Uncertainty quantification of granular computing-neural network model for prediction of pollutant longitudinal dispersion coefficient in aquatic streams.

Discharge of pollution loads into natural water systems remains a global challenge that threatens water and food supply, as well as endangering ecosystem services. Natural rehabilitation of contaminated streams is mainly influenced by the longitudinal dispersion coefficient, or the rate of longitudinal dispersion (Dx), a key parameter with large spatiotemporal fluctuations that characterizes pollution transport. The large uncertainty in estimation of Dx in streams limits the water quality assessment in natural streams and design of water quality enhancement strategies. This study develops an artificial intelligence-based predictive model, coupling granular computing and neural network models (GrC-ANN) to provide robust estimation of Dx and its uncertainty for a range of flow-geometric conditions with high spatiotemporal variability. Uncertainty analysis of Dx estimated from the proposed GrC-ANN model was performed by alteration of the training data used to tune the model. Modified bootstrap method was employed to generate different training patterns through resampling from a global database of tracer experiments in streams with 503 datapoints. Comparison between the Dx values estimated by GrC-ANN to those determined from tracer measurements shows the appropriateness and robustness of the proposed method in determining the rate of longitudinal dispersion. The GrC-ANN model with the narrowest bandwidth of estimated uncertainty (bandwidth-factor = 0.56) that brackets the highest percentage of true Dx data (i.e., 100%) is the best model to compute Dx in streams. Considering the significant inherent uncertainty reported in the previous Dx models, the GrC-ANN model developed in this study is shown to have a robust performance for evaluating pollutant mixing (Dx) in turbulent environmental flow systems.

Effect of COVID-19 on transportation air pollution by moderation and mediation analysis in Queens, New York.

The outbreak of the COVID-19 virus in 2020 has left many changes in the quality of life and environment, including air quality in different parts of the world. As a result of lockdown conditions, the level of air pollution has been changed considerably due to topographic, geographical, and cultural conditions as well as traffic restrictions. Thus, this study aimed to investigate the effect COVID-19 outbreak on improving air quality as a result of changes in traffic volume and traffic patterns in Queens, New York, using the moderation and mediation analysis model structure. In this model, COVID-19 outbreak periods were defined as a moderating variable, traffic volume (number of daily vehicles) as an independent variable and mediator, and air pollution concentration parameters (NOx, PM2.5, and O3) individually as dependent variables. Three-time periods were selected, each representing the duration and severity of traffic restrictions and prohibitions, and these three periods corresponded to 1 February-4 March, 5 March-21 March, and 22 March-15 May. They represented the normal, aware, and lockdown periods, respectively. The result of the study showed that in 2020 compared to the last five consecutive years, PM2.5 and NOx pollutants decreased by 39.2% and 35.8% as a result of the traffic ban due to the COVID-19, but an increase of 15.1% in O3 pollutant was observed in the mentioned period. Although traffic restrictions reduced total traffic volume compared to the same period last year, there has been no significant reduction in the air quality index (AQI). The reduction in NOx concentration leads to more O3 ground levels, and this caused the AQI not to decrease significantly. Finally, the moderation and mediation model results showed that the COVID-19 almost has no significant effect on the correlation between daily traffic and the concentration of NOx, PM2.5, and O3 pollutants as moderator. However, the COVID-19 has a significant correlation with O3 and PM2.5 concentration, and the traffic volume mediation effect is negligible. Therefore, the statistical analysis and models show that the COVID-19 pandemic is an effective traffic volume and air quality parameter.

Using a deep convolutional network to predict the longitudinal dispersion coefficient.

Given the interest in accurately predicting the Longitudinal Dispersion Coefficient (Dx) within the fields of hydraulic and water quality modeling, a wide range of methods have been used to estimate this parameter. In order to improve the accuracy of Dx predictions, this paper proposes the use of a Deep Convolutional Network (DCN), a sub-field of machine learning. The proposed deep neural network architecture consists of two parts; first, a one-dimensional convolutional neural network (CNN) to build informative feature maps, and second, a stack of deep, fully connected layers to estimate pollution dispersion (as Dx) in streams. To accurately predict Dx the developed model draws upon a large and diverse array of datasets in the form of three dimensionless parameters: Width/Depth (W/H), Velocity/Shear Velocity (U/u*), and Longitudinal Dispersion Coefficient/(Depth * Shear Velocity) (Dx /Hu*). The model's accuracy is compared to that of several empirical models using a number of statistical measures. In addition, the DCN model results are compared with artificial neural network (ANN) and support vector machine (SVM) models implemented in this research and also similar studies applying various machine learning models (ML) towards Dx prediction. The statistical evaluation indicates that the DCN model outperforms the tested empirical, ANN, SVM and ML models with a significant difference. Additionally, five-fold cross-validation is performed to analyze the sensitivity and dependency of the DCN model's results on dataset selection, which shows that the dataset selection process does not significantly affect the model's accuracy. Since both ML and empirical models are, in general, poor predictors of the upper and lower ranges of Dx values, the DCN model's predictions of Dx in six different extreme-value ranges are assessed. The DCN model shows excellent accuracy in estimating Dx over the full possible range of data. In comparison with the empirical and ML models mentioned above, the DCN model more accurately predicts Dx values from river geometry and hydraulic datasets, with low errors across all ranges of Dx. The most significant advantage of DCN is that it tries to learn high-level features from data in an incremental manner.

Co-transport of chromium(VI) and bentonite colloidal particles in water-saturated porous media: Effect of colloid concentration, sand gradation, and flow velocity.

The transport of pollutants inside the groundwater system is profoundly affected by absorption and transmission via colloid or soil particles. Therefore, it is essential to investigate the significant pollutants (Such as hexavalent chromium (Cr(VI))) transfer in the presence of colloid particles that can facilitate or retain this transfer. For this purpose, an experiment is carried out in a saturated porous media column to study the bentonite concentration, flow velocity and sand grain size effects on co-transport of Cr(VI) with bentonite. The results of this study demonstrated that the colloid particles facilitate the transfer of Cr(VI) by 30% in 200 mg/l bentonite colloids concentration. The amount of transmitted Cr(VI) is decreased by increasing the bentonite colloids concentration from 200 mg/l to 300 mg/l. As the flow velocity increased from 2 cm/min to 3.3 cm/min, the amount of transferred Cr(VI) increased by 7%. The results show that with reducing the sand grain size, the amount of transmitted bentonite and Cr(VI) is reduced that this effect is more sensible in bentonite transport. As a result, it can be noted that the bentonite colloidal particles according to its concentration and experimental conditions, may facilitate or retain the Cr(VI) transport and sand gradation has a significant impact on colloid and pollutant transmission.

Effect of bentonite particles' presence on two-dimensional chromium transmission.

The co-transport of pollutants with colloidal particles to lower depths of groundwater and porous environments has been demonstrated in many studies in recent three decades. Despite the numerous researches, all experimental and numerical studies of pollutant transfer in the presence of colloidal particles have been carried out in one dimension, which causes significant errors in this phenomenon. In this study, the two-dimensional transfer experiment of chromium in the presence of bentonite colloidal particles is done in saturated porous media. In order to conduct the experiment in two-dimensional conditions, the sampling was done in central and lateral of the last experiment column section. The results have been demonstrated that the transmission along the longitudinal direction is higher than lateral in the three tests of the transfer of chromium, bentonite, and chromium in the presence of bentonite colloidal particles at the beginning of the experiment, and due to completed mixing in the section, it reached to a constant value as lateral samples. While the presence of bentonite colloidal particles facilitates the transfer of chromium in both longitudinal and lateral directions, increasing the bentonite colloidal particle concentration causes more getting stuck of colloid particles between the sand grains and reduction of the chromium transfer in both longitudinal and lateral directions. So, it can be concluded that transfer in the lateral direction is lower in bentonite colloidal particles compared with chromium, and the reason is the bentonite colloidal particles getting stuck between sand grains, which is exacerbated by increasing the concentration of the bentonite. Also, due to the chromium co-transport with colloid particles in the fraction of chromium total transport, increasing the bentonite concentration causes decreasing the chromium lateral transfer.

Granular computing-neural network model for prediction of longitudinal dispersion coefficients in rivers.

Successful application of one-dimensional advection-dispersion models in rivers depends on the accuracy of the longitudinal dispersion coefficient (LDC). In this regards, this study aims to introduce an appropriate approach to estimate LDC in natural rivers that is based on a hybrid method of granular computing (GRC) and an artificial neural network (ANN) model (GRC-ANN). Also, adaptive neuro-fuzzy inference system (ANFIS) and ANN models were developed to investigate the accuracy of three credible artificial intelligence (AI) models and the performance of these models in different LDC values. By comparing with empirical models developed in other studies, the results revealed the superior performance of GRC-ANN for LDC estimation. The sensitivity analysis of the three intelligent models developed in this study was done to determine the sensitivity of each model to its input parameters, especially the most important ones. The sensitivity analysis results showed that the W/H parameter (W: channel width; H: flow depth) has the most significant impact on the output of all three models in this research.