جستجوی مقالات مرتبط با کلیدواژه « خوشه ‏بندی » در نشریات گروه « آب و خاک »

Identification of homologous hydrological groups is one of the fundamental topics in hydrology in both applied and research dimensions. One of the common methods to achieve homogeneous hydrological zones for estimating flood zones is the use of clustering methods. In this study, the using, evaluation and comparison of statistical methods and methods based on artificial intelligence for clustering Karkheh catchment have been investigated. SOM, K-meansand hierarchical clustering were used for the study area. In the following, the study of hydrological homogeneity of the obtained areas was evaluated using the linear torque method and the heterogeneity adjustment was evaluated using the methods proposed by Husking and Wallis. The results show that the study area can be converted into two clusters. The values ​​of homogeneity statistics for the first and second clusters were calculated to be 0.33 and 0.17, respectively, which indicates the homogeneity of each region. Due to the short statistical period in some stations, statistical shortcomings and the reliability of the results of regional frequency analysis, this method should be used in estimating floods in other catchments.

Nitrate is one of the pollutants that affects groundwater in most aquifers of the country. In order to investigate the behavior of hydrogeochemical parameters of groundwater in the Ardabil plain aquifer, the results of qualitative data of 139 wells that were measured in both wet and dry seasons during the water year 2012-2011 were used. Using XLSTAT software, contamination was identified and samples were classified using principal factor analysis (PCA) and hierarchical clustering (HCA). Factor analysis method led to the extraction of five factors affecting groundwater quality with a total variance of 68.09 percent. Also, the results of factor analysis showed that the first, fourth and fifth factors are due to geogenic processes and the second and third factors are due to anthropogenic processes. The results of the hierarchical clustering method divided the samples into three groups, each group having two subsets. The set related to cluster one is spread in the northern, eastern, southern and parts of the center, west by mix(agriculture-garden), medium pasture and dry farming as well as urban areas. The set of the second cluster is often spread in the central, western and urban areas of the aquifer, with mix(agriculture-garden) and residential areas. The set of third cluster has been expanded to southwestern and urban areas. Generally, the existence of dissolution and ion exchange processes on the one hand, as well as leaching of fertilizers and incomplete development of the sewage system in the aquifer on the other hand has provided the groundwater pollution.

Knowing the intensity and duration of rainfall can be useful in many environmental analyses, including the estimation of rain erosivity and soil erosion. There are various devices to record the intensity and duration of rainfall, but purchasing and maintaining them are costly and often requires an operator to take care of them. The present research deals with the feasibility of using the analysis of sound signals caused by the collision of droplets with surfaces and objects in nature to determine the intensity and duration of rainfall. For this purpose, in the laboratory of the Department of Soil Science, Faculty of Agriculture, University of Tabriz, in 2022, rain simulators were designed to produce rains of different intensities, then, the sound signals caused by the impact of raindrops with the metal tray that was placed under the rain were recorded and transferred to the computer for processing. Then, the frequency size of audio files was extracted in MATLAB software. The results showed that with the increase in rainfall intensity, the audio amplitude and frequency size of the audio signals increased. Then, the frequency measurements were automatically placed in two clusters in SPSS software using the two-stage clustering method. Then the mean and standard deviation of each cluster were calculated and according to the correlation of each with each other and with the intensity of rainfall, and in order to avoid the multi-collinearity phenomenon, only the average of the second cluster was used as the input of gene expression programming and linear regression models. In order to test the accuracy and correctness of the results obtained from the models, the coefficient of determination (R2), root mean square error (RMSE), geometric mean of error ratio (GMER), geometric standard deviation of error ratio (GSDER) statistics were used. The values of R2, RMSE (mm/h), GMER(mm/h) and GSDER (mm/h) for the gene expression programming model in the training series data were 0.97, 1.85, 1.11 and 1.09 respectively and for the validation series data were 0.96, 2.05, 1.14 and 1.12 respectively. While the values of the above criteria in the regression model were 0.94, 2.74, 1.25 and 1.34 respectively for the training series data and 0.92, 2.91, 1.28 and 1.37 respectively for the validation series data. The results of the above statistics indicate that the gene expression programming model is relatively more accurate than the regression and overestimation model, and the estimated data of the regression model is relatively more spread than the gene expression programming model.

Precipitation is one of the most significant climatic parameters; its distribution and values in different areas is the result of complex linear and nonlinear relationships between atmospheric elements-climatic processes and the spatial structure of the earth's surface environment. Classification of data and placing them in small and homogeneous zones can be effective in improving the understanding of these complex relationships and their results. In the present study, zoning and analyzing the distribution of rainfall in Iran concerning environmental factors was performed using the annual precipitation data of 3423 synoptic, climatological, and gauge stations in the country during the period from 1961 to 2015 and the altitude, slope, aspect, and station density data. After standardization and preparation of the data matrix, the optimal number of clusters was determined and the data set was entered into the neural-fuzzy network model (ANFIS-FCM). The results showed that the values of R2  and MAE  indices were 0.76 and 0.23, respectively which indicate the appropriate accuracy of the model. It was also found that in the four output zones of the model, environmental factors have a high impact on the spatial distribution of precipitation. In the first and third zones, the combination of high altitude and slope factors along with geographical proximity to precipitation systems has caused the average annual rainfall in these zones to be 318 and 181 mm, respectively. The mean annual rainfall has decreased to about 100 mm by the weakening of the role of environmental factors in the second and fourth clusters.

The water quantity and quality has always been one of the main challenges in the issue of allocating water resources for different uses. Water quality management requires the collection and analysis of large amounts of water quality parameters that will be evaluated and concluded. Many tools have been found to simplify the evaluation of water quality data, and the water quality index (WQI) is one of these widely used tools. In summary, the WQI can be defined as a number obtained from the combination of several quality parameters based on standards for its extraction. The aim of this study was to develop and introduce the new Surface water Drinking Water Quality Index (SDWQI) adopt the water quality parameters measured on hydrometric stations of Iran. In developing this index, criteria such as the availability of required parameters in most rivers and simple and accurate methods have been considered. Also, the ability to calculate with the minimum general parameters of water quality, simple calculations and in terms of the international standard WHO for drinking is one of the advantages of the introduced index.

For this purpose, 12 water quality parameters including Total Dissolved Solids (TDS), Electrical Conductivity (EC), Total Hardness (TH), pH, Chloride (Cl-), Sulfate (SO42-), Carbonate (CO32-), Bicarbonate (HCO3-), Magnesium (Mg2+), Sodium (Na+), Calcium (Ca2+) and Potassium (K+) have been used from Rudbar and Astaneh hydrometric stations located on Sefidroud river. Then initial preprocessing on data e.g. correlation analysis, and multivariate statistical methods including cluster analysis (CA) and principal components analysis (PCA) are used to selecting and weighting of water quality parameters using the “clustering” and “factoextra” packages in R 4.1.1. In order to develop the SDWQI were performed four steps including, parameter selection, sub-indexing, weighting and aggregation of the index. Also, in order to evaluate the index of the present research, the results of the SDWQI have been compared with the WHO drinking water quality index and Schoeller drinking water quality classification.

Correlation analysis between water quality parameters shows a significant correlation between TDS, EC and TH parameters and also with Cl-, Ca2+ and Mg2+ parameters at the level of 1% in both Astaneh and Rudbar stations. On the other hand, the lowest values of Pearson correlation coefficient are related to pH and CO32- parameters with other quality parameters. The results of CA indicate that most of the water quality parameters are located in separate clusters. So only the parameters TDS, EC, Cl- and Na+ in both Rudbar and Astaneh stations are in the same cluster. The weights of the parameters showed that TDS and K+ are assigned with the highest and lowest weights equal to 0.163 and 0.031 based on PCA method. Also, PCA results show that first and second principal components covered 59.3% and 67.6% of the total variance of measured water quality parameters in Rudbar and Astaneh stations, respectively. Water quality classification results indicate that (40.5%, 16.4% and 23.7%) and (90.1%, 73.1% and 57.3%) of data in Rudbar and Astaneh stations, respectively, fell into the excellent and good categories for drinking purposes based on Schoeller classification, WHOWQI and SDWQI.

Generally, the comparison of the SDWQI with the WHO index and the Schoeller classification shows the rigidity of the new index in the classification of water quality for drinking purposes. Each water quality index developed in order to evaluate the uncertainty of results, should be tested for data with different characteristics in terms of the range of variation with different limit values​​ (minimum and maximum). The index developed in the present study is no exception to this rule and in order to better evaluate the results, it is suggested that to be evaluated and analyzed with data from other hydrometric stations. Another important points that should be considered in using any water quality index, including the present research index, is to examine the allowable limits of water quality parameters that are not considered in these indicators. The results of the study indicated that, two most important steps in the development of a quality index that have a great impact on its results are sub-indexing and weighting of parameters. According to the results, two ideas recommended for future research. One, choosing an appropriate method such as non-deterministic (fuzzy) and intelligent (machine learning) methods to sub-index the parameters and two, to weigh the parameters more effectively, multivariate statistical methods such as clustering, factor analysis and principal component analysis should be used.

One of the most important crises that most countries are currently ‎facing is the issue of reducing the quality of water resources. Reducing groundwater resources and ‎increasing pollution, reduced the potential for using groundwater for various uses. One of the main ‎reasons for the decline in groundwater quality is the impact of agricultural drainage due to the ‎excessive use of fertilizers and pesticides. Also, industries that in many cases have contaminated ‎groundwater with chemical and hydrocarbon contaminants.‎‏ ‏In addition to these factors, the disposal ‎of sewage in cities and villages through absorbent wells made up of viruses and bacteria also ‎contributes to contaminants‏.‏‎ Groundwater quality monitoring program can ensure the proper quality ‎of water resources for different uses‏.‏‎ Without monitoring, continuous reporting on the quality of ‎the water supply, its evolution, planning for optimal allocation for different uses, assessing the ‎impact of new developments, and designing and implementing management plans is not feasible.‎‏ ‏Identification of homogeneous regions in terms of groundwater quality in Golestan province of ‎northern Iran using Fuzzy Clustering Method combination with genetic algorithm (GA-FCM) was ‎performed on 14 parameters in a 5- year- time step in 2006, 2011 and 2016.‎

To determine the homogeneous regions for each year, the optimal number ‎of clusters was initially obtained. After data clustering in Matlab software, the results of clustering ‎were evaluated qualitatively with Schuler and Wilcox diagrams. For better representation of ‎homogeneous regions, classification maps for the study area were presented.‎

The results showed that the optimum numbers of clusters in 2006, 2011, ‎and 2016 were 6, 5, and 6, respectively. Analysis of groundwater quality classification maps ‎showed that in 2006, cluster no. 6, including 2.7% of the studied wells located within the city of ‎Kalaleh, is poor in terms of drinking and farming groundwater quality. Also, based on the results, it ‎can be seen that 36.8% of the wells across the province were in good condition in terms of quality ‎of drinking and agricultural parameters in 2011. Likewise, 33.33% of the wells are in a moderate ‎condition in terms of drinking quality, and the status of their groundwater has improved in terms of ‎quality since 2006. Also, the results of NSGA- FCM in 2016 showed that most of the parameters ‎‎(5.55% of the wells in the province) in the cluster 3 have a moderate quality.‎

The findings of this study showed that the groundwater quality in the province in ‎‎2016 is lower than in 2011, so appropriate management plans should be adopted. Moreover, it was ‎observed that the fuzzy clustering method is a suitable method for assessing and identification of ‎critical region of the quality of groundwater resources, since it considers the uncertainty conditions ‎in the classes of the classification system.‎

In drought risk management, the regional analysis of drought is significant. In this paper, this important issue is investigated by presenting the new hydrological regional drought index (RDI). For this purpose, the Ajichai basin was selected as the study area. First, the time series of the streamflow drought index (SDI) was calculated for each of the hydrometric stations in the basin f regional analysis of hydrological drought. Then, to determine the homogeneous regions in terms of hydrological drought, the k-means method was used for clustering analysis. Based on the clustering results, 6 Homogeneous regions were identified in the basin. For each of these regions, the time series of the RDI index was calculated from 1365 to 1393. The results showed that during the study period in each of the regions 1, 2, 3, 4, 5, and 6, mild Wet and mild drought has occurred at 82.1, 80.1, 78.9, 83.3, and 84.3 percent of regions, respectively. Also, the total percentage of drought events (moderate and high) is higher than the total percentage of wet events (moderate and high) in all regions. So, during the study period, the total percentage of drought events (moderate and high) is more than twice the total percentage of wet events (moderate and high) in regions 2 and 3.

Non-Revenue Water(NRW) is one of the most important problems that beneficiaries companies are trying to reduce. Most of researches have been done in this subject, focused on actual losses and leakage in water conveyance and distribution systems and less attention has been paid to the second component of NRW and Precision of Measurement Equipment (PME) yet. In this research, due to the lack of a clear scientific knowledge about the replacement of Water Meters(WM) in national water and wastewater engineering companies, field studies have been carried out on the PME and errors of subscribers' WM and the performance range of the WMs was compared using a new approach. At first, the identification of effective parameters on the accuracy of the WM was investigated and then by clustering method, the WM were compared and the effect of each parameter on the accuracy of the WM was determined. In order to predicting the performance and accuracy of the WM, Artificial Neural Network(ANN) algorithm was used and for optimizing the weights and biases matrix, Genetic Algorithm(GA) was used too

Mapping and spatial analysis of wheat fields are very important in studying macroeconomic and social issues, including agricultural management. Highly variable crop pattern maps and its preparation using terrestrial data are associated with many problems. The purpose of this study is to implement a practical method for extracting wheat fields by using changes in vegetation index and spatial analysis of wheat fields in western of Iran. Investigation of the changes curve of vegetation index of wheat typic farms showed that the highest amount of reflection index of wheat farms is in June and early July and after harvest the reflectance index decreases extremely. In this regard, Sentile sensor data was processed in the Earth Engine system and the 12-month vegetation index of 1398 was extracted as a data set. By introducing training data to the data set created by the support vector machine classification method, the land use of the study area was obtained in five classes. By applying altitude filter and removing the extracted fields above 3,000 meters, the distribution map of wheat fields was verified with the remaining 48 ground data. The total accuracy and the kappa coefficient were obtained 0.86 and 0.79, respectively. Since in the proposed method, more training data are given to the algorithm, the overall accuracy of the classification is increased. The spatial pattern of wheat fields with the mean function of the nearest neighbor and P_value <0.05 indicating the cluster dispersion of the fields and the Caripley function indicating the non-random scattering of wheat fields up to distances of 21,000 meters. The results of this research and its output maps can be used to obtain information for agricultural planning as well as the allocation and spatial distribution of resources and facilities.

In this study, the new method of self-organized maps (SOM) was used to cluster the piezometers in the Urmia aquifer and identify piezometers with similar behavior, in consequence the optimal groundwater monitoring network was determined. To achieve this goal, at first the normalization of monthly groundwater level data over a period of 13 years (2002-2015) was done and then the K-means non-hierarchical clustering algorithm was used to determine the number of optimal clusters. Then, using the SOM model which is one of the most widely used neural network models in the clustering techniques, spatial preprocessing operations were performed for all piezometers and the clusters were detected using a groundwater depth map. Also, the representative piezometers of each cluster were determined using the piezometers Euclidean distance. According to the results of clustering and its good compatibility with changes in groundwater level at each piezometer, it should be noted that the SOM clustering algorithm has a high capability for clustering. Results show that this type of network is able to detect the appropriate data patterns that can help identify the characteristics of members of each cluster. Therefore, it seems that the SOM clustering algorithm can be used to analyze the quantitative behavior of the aquifer (groundwater level changes) and find a few representative piezometers, thus providing an acceptable assessment of aquifer behavior in a short time and at the lowest cost.

Grouping of catchments based on their climatic factors and physiographic characteristics is a prerequisite for regional analysis of runoff and its use for estimating discharge of catchments without discharge measurement station. In this study, catchments located in Ardabil province were separated into homogeneous hydrological zones using discharge, mean annual precipitation and physiographic characteristics of catchments and hierarchical clustering method. Due to the large number of parameters, by using principal component analysis, the first four components with 83.6% of total variance were selected as inputs for cluster analysis. Then, the optimal number of clusters was determined by using hierarchical method and drawing the tree diagram, and finally the final clustering was done by K-means method. Subsequently, the sub-catchments that followed a hydrological process were identified using the Dalrymple uniformity test. The results of the uniformity test showed that by excluding stations outside the confidence limits of each cluster, catchments that were similar in terms of annual discharge and other physiographic and meteorological parameters were clustered. Therefore, the number of sub catchments located in clusters 1, 2, 3 and 4 were 8, 4, 9 and 9 catchments, respectively, which can be used in regional analysis to estimate runoff and floods in catchments without discharge measurement data.

Long-term planning and proper management of groundwater resources utilization are essential to ensure a reliable supply of water to countries, especially in arid and semi-arid regions. Therefore, it is necessary to employ appropriate models to predict the spatial and temporal fluctuations of aquifers and their future behavior. This study aimed to apply zoning strategies to Miandoab aquifer and predict its spatial and temporal groundwater level using an artificial neural network. First, the six parameters of transmissivity coefficient, groundwater level, ground elevation, withdrawal, rainfall, and discharge were spatially clustered to identify their effect on the simulation model. Three clustering approaches of single-parameter, three-parameter and integrated-parameter were evaluated using some statistical indices. The number of suitable clusters was determined using silhouette width. Groundwater level data (2002-2012) from 77 observational wells were used for model training and validation. Results showed that the correlation clustering approach performs better than the other methods. Precipitation, aquifer recharge, aquifer discharge, and groundwater level of the previous month were inputs to the back-propagation artificial neural network (ANN) for predicting a two-year period of groundwater level. The results showed that the correlation coefficients of variation in 6 clusters were 0.71- 0.97, and the RMSE variations were 0.19 - 0.58, indicating appropriate accuracy of this approach for predicting groundwater level.

Rainfall is among the most important climatic factors affecting the rainfed cultivation. Thus, in order to maintain water consumption in current agriculture, with the view of water resources management, the country needs to convert some irrigated land areas to rainfed cultivation in the near future. Indeed, it is necessary to conduct an analytical study on rainfed agriculture and identify appropriate areas for rainfed agriculture in the country, especially in Urmia Lake basin. Principal component analysis (PCA), K-Means and Ward have been already used to assess climate regionalization in different regions such as Spain (Diaz and Rodrigo, 2004), Greece (Kitsara et al, 2005), central-northeastern region of Mexico (Pineda-Martinez et al, 2007), Luanhe basin (Hassan and Ping, 2012) and Iberian Peninsula (Parracho et al, 2015). This study was, thus, intended to study the regionalization of the eastern part of Urmia Lake basin on the basis of the precipitation and yield of rainfed wheat using PCA, K-Means and Ward methods. To that end, the maps were drawn in the GIS environment and three methods of clustering were compared. Finally, using the clustering of precipitation and rainfed yield, wheat cultivability was investigated in the eastern part of Urmia Lake. To that end, the daily rainfall dataset of 26 rain gauge stations were used and the yield of rainfed wheat was considered during the period. Then, PCA, K-Means and Ward clustering were performed and the results were compiled. The homogenousity of the resulting clusters were analyzed by H and S statistical tests and homogeneous clusters were drawn in the GIS environment. The analytical factor coefficients to the main components, through K-Means clustering method, showed that the clusters point of view, precipitation and rainfed yield were more consistent and the results were close to each other.

Considering the importance of precipitation and scarcity of water resources, spatial analysis of daily rainfall together with corresponding time intervals is one of the requirements. Among the atmospheric factors affecting the rainfed farming, precipitation is the most important factor in numerous studies about the rainfed cultivation. In order to reduce the destruction threat of water resources and resolve the future’s food needs of the people, the importance of rainfed agriculture would be inevitable. In this regard, one of the effective measurements that can be done to find rainy areas with the possibility of rainfed cultivation. The greatest water consumption is corresponding to the agricultural sector. According to the water crisis in the country, decline in the groundwater levels and the importance of agriculture in job creation, the water resource management is essential. Therefore, it will have a value of higher management to determine for the fertile lands in the current situation. Despite the reduction in rainfall, it is also necessary to analytically study the rainfed and determine the fertile lands in the country, especially in the Urmia Lake basin. This study aimed to introduce an index for identification of the suitable areas from the viewpoint of rainfed ability, by using daily rainfall in corresponding with the time intervals for each year in the east part of the Urmia Lake basin. First a program is written in Fortran, using the rainfall data, the proportion of rainfall intervals is determined, then the Fortran program is developed for this propose, based on the concept of the temporal intensity precipitation and their related mechanistic, an index is defined which is the main outcome of the Fortran programming. The index was a criterion for analysis and disintegration of the rainfed quantity. The idea of introducing an index is based on the ratio of rainfall to the interval corresponding. If the ratio quantity was high, it represents the water supply, in other words, the rainfed capability will be great which can play the role of an index rating. In this index, the rainfall depth has positive role and intervals have negative role in the assessment. When the index amount is high, the rainfed capability will increase. Because the amount of precipitation must be bigger (which is the sign of the large amount of rainfall or the smaller interval time) an increase in the amount of rainfall would be followed, or in compound, causes an increase in the proportion, which shows the increase of rainfed ability. The statistical period was 1992-2014 for 23 rainfall stations. The index values were calculated for each year of all stations. Clustering is one of the most useful classification methods. In the cluster analysis, one attempts to actual observations of each cluster, which have the most similarity in terms of variables together. In this study, the clustering methods of the K-Means and Ward were carried out to deline the homogeneous regions based on the developed index values. Clustering in the rainfall stations was considered as a variable, the correlation matrix 23 * 23, where 23 is related to the number of the years stations used. The homogeneity of the clusters were checked through the H-Statistics method and the homogeneous clusters were shown in the GIS environment. In order to use the K- Means method, the whole study area was considered as two clusters and the results of the H-Statistics homogeneity test showed that the two clusters are homogeneous. The first cluster has a mild homogeneity while the second one is completely homogeneous. In the clustering index using the Ward method, the study area was divided into two clusters. The observation of thehomogeneous clusters with the H-Statistics homogeneity test showed that the both clusters are mild homogeneous. On the other hand, from the viewpoint of spatial variation, Iso-Index lines were drawn over the study area. Identifing the rainy areas was conducted by examining the lines of these areas w are suitable for the rainfed agriculture and have a better temporal rainfall distribution. The comparison of two maps regionalization with two methods and the Iso-Index lines show that the regionalization by K- means method, the first part of this regionalization with Iso-Index lines have the same incremental direction, that suggests more rain stations in this area in comparison with the second area. The second cluster of homogeneous region was obtained using the Ward method, having similarity with Iso-Index only in the limited stations. The comparison of the regions resulted from clustering methods with the spatial distribution of the Iso-Index lines over the study area implied that the K-Means method isolated the regions better than the Ward method. The results show that the south, northeast, northwest and the center parts of the study area were more eligible for the rainfed agriculture than the other parts. These areas, regarding to agriculture, temporal distribution of precipitation and richness of groundwater, are better.

Water shortage and limited water resources, reveals the importance of optimum utilization of irrigation water networks more than ever. In most of the previous studies performance improvement options are given based on individual networks condition, without considerations of providing common solutions for other networks. In this study similar Iranian irrigation networks are classified in different clusters based on the problems of control structures in the main canals, and common improvement options are given. The classical K-Mean clustering method is used for classification. For this purpose the criterion are analysed using AHP technique. The criterion with more than 7 percent rankings are selected for clustering. The most important criterion are network’s covered area, age and capacity of the networks, and No. of structures. 32 of Iranian irrigation networks are clustered in 7 clusters. 38 questionaires collected by Iranian National Committee of Irrigation and Drainage were analysed. The common problems of the networks were determined for each cluster. The most common problems are determined to be operational deficiencies, repairmen requirements, and farmers intervention and manipulation in operation. The common improvement options to resolve the identified problems for each cluster are given.

In this research work taxonomic distance and clustering performance for understanding of relation and correlation between calcareous, gypsiferous and saline soils of East Azerbaijan semi-arid and arid regions were studied based on soil taxonomy (ST) and world reference base (WRB) systems. The taxonomic distance by R software and Numerical Taxonomy in Excel and clustering using R and SPSS softwares were calculated and their relationship in two systems was investigated. Then using the average amount of physicochemical quantitative data, the results of calculation and clustering compared with each other and so with opinion of the experts, that there was a good quantitative correlation between salids and salonchaks, salonetz, gypsids and gypsisols, argids and luvisols, xerepts and calcisols and gypsisols. The obtained results of four methods were almost similar and identical with each other, but use of each one had its own limits and skills. The conceptual method was based on the dominant qualitative identifiers (soil-forming factors or processes, diagnostic horizons and in general dominant morphological properties of calcareous, gypsiferous and saline soils) and coding of them, that sufficient information or knowledge and skills are necessary to achieve a good accuracy in this method. The quantitative physicochemical data were used in the centroid-based approaches numerical taxonomy and clustering and results of these methods are more reliable. The easiest method was Numerical Taxonomy, which with simple options and without any programing, forms matrix and by normalizing data, the Mahalanubis equation makes the taxonomic distance easily available. In general, these tools and calculations are useful for finding the relationship between soils in different classification systems, and quantitatively numerical findings for their correlation and association in classification systems, especially ST and WRB, which are widely used in Iran are recommended.

The aquifers are about four percent of the water on the earth, but they are considered as the best and most accessible source of fresh water. In recent years, they have been faced with severe water withdrawal, therefore some plains was considered as forbidden plains that it means the water withdrawal from these aquifers is unauthorized. At some point, plains have been faced with land subsidence that showed the severity of the disaster. Given such a critical situation in aquifers, management of groundwater resources in the form of tools such as monitoring the level of the aquifers is essential. One of the plains in Iran that has a critical groundwater resource is Birjand plain which requires management measures to be protected from future water resources crisis. Prediction of groundwater level in future periods is a useful tool to enforce management measures before a crisis occurs. Thus, in this study, groundwater level was predicted in Birjand aquifer taking 10 monthly forecasting scenarios in a period of 5 years and in both crisp (continues) and clustering approaches using probabilistic Dynamic Bayesian Networks (DBNs). Nowadays, various tools are used to predict the aquifer level including mathematical models, artificial neural network, neuro-fuzzy, Bayesian networks, and time series and so on. In recent years, due to the flexible and simple structure, Bayesian networks have been used for predictions of different parameters, especially in forecasting of hydrological parameters. Bayesian network as a modern forecasting probabilistic method shows probabilistic relationships between a set of variables by graphical model. It represents the dependence structure among several factors, that affecting on each other, and is based on Bayesian theory. Dynamic Bayesian Networks have been extended from Bayesian Networks which are created for two first, as the cycle of dependency detector over the time, quite similar to Markov model; second, as the fixed process which is repeated in fixed-time interval. Another feature of Dynamic Bayesian network is their willingness to approximately structured changing. In this study, the input data (predictor parameters) of the model includes: temperature, rain, evaporation, monthly aquifer recharge in each Thiessen Polygon (Recharge), monthly withdrawals of groundwater in each aquifer Thiessen Polygon (discharge), groundwater levels in the current month and the groundwater level in the next month (predicted parameter). The first step in modeling by the dynamic Bayesian network is determining the dependent and independent data for calibration and validation. Model calibration data in both crisp and clustering approach include a 12-year period (1998 to 2009) and data validation include 5-year period (2010 to 2014), in monthly time step. Depending on the type of input data, both crisp and clustering approach was used. In the crisp approach, the probability functions were used and the predicted data were obtained by using the training data. In the clustering approach, by assigning each of the numbers to the right cluster, the modeling was done. A cluster is collection of objects that their relative distance to each other is low and relative to other members is high. In the first approach by using crisp data and NPC training structure in confidence level of 5%, the training was applied. In the second approach, the Silhouette index was calculated by using MATLAB software and by using the validation Silhouette index, and then, the number of clusters was determined. Finally, the clustering was performed by using k average method. Then, training was done in the second approach, using clustered data and NPC training structure in confidence level of 5%. Considering 10 different scenarios to predict groundwater level in the next month, the uncertainty of predicted parameters in both crisp and clustering approach were assessed. In fact, using these scenarios, sensitivity analysis was performed to check the accuracy of the model with respect to the existence or absence of different predictor parameters. In addition, the uncertainty of the model output is evaluated using dynamic Bayesian network probabilistic analysis. The results of the selected scenario in crisp approach showed the high prediction accuracy of Bayesian networks. For example, in piezometers 4 and 9, the coefficient of determination was estimated about 0.98. According to the results, crisp dynamic Bayesian networks approach predicted hydrograph aquifer more accurate than clustering approach. Due to the low efficiency of clustering approach in predicting groundwater hydrograph, to obtain the accuracy of 10 scenarios predictions in this approach, instead of coefficient of determination (R^2) and root-mean-square error (RMSE), the percentage of correct predicted clusters was determined. According to the results, the clustering approach predicts clusters with high accuracy for different piezometers. The scenario 6 had the best prediction which all predictor parameters except evaporation were used for the prediction. In this scenario, R^2 and RMSE were showed good accuracy as 0.9946 and 0.1275, respectively. The other scenarios had also the accuracy in their predictions very close to scenario 6, except scenario 9 which the groundwater levels were not used as input. Thus, in the crisp approach, the groundwater levels have a substantial impact on the accuracy of the prediction. Generally, in crisp approach, all predictor scenarios have acceptable accuracy of over than 90% except scenario 9. In clustering approach, by examining the accuracy of the scenarios in predicting clusters, most scenarios were accurate except scenario 9. Dynamic Bayesian Network model in clustering approach correctly predicted clusters but, unlike the crisp approach in predicting hydrograph, it was not able to present acceptable results. The results showed the ability of the proposed model in planning and management of groundwater resources, reducing the risk of aquifer level declining by applying short term management scenarios and predict its effects on rehabilitation. Moreover, this model can be used in the similar plains for aquifer management.

During the last decades, runoff decreasing is observed in our country as many dam reservoirs face water supply crisis even in normal periods. This decreasing trend is mainly due to the uncontrolled withdrawals, lack of supply and demand management as well as droughts. Using different flow prediction methods for surface water resources state analysis is important in water resources planning aspects. These methods can provide the possibility of planning for proper operation by using different factors to meet the needs of the region. Due to the stochastic nature of the hydrological processes, various models are used for prediction. Among these models, Bayesian Networks (BNs) probabilistic model has been considered by many researchers in recent years and it has shown the efficiency on these issues. Due to the growth of demand in different sectors and crises caused by drought of the water supply system that has put the basin under water stress, the water shortage has appeared in different sectors. Regarding to the strategic situation of Zayandeh Rood Dam in providing water resources for tap water, industry, agriculture and environmental water rights in Gavkhooni basin, this research presents the development of a model for prediction of Zayandeh Rood Dam annual inflow and hydrological wet and dry periods. Since the uncertainty of the predictions increase when the prediction horizon increases, this factor is the most important challenge of long-term prediction. Using Bayesian Network with reducing this uncertainty, provides the possibility of planning for water resources management, especially for optimal water allocation.

In this study for prediction of zayandeh Rood dam inflow five scenarios were defined by applying Bayesian Network Probabilistic approach. According to this, prediction of numerical annual dam inflow (scenario1), annual wet and dry hydrological periods (scenario 2, 3, 4) and range of annual inflow (scenario 5) were performed. For this purpose rainfall, runoff, snow, and discharge of transferred water to the basin from the first and the second tunnel of koohrang and Cheshmeh Langan tunnel were considered as predictor variables and the amount of Zayandeh Rood Dam inflow was selected as predictant for modeling and different conditions of input variable’s learning have been analyzed considering different patterns. Calibration and validation of the model have been done based on observed annual inflow data and the relevant predictors in scenario 1, by using SDI Hydrological drought index and long-term average of inflow to classify the runoff and clustering the other parameters in scenario 2, 3 and 4 and with classification of annual inflow data and other parameters by using clustering in scenario 5. To achieve this target, K-means method has been used for clustering and Davies-Bouldin and Silhouette Width has been used to determine optimal number of clusters.

The results of Bayesian Network modeling showed that the scenario 1 has a good potential to predict the dam inflow so that the best pattern of this scenario (considering discharge of first tunnel of Koohrang and Cheshmeh Langan tunnel, Zayandeh Rood natural inflow and rainfall with two years lag time as predictor variables), has had a correlation coefficient of 0.78 between observed and predicted dam inflow and relative error of 0.21 which shows an acceptable accuracy in prediction. Among scenarios 2, 3 and 4 for prediction of wet and dry hydrological periods, scenario 2 in which classification of runoff has been based on the long-term average, in the best pattern (with dam inflow with one-year lag predictor), is able to be predicted up to 75% accuracy. The analysis of the results showed that the scenario 5 is not very accurate in prediction of dam inflow’s range.

The results showed that the Bayesian Network model has a good efficiency to predict annual dam inflow numerically as well as hydrological dry and wet periods. Obtained results from prediction of hydrological dry and wet periods will be effective in better planning of water resources in order to considering possible ways of drought effect reduction. The overall results provide the possibility of water resources planning for the water authorities of this region. Systematic planning leads to optimal use of water and soil resources and helps considerably to analyze and modify the policy or rule curve of this dam for allocating water to downstream especially for agriculture and environment and industry sectors.

For reducing the destruction of water resources and for resolving the future populations, needs for food, the replacement of rainfed farming instead of water farming will be inevitable. According to the importance of the agricultural sector, determining indices to identify the fertile lands for cultivation of rainfed wheat is essential. For this purpose, to evaluate the yield of rainfed crop at the East part of Lake Urmia, two indices Interval Ratio and Partial Interval Ratio, for identification of the eligible areas for rainfed farming, were introduced using irrigation depth, irrigationintervals, and daily rainfall in corresponding with the time intervals. Clustering of the indices and yield of rainfed area were done with a K-means method. Regionalization of the obtained clusters for the yield and indices were done in GIS environment. Results of the regionalization showed that between the two introduced indices the Partial Interval Ratio index has the more compatibility to region reality, So the Partial Interval Ratio index selected for evaluating the rainfed regions. To identify the fertile rainfed regions of crop yield, the isoline of the yield of rainfed and index (Partial Interval Ratio index) was used. The results showed that in the northwest and central parts of the studied area the yield and Partial Interval Ratio index were increased, which indicated the good potential of rainfed farming in these areas. Therefor the fertile regions based on the introduced index were at the central and northwest parts of the studied area.

The purpose of this study was to cluster the observation well networks in Marageh Plain in East Azarbaijan and to predict the groundwater level by artificial neural networks. Primarily, by Hierarchical WARD clustering method, 20 observation wells of Maragheh Plain with over 15 years data period were clustered. Then, a cluster with 3 homogenous subclusters was selected and the representative of each subcluster was determined. Artificial neural networks with a multilayer perceptron structure utilizing back-propagation algorithm was used to simulate the representative groundwater level of each subcluster. The results indicated considering monthly temperature data as input for the artificial neural networks caused disorder of the network while considering lag time for the input data increased the accuracy of the estimated groundwater levels. Based on the results, the minimum and maximum RMSE between the observed and calculated values were 0.26 m and 0.63 m, respectively. Also the Maximum and minimum quantities of R2 were 0.86 and 0.82, respectively.