Monthly water quality forecasting and uncertainty assessment via bootstrapped wavelet neural networks under missing data for Harbin, China

In this paper, bootstrapped wavelet neural network (BWNN) was developed for predicting monthly ammonia nitrogen (NH⁴⁺–N) and dissolved oxygen (DO) in Harbin region, northeast of China. The Morlet wavelet basis function (WBF) was employed as a nonlinear activation function of traditional three-layer artificial neural network (ANN) structure. Prediction intervals (PI) were constructed according to the calculated uncertainties from the model structure and data noise. Performance of BWNN model was also compared with four different models: traditional ANN, WNN, bootstrapped ANN, and autoregressive integrated moving average model. The results showed that BWNN could handle the severely fluctuating and non-seasonal time series data of water quality, and it produced better performance than the other four models. The uncertainty from data noise was smaller than that from the model structure for NH⁴⁺–N; conversely, the uncertainty from data noise was larger for DO series. Besides, total uncertainties in the low-flow period were the biggest due to complicated processes during the freeze-up period of the Songhua River. Further, a data missing–refilling scheme was designed, and better performances of BWNNs for structural data missing (SD) were observed than incidental data missing (ID). For both ID and SD, temporal method was satisfactory for filling NH⁴⁺–N series, whereas spatial imputation was fit for DO series. This filling BWNN forecasting method was applied to other areas suffering “real” data missing, and the results demonstrated its efficiency. Thus, the methods introduced here will help managers to obtain informed decisions.     

Impacts of Meteorological Factors on PM10: Artificial Neural Networks (ANN) and Multiple Linear Regression (MLR) Approaches

In this study, prediction capacities of multi-linear regression (MLR) and artificial neural networks (ANN) onto coarse particulate matter (PM₁₀) concentrations were investigated. Different meteorological factors on particulate pollution were chosen for operating variables in the model analyses. Two different regions (urban and industrial) were identified in the region of Kocaeli, Turkey. All data sets were obtained from air quality monitoring network of the Ministry of Environment and Urban Planning, and 120 data sets were used in the MLR and ANN models. Regression equations explained the effects of the meteorological factors in MLR analyses. In the ANN model, backpropagation network with two hidden layers has achieved the best prediction efficiency. Determination coefficients and error values were examined for each model. ANN models displayed more accurate results compared to MLR.     

Prediction of dissolved oxygen concentration in hypoxic river systems using support vector machine: a case study of Wen-Rui Tang River,China

Accurate quantification of dissolved oxygen (DO) is critically important for managing water resources and controlling pollution. Artificial intelligence (AI) models have been successfully applied for modeling DO content in aquatic ecosystems with limited data. However, the efficacy of these AI models in predicting DO levels in the hypoxic river systems having multiple pollution sources and complicated pollutants behaviors is unclear. Given this dilemma, we developed a promising AI model, known as support vector machine (SVM), to predict the DO concentration in a hypoxic river in southeastern China. Four different calibration models, specifically, multiple linear regression, back propagation neural network, general regression neural network, and SVM, were established, and their prediction accuracy was systemically investigated and compared. A total of 11 hydro-chemical variables were used as model inputs. These variables were measured bimonthly at eight sampling sites along the rural-suburban-urban portion of Wen-Rui Tang River from 2004 to 2008. The performances of the established models were assessed through the mean square error (MSE), determination coefficient (R ²), and Nash-Sutcliffe (NS) model efficiency. The results indicated that the SVM model was superior to other models in predicting DO concentration in Wen-Rui Tang River. For SVM, the MSE, R ², and NS values for the testing subset were 0.9416 mg/L, 0.8646, and 0.8763, respectively. Sensitivity analysis showed that ammonium-nitrogen was the most significant input variable of the proposal SVM model. Overall, these results demonstrated that the proposed SVM model can efficiently predict water quality, especially for highly impaired and hypoxic river systems.     

Extreme learning machines: a new approach for modeling dissolved oxygen (DO) concentration with and without water quality variables as predictors

In this paper, several extreme learning machine (ELM) models, including standard extreme learning machine with sigmoid activation function (S-ELM), extreme learning machine with radial basis activation function (R-ELM), online sequential extreme learning machine (OS-ELM), and optimally pruned extreme learning machine (OP-ELM), are newly applied for predicting dissolved oxygen concentration with and without water quality variables as predictors. Firstly, using data from eight United States Geological Survey (USGS) stations located in different rivers basins, USA, the S-ELM, R-ELM, OS-ELM, and OP-ELM were compared against the measured dissolved oxygen (DO) using four water quality variables, water temperature, specific conductance, turbidity, and pH, as predictors. For each station, we used data measured at an hourly time step for a period of 4 years. The dataset was divided into a training set (70%) and a validation set (30%). We selected several combinations of the water quality variables as inputs for each ELM model and six different scenarios were compared. Secondly, an attempt was made to predict DO concentration without water quality variables. To achieve this goal, we used the year numbers, 2008, 2009, etc., month numbers from (1) to (12), day numbers from (1) to (31) and hour numbers from (00:00) to (24:00) as predictors. Thirdly, the best ELM models were trained using validation dataset and tested with the training dataset. The performances of the four ELM models were evaluated using four statistical indices: the coefficient of correlation (R), the Nash-Sutcliffe efficiency (NSE), the root mean squared error (RMSE), and the mean absolute error (MAE). Results obtained from the eight stations indicated that: (i) the best results were obtained by the S-ELM, R-ELM, OS-ELM, and OP-ELM models having four water quality variables as predictors; (ii) out of eight stations, the OP-ELM performed better than the other three ELM models at seven stations while the R-ELM performed the best at one station. The OS-ELM models performed the worst and provided the lowest accuracy; (iii) for predicting DO without water quality variables, the R-ELM performed the best at seven stations followed by the S-ELM in the second place and the OP-ELM performed the worst with low accuracy; (iv) for the final application where training ELM models with validation dataset and testing with training dataset, the OP-ELM provided the best accuracy using water quality variables and the R-ELM performed the best at all eight stations without water quality variables. Fourthly, and finally, we compared the results obtained from different ELM models with those obtained using multiple linear regression (MLR) and multilayer perceptron neural network (MLPNN). Results obtained using MLPNN and MLR models reveal that: (i) using water quality variables as predictors, the MLR performed the worst and provided the lowest accuracy in all stations; (ii) MLPNN was ranked in the second place at two stations, in the third place at four stations, and finally, in the fourth place at two stations, (iii) for predicting DO without water quality variables, MLPNN is ranked in the second place at five stations, and ranked in the third, fourth, and fifth places in the remaining three stations, while MLR was ranked in the last place with very low accuracy at all stations. Overall, the results suggest that the ELM is more effective than the MLPNN and MLR for modelling DO concentration in river ecosystems.     

A neural network-based approach for the prediction of urban SO2 concentrations in the Istanbul metropolitan area

A three-layer Artificial Neural Network (ANN) model was developed to forecast air pollution levels. The subsequent SO₂ concentration (24-hour averaged) being the output parameter of this study was estimated by seven input parameters such as preceding SO₂ concentrations (24-hour averaged), average daily temperature, sea-level pressure, relative humidity, cloudiness, average daily wind speed and daily dominant wind direction. After Backpropagation training combined with Principal Component Analysis (PCA), the proposed model predicted subsequent SO₂ values based on measured data. ANN testing outputs were proven to be satisfactory with correlation coefficients of about 0.770, 0.744 and 0.751 for the winter, summer and overall data, respectively.     

Source apportionment of polycyclic aromatic hydrocarbons in surface sediments of the Bohai Sea, China

A total of 112 surface sediment samples covering virtually the entire Bohai Sea were analyzed for polycyclic aromatic hydrocarbons (PAHs), in order to provide the extensive information of recent occurrence levels, distribution, possible sources, and potential biological risk of these compounds in this area. Surface sediment samples were collected from the Bohai Sea using a stainless steel grab sampler. Sixteen PAHs were determined by a Finnigan TRACE DSQ gas chromatography/mass spectrometry. Diagnostic ratios, cluster analysis, and principal component analysis (PCA) with multivariate linear regression (MLR) were performed to identify and quantitatively apportion the major sources of sedimentary PAHs in the Bohai Sea. Concentrations of total PAHs in the Bohai Sea ranged widely from 97.2 to 300.7 ng/g (mean, 175.7?±?37.3 ng/g). High concentrations of PAHs were found in the vicinity of Luan River Estuary-Qinhuangdao Harbor, Cao River Estuary-Bohai Sea Center, and north of the Yellow River Estuary. The three-ring PAHs were most abundant, accounting for about 37?±?5 % of total PAHs. The four-ring and five-ring PAHs were the next dominant ones comprising approximately 29?±?7and 23?±?3 % of total PAHs, respectively. Concentrations of acenaphthylene, acenaphthene, and dibenz[a,h]anthracene are higher than Canadian interim marine sediment quality guideline values at most of the sites in the study area. Contamination levels of PAHs in the Bohai Sea were low in comparison with other coastal sediments in China and developed countries. The distribution pattern of PAHs and source identification implied that PAH contamination in the Bohai Sea mainly originates from petrogenic and pyrogenic sources. Further PCA/MLR analysis suggested that the contributions of spilled oil products (petrogenic), coal combustion, and traffic-related pollution were 39, 38, and 23 %, respectively. Pyrogenic sources (coal combustion and traffic-related pollution) contributed 61 % of anthropogenic PAHs to sediments, which indicates that energy consumption could be a dominant factor in PAH pollution in this area. Acenaphthylene, acenaphthene, and dibenz[a,h]anthracene are the three main species of PAHs with more ecotoxicological concern in the Bohai Sea.     

Carrousel氧化沟系统出水COD预报的神经网络模型

以河南漯河市污水净化中心的氧化沟系统为考察对象,针对该系统进水水质复杂,控制滞后的难点,引入人工神经网络的理论和方法,对其进行模拟分析,建立了基于BP网络的氧化沟系统出水COD预报模型。模型性能检验和灵敏度检验表明,建成的模型准确度高,适应性强,可直接用于该系统出水COD预报。这为氧化沟工艺在线控制提供了一条简便的途径。     

Two-days ahead prediction of daily maximum concentrations of SO2, O3, PM10, NO2, CO in the urban area of Palermo,Italy

Artificial neural networks are functional alternative techniques in modelling the intricate vehicular exhaust emission dispersion phenomenon. Pollutant predictions are notoriously complex when using either deterministic or stochastic models, which explains why this model was developed using a neural network. Neural networks have the ability to learn about non-linear relationships between the used variables. In this paper a recurrent neural network (Elman model) based forecaster for the prediction of daily maximum concentrations of SO₂, O₃, PM10, NO₂, CO in the city of Palermo is proposed. The effectiveness of the presented forecaster was tested using a time series recorded between 1 January 2003 to 31 December 2004 in eight monitoring stations in urban area of Palermo (Italy). Experimental trials show that the developed and tuned model is appropriate, giving small values of root mean square error (RMSE) , mean absolute error (MAE) and mean square error (MSE). In addition, the related correlation coefficient ranges from 0.72 to 0.97 for each forecasted pollutant, underlying a small difference between the forecasted and the measured values. The above results make the proposed forecaster a powerful tool for pollution management systems.     

Spatial and temporal distribution of polycyclic aromatic hydrocarbons (PAHs) in surface water from Liaohe River Basin,northeast China

Liaohe River Basin is an important region in northeast China, which consists of several main rivers including Liao River, Taizi river, Daliao River, and Hun River. As a highly industrialized region, the basin receives dense waste discharges, causing severe environmental problems. In this study, the spatial and temporal distribution of aqueous polycyclic aromatic hydrocarbons (PAHs) in Liaohe River Basin from 50 sampling sites in both dry (May) and level (October) periods in 2012 was investigated. Sixteen USEPA priority PAHs were quantified by gas chromatography/mass selective detector. The total PAH concentration ranged from 111.8 to 2,931.6 ng/L in the dry period and from 94.8 to 2766.0 ng/L in the level period, respectively. As for the spatial distribution, the mean concentration of PAHs followed the order of Taizi River > Daliao River > Hun River > Liao River, showing higher concentrations close to large cities with dense industries. The composition and possible sources of PAHs in the water samples were also determined. The fractions of low molecular weight PAHs ranged from 58.2 to 93.3 %, indicating the influence of low or moderate temperature combustion process. Diagnostic ratios, principal component analysis, and hierarchical cluster analysis were used to study the possible source categories in the study area, and consistent results were obtained from different techniques, that PAHs in water samples mainly originated from complex sources, i.e., both pyrogenic and petrogenic sources. The benzo[a]pyrene equivalents (EBaP) characterizing the ecological risk of PAHs to the aquatic environment suggested that PAHs in Liaohe River Basin had already caused environmental health risks.     

β-Estradiol and ethinyl-estradiol contamination in the rivers of the Carpathian Basin

17β-Estradiol (E2) and 17α-ethinyl estradiol (EE2), which are environmental estrogens, have been determined with LC-MS in freshwater. Their sensitive analysis needs derivatization and therefore is very hard to achieve in multiresidue screening. We analyzed samples from all the large and some small rivers (River Danube, Drava, Mur, Sava, Tisza, and Zala) of the Carpathian Basin and from Lake Balaton. Freshwater was extracted on solid phase and derivatized using dansyl chloride. Separation was performed on a Kinetex XB-C18 column. Detection was achieved with a benchtop orbitrap mass spectrometer using targeted MS analysis for quantification. Limits of quantification were 0.05 ng/L (MS1) and 0.1 ng/L (MS/MS) for E2, and 0.001 ng/L (MS1) and 0.2 ng/L (MS/MS) for EE2. River samples contained n.d.–5.2 ng/L E2 and n.d.–0.68 ng/L EE2. Average levels of E2 and EE2 were 0.61 and 0.084 ng/L, respectively, in rivers, water courses, and Lake Balaton together, but not counting city canal water. EE2 was less abundant, but it was still present in almost all of the samples. In beach water samples from Lake Balaton, we measured 0.076–0.233 E2 and n.d.–0.133 EE2. A relative high amount of EE2 was found in river Zala (0.68 ng/L) and in Hévíz-Páhoki canal (0.52 ng/L), which are both in the catchment area of Lake Balaton (Hungary).     

Agricultural solid waste for sorption of metal ions,part II: competitive assessment in multielemental solution and lake water

Sugarcane bagasse and hydroponic lettuce roots were used as biosorbents for the removal of Cu(II), Fe(II), Mn(II), and Zn(II) from multielemental solutions and lake water, in batch processes. These biomasses were studied in natura (lettuce roots, NLR, and sugarcane bagasse, NSB) and chemically modified with HNO₃ (lettuce roots, MLR, and sugarcane bagasse, MSB). The results showed higher adsorption efficiency for MSB and either NLR or MLR. The maximum adsorption capacities (q_max) in multielemental solution for Cu(II), Fe(II), Mn(II), and Zn(II) were 35.86, 31.42, 3.33, and 24.07 mg/g for NLR; 25.36, 27.95, 14.06, and 6.43 mg/g for MLR; 0.92, 3.94, 0.03, and 0.18 mg/g for NSB; and 54.11, 6.52, 16.7, and 1.26 mg/g for MSB, respectively. The kinetic studies with chemically modified biomasses indicated that sorption was achieved in the first 5 min and reached equilibrium around 30 min. Sorption of Cu(II), Fe(II), Mn(II), and Zn(II) in lake water by chemically modified biomasses was 24.31, 14.50, 8.03, and 8.21 mg/g by MLR, and 13.15, 10.50, 6.10, and 5.14 mg/g by MSB, respectively. These biosorbents are promising and low costs agricultural residues, and as for lettuce roots, these showed great potential even with no chemical modification.     

Agricultural solid waste for sorption of metal ions: part I—characterization and use of lettuce roots and sugarcane bagasse for Cu(II), Fe(II), Zn(II), and Mn(II) sorption from aqueous medium

Sugarcane bagasse and hydroponic lettuce roots were used as biosorbents for Cu(II), Fe(II), Zn(II), and Mn(II) removal from monoelemental solutions in aqueous medium, at pH 5.5, using batch procedures. These biomasses were studied in natura (lettuce roots, NLR, and sugarcane bagasse, NSB) and modified with HNO₃ (lettuce roots, MLR, and sugarcane bagasse, MSB). Langmuir, Freundlich, and Dubinin-Radushkevich non-linear isotherm models were used to evaluate the data from the metal ion adsorption assessment. The maximum adsorption capacities (q_max) in monoelemental solution, calculated using the Langmuir isothermal model for Cu(II), Fe(II), Zn(II), and Mn(II), were respectively 24.61, 2.64, 23.04, and 5.92 mg/g for NLR; 2.29, 16.89, 1.97, and 2.88 mg/g for MLR; 0.81, 0.06, 0.83, and 0.46 mg/g for NSB; and 1.35, 2.89, 20.76, and 1.56 mg/g for MSB. The Freundlich n parameter indicated that the adsorption process was favorable for Cu(II) uptake by NLR; Fe(II) retention by MLR and MSB; and Zn(II) sorption by NSB, MLR, and NSB and favorable for all biomasses in the accumulation of Mn(II). The Dubinin-Radushkevich isotherm was applied to estimate the energy (E) and type of adsorption process involved, which was found to be a physical one between analytes and adsorbents. Organic groups such as O–H, C–O–C, CH, and C=O were found in the characterization of the biomass by FTIR. In the determination of the biomass surface charges by using blue methylene and red amaranth dyes, there was a predominance of negative charges.     

Water quality modelling of Lis River, Portugal

The aim of the study was to predict the impact of flow conditions, discharges and tributaries on the water quality of Lis River using QUAL2Kw model. Calibration of the model was performed, based on data obtained in field surveys carried out in July 2004 and November 2006. Generally the model fitted quite well the experimental data. The results indicated a decrease of water quality in the downstream area of Lis River, after the confluence of Lena, Milagres and Amor tributaries, as a result of discharges of wastewaters containing degradable organics, nutrients and pathogenic organisms from cattle-raising wastewaters, domestic effluents and agricultural runoff. The water quality criteria were exceeded in these areas for dissolved oxygen, biochemical oxygen demand, total nitrogen and faecal coliforms. Water quality modelling in different scenarios showed that the impact of tributaries on the quality of Lis River water was quite negligible and mainly depends on discharges, which are responsible by an increase of almost 45, 13 and 44 % of ultimate carbonaceous biochemical oxygen demand (CBOD_u), ammonium nitrogen and faecal coliforms, for winter simulation, and 23, 33 and 36 % for summer simulation, respectively, when compared to the real case scenario.     

Tropical surface water quality studies: Implications for the aquatic fate of N-methyl carbamate pesticides

Water quality assessment was conducted on the Ruiru River, a tributary of an important tropical river system in Kenya, to determine baseline river conditions for studies on the aquatic fate of N-methyl carbamate (NMC) pesticides. Measurements were taken at the end of the long rainy season in early June 2013. Concentrations of copper (0.21–1.51 ppm), nitrates (2.28–4.89 ppm) and phosphates (0.01–0.50 ppm) were detected at higher values than in uncontaminated waters, and attributed to surface runoff from agricultural activity in the surrounding area. Concentrations of dissolved oxygen (8–10 ppm), ammonia (0.02–0.22 ppm) and phenols (0.19–0.83 ppm) were found to lie within normal ranges. The Ruiru River was found to be slightly basic (pH 7.08–7.70) with a temperature of 17.8–21.2°C. The half-life values for hydrolysis of three NMC pesticides (carbofuran, carbaryl and propoxur) used in the area were measured under laboratory conditions, revealing that rates of decay were influenced by the electronic nature of the NMCs. The hydrolysis half-lives at pH 9 and 18°C decreased in the order carbofuran (57.8 h) > propoxur (38.5 h) > carbaryl (19.3 h). In general, a decrease in the electron density of the NMC aromatic ring increases the acidity of the N-bound proton removed in the rate-limiting step of the hydrolysis mechanism. Our results are consistent with this prediction, and the most electron-poor NMC (carbaryl) hydrolyzed fastest, while the most electron-rich NMC (carbofuran) hydrolyzed slowest. Results from this study should provide baseline data for future studies on NMC pesticide chemical fate in the Ruiru River and similar tropical water systems.     

The use of artificial neural network (ANN) for the prediction and simulation of oil degradation in wastewater by AOP

The application of advanced oxidation process (AOP) in the treatment of wastewater contaminated with oil was investigated in this study. The AOP investigated is the homogeneous photo-Fenton (UV/H₂O₂/Fe⁺²) process. The reaction is influenced by the input concentration of hydrogen peroxide H₂O₂, amount of the iron catalyst Fe⁺², pH, temperature, irradiation time, and concentration of oil in the wastewater. The removal efficiency for the used system at the optimal operational parameters (H₂O₂?=?400 mg/L, Fe⁺²?=?40 mg/L, pH?=?3, irradiation time?=?150 min, and temperature?=?30 °C) for 1,000 mg/L oil load was found to be 72 %. The study examined the implementation of artificial neural network (ANN) for the prediction and simulation of oil degradation in aqueous solution by photo-Fenton process. The multilayered feed-forward networks were trained by using a backpropagation algorithm; a three-layer network with 22 neurons in the hidden layer gave optimal results. The results show that the ANN model can predict the experimental results with high correlation coefficient (R ²?=?0.9949). The sensitivity analysis showed that all studied variables (H₂O₂, Fe⁺², pH, irradiation time, temperature, and oil concentration) have strong effect on the oil degradation. The pH was found to be the most influential parameter with relative importance of 20.6 %.     

Development of a multi-factor model for predicting the effects of ambient ozone on the biomass of white clover

Results are presented from the UN/ECE ICP Vegetation (International Cooperative Programme on effects of air pollution on natural vegetation and crops) experiments in which ozone(O(3))-resistant (NC-R) and -sensitive (NC-S) clones of white clover (Trifolium repens cv. Regal) were exposed to ambient O(3) episodes at 14 sites in eight European countries in 1996, 1997 and 1998. The plants were grown according to a standard protocol, and the forage was harvested every 28 days for 4-5 months per year by excision 7 cm above the soil surface. Biomass ratio (NC-S/NC-R) was related to the climatic and pollutant conditions at each site using multiple linear regression (MLR) and artificial neural networks (ANNs). Twenty-one input parameters [e.g. AOT40, 7-h mean O(3) concentration, daylight vapour pressure deficit (VPD), daily maximum temperature] were considered individually and in combination with the aim of developing a model with high r(2) and simple structure that could be used to predict biomass change in white clover. MLR models were generally more complex, and performed less well for unseen data than non-linear ANN models. The ANN model with the best performance had five inputs with an r(2) value of 0.84 for the training data, and 0.71 for previously unseen data. Two inputs to the model described the O(3) conditions (AOT40 and 24-h mean for O(3)), two described temperature (daylight mean and 24-h mean temperature), and the fifth input appeared to be differentiating between semi-urban and rural sites (NO concentration at 17:00). Neither VPD nor harvest interval was an important component of the model. The model predicted that a 5% reduction in biomass ratio was associated with AOT40s in the range 0.9-1.7 ppm x h (microl l(-1) h) accumulated over 28 days, with plants being most sensitive in conditions of low NO(x), medium-range temperature, and high 24-h mean O(3) concentration.     

Occurrence, distribution, and sources of six phenolic endocrine disrupting chemicals in the 22 river estuaries around Dianchi Lake in China

The objectives of this study are to track the occurrence, distribution, and sources of phenolic endocrine disrupting compounds (EDCs) in the 22 rivers around Dianchi Lake in China, to estimate the input and output amounts of phenolic EDCs in the water system, and to provide more comprehensive fundamental data for risk assessment and contamination control of phenolic EDCs in aquatic environment. Six phenolic EDCs were systematically evaluated in water and surface sediment in the estuaries of those rivers. The water and sediment samples were preconcentrated by solid-phase extraction system and microwave-assisted extraction system, respectively. Phenolic EDCs were analyzed by GC-MS (Thermo Fisher Scientific, USA) after derivatization. Phenolic EDCs were found ubiquitously in the aquatic environment. The total concentrations ranged from 248 to 4,650 ng/L in water, and 113 to 3,576 ng/g dry weight in surface sediment. The residue amount of phenolic EDCs in Dianchi Lake was 258 kg/a. Concentrations of the phenolic EDCs in the Lake decreased with increase in distance to the estuaries of those rivers which run through urban and industrial areas. The rivers seriously contaminated by phenolic EDCs were Xin River, Yunliang River, Chuanfang River, Cailian River, Jinjia River, Zhengda River, and Daqing River which run through the old area of Kunming City. Satisfying correlations were observed between the concentrations of the target compounds in water and in surface sediment. NP1EO, NP2EO, and BPA were identified as the three predominant phenolic EDCs. There were significant correlations between phenolic EDCs and many basic water quality parameters. Urban and industrial areas are the major contributors for phenolic EDCs, especially in Kunming City. Compositional profiles of phenolic EDCs in surface sediment were similar to those in river water. The concentrations of phenolic EDCs in the rivers located in the northwest part of the valley were very high, and posed a potential risk to aquatic organisms and even human. The concentrations of NP2EO, NP1EO, and BPA were at moderate levels of other areas. The basic water quality parameters (TOC, TN, DO, and pH) play important roles on the distribution, fate, and behavior of phenolic EDCs in the valley.     

Predicting hourly air pollutant levels using artificial neural networks coupled with uncertainty analysis by Monte Carlo simulations

Recent progress in developing artificial neural network (ANN) metamodels has paved the way for reliable use of these models in the prediction of air pollutant concentrations in urban atmosphere. However, improvement of prediction performance, proper selection of input parameters and model architecture, and quantification of model uncertainties remain key challenges to their practical use. This study has three main objectives: to select an ensemble of input parameters for ANN metamodels consisting of meteorological variables that are predictable by conventional weather forecast models and variables that properly describe the complex nature of pollutant source conditions in a major city, to optimize the ANN models to achieve the most accurate hourly prediction for a case study (city of Tehran), and to examine a methodology to analyze uncertainties based on ANN and Monte Carlo simulations (MCS). In the current study, the ANNs were constructed to predict criteria pollutants of nitrogen oxides (NO_x), nitrogen dioxide (NO₂), nitrogen monoxide (NO), ozone (O₃), carbon monoxide (CO), and particulate matter with aerodynamic diameter of less than 10 μm (PM₁₀) in Tehran based on the data collected at a monitoring station in the densely populated central area of the city. The best combination of input variables was comprehensively investigated taking into account the predictability of meteorological input variables and the study of model performance, correlation coefficients, and spectral analysis. Among numerous meteorological variables, wind speed, air temperature, relative humidity and wind direction were chosen as input variables for the ANN models. The complex nature of pollutant source conditions was reflected through the use of hour of the day and month of the year as input variables and the development of different models for each day of the week. After that, ANN models were constructed and validated, and a methodology of computing prediction intervals (PI) and probability of exceeding air quality thresholds was developed by combining ANNs and MCSs based on Latin Hypercube Sampling (LHS). The results showed that proper ANN models can be used as reliable metamodels for the prediction of hourly air pollutants in urban environments. High correlations were obtained with R ² of more than 0.82 between modeled and observed hourly pollutant levels for CO, NO_x, NO₂, NO, and PM₁₀. However, predicted O₃ levels were less accurate. The combined use of ANNs and MCSs seems very promising in analyzing air pollution prediction uncertainties. Replacing deterministic predictions with probabilistic PIs can enhance the reliability of ANN models and provide a means of quantifying prediction uncertainties.     

Potential source apportionment of polycyclic aromatic hydrocarbons in surface sediments from the middle and lower reaches of the Yellow River,China

In this work, principal component analysis/multiple linear regression (PCA/MLR), positive matrix factorization (PMF), and UNMIX model were employed to apportion potential sources of polycyclic aromatic hydrocarbons (PAHs) in surface sediments from middle and lower reaches of the Yellow River, based on the measured PAHs concentrations in sediments collected from 22 sites in November 2005. The results suggested that pyrogenic sources were major sources of PAHs. Further analysis indicated that source contributions of PAHs compared well among PCA/MLR, PMF, and UNMIX. Vehicles contributed 25.1–36.7 %, coal 34.0–41.6 %, and biomass burning and coke oven 29.2–33.2 % of the total PAHs, respectively. Coal combustion and traffic-related pollution contributed approximately 70 % of anthropogenic PAHs to sediments, which demonstrated that energy consumption was a predominant factor of PAH pollution in middle and lower reaches of the Yellow River. In addition, the distributions of contribution for each identified source category were studied, which showed similar distributed patterns for each source category among the sampling sites.     

Trace metals in Ganges soft-shell turtle (Aspideretes gangeticus) from two barrage: Baloki and Rasul, Pakistan

The concentration of nine metals was measured in liver, kidney, heart, muscle, plastron, and carapace of Aspideretes gangeticus from Rasul and Baloki barrages, Pakistan. The results indicated that metal concentration were significant different among tissues of Ganges soft-shell turtles. However, higher concentrations of Co (5.12 μg/g) and Ni (1.67 μg/g) in liver, Cd (0.41 μg/g) in heart, Fe (267.45 μg/g), Cd (2.12 μg/g) and Mn (2.47 μg/g) in kidney, Cd (0.23 μg/g), Cu (2.57 μg/g), Fe (370.25 μg/g), Mn (5.56 μg/g), and Pb (8.23 μg/g) in muscle of A. gangeticus were recorded at Baloki barrage than Rasul barrage. Whereas mean concentrations of Pb (3.33 μg/g) in liver, Co (1.63 μg/g), Cu (11.32 μg/g), Pb (4.8 μg/g) and Zn (144.69 μg/g) in heart, Co (4.12 μg/g) in muscle, Ni (1.31 μg/g), Pb (2.18 μg/g), and Zn (9.78 μg/g) in carapace were recorded higher at Rasul barrage than Baloki barrage. The metals followed the trend Fe > Zn > Ni > Cu > Mn > Pb > Cr > Co > Cd. Metals of toxicological concern such as Cr, Pb, and Cd were at that level which can cause harmful effects to turtles. The results provide baseline data of heavy metals on freshwater turtle species of Pakistan.