artificial neural network

In recent years and decades, due to minor changes and challenges in sufficient and high-quality water resources, sustainable water resources have been the subject of various studies and research. The lack of safe and high-quality water resources is a major obstacle to sustainable development. For this reason, understanding the processes of the water cycle is very important and requires accurate information about hydrological phenomena. Runoff from water transfer is one of the main sources meeting various water demands, including agriculture, industry, and domestic use. The allocation of water resources to these sectors is planned based on runoff data at different times. A significant portion of precipitation in the hydrologic cycle is converted into runoff due to watershed characteristics. Considering the issue that the Lake Urmia Basin is shrinking, identifying the water resources of this basin and its sub-basins is crucial.

The Ajichai Basin is one of the sub-basins of Lake Urmia. In this study, rainfall data from the Tabriz synoptic station and runoff data from the Nahand hydrometric station were used. The aim of this research is to model the daily rainfall-runoff of the Ajichai Basin using intelligent machine learning models, including Artificial Neural Network (ANN), Support Vector Machine (SVM), Gene Expression Programming (GEP), and Random Forest (RF). Seventy percent of the data was used for training, and 30% was used for testing the models. Statistical measures such as the Coefficient of Determination (R²), Root Mean Square Error (RMSE), Nash-Sutcliffe Efficiency (NSE), and Willmott Index (WI) were used to evaluate the performance of the models.

The results of this research showed that all models performed very well in simulating rainfall-runoff in the Ajichai Basin. According to the obtained results, the GEP model, with R² equal 0.84, RMSE equal 0.024 m³/s, NSE equal 0.864, and WI equal 0.968, was the most accurate in modeling rainfall-runoff in the Ajichai Basin. Based on scatter plots and time series analysis, the GEP model demonstrated higher accuracy than other models in predicting rainfall-runoff values with a high correlation.

According to the results, all the investigated models showed good capability in modeling daily rainfall-runoff in the Ajichai Basin. The findings of this research highlight the strong performance of machine learning models in rainfall-runoff modeling. In general, due to the high accuracy of intelligent models, particularly the GEP model, in predicting daily rainfall-runoff, it is recommended to apply these methods to hydrological problems. Additionally, for future research, it is suggested that intelligent methods and data mining techniques be used to model the precipitation-runoff process in different basins separately for drought-affected and wet years.

In this study, the air drying of cumin seeds was boosted by cold plasma pre-treatment (CPt) followed by high-power ultrasound waves (USp). To examine the impact of included effects, different CP exposure times (0, 15, and 30 s), sonication powers (0, 60, 120, and 180 W), and drying air temperatures (30, 35, and 40 ºC) were selected as input variables. A series of well-designed experiments were conducted to evaluate drying time, effective moisture diffusivity, and energy consumption, as well as color change and rupture force of dried seeds for each drying program. Numerical investigations can effectively bypass the challenges associated with experimental analysis. Therefore, the wavelet-based neural network (WNN), the multilayer perceptron neural network (MLPNN), and the radial-basis function neural network (RBFNN), as three well-known artificial neural networks models, were used to map the inputs and output data and the results were compared with the Multiple Quadratic Regression (MQR) analysis. According to the results, the WNN model with an average correlation coefficient of R2 > 0.92 for the train data set, and R2 > 0.83 for the test data set provided the most beneficial tool for evaluating the drying process of cumin seeds.

The expansion of villages in the last few decades has caused significant changes in the quality and quantity of rural waste. Rural waste not only leads to diseases but also results in unpleasant odors, unsightly rural landscapes, and pollution of water, soil, and air in the village environment, adversely affecting rural residents. According to the Demographic Data Organization statistics, Iran ranked 17th globally in waste production in 2021. The per capita waste production in Iran is approximately double the global average. Despite the Public Service Department of Rural Affairs of Iran reporting that 10,000 tons of regular wastes are generated daily in the country's villages, half of which cannot be recycled, emphasizing the importance of optimal rural waste management and identifying effective components in this area.

This study evaluated the impact of social capital components on rural waste management in villages of the Padena district (Semirom County). Data were collected using a questionnaire in 2024. The questionnaire's validity was confirmed by a panel of experts, and its reliability was assessed using Cronbach's alpha test. Pearson's correlation coefficient was employed to analyze the relationship between social capital components and rural waste management behavior in the study area. The artificial neural network was utilized to identify the most crucial social capital component in rural waste management. Lastly, a stepwise multivariate regression method was used to assess the statistical significance of social capital components in influencing waste management behavior and identify key predictor components.

The Pearson correlation analysis revealed a positive correlation between various dimensions of social capital and waste management at a significant confidence level, indicating a strong relationship between social capital and effective rural waste management in the region. The study found that the social ethics component, with an importance coefficient of 0.22 and a relative importance of 100%, was relatively more critical in rural waste management behavior than other social capital components. Social trust was identified as the second most important component in waste management behavior. Among the social capital components, only ethics and social participation had a positive and significant impact on rural waste management in the study area. These two components should be considered as predictive variables in waste management behavior, with each unit increase in social ethics leading to a 1.41 unit improvement in waste management behavior, and each unit increase in social participation contributing 0.75 units to improved waste management behavior.

The study highlights the importance of adhering to social ethics and positive characteristics such as respecting others' rights, social responsibility, empathy, solidarity, and participation in social activities in rural waste management. The results emphasize the need for planning and policy-making to enhance ethics and social participation in optimal rural waste management through educational workshops and financial incentives in rural areas.

Iron is an essential element in the growth process of plants and plays a crucial role in chlorophyll production. Iron deficiency is a serious limitation in vineyards that can significantly affect both the yield and the quality of the crop. The use of modern methods such as digital image processing not only increases precision but also reduces the need for costly and time-consuming laboratory testing, thereby lowering costs and speeding up data-driven decision-making processes in orchard management. The aim of this study is to develop a system based on image processing and neural networks to estimate the active iron content in grape leaves. For this purpose, 55 leaf samples with different levels of iron deficiency were collected and analyzed from vineyards around Urmia. The total and active iron content in the samples was measured using atomic absorption spectroscopy and the leaves were photographed and processed under controlled light conditions. Statistical features were extracted from the images and their correlation with active and total iron content was analyzed. Finally, the best features were used to predict iron content using a multilayer artificial neural network. The results of the linear regression show that active iron correlates with the R, G, H, and S color channels with coefficients of 0.64, 0.58, 0.54, and 0.45, respectively, and that total iron does not correlate with the changes in leaf color. The neural network with an optimized structure of 8-9-1 was able to predict the data from the atomic absorption device with an accuracy of 0.83, 0.88, and 0.84 for training, test, and all data, respectively. In summary, image processing can be effectively and reliably used as a tool for optimal plant nutrition management and rapid diagnosis of iron deficiency.

The use of geospatial techniques for mapping soils is broadly covered by the term digital soil mapping (DSM). Soil maps have considerable significance as basic maps in many environmental and natural resources studies. Digital soil maps are based on the relationship between environmental variables and soil properties. With the development of computers and technology, digital and quantitative approaches have been developed. Continuous utilization of agricultural lands regardless of the land suitability caused soil destruction. Also, incompetency in custom methods, invention geographic information system (GIS), and remote sensing (RS) techniques cause erupt and use of digital soil mapping.

The study area is approximately 5000 ha which is located in the west of Heris region of East Azerbaijan province, Iran. In the first study, the potential of different models to predict soil classes at different taxonomic levels was investigated. According to semi-detailed soil, survey and using stratified random sampling method, 50 pedons and 50 augers with an approximate distance of 1000 m were excavated, described and soil samples were taken from different genetic horizons. Based on the pedon descriptions and soil analytical data, pedons were classified up to the family level. Different machine learning techniques, namely boosted regression tree (BRT), random forest (RF), artificial neural networks (ANNs), and multinomial logistic regression (MLR) were used to test the predictive power for mapping the soil classes. After preparing the soil properties maps and checking their accuracy, these maps were used along with auxiliary parameters for estimating soil classes using an artificial neural network model in the R software. Finally, the accuracy and uncertainty of the model were evaluated by overall accuracy and confusion index, respectively.

Results showed that the different models had the same ability for prediction of the soil classes across all taxonomic levels but a considerable decreasing trend was observed for their accuracy at subgroup and family levels. The terrain attributes were the most important auxiliary information to predict the soil classes up to the family level. The main goal of the second study was to predict soil surface properties (pH, electrical conductivity, gypsum, organic carbon, calcium carbonate equivalent, coarse fragments, and particle size distribution) using ANNs, BRT, generalized linear model (GLM), and multiple linear regression (MLR). Among the studied models, GLM showed the highest performance to predict most soil properties whereas the best model is not necessarily able to make an accurate estimation. Also, the terrain attributes were the most important environmental covariates to predict the soil classes in all taxonomic levels, but they could not display the soil variation entirely. This shows that the unexplained variations are controlled by unobserved variations in the environment, which can be due to the management over time. Results suggested that the DSM approaches have not enough prediction accuracy for the soil classes at lower taxonomic levels that focus on the soil properties affecting land use and management. Results showed that the entry of more details in the soil classification at the lower levels of the Soil Taxonomy system while increasing the number of classes, leads to decreasing the overall accuracy and increasing uncertainty. It is noticeable that the ANNs model has a good accuracy up to the great group level through the acceptable level of overall accuracy (i.e., 75 %), hence it has a high degree of uncertainty. Therefore, the accuracy of the model could not be effective in its selection through the modeling process; however, paying attention to its uncertainty is also very important along with the model error.
 

Terrain attributes were the main predictors among different studied auxiliary information. The accuracy of the estimations with more observations is recommended to give a better understanding about the performance of DSM approach over low-relief areas. Further studies may still be required to distinguish new environmental covariates and introduce new tools to capture the complex nature of soils. Accordingly, we suggest using the other methods of soft computing for modeling in plain areas or low relief regions. Finally, the use of DSM methods is increasing over time and will eventually be considered as distinct and novel techniques.

Nutritional problems according to high amounts of fat in some cheeses have caused the use of fat substitutes to be investigated. Modeling the hardness, whey and pH of UF low-fat feta cheese with three levels of lecithin at zero, 1 and 2 g/kg, three levels of zero, 10 and 20 g/kg of whey powder, xanthan gum. in amounts of zero, half and 1 g/kg and additional starter of Lactobacillus paracasei in two amounts of 1 and 3 g/l with artificial neural network in order to determine the best type of transfer function, type of learning law and percentage of data used for training, evaluation and the test was performed based on the lowest error and the highest correlation coefficient. The results showed that the best model for predicting cheese hardness changes was an algorithm with a hidden layer and the number of 7 neurons, under the sigmoid transfer function with Levenberg's learning law, which could show a good correlation coefficient (0.985). For whey, a model with a hidden layer, the number of 3 neurons and the tangent transfer function and Levenberg's learning law created the best algorithm with a correlation coefficient of 0.908. Also, the pH of cheese was predicted by a model with a hidden layer, the number of 8 neurons and the sigmoid transfer function under Levenberg's learning law and the correlation coefficient was 0.8493. The best percentage of data for education, evaluation and testing of hardness, whey and pH values were obtained 5.35.60, 30.10.60 and 45.20.35 respectively

In many countries of the world, the camel has played an important role in the lives of people in arid and semi-arid regions in terms of providing milk and meat products. Based on phenotypic characteristics, the camel has been adapted to the deserts from physiological, anatomical, and behavioral points of view. Due to limited water resources, the south Khorasan province of Iran has not been an appropriate geographical region for a great number of agricultural products. Weighing plays an important role in the management of camel breeding to adjust the nutritional needs and also the annual evaluation of the animals. One of the main challenges of camel breeding is the difficulty of recording and the lack of records due to the wild nature and also the large size, especially in adults. Due to the many difficulties and risks, camel breeders usually use various potential alternative tools such as apparent estimation or weighing tape to estimate the weight of camels at different ages. For farm animals, significant correlations between body measurements and body weight can be used as a tool to estimate the weight of animals through a mathematical equation. The main objective of this study was to compare the accuracy of artificial neural network (ANN) and multiple linear regression model (MLR) in estimating the body weight of a humped camel.
In the present study, the data of a total number of 177 one-humped camels in four groups  (including 1. 63 adult female Pakistani camels aged between 9 and 12 years, 2. 21 adult female Baluchi camels aged between 9 and 12 years, 3. 93 male and female camels less than 2 years of age of Pakistani and Baluchi breeds, and 4. total camels) collected at the South Khorasan camel breeding station in 2019 were used. Morphological characteristics of camels include: neck length, neck girth, hand length, foot length, shoulder height to the ground, hump height to the ground, hip height to the ground, chest girth, chest width, abdomen girth, hip width, tail length, breast height, and breast girth were measured. After measuring the body dimensions, the evaluated camels were weighed using a 1000 kg digital scale. To estimate the weight of camels from their body dimensions, the data were analyzed using the MLR model and ANN with the nnet package in R software. For ANN analysis, 80% of the data were considered for network training and 20% for testing. The accuracy of MLR and ANN for camel body weight estimation was compared using the coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE).
The results of the MLR model showed that body dimensions of shoulder height to the ground, chest girth, abdominal girth, neck girth, and hand length had a significant effect on body weight (P<0.05). In the ANN model, chest girth, abdominal girth, and shoulder height to the ground were the most important variables in estimating the body weight of camels of the whole population. The MLR and ANN models had acceptable accuracy in weight estimation. However, compared to the MLR model, the ANN model had a higher R2 and a lower error in estimating the weight of camels. In the first group, R2 and RMSE were found to be 0.996 and 6.852, respectively, for ANN while the corresponding values were 0.979 and 22.955, respectively, for MLR.  In the second group, R2 and RMSE were found to be 0.995 and 3.525, respectively, for ANN while the corresponding values were 0.989 and 5.377, respectively, for MLR.  In the third group, R2 and RMSE were found to be 0.896 and 13.959, respectively, for ANN while the corresponding values were 0.849 and 17.549, respectively, for MLR. In the fourth group, R2 and RMSE were found to be 0.929 and 20.248, respectively, for ANN while the corresponding values were 0.903 and 38.505, respectively, for MLR. The results showed that ANN is more accurate compared to MLR in predicting the body weight of camels. 
In terms of goodness of fit (including R2 and RMSE), the results of the present research suggest that both MLR and ANN methods have high acceptable accuracy for predicting body weight in camels. ANN was more suitable compared to MLR, suggesting that it could be used to predict camel body weight. Furthermore, grouping the camels by age and breed could also lead to higher precision and lower prediction error.

Satellite remote sensing images are widely used to monitor the earth's surface phenomena changes at various periods. For accurate change detection, spatial features can be used as the complement information of spectral features. Hand-craft spatial features such as the co-occurrence matrix features are inefficient in detecting the changes due to the complex structure of satellite images. In the present study, a deep learning-based model is proposed as the alternative to address the problems of classical change detection methods. deep neural networks are mainly developed for images and hierarchically extracting spatial-spectral features. In this study, Landsat-8 images between 2013 and 2021 were used to evaluate the changes in Sahand city using the proposed deep network. Pre- and post-classified Landsat-8 images are produced using a deep neural network in the first stage. In the second stage, for producing the change maps, the post-classification approach is used in that change maps are produced based deference of classified images. Finally, the majority voting technique eliminates the noises in change maps. The proposed method results are compared with those obtained by two classical machine learning methods, random forest, and artificial neural networks. According to the change detection results, the proposed deep learning network improves detection accuracy by 13.88% and 12.80% compared with artificial neural networks and random forests. Compared to the random forest and artificial neural networks, the proposed network has improved the overall accuracy of the from-to-change maps by 57.81% and 65.7%, respectively. Final results demonstrate that although Random forest and artificial neural networks have been able to identify the location of changes, they perform poorly in detecting the from-to changes

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.

Dust is one of the environmental issues that have adverse effects in all sectors of agriculture and natural resources. The aim of this research is to predict the concentration of dust in the air. A laboratory system for acquiring images of dust was implemented including a glass chamber, a blower, a dust meter, an imaging camera, and a personal computer. Using clay soil, dust storms with different concentrations of 0, 275, 1289, 1896, 2316, 2585, and 2750 µg/m3 were created inside the glass chamber. For each studied dust concentration, 15 images were obtained and after their preprocessing, the mean values of different image channels in various color spaces were extracted. The features of the images were used to predict dust concentration using artificial intelligence technology. The data were divided into three groups, 60% of the data were used for training, 20% for validation, and 20% for testing the network. Different models of multilayer perceptron artificial neural networks were investigated and 10-8-1 structure with tansig activation function in hidden and output layers has the highest accuracy (93.81%). The findings of the present research show the high capability of image processing and artificial intelligence technologies in predicting dust concentration with high accuracy and low cost.

In this study artificial neural network (ANN) used to predict milk test day records at 4th, 5th, 10th months of lactation duration and 305-day milk yield in Holstein dairy cows. Primary data source was consisting of 274025 milk production records of 7201 primiparuse to fourth birth Holstein cows, from two herd. Final source of data obtained from milk production records was consist of 87980 monthly milk test day records in 8798 rows which each row contains number of animal, herd, age, lactation, month of production, first to tenth monthly milk production records and 305-day milk yield. From the total of data, 50% was considered for neural network training, 20% for validation and 30% for testing. A multilayer perceptron (MLP) network with back propagation of error learning mechanism (BP) was used through different artificial neural network (ANN) structures to predict milk production. In order to optimize artificial neural network (ANN) structure three activation functions (hyperbolic tangent axon, sigmoid axon or linear hyperbolic tangent axon) and three back propagation algorithms viz. momentum, conjugate gradient (CG) and Leven-berg–Marquardt (LM) Training algorithms used in the hidden layer as well as in the output layer. Coefficient of determination, root of mean square error and mean absolute error were used to compare algorithms. In prediction of milk production of 4th and 5th monthly test day records, LM algorithm with sigmoid axon activation functions and LM Training algorithm, with hyperbolic tangent Axon functions had the best performance between network structures respectively. In these net work structures R2 were highest (0.725 and 0.642 respectively), RMSEs were lowest (4.785 and 5.345 respectively) and MAEs were lowest (3.715and 4.057 respectively). In prediction of 10th monthly test day milk production through three or four monthly test day records, obtained from the same lactation period, none of the structures had ability to predict milk production successfully. In prediction of 305-day milk yield, LM algorithm and hyperbolic tangent activation function had the best prediction through 3 test day records and R2, RMSE and MAE as performance criteria were 0.799, 984.14 and 790.21 respectively. Also the same structure of the network had the best performance to predict 305-day milk yield through four or five initial test day records and performance criteria, Coefficient of determination, root of mean square error and mean absolute error were 0.856, 850.98 and 653.33 respectively, in ANN with four test day record as input variables and 0.904, 706.59 and 548.69 respectively, in ANN with five test day record as input variables, respectively. The artificial neural network designed in this study was able to predict the milk production of animals in the fourth month of lactation with a correlation coefficient of 0.84. On the other hand, the designed neural network was able to predict the total milk production of the animal in a lactation period of 305 days with appropriate accuracy. So that the correlation coefficients in using the first three, four and five monthly records of livestock for prediction were 0.89, 0.92 and 0.95 respectively.

Life on Earth is influenced by precipitation. Precipitation is one of the most significant factors that affect the hydrological cycle. Considering that precipitation is non-linear, complex, and can be changed according to spatial and temporal, estimating the amount of this important atmospheric factor in each month or year for each region and watershed is particularly important in managing and optimizing water resources. Various optimization models and algorithms have been proposed for modeling hydrological systems in recent decades. These algorithms have significantly reduced errors and increased accuracy. Still, since hydrological systems rely on random events, none of the methods can be completely and accurately selected as a superior model for modeling and estimating. The honey badger algorithm is an innovative algorithm that requires a few iterations to achieve an optimal solution, and this increases the speed of reaching the desired results. In current study investigates the performance of three models, including multiple linear regression (MLR), artificial neural network (ANN), and hybrid artificial neural network with honey badger optimization algorithm (HBA-ANN) for modeling the temporal and spatial precipitation in East Azarbaijan province. The best-developed model was selected by evaluation criteria such as R, RMSE, NRMSE, MBE, and NSE, the best model is selected.

The MLR model is one of the methods to analyze and investigate several variables. In this method, the model has one dependent variable and several independent variables, so that a linear equation is generated between the independent variables called X1, X2, ..., Xn and the dependent variable Y. ANN is a black box model of neural networks in the human brain. One of the most used methods is the BP method, which includes two stages. In the first stage, which is entitled feed-forward, the error value is calculated, after comparing output and objective values. In the second stage, which is labeled the back-propagation, the error value calculated in the previous step is corrected. The mentioned two stages continue until the output of the model approaches the desired output. The HBA is a new algorithm that simulates the honey-seeking behavior of a creature called the honey badger. The HBA includes two stages. In the first phase, the locations of this creature are calculated, and in the second phase, the exact distance between the HBA and the prey (dj) is calculated based on the honey intensity (S) and the honey smell intensity (Ij), as well as its new and optimal location for the prey Xnew. In the HBA-ANN model, the HBA algorithm is used to determine the most optimal output value in the ANN and increase performance in modeling. Therefore, the developed hybrid model can have the characteristics of both ANN and HBA methods.

In this study, in the first stage, the temporal modeling, and in the second stage, the spatial modeling of the monthly precipitation of 18 stations in East Azarbaijan province during the period of 1996-2022 using MLP, ANN, and HBA-ANN models has been paid. For temporal modeling of precipitation, one and two-month precipitation delay steps of the stations were used as input parameters. The first 70% of the dataset was selected for the training phase and the last 30% of the dataset was selected for the testing phase. Based on the results obtained from evaluation criteria and graphic diagrams, it can be concluded that the HBA-ANN model indicated significant accuracy compared to other models in the temporal modeling of precipitation. Also, by comparing the results of the stations in the HBA-ANN model, the Heris station with R =0.94, RMSE=2.25, NSE=0.79, NRMSE=0.04, and MBE=1.06 in the testing stage performed better compared with other stations. For spatial modeling of precipitation, the geographic coordinates of the stations, which include longitude, latitude, and altitude, are used as input parameters, and average monthly precipitation is used as the output parameter. From eighteen stations, 70% of the stations were selected for the training phase and 30% of the stations were selected for the testing phase. Based on the results obtained from R=0.95, RMSE=1.03, NSE =0.92, NRMSE = 0.03, and MBE = -0.81 and graphical diagrams, it can be concluded that the HBA-ANN model revealed significant accuracy compared to other models in spatial modeling of precipitation.

Precipitation is one of the most important factors that significantly change the hydrological cycle. Therefore, modeling and estimating this parameter is vital. In this study, the performance of multiple linear regression (MLR), artificial neural network (ANN), and hybrid ANN using honey badger algorithm (HBA-ANN) models were used for the spatial and temporal modeling of precipitation in East Azarbaijan province. For spatial modeling, the time delay steps of one and two months of station precipitation were selected as input parameters. Also, for temporal modeling, the longitude, latitude, and altitude parameters were used. The mentioned models were evaluated by R, RMSE, NSE, NRMSE, and MBE assessment criteria. According to the results of temporal modeling, the HBA-ANN model for all stations, especially Heris station with R equal to 0.94, RMSE equal to 2.25, NSE equal to 0.79, NRMSE equal to 0.04, and MBEequal to 1.06 is selected as the superior model. Also, based on the results obtained from spatial modeling, the HBA-ANN model with R equal to 0.95, RMSE equal to 1.03, NSE equal to 0.92, NRMSE equal to 0.03, and MBEequal to -0.81 was selected as the best model. The MLR and ANN models, respectively, presented a relatively poor performance compared to the developed hybrid model.

Today, the use of intelligent models in simulating runoff has been widely used in water resources management. In this study, in order to predict the daily flow time series of the Morghak hydrometric station in Karun basin, an intelligent model of artificial neural network combined with wavelet analysis has been used. For this purpose, the ERA-INTRIM observational and analytical precipitation time series for 16 years (1378-1382) was decomposed by wavelet transform into frequency subsets, then each subset separately as input data to the artificial neural network model was introduced. The results showed that the analytical data have a high ability to simulate runoff precipitation models and can be a good alternative to observation data of rainfall stations. Also, according to the results of the wavelet transform technique, it can be effective in improving the performance of the simple ANN model for the Bazoft basin by 38% on a daily scale and 72% on a monthly scale.

Biological indicators play a pivotal role in enhancing our comprehension of the ecological status and health of wetland ecosystems. This study investigated the ecological status of Chaghakhor International Wetland utilizing the biological value index Z at 10 water and sediment sampling sites. The Z index results revealed a Beta-mesosaproby status (Z = 2.30 ± 0.03) with moderate pollution. The findings from the artificial neural network model, displaying a determination coefficient of 0.943, indicated that elevated levels of parameters such as sediment electrical conductivity (553.97 ± 18.43), chemical oxygen demand (15.18 ± 0.44), biological oxygen demand (8.23 ± 0.52), and total dissolved solids (173.66 ± 7.63) in water significantly influenced changes in the biodiversity index Z. Human activities surrounding the wetland, particularly the expansion of agricultural lands, appear to have exerted a substantial impact on the present conditions. Consequently, improving agricultural land management, specifically through agricultural system reform, and preventing agricultural runoff from infiltrating the wetland environment are indispensable strategies for ecosystem preservation. Additionally, continuous monitoring of benthic communities in Chaghakhor Wetland and assessment of seasonal variations are crucial for formulating necessary conservation programs.

Increasing consumers demand for natural and additive-free foods and high volumes of food industry wastes, are stimulating the use of these resources in other food industries. Tomato pomace is one of the food factory wastes is the resulting by-product of tomato paste and sauce factories. The aim of this study was to evaluate the effect extraction method and microwave pretreatment of tomato seeds on the physicochemical characteristics of their extracted oil. The seeds were treated with microwaves using various power levels (0, 200 and 500 W) and different process times (0, 1, 3 and 5 min) and their oil was extracted by Soxhlet and press methods. Fatty acids composition of oils was determined by gas chromatography. Some physicochemical characteristics of extracted seed oil including oil yield, viscosity, acid value, peroxide value, and color index (L, b, a values) were evaluated. Data was analyzed with factorial treatment structure in a Completely Randomized Design in three replications. The experimental data was modeled by artificial neural network with 3 inputs (extraction method, microwave power and pretreatment time) and 7 outputs (oil yield, acid value, peroxide value, viscosity, L value, b value and a value). The results of artificial neural network modeling showed that the network with a 3-8-7 structure and using the Hyperbolic tangent activation function can predict the oil yield, acid value, peroxide value, viscosity, L value, b value and a value of tomato seed oil with high correlation coefficient and low error. Based on the results of the sensitivity analysis, the extraction method compared to the power and time of microwave assisted pretreatment of seeds was determined as the main factor.

Fish are cold blooded animals and their metabolism, growth and feeding are strongly dependent on water temperature. Temperature changes in fish breeding pools cause stress and disease outbreaks occur especially above the tolerance thresholds. The aim of this study is predicting pool water temperature from observed air temperate using several machine learning approaches, namely artificial neural network, gradient boosting and random forest in Gilan province.Maximum and minimum air temperature data of Rasht Agrometorological station for the period of June 2016 to November 2018 were collected and used for prediction of corresponding data of fish breeding pond .The obtained results showed that for prediction of the minimum temperature, the neural network model (with a root mean square of 1.93 and a correlation of 0.92) and for the pool water maximum temperature, the random forest model (with a root mean square of 1.61 and a correlation of 0.95) did a better job comparing to other two approaches. These selected models can be applied for prediction of water temperature using air Tmax and Tmin for improved management options under changing conditions.

Artificial neural networks (ANN) have been widely used for both prediction and classification tasks in many fields of knowledge; however, few studies are available on animal science. The objective of this study was to prediction of breeding values of weaning weight in Markhoz goats based on the Mixed Model Equation (MME) and Artificial Neural Networks (ANNs). Quality control and calculation of descriptive statistics was performed using the GLM procedure of the SAS statistical package. The pedigree file included 5541 kids produced by 261 bucks and 1616 does. In the first step, genetic evaluations and Best Linear Unbiased Prediction (BLUP) of breeding values for weaning weight was computed with the mixed model equations, animal model by DMU program. Later, unique dataset was introduced to the ANN models by the R statistical program. A variety of models including, multilayer perceptron (MLP), radial basis function (RBF) and Support Vector Regression (SVR) were used to build the neural models. The artificial neural networks were trained and several networks (different hidden layers and nodes/ neurons) were evaluated. In artificial neural networks, the data were randomly divided to two parts (75% training and 25% for test/validation). Best architecture was selected according to the mean square of error and correlation. Correlation between true breeding value (BVs predicted by MME) and estimated breeding value (BVs predicted by ANNs) for MLP, RBF and SVR models were 0.72, 0.49 and 0.73, respectively. Analysis of farm data showed that the MLP and SVR models have higher performance than RBF for prediction of breeding values or ranking of individuals.

Considering the importance of fresh water for human life and the vulnerability of groundwater sources to all kinds of pollution and the possibility of transferring pollutants to other surface and groundwater sources, as well as the location of Iran in the arid and semi-arid belt, protecting this valuable and rare element is imperative and its continuous monitoring must be one of the priorities of water resources managers. Therefore, in the current research, nitrate pollution in the eastern plains of Mazandaran province was discussed and relevant issues were investigated, and an efficient and optimal model for predicting nitrate concentration was presented. In this research, three machine learning models including decision tree, logistic regression and artificial neural network were compared. The physical and chemical data measured during the years 1985 to 2020 were used and entered as the input variables of the models. The variables include temperature, water level, pH, EC, HCO3-, CL-, SO24-, Na+, K+, Mg2+, Ca2+, TH and TDS; The amount of nitrate contamination of the groundwater, was predicted by dividing 70% of the dataset as training and 30% as testing data. The R2, RMSE, NSE and PBIAS indexes were applied for model evaluation. The results indicated that the Decision Tree model had the best performance with a large difference compared to the other two models (R2 = 0.957 and RMSE = 0.297, NSE = 0.95 and testing acc = 0.907). After That, logistic regression and artificial neural network had much weaker performances than the lead model. It is suggested to conduct another research with other machine learning models by changing the input variables and add some extra ones such as land-use and compare the results with the current research.