It is more expensive to remove pollution from groundwater than to prevent it. Delineation areas that arevulnerable to surface pollutants is one of methods to prevent pollution of groundwater resources. Focusing on this issue, DRASTIC model was used for evaluation of vulnerability of Tabriz-plain aquifer to pollution and the aquifer vulnerability map was prepared. The study shows that main zone of the aquifer’s groundwater is low tomodrate vulnerability to pollution (DRASTIC Index of 120-40) that consist of about 55.84% and areas with low, moderate to high, and high risk zones comprise 21.81, 22.08.% and 0.26% of the studied area, respectively Two tests of sensitivity analyses were carried out: the map removal and the single-parameter sensitivity analyses. Based on the characteristics of the studied area, the results from both map removal and single-parameter sensitivity analyses showed that the depth to water table has the most significant impact on the vulnerability risk zone. By overlaying of the vulnerability and landuse maps the areas where are subjected to potential release of pollutants from the agricultural activities were determined. Nitrate ion concentration and SINTACS model confirms the results of the vulnerability assessment.

Currently, uncontrolled extraction of groundwater resources has led to a decline in the water table of aquifers, especially in arid and semi-arid regions. Today, one of the ways to improve the water table of aquifers is the use of artificial recharge plans. This study has been done to numerically simulate the artificial recharge plan of Birjand plain through the construction of injection wells using the numerical model of Isogeometric analysis. In this regard, first a standard example with two wells with discharges of 1142. 85 and 1428. 57 m3/day was used to simulate the water table for 210 days. After evaluating the accuracy of the model, the water table before injection was simulated based on the existing conditions and after injection using 20 injection wells with a constant flow of 7638. 44 m3/day in 12 time steps in the Unconfined aquifer of Birjand plain. Comparison of the results of Isogeometric analysis and analytical solution model in standard aquifer with estimated evaluation criteria equal to ME=-0. 0096, MAE=0. 0111 and RMSE=0. 0146 and also in Birjand plain Unconfined aquifer with evaluation criteria of ME=-0. 033, MAE=0. 372 and RMSE=0. 229 shows the model accuracy in simulating the water table before injection. Also, after running the numerical code of the injection wells construction plan, by interpolating the water table obtained at the control points in the location of 10 existing observation wells, the results at the end of the simulation period shows increasing the water table maximum 60. 53 cm in observation well number 2 and minimum 1. 25 cm in observation well number 9.

The water movement in the aquifer is studied using the basic equations of flow, i. e. Darcy, continuity and mass survival equations, which are differently analyzed in the case of confined and Unconfined aquifers. In these equations, hydrodynamic coefficients and other characteristics of the aquifer are used. Therefore, the determination of hydrodynamic coefficients of aquifer is necessary for groundwater studies. The most important characteristics of the aquifer i. e. hydrodynamic coefficients can be determined across the aquifer using modeling. If the aquifer model is properly provided, it can be a useful tool for better recognition of aquifer characteristics and also predicting the impact of different management plans on the water resources of the region. In this study, groundwater modeling was performed to determine the hydrodynamic coefficients in the Unconfined aquifer of Kermanshah plain using GMS 7. 1 software. The results showed that the amount of hydraulic conductivity in the Kermanshah plain aquifer varies from 0. 1 to 30 meter per day and the amount of specific yield varies from 0. 02 to 0. 4. The amounts of computed water table by the model and the observed water table in the piezometers have high correlation and the root mean square error value is always less than one, which indicates the acceptable accuracy of the model for simulating the aquifer status.

Having a pervasive knowledge about the groundwater balance conducts the engineers and managers to have an optimal management about these resources. In the present study, with using meshless local Petrov-Galerkin, one of the meshless methods in the field of fluid dynamic, groundwater balance is computed for a real field aquifer. The independency of this method from meshing the domain, removes the drawbacks and errors come from meshing state and improves the accuracy of results. The case study is Birjand Unconfined aquifer located in South Khorasan province, east of Iran. After groundwater flow simulation and model calibration in one year 2011-2012 with monthly time step, the groundwater head is computed for each month and compared with the results of finite difference method (FDM). RMSE, MAE and ME criterion are calculated for both methods. They are 0. 757m, 0. 573m,-0. 08m for MLPG method and 1. 197m, 1. 434m, 0. 159m for FDM, respectively. The higher results of MLPG shows more correspondence of this method to real condition of aquifer. Finally, the input and output volume of water are computed for the aquifer. They are almost 91 and 88 MCM respectively. Therefore, the value of groundwater balance is computed. It was 3 MCM with minus sign which indicated the higher amount of output regarding to inputs.

In this paper, an Artificial Neural Network (ANN) is designed for the determination of Unconfined aquifer parameters: transmissibility, storage coefficient, specific yield, and delay index. The network is trained for the well function of Unconfined aquifers by the back propagation technique and adopting the Levenberg-Marquardt (LM) optimization algorithm. By applying the principal component analysis (PCA) on the training data sets the topology of the network is reduced and fixed to [3×6×3] regardless of number of records in the pumping test data. The network generates the optimal match point coordinates for any individual real pumping test data set. The match point coordinates are then incorporated with Boulton analytical solution (1963) and the aquifer parameter values are determined. The generalization ability and performance of the developed network is evaluated with 100/000 sets of synthetic data and its accuracy is compared with that of the type curve matching technique by two sets of real field data. The proposed network is recommended as a simpler and more reliable alternative for the determination of Unconfined aquifer parameters compare to the conventional type-curve matching techniques.

Exploitation, management and protection of water resources, especially groundwater resources, are very important. The management of water resources requires accurate knowledge of the characteristics of the aquifer, the tension to it and finally the groundwater balance. Simulation models help planning managers to determine the amount of aquifer utilization for long-term and sustainable use. For this purpose, changes in the water level of the Kashan plain aquifer in the statistical period of 1387-1397 were simulated by the GMS model for 125 stress periods. The model was calibrated in two permanent and non-permanent modes for a period of 94 months (1387-1395) and validated for a period of 31 months (1395-1397).The result was a drop of 4.4 meters of groundwater during the 10-year study period. To predict the behavior of the aquifer in the near future, the reservoir level was modeled until 1407. Due to the location of this plain in the climatic dry belt and the occurrence of climate change and the lack of effective rainfall to properly charge the aquifer, in this study the focus was on the management of extraction from the aquifer and according to these three scenarios, extraction with the current trend,15% increase and 15% reduction definition and simulation was done. The simulation results of these three scenarios showed a drop of 9.866, 11.895 and 6.993 meters at the end of 1407 respectively.

Solving the governing equations of a system is the most important issues that is always discussed in science and engineering fields. Since there are few equations that have analytical solution, many numerical methods have been proposed for solving the equations that have no analytical solutions. Numerical methods are developed by the advent of computers. Today, with using computers and these methods together, complicated equations in diverse areas can be resolved. Several numerical methods such as finite element method (FEM), finite difference method (FDM) and meshless (MFree) method have been suggested for solving partial differential equations. In this study, Isogeometric analysis method is engaged as a numerical method. Isogeometric analysis was developed by Hughes in 2005 in order to eliminate the gap between the world of finite element analysis and computer modeling. This method uses the same basis functions, in the process of modelling. Isogeometric method provides the possibility of simulation in irregular and complex geometry domains and also removes errors due to the multiple elements. Two variable NURBS basis functions are defined by B-spline basis functions. B-spline basis functions are calculated by the Cox– de Boor recursion. In this study, Birjand aquifer is modeled in two dimensions by the Isogeometric analysis using four-point Gauss integration method. After creating the geometry of the aquifer by control points and knot vector, NURBS basis functions and their derivatives were calculated. Then, with using input information, such as hydraulic conductivity coefficients, boundary conditions, precipitation rates and the sources and sinks, water table is computed. In order to allocate hydraulic conductivity coefficients of the aquifer, the domain is divided by the GIS software to multiple homogeneous Thiessen. According to the location of NURBS elements in the aquifer, a value has been assigned to NURBS elements. In Birjand aquifer there are boundary conditions with constant head. There are 190 wells in the Birjand aquifer, the Extracted water from the wells were used as the discharge rate in the model. Also, 15 percent of the amount of rainfall was considered as the recharge rate in 2011-2012 period, the value of recharge rate is 0. 0000727 m/day based on rain gauges. In order to ensure the accuracy of modeling the results of Isogeometric method is compared with finite difference method solutions and observation data, the relative mean error of Isogeometric method is 0. 000256. In order to evaluate the model, three criteria is calculated. Mean error (ME), mean absolute error (MAE) and root mean square error (RMSE) whose values are 0. 09, 0. 34, and 0. 459 respectively. The values of the error and computation time has shown the power of this method in modeling of groundwater flow. Finally, Birjand aquifer groundwater balance was calculated using the input values, extracted water and water storage in plain. By studying the model balance and actual balance of aquifer and comparing them with each other, it is determined that the change in the volume of the aquifer in the time period considered is close to that of the aquifer, which indicates the accuracy of the model.