Journal of Environmental Engineering

This paper presents a methodology for assessing the effects of source-, chemical-, and aquifer-related parameter uncertainty on the response of a semianalytical transport model using first- and second-order reliability methods. A probabilistic model is developed by coupling the deterministic transport model with a general-purpose probability analysis program. Ground-water contamination risk is addressed by evaluating the probability that a given contaminant exceeds the regulated standards at a well downgradient from a waste source. The general applicability of the methodology is demonstrated on two simple hypothetical case studies of transport of nonreactive and reactive solutes in the subsurface. Sensitivities of the probabilistic outcome to the basic uncertainties in the input random variables are provided through importance factors. The reliability-method results are checked against those obtained using the classical Monte Carlo–simulation method, and the results are in good agreement except for very low probability events in which the reliability methods provide accurate results much more efficiently than the Monte Carlo method. The need for a careful trade-off analysis between accuracy and computer time is highlighted.

The maximum likelihood (ML) method for regression analyses of left-censored data is improved for general acceptance by considering the censored observations to be doubly censored. The existence of a lower bound (i.e., the concentration of a pollutant cannot be negative) is included; the improved ML method utilizes this information in the formulation of a likelihood function. The improved ML method has been translated into an equivalent least squares (LS) method, and an iterative algorithm is presented to estimate the statistical parameters from this LS translation. The LS translation is easy to explain to nonstatisticians, and computational requirements for implementing the LS method are minimal. The methodology is applied to a mechanistic model for air transport and deposition of polycyclic aromatic hydrocarbons (PAH) to a snow surface. For a censored data set, parameter estimates of the model, namely, dry deposition velocities and washout ratios, were obtained for various PAH species by using the following three procedures: (1) the NAG-15 routine for maximization of a likelihood function; (2) the proposed algorithm for the equivalent LS method; and (3) the modified iterative least squares method.

The effect of uncertainty in the system parameters on the reliability of domestic-waste biofilm reactors is investigated. Based on typical ranges and default values reported in published literature, a representative set of statistics describing the uncertainty in the parameters is selected for this study. Monte Carlo simulation runs are performed, from which the variability in the reactor height required to meet a specified effluent concentration level is estimated. The parameters contributing significantly to this variability are also established. From the results, a preliminary set of factors corresponding to various performance reliability levels is proposed by which the reactor height from deterministic analysis should be multiplied to arrive at the design reactor height. A similar set of partial factors that one can use to factor the values of the system parameters prior to the deterministic analysis to compute the reactor height is also proposed as an alternative means of accounting for the effect of uncertainty. From an operational point of view, the variability in the effluent quality resulting from the uncertainty in the system parameters is also investigated for a reactor with a fixed height.

Trichloroethylene (TCE) and cis-1,2-dichloroethylene (DCE) cometabolic degradation by a filamentous, phenol-oxidizing enrichment from a surface-water source were investigated in batch tests. No intermediate toxicity effects were evident during TCE or DCE degradation for loadings up to 0.5 mg TCE/mg VSS or 0.26 mg DCE/mg VSS. Phenol addition up to 40 mg/L did not inhibit TCE or DCE degradation. TCE specific degradation rates ranged from 0.28 to 0.51 g TCE/g VSS-d with phenol present, versus an average endogenous rate of 0.18 g TCE/g VSS-d. DCE specific degradation rates ranged from 0.79 to 2.92 g DCE/g VSS-d with phenol present, versus 0.27 to 1.5 g DCE/g VSS-d for endogenous conditions. There was no inhibition of DCE degradation rates at concentrations as high as 83 mg/L. TCE degradation rates declined between 40 and 130 mg/L TCE. Perchloroethylene, 1,1,1-trichloroethane, and chloroform were not degraded.

A two-stage bioreactor system was used to reduce Cr(VI). Escherichia coli ATCC 33456 cells were aerobically grown in a first-stage chemostat, and then pumped to a second-stage plug-flow reactor where anaerobic Cr(VI) reduction occurred. Experimental results demonstrated that near complete removal of Cr(VI) was achieved in the plug-flow reactor under appropriate operating conditions. The removal efficiency in the plug-flow reactor was significantly affected by the influent Cr(VI) concentration, Cr(VI) loading rate, liquid detention time, and a consumed Cr(VI) reduction capacity factor. The consumed Cr(VI) reduction capacity factor was directly related to the toxicity effect of Cr(VI) on cells. To illustrate the influence of the consumed reduction capacity factor on Cr(VI) reduction in the plug-flow reactor, a mathematical model was developed by incorporating Cr(VI) reduction kinetics into the mass balance relationship of Cr(VI) for a plug-flow reactor. Analyses of Cr(VI) concentration profiles along the plug-flow reactor using both model simulations and experimental data indicated that the rate of Cr(VI) reduction decreased with the depletion of the reduction capacity of cells and Cr(VI) reduction ceased after the reduction capacity was exhausted.

The in-situ remediation of a lead-contaminated soil (silt loam, K _H = 5 × 10 ⁻⁸ cm/s, soil Pb = 1,000 mg/kg) by electrokinetic (EK) soil flushing [60 V (DC)] was studied. Research focused on the chemical conditioning of the electrode reservoirs with either 500 μS/cm (as NaNO ₃ , baseline behavior), acetic acid (HAc), HCl, or EDTA. For baseline tests there were significant amounts of lead transported through the soil, but the Pb precipitated or was readsorbed on the soil adjacent to the cathode because of the high soil pH in that region. The addition of 1 M HAc to the cathode reservoir prevented the formation of the basic conditions in the soil, and about 65% of the Pb was transported into the cathode. When HCl was added to the anode and HAc was added to the cathode, more than 75% of the lead resided in the cathode. Pb removals in the EDTA experiments were greater than those observed in the baseline experiments and were similar to those observed in the HCl-HAc experiments. A low anode reservoir pH resulting from a high current was the most likely reason.

Coal piles stored directly on soil surfaces may pose a risk of ground-water contamination, particularly when extremely acidic coal-pile leachates, which contain elevated concentrations of heavy metals, are generated. Soil-column experiments were run to evaluate the ability of a sand, a silt loam, and a silt loam amended with 4% CaCO ₃ to buffer dilute H ₂SO ₄ solutions (pH 3.5 and 1.5) and a synthetic coal-pile leachate (a pH 1.5 H ₂SO ₄ solution containing metal sulfate salts). Both silt-loam soils were able to buffer the pH 3.5 H ₂SO ₄ solution for the entire leaching period (more than 50 pore volumes), but the sand was not. The buffering capacities of all three soils were insufficient to neutralize pH 1.5 acid and synthetic coal-pile leachates. The sand exhibited the least buffering, and the silt loam at a low flow rate and the silt loam amended with CaCO ₃ showed the greatest buffering. The buffering capacities of both silt loam soils were exhausted more rapidly when leached with synthetic coal-pile leachate rather than pH 1.5 H ₂SO ₄ solution. This suggests that soil-buffering studies using pure acid solutions may underpredict contamination due to coal-pile leachates. The results indicate that there is a potential for ground-water contamination due to coal-pile storage, particularly where leachates are acidic and coarse textured soils exist. Coal storage practices should be adopted that protect soils and ground water from acidification and metals contamination.

Sewer headspace H ₂ S reduction by precipitating dissolved sulfide in wastewater was investigated using iron salt (FeCl ₃ and FeCl ₂ ). Full-scale experiments were conducted in a 40-km (25 mi) sewer with an average flow of 8.7 m ³ /s (200 mgd). Results were sensitive to total Fe dosages and Fe(III)/Fe(II) blend ratios injected. A concentration of 16 mg/L total Fe and a blend ratio of 1.9:1 [Fe(III):Fe(II)] reduced dissolved sulfide levels by 97%. Total sulfide and headspace H ₂ S were reduced by 63% and 79%, respectively. Liquid and gas-phase sulfide reductions were largely due to the effective precipitation of sulfide with Fe(III) and Fe(II) and the limited volatilization of H ₂ S, respectively. Oxidation of sulfide in the presence of Fe(II) and minute amounts of O ₂ may have occurred. A combination of Fe(III) and Fe(II) proved more effective than either salt alone. By using excess Fe(III), dissolved sulfide can be reduced to undetectable levels. No specific relation between the concentration of Fe or Fe(III)/Fe(II) blend ratio and sewer crown pH was inferred. Iron salts may retard crown corrosion rates by precipitating free sulfide and reducing its release to the sewer headspace as H ₂ S. A mechanism to inhibit certain responsible bacteria was not established in the 40-km (25 mi) sewer.

A series of dye-tracer experiments was performed in and around a river plume in Kootenay Lake, British Columbia. River and stream inflows are often considered to be a method of nutrient or waste introduction and dispersal; in this paper we show that they can be unpredictable and highly variable in their behavior over short time scales. Consequently, diffusivities observed over small length scales and short time scales are likely to differ greatly from those that might be attributed to average flow conditions in the absence of detailed observations. The observed presence of velocity shears associated with wind stress, river inflow, and basin geometry suggests that the phenomenon of shear dispersion augments the stirring action of local turbulence. The densimetric Froude number indicates that the river behaves as a mixing layer and values of the apparent horizontal diffusion coefficient in and near the river plume are of the order of 2–5 m ² s ⁻¹ and the dispersion in the direction of the river inflow is separated from the transverse turbulent diffusion.

Although some state regulations allow a mixing zone near a discharge outfall, many National Pollutant Discharge Elimination System (NPDES) permits still require that ambient water quality criteria be met at the end of the pipe. This paper presents a conservative approach to analyzing mixing zones for discharges with minimum initial dilution. An application of this approach to the effluent discharged from the Falling Creek wastewater-treatment plant to the James Estuary is presented. Model results have been approved by the regulatory agencies to lift an overly stringent effluent-toxicity limit while preventing adverse ecological impact on the receiving water.