficiently identified the local gullies using the mean shift algorithm, while the random forest algorithm excelled in predicting areas prone to gully erosion. Key factors contributing to gully erosion were identified, including rainfall, distance from the river, soil HAND index, and distance from roads and valleys. The findings from this study provide valuable insights for managing and preserving basin resources. Implementing the recommendations from this research could help mitigate the impacts of gully erosion in the future and ensure the sustainability of the Boustan Dam and its surrounding ecosystem.

High-resolution satellite imagery data is widely utilized for Land Use/Land Cover (LULC) mapping. Analyzing the patterns of LULC and the data derived from changes in land use caters to the increasing societal demands, improving convenience, and fostering a deeper comprehension of the interaction between human activities and environmental factors. Although numerous studies have focused on remote sensing for LULC‎ mapping, there is a pressing need to improve the quality of LULC maps to achieve sustainable land management, especially in light of recent advancements made. This study was carried out in an area covering approximately 8000 hectares, characterized by diverse conditions in LULC, geomorphology and pedology. The objective was to investigate the potential for achieving maximum differentiation and accurate mapping of land features related to LULC. Additionally, the study assessed the impact of various spectral indices on enhancing the results from the classification of Landsat 8 imagery, while also evaluating the efficacy of support vector machine (SVM) and maximum likelihood algorithms in producing maps with satisfactory accuracy and precision.

As an initial step, LULC features were identified through fieldwork, and their geographic coordinates were recorded using GPS. These features included various types of LULC, soil surface characteristics, and landform types. Following the fieldwork, 12 types of LULC units were identified. Subsequently, the LULC pattern in the study area was classified using the RGB+NIR+SWIR1 bands of Landsat 8, employing both SVM and maximum likelihood classifiers. To assess the impact of various spectral indices on improving the accuracy of the LULC maps, a set of vegetation indices (NDVI, SAVI, LAI, EVI, and EVI2), bare soil indices (BSI, BSI3, MNDSI, NBLI, DBSI, and MBI), and integrated indices (TLIVI, ATLIVI, and LST), and digital elevation model of study area were successively incorporated into the classification algorithms. Finally, the outcomes from the two classification algorithms were compared, taking into account the influence of the applied indexes. The classification process continued with the selected classifier and indices until reaching the maximum overall accuracy and kappa coefficient.

Field observations revealed that the study area could be categorized into 12 primary LULC units, including irrigated farms, flow farming, dry farming, traditional gardens (with no evident order observed among planted trees), modern gardens (featuring regular rows where soil reflectance is visible between tree rows), grasslands, degraded grasslands, highland pastures (covered by Astragalus spp., dominantly), lowland pastures (covered by halophyte plants), salt domes (with no or very poor vegetation), outwash areas (River channel with many waterways), and resistant areas. The results of image classification indicated that the performance of the SVM algorithm across different band combinations is superior to that of the maximum likelihood method. Using SVM resulted in an increase in overall accuracy and Kappa coefficient by 3-8% and 0.03-0.08, respectively. For the map generated using RGB+NIR+SWIR1 bands and employing SVM, overall accuracy and Kappa coefficient were determined to be 76.6% and 0.72, respectively. Among the vegetation indices used in the SVM algorithm, LAI had the most significant impact, increasing the classification accuracy by 2.64%. Among the soil indices, BSI and MBI indices demonstrated the best performance; with BSI increasing the classification accuracy by 1.95% and MBI by 1.64%. Among the integrated indices, LST and ALTIVI enhanced the classification accuracy by 2.75% and 2.35%, respectively. It should be noted that the inclusion of the digital elevation model did not significantly improve the classification accuracy when using the support vector machine algorithm; in fact, it led to a decrease in accuracy when applied to the maximum likelihood classification. The probable reason for this issue is the different nature of DEM data compared to the other input data, as well as the limitations of parametric statistical approaches to effectively integrating data from diverse sources. Finally, the classification process was executed using the three visible bands, NIR, and SWIR1, in conjunction with selected indices (LAI, BSI, MBI, LST, and ALTIVI). Results indicated that using these spectral indices significantly improved classification accuracy, particularly for the DF, DGL, MG, O, and IF land cover/use classes. The calculated accuracies for these classes increased by 11.62%, 18.57%, 20.06%, 29.39%, and 33.19% respectively. Consequently, the accuracy of the classification and the Kappa coefficient (using support vector machine algorithm) increased to 85.24% and 0.82, respectively.

In this research, we aimed to accurately map various land use/land covers by utilizing Landsat 8 imagery and incorporating three group of spectral indexes. Despite spectral interferences and overlaps among various phenomena related to LULC, the utilization of different spectral indices resulted in significant differentiation among LULC classes. Finally, considering the limitations of modelling in ENVI software, it is recommended to investigate the effectiveness of other models for classification in more specialized software, such as R.

The use of remote sensing, due to the provision of timely data and the high capability of image analysis, as well as the possibility of studying in a wide range with acceptable accuracy, is of great help to the planners and implementers of the agricultural sector. Remote sensing is one of the effective tools for monitoring, studying and determining the level of cultivation of agricultural and horticultural products, especially in large areas. Therefore, Annual Landsat images are valuable resources that enable crop monitoring in issues related to diagnosis, crop yield prediction, and crop-cultivation pattern studies.

Landsat 8 and OLI satellite images related to 2018 were used to estimate the overall level of crops in the area and to separate the agricultural area from other areas. These images were downloaded from the site http://earthexplorer.usgs.gov, which is related to row 37 and path 166. In order to collect additional information used in this research, Google Earth images and ground control points taken with a GPS device, Garmin model 64s, were used. Finally, the area under cultivation of the products was calculated. The statistics of the Ministry of Agricultural Jihad in the crop year 2016-2017 were used to evaluate the results. The classification steps are done to classify different classes.

For this purpose, in this research, to estimate the monitoring of main crops in the area and to classification agricultural crops, satellite images of Landsat 8 sensors with OLI sensor and NDVI index related to the time series of one-year images were used. The area of wheat and barley cultivated land in the region was estimated to be 10639 hectares, which shows an error of about 6.8% compared to the results of Jihad Keshavarzi statistics, which is 9956.56 hectares. After classifying the satellite images, using teaching samples that were not involved in the classification process, the accuracy of the classified image was evaluated. In this research, after calculating the cultivated area in GIS, the results were validated by comparing with the available statistics of Jihad-Agriculture and regional service centers and time control points by GPS, kappa coefficient and general accuracy coefficient. Supervised classification method with support vector machine algorithm and artificial neural network was used to separate farms. The crop classification maps were divided into 5 categories, wheat and barley, summer crops, rice crops, irrigated crops and non-agricultural lands. The lowest amount of commission error in both classification methods, 8.1 %, is related to summer cultivation, and the emission error in non-agricultural lands was 0.5 % in support vector machine method and 2.5% in artificial neural network method. Sentinel-2 images with a spatial resolution of 10 meters were used to prepare the classification map.

Conclusion :

After classification, the maps produced by satellite images were compared with reference data and ground facts, as well as with the help of satellite images available in Google Earth software. Then the error matrix was formed. Kappa coefficient and overall accuracy were obtained in the vector machine method as 0.74 and 68%, respectively, and in the neural network method as 0.70 and 76% respectively.

River flow prediction is one of the key issues in the management and planning of water resources, in particular the adoption of proper decisions in the event of floods and droughts. To predict the flow rate of rivers, various approaches have been introduced in hydrology, the most important of which are the intelligent models. In this study, a hybrid, model Firefly Algorithm- support vector machine, was applied to estimate the discharge of Dez river basin based on the daily discharge statistics provided by the hydrometric stations located at the upstream of the dam during the statistical period (2002-2022) and its performance was compared with the support vector machine model. The correlation coefficients, root mean square error, and mean absolute error was used for evaluation and a comparison of the performance of models. The results showed that the hybrid structures presented acceptable outcomes in the modeling of river discharge. A comparison of models also showed that the hybrid model of Firefly Algorithm-support vector machine has a better performance in forecasting the flow. In conclusion, the use of the FA-SVM model could be effective in estimating flood peak discharge.

Nonlinear relationships, inherent uncertainty, and the need for a lot of climate information in estimating evapotranspiration have made researchers use data-mining methods to estimate evapotranspiration in recent decades. The purpose of this research is to investigate the efficiency of data mining methods of support vector machine, decision tree, random forest and Gaussian process regression in forecasting the daily reference evapotranspiration of coastal strip stations in the south of the country. To do the work, daily reference evapotranspiration was calculated using 20year climatic data (2001-2021) using the Fao-Penman-Monteith method. Then, using these data as output data, 6 combined scenarios were evaluated based on the correlation between meteorological variables and reference evaporation-transpiration using data mining methods. The results of the investigations showed that all four data mining methods were able to estimate the reference evaporation-transpiration values in the studied areas.In all four stations, the Gaussian process regression method with the highest R2 value and the lowest RMSE and MAE values had a better estimate of the reference evapotranspiration values, and random forest, decision tree, and support vector machine methods were in the next ranks respectively. Gaussian process regression model in estimating reference evapotranspiration, this method is recommended for estimating reference evapotranspiration

Soil properties play a crucial role as they determine the soil's suitability for different types of plant growth, ecosystems, and biota functioning. They have a significant impact on nutrient cycling, carbon sequestration, and soil management. Digital Soil Mapping (DSM) is a process aimed at delineating soil properties. Soil sampling for DSM serves as  a fundamental  step in improving prediction accuracy and is crucial for incorporating variability in terms of environmental covariates. Conditioned Latin Hypercube (CLH) sampling is a technique utilized to generate a sample of points from a multivariate distribution conditioned on one or more covariates. Numerous researchers (Ramirez-Lopez et al., 2014; Adhikari et al., 2017; Zhang et al., 2022) have endorsed this approach in their studies, following its inception by Minasny and McBratney in 2006. However, there has been limited research to date on the impact of the Latin hypercube method's random sample selection process on the accuracy of resulting maps. Hence, the central question remains: Is the Latin hypercube sampling method, which is currently widely adopted, always a dependable approach in this field?

The study area covers longitudes 50°35'47'' to 51°29'' east and latitudes 31°36''31'' to 32°15'48'' north in Borujen city, Chaharmahal, and Bakhtiari Province. The region, with an average elevation of 2338 meters above sea level, receives an annual rainfall of 250 millimeters and maintains an average temperature of 11.5 degrees centigrade. In this investigation, inherited data from soil studies were utilized, consisting of 250 samples distributed across the study area. In this research, the studied characteristics included percentage of equivalent calcium carbonate, clay, and soil organic carbon at a depth of 0 to 30 cm. Land component variables were extracted using the Alus Palsar digital elevation model with a spatial resolution of 12.5 meters. In the initial stage, digital maps of equivalent calcium carbonate, clay, and soil organic carbon were generated using the support vector machine method. The modeling process proceeded until a highly accurate model was achieved, with the root mean square error percentage (RMSE%) being less than 40. The Latin hypercube approach was utilized for sample design, with 500 repetitions in this study. After selecting sampling points for each run using the Latin hypercube method, these points were mapped onto a detailed map, and the corresponding feature values were retrieved. The final map was created based on the extracted points. Subsequently, the latin hypercube approach was employed to generate soil property maps for each selected dataset. Validation was conducted using criteria such as the coefficient of explanation, root mean square error, and root mean square error in multiple iterations to ensure the accuracy of the generated maps.

The results distinctly illustrates the varied selection of sampling positions with each implementation of the Latin hypercube method. It is important to note that there may be some overlaps in different implementations. Consequently, the primary question arises: Is a one-time execution of the Latin hypercube sufficient for selecting study points? The findings indicate that the support vector machine model achieves satisfactory accuracy for all the examined characteristics. In the studied area, the environmental factors such as slope and elevation were identified as a significant predictors for estimating percentage of equivalent calcium carbonate.

In the present study, the accuracy of the latin hypercube method was assessed for selecting sampling location for digital soil mapping endeavors in Chaharmahal and Bakhtiari Province. Given the impracticality of collecting numerous field samples to evaluate the soil sampling method, this research aimed to employ simulation methods based on highly accurate maps for this purpose. The results indicate that the different outputs of the Latin hypercube method influence the accuracy of modeling, although this effect is also influenced by the specific feature under investigation and the extent of its variability within the study area.  Considering that the Latin hypercube method is based on the principle that samples are randomly selected in each class of environmental parameters, it is suggested that future studies using this method should account for this principle. Adequate consideration should be given, and the selection of sampling locations should rely on multiple implementations of the Bhattacharya distance method to ensure robustness and reliability.

Fire is one of the most important natural hazards that has a great impact on the structure and dynamics of natural ecosystems. Due to Iran's location in the arid and semi-arid belt of the world, a large number of human-made and natural fires occur in different regions of the country every year. Therefore, determining sensitive areas to fire occurrence plays an important role in fire management in natural resources. To do so, the current study aims to identify fire-prone areas in Dere Shahr city in Ilam province using two machine learning of random forest (RF) and support vector machine (SVM) and 2024 fire occurrence points. Environmental factors were prepared in categories including topographical factors (altitude, slope direction, slope anlgle), climatic factors (rainfall, relative humidity, wind, temperature), biological factors (vegetation and soil moisture) and man-made factors (distance from residential areas, distance from road, distance from agricultural land, distance from river). The model’s accuracy was evaluated using the area under the curve (AUC) in the ROC curve and cross-validation statistics. Examining the AUC index showed that both models had good accuracy, although the RF model (AUC = 0.97) had higher accuracy than the support vector machine model (AUC = 0.86). According to the results of RF model, about 60% are in the low-risk class and about 20% are in the high fire risk class. Investigating the contribution of the factors affecting the occurrence of fire showed that man-made factors (distance from residential areas) and climatic factors (temperature) played a more important role in areas with a history of fire. Therefore, increasing public culture and reducing dangerous behaviors in nature can reduce the occurrence of fire in this area and contribute greatly to the protection of the environment and preservation of natural resources.

Due to the location of Iran in arid and semi-arid regions and according to the quantitative and qualitative limitations of water resources, optimal management and volumetric delivery of water is important in irrigation and drainage networks. In this regard, it is necessary to estimate the water requirement of crops accurately and provide adequate water to farmers. Remote sensing technology provides facilities that can be used to obtain different layers of information at the lowest cost in the fastest time. Accordingly, many researchers have used remote sensing data to monitor vegetation cover, provide land use maps, estimate crop evapotranspiration and have declared this technology as appropriate tool for such studies. Based on the previous studies, it is observed that low researches has been conducted to investigate the crop evapotranspiration considering the crop water requirement. Therefore, the most important objectives of this study are: provide the cropping pattern and land use maps using Sentinel 2 satellite images, determination of the water requirement for the delivery points of irrigation network, determination of the actual evapotranspiration of the crop cover using SEBAL algorithm and Landsat 8’s images and finally evaluation of the water supply and management in the Mahabad irrigation and drainage network. In order to determine the cropping pattern of the Mahabad irrigation and drainage network, Sentinel 2 images have been used related to the 2018-2019 crop year. The images were examined in terms of the region of syudy and the percentage of cloudiness and after selecting the appropriate images, pre-processing operations including radiometric and atmospheric corrections were applied on them. Then, the NDVI index was calculated based on selected images. On the other hand, after determination of the classification classes, the phenological cycle of crops were examined for each class and spectral pattern of crops was determined during the growing season. Training samples were selected for supervised classification using the existing maps, Google Earth images, creating images with false color composites and considering the growth pattern and some of them were also considered for validation of the classified map. Then, the cropping pattern map was obtained by using the SVM classification algorithm. After generating the crop classification map, the water requirement of the different classes was determined based on the Penman-Montith evapotranspiration method, applying plant coefficients and irrigation application efficiency at the volumetric water delivery points. Finally, the actual evapotranspiration rate of the study area calculated based on the SEBAL algorithm and compared with the net water requirement map. Based on the results, kappa coefficient and overall accuracy of the classified map were determined to be 0.953 and 91%, respectively. The area of the planted agricultural farms was equal to 10594 hectares and 1576 hectares of farms were without planting. The area of orchard farms was equal to 6786 hectares and the area of sugar beet, wheat, alfalfa and corn lands were obtained to 998, 1839, 693 and 278 hectares, respectively. Thus, the net irrigation water requirement was equal to 71 million cubic meters and the gross irrigation water requirement was calculated equal to 161.36 million cubic meters, considering the irrigation efficiency of 44%. On the other hand, the evaluation of the SEBAL evapotranspiration maps during the growing season indicated that the total amount of evapotranspiration was equal to 79.78 million cubic meters, and this amount was 14% higher than the net irrigation water requirement. Finally, according to the crop classification map and based on the comparison of the net irrigation water requirement and evapotranspiration maps, the water consumption in the Mahabad irrigation and drainage network was evaluated. It turned out that in the upstream farms of the network or close to the Mahabad River, the Water consumption was more than net water requirement and downstream areas were faced to deficit irrigation due to lack of sufficient water.Finally, based on the results of this study, it was observed that by using the capabilities of satellite images and remote sensing, it is possible to monitor and evaluate the condition of agricultural farms on a large scale with acceptable accuracy. Also it is possible to improve the management of water supply and water use efficiency in irrigation and drainage networks by creating up-to-date land use maps, determining net and gross irrigation water requirment and comparing with actual evapotranspiration maps.

The aim of this study is to investigate the performance of various algorithms in enhancing the accuracy of land cover maps using Sentinel images.

The study area is a sheet (1:25000) encompassing an area of 15782.6 hectares. The land cover map was created using a range of algorithms - Mahalanobis distance, maximum likelihood, minimum distance, neural network, parallelepiped, support vector machine, spectral angle mapper, spectral information divergence, and binary encoding - applied to the optimal band composition derived from 12 bands (2, 3, 4, 5, 6, 7, 8, 8a, 11, 12, NDVI, SAVI). The training area was obtained from field information and satellite images. 70% of the samples were used for classification and 30% for evaluating the accuracy of the classified maps.

The results indicate that the support vector machine, neural network, and maximum likelihood classifications have the highest accuracy. The support vector machine classifier is slightly more accurate than the other two classifiers. To improve the classification, support vector machine kernels (linear, polynomial, radial basis function, and sigmoid), spectral angle mapper settings (6 modes), and parallelepiped (2 modes) were utilized. The results show that the support vector machine classifier, using the 6th degree polynomial function method, has the highest accuracy. Following this, the map was post-processed with the highest accuracy, resulting in an overall accuracy of 96.63 and a kappa coefficient of 0.9393.

An examination of 21 classification modes on Sentinel images in this study provides a comprehensive review of classification algorithms compared to previous studies.

Rivers are always faced with erosion and sediment transport. Sediment transport in rivers is one of the most complex topics in river engineering and is always the focus of experts and water engineers. This phenomenon is one of the important hydrodynamic processes that affect many hydraulic systems and water facilities and is considered one of the basic problems in the exploiting surface water resources globally. Estimating the sediment load of rivers is one of the important and practical issues in the studies and design of water engineering projects, such as the design and development of irrigation and drainage networks, water extraction from rivers, etc. Sediment concentration can be calculated by direct or indirect methods, which are usually expensive and time-consuming direct methods. Various factors affect this phenomenon, which makes their analysis difficult. Therefore, they cannot model the sedimentation phenomenon with acceptable accuracy. Hydraulic models cannot always be trusted due to the need for a lot of data, unavailability of the required data, and the inaccuracy of the data due to human error for simulating sediments. Nowadays, fuzzy and neural intelligent conductor systems, due to their ability to solve complex and nonlinear phenomena, have found many applications in various water engineering problems, including sedimentation. The purpose of this research is to evaluate and compare adaptive neural fuzzy models (ANFIS), support vector machine (SVM), gene expression programming (GEP), and group model of data handling (GMDH) in estimating the sediment load of Tirah River, Markazi Province.

In this research, first, the long-term daily statistics of temperature, rainfall, average flow rate, and sediment concentration of Hasan Abad hydrometric and sediment measuring station located on the main branch of the Tirah River were collected. Then, the data sufficiency test for analysis, checking the correlation between parameters of river discharge, precipitation, temperature with sediment discharge, and determining the long-term average of suspended sediment in the studied stations were performed. In the next step, a suitable combination of input variables was selected. The design of the input parameter pattern can be based on the relationship between flow and sediment flow parameters, rainfall, temperature, flow, and sediment flow. Of course, considering that the mentioned parameters have a historical course, therefore, the design of the input patterns of soft computing models should be done based on time delays (like what is discussed in the analysis and forecasting of time series). Determining the most appropriate time delay of the input parameters in the modeling of discharge, sediment, temperature, and rainfall, then the appropriate design of the structure of the used soft calculation models was done. In the next step, the estimation of sediment discharge using an SVM, GEP, and ANFIS group method of GMDH data control and comparison of three data mining methods, and also with the sediment rating curve and observational data. About 70 % of the research data was used as training and between 20 to 30 % for validation and testing.

Based on the statistical indicators of optimal model selection, the best performance of the SVR model has been obtained for model number one. In this model, the R2 and RMSE obtained from the model are 0.96 and 0.0047, respectively. Besides, the R2 and the RMSE error of the models in predicting suspended sediment values in the test stage are 0.95 and 0.014, respectively for the ANFIS model, and 0.50 and 4.97, respectively for the GEP model. The best performance of the ANFIS model has been obtained for model number one. In this model, the R2 and the RMSE obtained from the model are 0.95 and 0.014. The R2 and RMSE of the models in predicting suspended sediment values in the test stage are 0.96, 0.0047 for the SVR model, and 0.50, 4.97 for the GEP model, respectively. The best performance of the GEP model has been obtained for pattern number nine. In this model, the R2 and RMSE obtained from the model are 0.99 and 0.010, respectively. The R2 and the RMSE of the models in predicting the amount of suspended sediment in the test stage are respectively equal to 0.70, 0.015 for the ANFIS model and 0.78, 0.0185 tons respectively for the SVR model.

It can be seen that the performance of the GEP model was better compared to other models. SVR and ANFIS models are ranked second and third. In the next step, the best-selected pattern of ANFIS, SVM, and GEP models was used as the input of the GMDH model. First, input pattern one, which was selected as the best pattern for ANFIS and SVM models, was introduced as the input of the GMDH model. In the training and test, the values of R2 statistical indices are 0.94 and 0.99, respectively, the RMSE error value is 0.0079 and 0.0038, respectively, the MSE value is 0.000062 and 0.000015, respectively, and the MAPE values are respectively 0.007 and 0.003. In the next step, input pattern nine, which was selected as the best pattern for the GEP model, is introduced as GMDH input. In the training and test steps, the value of R2 is equal to 0.95 and 0.98 respectively, the RMSE error value is equal to 0.0077 and 0.0045 respectively, and the MSE value is equal to 0.0006 and 0.00002 respectively, and MAPE value is equal to 363 and 502. The results showed the acceptable performance of the GMDH model with the highest R2 equal to 0.99 and 0.98 and the lowest RMSE equal to 0.0038 and 0.0045, respectively.