Journal of Environmental Engineering

The use of artificial neural networks (ANN) for modeling complex processes is an attractive approach that has been successfully applied in various fields. However, in many cases the use of an ANN alone may be inadequate and inaccurate when data are insufficient, because the ANN black-box model relies completely on the data. As a result, a hybrid model consisting of a simplified process model (SPM) and a neural network (residual model) is used in the present study for developing a dynamic model of sequencing batch reactor systems. The implemented SPM model consists of only five discrete rate equations and an ANN is added to the SPM in a parallel connection. Both the SPM and the ANN receive influent chemical oxygen demand (COD), total kjeldahl nitrogen (TKN), ${\mathrm{PO}}_4^{3-}$ and ${\mathrm{NH}}_4^{+}$ data and timer output signals (for phase control) as inputs. The SPM output provides a preliminary prediction of the dynamic behavior of the ${\mathrm{PO}}_4^{3-}$ and ${\mathrm{NO}}_x^{-}$ concentrations. The outputs of the trained ANN compensate for the output errors of the SPM model. The hybrid model output of the final predictions of the process states is obtained by summing the outputs from both the SPM and ANN. Successful application of such a hybrid model is demonstrated.

Wastewaters containing benzoate and sulfate were treated in two upflow anaerobic sludge blanket (UASB) reactors at 34–37°C for 320 d. The sulfate concentration was increased stepwise in Reactor-A up to 7,500 mg/L, and was kept mostly constant at 3,000 mg/L in Reactor-B. Both reactors removed over 98% of organic chemical-oxygen demand (COD) for sulfate up to 6,000 mg/L, despite the fact that the mixed liquor contained up to 769 mg S/L of total sulfides and up to 234 mg S/L of dissolved H₂S. Sulfate-reducing efficiency decreased with the increase in sulfate concentration, but increased with time at each sulfate concentration. Reactor-B consistently reduced 89% of sulfate. However, both organic COD removal and sulfate-reducing efficiencies of Reactor-A dropped drastically at 7,500 mg ${\mathrm{SO}}_4^{-2}$/L, and showed no sign of recovery after 50 d. The system failure was likely due to the increased sulfate, instead of sulfide, toxicity. From the COD balance, 93.4% of COD removed was converted to methane instead of sulfides, with a net sludge yield of 0.047 g volatile suspended solids (VSS)/g COD. The sulfur balance was over 97%.

Three schemes proposed in the literature for analyzing joint toxic effects of multicomponent mixtures on fish, namely the additivity index (AI), the mixture toxicity index (MTI), and the similarity parameter (λ) are evaluated in this study for microbial toxicity. A new approach is proposed to establish acceptance limits for the similarity parameter, λ, based on experimental errors and uncertainties. Quantitative structure activity relationship (QSAR) techniques are then used to develop a model to predict the concentrations of components in mixtures that would jointly cause 50% inhibition of microbial respiration. The application of this approach is demonstrated on the experimental toxicity data of six eight-component organic chemical mixtures on microorganisms.

High-energy radiation has been investigated as an alternative method for sludge conditioning. Since the efficiency of sludge disinfection and toxic organic chemical destruction due to irradiation is well documented, this study concentrated on theoretical and experimental aspects of sludge dewatering enhancement (conditioning) due to irradiation. The primary parameters used for evaluation of conditioning were specific resistance of filtration, bound water current, and zeta potential of sludge. Experimental results showed that the scavenging power of sludge systems was higher at larger absorbed radiation doses. When the scavenging power exceeded 10⁹ (s⁻¹), the scavenger (KCNS), could completely eliminate the effect of conditioning of liquid sludge by radiation-absorbed doses up to 2,000 krads. It was observed that bound water content was reduced by irradiation processes, and sludge zeta potential became more positive with increased doses. The experimental results also showed that larger absorbed radiation doses yielded better treatment results, but generally with a lower efficiency within the tested dose range. These findings confirm that irradiation can reduce sludge bound-water content and enhance agglomeration by altering sludge particle surface charges.

The relationship between filtrate flow rate, cake liquid content, and cake structure compression is investigated for compressible structures in cake filtration. It has been observed that when the pressure is above a critical value, the filtrate flow rate is nearly independent of filtration pressure for highly compressible cake structures. In this study the specific resistance of waste-activated sludge is investigated and the critical filtration pressure, above which filtrate flow rate is independent of pressure, is found to be 0.02 bar, a value well below that used in dewatering processes. Local values of any point within the cake of specific resistance and liquid content are important to characterize the behavior of compressible structures in cake filtration. It is, however, difficult to make local measurements in the cake structure. Therefore, the local properties are often determined from overall sludge filtration parameters by calibration of mathematical models for cake filtration. These models require, however, knowledge of local functional relationships between structure stress, flow resistance, and liquid content. This requirement is a problem in the calibration of the models. The methods developed in this study avoid the requirement of local relationships but allow the determination of relationships between local structure stress, specific resistance, and liquid content from overall measurements. The values of local specific resistance and void ratio are determined for waste-activated sludge using the relationships. The relationships determined for waste-activated sludge show the logarithmic local specific resistance to be proportional to the logarithmic structure stress. The local void ratio is shown to decrease rapidly for increasing structure stress which corresponds to an extremely compressible behavior.

Cosolvents have been recognized to have a significant impact on the redistribution and movement of hydrophobic organic compounds (HOCs) in the environment. The cosolvent effect of oxygenated compounds (methanol, ethanol, and methyl tert-butyl ether) on the solubility of polynuclear aromatic hydrocarbons (PAHs) was examined by laboratory experiments. Coal-tar–contaminated sediment was used to study the redistribution and facilitated solubilization of 18 PAH compounds in both sediment and aqueous phases caused by cosolvent effects when oxygenated fuel spills occur. A batch dilution experiment was designed to represent the dilution effect of the cosolvent in a simulated river system. The results indicate that PAH solubility increased essentially in a log-linear manner with an increased volume fraction of methanol and ethanol. Deviations from a log-linear relationship were observed for methyl tert-butyl ether due to its limited aqueous solubility, and the cosolvent effect was seen only at the simulated fuel spill site. A linear relationship between the cosolvency power and logarithm of octanol/water partition coefficient was observed.

Laboratory tests were conducted to determine the effect of soil gas oxygen concentration on the degradation and mineralization of spiked ¹⁴C-pentachlorophenol and nonlabeled pentachlorophenol (PCP) present in soil taken from a prepared-bed land treatment unit at the Champion International Superfund Site in Libby, Mont. This soil was contaminated with wood preserving wastes including creosote and PCP. Degradation rates of ¹⁴C-PCP and nonlabeled PCP were found to be enhanced under soil gas oxygen concentrations between 2 and 21% in the contaminated soil. Between 48 and 64% of ¹⁴C-PCP spiked onto the soil was mineralized after 70 days at soil gas oxygen levels between 2 and 21%. No statistically significant mineralization of PCP was found to occur at 0% oxygen concentrations. Mineralization of ¹⁴C-PCP in contaminated soil poisoned with mercuric chloride was determined to be less than 0.2%. Degradation of indigenous nonradiolabeled PCP in the nonpoisoned soil was statistically significantly greater than in poisoned soil. These results indicated that degradation of PCP was biological and would occur under low oxygen concentrations. Soil gas oxygen concentrations necessary for PCP biodegradation (2–5%) could be maintained, for example, using bioventing technology in order to achieve continued treatment of buried lifts of soil while new lifts are added, thus decreasing the total time for soil remediation of the prepared bed.

Nitrate concentrations in surface water and especially in ground water have increased in Canada, the United States, Europe, and other areas of the world. This trend has raised concern because nitrates cause methemoglobinemia in infants. Several treatment processes including ion exchange, biological denitrification, chemical denitrification, reverse osmosis, electrodialysis, and catalytic denitrification can remove nitrates from water with varying degrees of efficiency, cost, and ease of operation. Available technical data, experience, and economics indicate that ion exchange and biological denitrification are more acceptable for nitrate removal than reverse osmosis. Ion exchange is more viable for ground water while biological denitrification is the preferred alternative for surface water. This paper reviews the developments in the field of nitrate removal processes.

In this study, a nonlinear programming approach using the successive quadratic programming optimization technique is developed for the optimal design of a pipeline network for water supply systems. The proposed method eliminates the equality constraints describing the hydraulics by a suitable choice of dependent and independent variables. The dependent variables are chosen based on graph theoretic decomposition of the network structure. This makes it possible to compute analytically the reduced constraints, objective function gradients, and reduced Hessian in a very efficient manner. This method of decomposition ensures that the nodal and loop balances are exactly satisfied and is robust for any initial starting point, able to handle incorrect initial flow directions. The method gives solutions comparable to the previous optimal solutions for the design of new as well as expansion of existing water distribution networks.

The anaerobic fluidized bed (AFB) reactor, using spherical granular activated carbon (GAC) particles as the media, is evaluated for the pretreatment of hog wastewater. A laboratory investigation is performed to assess the effects of organic loading-rate on organic removal and methane production. The experimental evidence indicates that good removal of five-day biochemical oxygen demand (BOD₅) (filtered and unfiltered) and filtered chemical oxygen demand (COD) (i.e., >60%) can be achieved at loadings as high as 10.4 g unfiltered COD L/d. On the other hand, the removal of unfiltered COD was unsatisfactory because of the presence of high volatile suspended solids (VSS) concentrations in AFB effluents. The observed specific methane yield decreases with increasing loadings; however, the biogas produced is rich in methane, especially at low loadings. Attached biomass densities ranging from 0.01 to 0.024 g volatile solids (VS)/g carbon are attainable, which accounts for >99% of total reactor VS. Therefore, the GAC particles used are conducive for buildup of attached biomass in AFB reactors.

The air pollution monitoring data records are not entirely complete, making it difficult to estimate the required statistical parameters with an acceptable error and level of confidence. An intercomparison of various strategies for interpolating daily values over the Madrid urban air pollution monitoring (UAPM) network has been undertaken for the following set of air pollutants: SO₂, NO₂, NO_x, CO, and suspended particulate matter (SPM). The methodology accounts for either the algorithms [distance, spatial correlation (or Alexandersson's) and multiple regression] or the data input (original, log-transformed and log-standardized). The well-known Alexandersson's test has been implemented as a spatial correlation method in order to interpolate missing daily concentration data. In general, the distance method and the spatial correlation method are greatly superior to multiple regression regardless of the pollutant considered; they provide good estimates when applied to log-standarized and original data, respectively. Given that the distance method does not lead to a parsimonious model and is difficult to apply (it is not clear and is not intuitive), the spatial correlation method is proposed as the best interpolating algorithm.

Ambient air samples were collected using Tenax-TA adsorbents at sites of tunnels and urban areas in Tainan, Taiwan, and a remote recreation area and national park in central Taiwan. A gas chromatograph equipped with mass spectrometry was then used to analyze volatile aromatics in the samples. Experimental data indicate that the airborne concentration at the national park was very low, as expected. The concentrations in motorcycle and car tunnels were significantly higher than those in the background by two orders of magnitude. To further explore the impact of vehicle exhaust on the airborne aromatic contents in Tainan, characteristic ratios were used to evaluate the experimental data. The characteristic ratios for the tunnels, urban area, and the national park were found to be different.