JWSS - Journal of Water and Soil Science

Statistical analysis and forecast discharge data play an important role in management and development of water systems. The most fundamental issues of statistical analysis and forecast discharge in Iran are lack of data in long term period and lack of stream flow data in gauging stations. Considering the issues mentioned in this study, we tried to estimate the daily data flow (runoff) of Santeh gauging station in Kordestan province using the nearby hydrometric and meteorological stations data. This estimation occurred based on the sixteen different input combinations, including data of daily flow of hydrometric stations Safakhaneh and Polanian and daily runoff in Santeh precipitation gauging station. In this research, the daily flow estimation of the Santeh station in each of the months of the year was evaluated for sixteen different combinations and artificial neural network models and multiple linear regressions. The performance of each model was evaluated with the indicators RMSE, CC, NS and t-student statistic. The results showed good performance of both models but the performance of the artificial neural network model was better than the regression model in estimation of the daily runoff in the most months of the year. Mean error of artificial neural network and multiple linear regression models was respectively estimated as 6.31 and 8.07 m³/s in the months of the year. It should be noted that the artificial neural network, for each sixteen combination used, had better result than the regression model.

Occurrence of local scour is one of the most significant causes of damage to the pipes. Therefore, safe and economical design of pipes in the flow path requires a good estimate. In this study, based on the important and effective parameters in the scouring phenomenon, in order to develop educational patterns according to the data obtained in the laboratory of Ahvaz Islamic Azad University, models based on artificial neural networks were created with the NeuroSolution5 software. MLP, GFF and RBF were the models used in this study; after comparing, MLP was selected as the basis for our study. Finally, the effect of each parameter on scouring was determined using the  artificial neural networks technique, based on which the  shields parameter with a very high effect (more than 95 percent) was determined as one of the most effective causes of the local scour.

Expert aquatic designers face many problems; among these, in hydrology, defective occurrences in time-series can cause errors in the ultimate results of the study. This more often happens in the regions where the number of hydrometric and rain gauge stations is limited. In addition, assessing, developing and maintaining the use of water resources require accessible long-term and high-quality quality hydrological time-series. Thus, this necessitates correcting the statistical flaws and magnifies the importance of how to deal with the problems in the hydrological analyses. Statistical methods are, currently, used to infill data and statistical gaps. In this study, in order to introduce a multivariate method for estimating the missing data on rainfall and runoff, in a hydrologic homogeneous region in the Mazandaran province, self-organizing map methods were examined under two scenarios and some reliable estimates were obtained. In this regard, the correlation coefficients between the observational data and the model output were calculated for the precipitation data up to 0.92 and up to 0.95 for the runoff data. Therefore, to avoid the reduction of uncertainty caused by the inadequate data in water resource management, this method could be used.

In this study, we used the ARIMA time series model, the fuzzy-neural inference network, multi-layer perceptron artificial neural network, and ARIMA-ANN, ARIMA-ANFIS hybrid models for the modeling and prediction of the daily electrical conductivity parameter of daily teleZang hydrometric station over the statistical period of 49 years. For this purpose, the daily data for the 1996-2004 period were used for model training and data for the 1996-2006 period were applied for testing. In order to verify the validity of the fitted ARIMA models, the residual autocorrelation and partial autocorrelation functions and Port Manteau statistics were used. PMI algorithm were   then used to model and predict electrical conductivity for selecting the effective input parameter of the neural fuzzy inference network and the artificial neural network. The daily parameters of magnesium (with two days delay) and sodium (with one day delay), heat (with one day delay), flow rate (with two months delay), and acidity (with one day delay) were obtained with the lowest values of Akaike and highest values of hempel statistics as the input of the neural fuzzy inference network and the artificial neural network for modelling daily electric conductivity predictions; then predictions were made. Also, models evaluation criteria confirmed the superiority of the ARIMA-ANFIS hybrid model with the trapezoidal membership function and with two membership numbers, as compared to other models with a coefficient of determination of 0.86 and the root mean square of 29 dS / m. Also, the Arima model had the weakest performance. So, it could be applied to modeling and forecasting the daily quality parameter of the tele Zang hydrometer station.

The simulation of the rainfall-runoff process in the watershed has particular importance for a better understanding of hydrologic issues, water resources management, river engineering, flood control structures, and flood storage. In this study, to simulate the rainfall-runoff process, rainfall and discharge data were used in the period 1997-2017. After data qualitative control, rainfall, and discharge delays were determined using the coefficients of autocorrelation, partial autocorrelation, and cross-correlation in R Studio software. Then, the effective parameters and the optimum combination were determined by the Gamma test method and used to implement the model under three different scenarios in MATLAB software. Gamma test results showed that today's precipitation parameters, precipitation of the previous day, discharge of the previous day, and discharge of two days ago have the greatest effect on the outflow of the basin. Also, the P_t Q_t-1 and P_t P_t-1 Q_t-1 Q_t-2 Q_t-3 combinations were selected as the most suitable input combinations for modeling. The results of the modeling showed that in the support vector machine model, the Radial Base kernel Function (RBF) has a better performance than multiple and linear kernels. Also, the performance of the Artificial Neural Network model (ANN) is better than the Support Vector Machine model (SVM) with Radial Base kernel Function (RBF).

Desertification is a phenomenon that has more destructive effects in arid, semi-arid, and semi-humid regions than in other regions. This paper tries to provide a map of the future of desertification in Tehran Province, for futurism in the face of land degradation and desertification. The IMDPA model was used to evaluate land degradation and desertification. To use this model and evaluate desertification, three criteria of groundwater including groundwater depletion, electrical conductivity, and sodium adsorption ratio indices, climate criterion including precipitation, aridity, and drought indices, and land use criteria were selected as key criteria effected on desertification according to regional conditions. Land use index map with IGBP standard and zoning map of other indicators were prepared by IDW method for 2011 and 2016. The maps of land use index and other indices were predicted using the CA-Markov model in TerrSet software, and using the RBF method in artificial neural network toolbox, respectively. Scoring based on the IMDPA model, the maps of indices and criteria maps were prepared for 2011, 2016, and 2021. Finally, the desertification intensity map was calculated by geometric averaging for all three criteria for all three time periods. The results showed that 59.78% and 40.22% of the area of Tehran Province were in the low and medium classes, respectively. However, in 2016, the area of the medium class has increased to a 44.8%, and it is predicted that this increase will continue until 2021 so that 47.65% of the area of Tehran Province will be in the medium class. In addition, in this year, about 1% of the area of Tehran Province will be allocated to the high class in the western regions, which did not exist in the previous two periods. In general, due to human activities, the intensity of desertification in the western and southern parts of the province is higher than in the eastern and northern regions.

An accurate and reliable prediction of groundwater level in a region is very important for sustainable use and management of water resources. In this study, the generalized feedforward (GFF) and radial basis function (RBF) of artificial neural networks (ANNs) have been evaluated for monthly predicting groundwater levels in the Dezful-Andimeshk plain in southwestern Iran. The partial mutual information (PMI) algorithm was used to determine efficient input variables in ANNs. The results of using the PMI algorithm showed that efficient input variables for monthly predicting groundwater level for piezometers affected by water discharge and recharge include only water level in the current month. Also, efficient input variables for predicting the water level for piezometers affected only by water discharge include the water level in the current month, the water level in the previous month, the water level in the previous two months, transverse coordinates of piezometers to UTM, the water level in the previous three months, the water level in the previous four months, the water level in the previous five months and longitudinal coordinates of piezometers to UTM. In addition, efficient input variables of monthly predicting groundwater level for piezometers neither affected by water discharge nor water recharge, respectively, include the water level in the current month, the water level in the previous month, the water level in the previous two months, the water level in the previous three months, the water level in the previous four months, the water level in the previous five months, the water level in the previous six months, transverse coordinates of piezometer to UTM and longitudinal coordinates of piezometer to UTM. The results indicated that the GFF network is more accurate than the RBF network for monthly predicting groundwater level for piezometers including water discharge and recharge and piezometers including only water discharge. Also, the RBF network is more accurate for monthly predicting groundwater levels for piezometers that include neither water discharge nor recharge than the GFF network.

Unpredictable settlement of earth dams has led researchers to develop new methods such as artificial neural networks, wavelet theory, fuzzy logic, and a combination of them. These methods do not require time-consuming analyses for estimation. In this research, the amount of settlement in rockfill dams with a central core has been estimated using artificial intelligence methods. The data of 35 rockfill dams with a central core were used to train and validate the models. The artificial neural network, wavelet transform model, and fuzzy-neural adaptive inference system are the proposed models which were used in the present study. According to the results, the best model for an artificial neural network had two hidden layers, the first layer of 18 neurons and the second layer of 7 neurons, with the Tansig-Tansig activation function, with a coefficient of determination R2=0.4969. The best model for the fuzzy-neural inference system had the ring function (Dsigmoid) as a membership function, with three membership functions and 142 repetitions with a coefficient of determination R2=0.2860. Also, combining wavelet-neural network conversion with the coif2 wavelet function due to the more adaptation this function has to the input variables, the better the performance, and this function, with a coefficient of determination R2=0.9447, had the highest accuracy compared to other models.

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.