Application of wavelet-artificial intelligence hybrid models for water quality prediction: a case study in Aji-Chay River,Iran

The accuracy of Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), wavelet-ANN and wavelet-ANFIS in predicting monthly water salinity levels of northwest Iran’s Aji-Chay River was assessed. The models were calibrated, validated and tested using different subsets of monthly records (October 1983 to September 2011) of individual solute (Ca²⁺, Mg²⁺, Na⁺, SO₄ ^2? and Cl^?) concentrations (input parameters, meq L^?1), and electrical conductivity-based salinity levels (output parameter, µS cm^?1), collected by the East Azarbaijan regional water authority. Based on the statistical criteria of coefficient of determination (R²), normalized root mean square error (NRMSE), Nash–Sutcliffe efficiency coefficient (NSC) and threshold statistics (TS) the ANFIS model was found to outperform the ANN model. To develop coupled wavelet-AI models, the original observed data series was decomposed into sub-time series using Daubechies, Symlet or Haar mother wavelets of different lengths (order), each implemented at three levels. To predict salinity input parameter series were used as input variables in different wavelet order/level-AI model combinations. Hybrid wavelet-ANFIS (R² = 0.9967, NRMSE = 2.9 × 10^?5 and NSC = 0.9951) and wavelet-ANN (R² = 0.996, NRMSE = 3.77 × 10^?5 and NSC = 0.9946) models implementing the db4 mother wavelet decomposition outperformed the ANFIS (R² = 0.9954, NRMSE = 3.77 × 10^?5 and NSC = 0.9914) and ANN (R² = 0.9936, NRMSE = 3.99 × 10^?5 and NSC = 0.9903) models.     

Hydrogeologic framework of the Maku area basalts, northwestern Iran

The Maku area in northwestern Iran is characterized by young lava flows which erupted from Mount Ararat in Turkey. These fractured volcanic rocks overlie alluvium associated with pre-existing rivers and form a good basalt-alluvium aquifer over an area of 650 km². Groundwater discharge occurs from 12 large springs, ranging from 20 to 4,000 L s^?1, and from some extraction wells. Permian and Oligo-Miocene age limestones along the northern boundary of the Bazargan and Poldasht Plains basalts are intensively karstified and groundwater from these high lands easily enters the basalt-alluvium aquifers. The transmissivity of the basalt-alluvium aquifer ranges from 24 to 870 m² d^?1, indicating heterogeneity. Groundwater of the aquifer is a sodium-bicarbonate and mixed cation-bicarbonate type and the concentration of fluoride is higher than the universal maximum admissible concentrations for drinking. In order to determine the chemical composition and identify the source of the high fluoride concentrations in the groundwater of the basaltic area, water samples from the springs, wells and rivers were analyzed. The results indicate that the high fluoride water enters the study area from the Sari Su River.     

An Investigation into Ridge-Top Turbulence Characteristics During Neutral and Weakly Stable Conditions: Velocity Spectra and Isotropy

An investigation into high Reynolds number turbulent flow over a ridge top in New Zealand is described based on high-resolution in-situ measurements, using ultrasonic anemometers for two separate locations on the same ridge with differing upwind terrain complexity. Twelve 5-h periods during neutrally stratified and weakly stable atmospheric conditions with strong wind speeds were sampled at 20 Hz. Large (and small) turbulent length scales were recorded for both vertical and longitudinal velocity components in the range of 7–23 m (0.7–3.3 m) for the vertical direction and 628–1111 m (10.5–14.5 m) for the longitudinal direction. Large-scale eddy sizes scaled to the WRF (Weather Research and Forecasting) numerical model simulated boundary-layer thickness for both sites, while small-scale turbulent features were a function of the complexity of the upwind terrain. Evidence of a multi-scale turbulent structure was obtained at the more complex terrain site, while an assessment of the three-dimensional isotropy assumption in the inertial subrange of the spectrum showed anisotropic turbulence at the less complex site and evidence of isotropic turbulence at the more complex site, with a spectral ratio convergence deviating from the 4/3 or unity values suggested by previous theory and practice. Existing neutral spectral models can represent locations along the ridge top with simple upwind complexity, especially for the vertical wind spectra, but sites with more orographic complexity and strong vertical wind speeds are often poorly represented using these models. Measured spectra for the two sites exhibited no significant diurnal variation and very similar large-scale and small-scale turbulent length scales for each site, but the turbulence energy measured by the variances revealed a strong diurnal difference.     

Burden prediction in blasting operation using rock geomechanical properties

Burden prediction is a vital task in the production blasting. Both the excessive and insufficient burden can significantly affect the result of blasting operation. The burden which is determined by empirical models is often inaccurate and needs to be adjusted experimentally. In this paper, an attempt was made to develop an artificial neural network (ANN) in order to predict burden in the blasting operation of the Mouteh gold mine, using considering geomechanical properties of rocks as input parameters. As such here, network inputs consist of blastability index (BI), rock quality designation (RQD), unconfined compressive strength (UCS), density, and cohesive strength. To make a database (including 95 datasets), rock samples are used from Iran’s Mouteh goldmine. Trying various types of the networks, a neural network, with architecture 5-15-10-1, was found to be optimum. Superiority of ANN over regression model is proved by calculating. To compare the performance of the ANN modeling with that of multivariable regression analysis (MVRA), mean absolute error (E _a), mean relative error (E _r), and determination coefficient (R ²) between predicted and real values were calculated for both the models. It was observed that the ANN prediction capability is better than that of MVRA. The absolute and relative errors for the ANN model were calculated 0.05 m and 3.85%, respectively, whereas for the regression analysis, these errors were computed 0.11 m and 5.63%, respectively. Moreover, determination coefficient of the ANN model and MVRA were determined 0.987 and 0.924, respectively. Further, a sensitivity analysis shows that while BI and RQD were recognized as the most sensitive and effective parameters, cohesive strength is considered as the least sensitive input parameters on the ANN model output effective on the proposed (burden).     

Application of spectrum–area fractal model to identify of geochemical anomalies based on soil data in Kahang porphyry-type Cu deposit,Iran

The aim of this study is to identify geochemical anomalies using power spectrum–area (S–A) method based on the grade values of Cu, Mo and Au in 2709 soil samples collected from Kahang porphyry-type Cu deposit, Central Iran. S–A log–log plots indicated that there are three stages of Cu, Mo and Au enrichment. The third enrichment was considered as the main stage for the presence of Cu, Mo and Au at the concentrations above 416 ppm, 23 ppm and 71 ppb, respectively. Elemental anomalies are positively associated with monzo–granite–diorite and breccias units which are in the central and western parts of the deposit. The anomalies are located within the potassic, phyllic and argillic alteration types and also there is the positive correlation between the anomalies and nearing faults in the studied area. The results obtained via fractal model were interpreted accordingly to incorporate the information for the mineralized areas including detailed geological map, structural analysis and alterations. The results show that S–A multifractal modeling is applicable for anomalies delineation based on soil data.     

Characterization of hydrogeologic properties of the Tabriz plain multilayer aquifer system,NW Iran

A detailed hydrogeological investigation was carried out in the Tabriz plain in Iran using conventional hydrogeological field investigations and hydrochemistry. The study was carried out because the aquifers are of particular importance as they are more or less the only source of water supply available to the rural population and for agricultural and industrial activities. Analytical and numerical methods were applied to the constant rate pumping test data from the Tabriz airport and the Tabriz Power Station well fields. Two types of aquifers of different water quality were identified in the study area: an unconfined aquifer that extends over the plain and confined aquifers that are found in the deeper layers of the multilayered sediment terraces of the Aji-Chay River course. Therefore, the central part of the Tabriz plain contains both unconfined and confined aquifers, while close to the highlands, there is only an unconfined aquifer. There was evidence of minor leakage in the confined aquifers when the numerical method was used for analysis. The groundwater in the area can be identified by three main geochemical facies: Na-Cl, Ca-HCO₃, and mixed Ca-Mg-Cl-SO₄. The processes responsible for the hydrochemical evolution in the area fall into five categories: dissolution of evaporate minerals, precipitation of carbonate minerals, evaporation, ion exchange, and anthropogenic activity.     

Assessing the hydrogeochemistry and water quality of the Aji-Chay River,northwest of Iran

Aji-Chay River is one of the most important surface reservoirs of northwest of Iran, because it passes through Tabriz city and discharges to Urmia Lake, one of the largest permanent salty lakes in the world. The main objectives of the present study are to evaluate its overall water quality and to explore its hydrogeochemical characteristics, including the potential contamination from heavy metals and metalloids such as Co, Pb, Zn, Cd, Cu, Cr, Al and As. For this purpose, 12 water samples were collected from the main river body and its tributaries within Tabriz plain. The Piper diagram classified water samples mainly into Na–Cl and secondary into Ca–HCO₃ and mixed Ca–Mg–Cl types, denoting a profound salinization effect. The cross-plots showed that natural geochemical processes including dissolution of minerals (e.g., carbonates, evaporites and silicates), as well as ion exchange, are the predominant factors that contribute to fluvial hydrogeochemistry, while anthropogenic activities (industrial and agricultural) impose supplementary effects. Cluster analysis classified samples into two distinct clusters; samples of cluster B appear to have elevated electrical conductivity (EC) values and trace metals concentrations such as Co, Pb and Cd, while SiO₂ and Zn are low in comparison with the samples of the cluster A. The main processes controlling Aji-Chay River hydrogeochemistry and water quality were identified to be salinization and rock weathering. Both are related with geogenic sources which enrich river system with elevated values of Na⁺, Cl^?, Ca²⁺, Mg²⁺, K⁺, SO₄ ^2? and EC as a direct effect of evaporites leaching and elevated values of Pb and Cd as an impact from the weathering process of volcanic formations. According to the US salinity diagram, all of the water samples are unsuitable for irrigation as having moderate to bad quality.