Journal of Environmental Engineering

Analytical procedures are proposed for estimating probability distributions of travel times for chemical waste loads to groundwater. Travel distance of the chemical is treated as a renewal process, and travel time is given by the number of annual chemical displacements or renewals required for passage through the soil unsaturated zone. The methods were evaluated for two organic chemicals for a range of site conditions near Lexington, Kentucky, and Des Moines, Iowa. It was found that groundwater contamination risk is very dependent on site‐specific soil and hydrologic conditions. Probability distributions based on both hydrologic and soil spatial variability differed markedly from those that considered only hydrologic variability. The effect of soil spatial variability is to increase travel‐time variability and risk of groundwater contamination. Travel‐time probability distributions were combined with biochemical degradation rates to estimate the expected fraction of chemical mass reaching the water table. This fraction serves as a screening criterion to identify and rank combinations of chemicals and sites that are likely to produce groundwater pollution.

Measurements of physical and chemical solute transport properties of soil/waste systems are necessary to predict pollutant migration in soils. The sorption distribution coefficient is most often determined in the laboratory by performing column studies or batch equilibrium tests. Both of these methods suffer from disadvantages. For fine‐grained soils, which have small permeabilities, column studies require long periods of time to conduct and batch tests usually overestimate the sorption distribution coefficient because of the solids effect. An alternative method is developed that uses a constant head permeameter spinning in a centrifuge. As the permeameter spins, larger confining stresses are imposed on the soil sample and pore water velocities are increased relative to ordinary column studies. The utility of the method was demonstrated by conducting miscible displacement experiments for a range of column sizes and inertial accelerations. Sorption distribution coefficients, based on the local equilibrium assumption, were computed from experimental breakthrough curves using an analytical solution to the solute transport equation.

A new geophysical tool for high resolution subsurface imaging has been developed which has numerous applications in the field of hazardous waste management. The primary steps of the technique are: (1) An acoustic wave is projected through a subsurface cross section to a hydrophone array located vertically in a borehole; (2) the amplitude versus time signal of the propagating wave is collected at each hydrophone; and (3) the reduced wave equation and theory of backpropagation tomography are invoked to reconstruct the subsurface features from the collected data. The development of this emerging technology is highlighted beginning with its theoretical conception, through system development and field testing, to an examination of its future applications in environmental engineering.

Experimental data and hypothetical process behavior are analyzed in an attempt to elucidate the kinetics of aqueous phase ozone decomposition reactions. Three commonly accepted mathematical techniques (integral, differential, van't Hoff) for the analysis of rate data are employed. The results indicate that when single‐order dependencies are used to model autocatalytic reactions, timevariable behavior of the rate constant and order is exhibited. This can partially explain existing controversy within the scientific community regarding the apparent order of ozone decomposition reactions. A simple autocatalytic rate expression is developed that accurately describes the decomposition rate data. The conclusions from the work have implications not only for development of an understanding of ozone decomposition kinetics, but for the analysis of rate data derived from nonelementary reactions in general.

Several thermodynamic models that may be used to predict single and binary adsorption equilibria for organic vapors and competitive interactions between water vapor and a single organic vapor were developed and tested. The low organic vapor concentrations and humidities that were examined are similar to those that would be encountered in air stripping tower air emissions. The Dubinin‐Radushkevich (D‐R) equation was shown to predict single‐component adsorption equilibria for trichloroethene (TCE) and tetrachloroethene (PCE) using a reference isotherm which was conducted on toluene and physical properties of TCE and PCE. Both the Polanyi potential theory and ideal adsorbed solution theory (IAST) were shown to predict binary adsorption equilibria for TCE and PCE using single component isotherm data. The competitive interactions between water vapor and TCE were predicted using TCE isotherm data (which was conducted at low relative humidities), a water vapor isotherm, and the pore size distribution of the carbon.

Radon, a naturally occurring radioactive gas, is drawn from the soil into a house when low air pressure exists in the house. This is a commonplace environmental hazard in the United States, Canada, and northern Europe. The U.S. Environmental Protection Agency (EPA) is developing and demonstrating procedures to use in reducing the radon concentrations in a variety of house types. Until recently, research has focused on basement houses because of their great potential for radon entry; however, other housing substructures also present unique radon problems. Several radon reduction alternatives for crawl space houses are noted, and the successful demonstration of one of these alternatives, subplastic suction, is described in detail. The findings of this study need to be confirmed and supplemented with more measurements, preferably on larger and more complex crawl space houses, and more houses need to be mitigated to provide a statistical test of the subplastic suction technique.

A time series analysis approach was applied to model 21 years of mean monthly water quality data in the Guangzhou reach of the Pearl River in southern China. The basic properties of the water quality data time series were determined, time‐ and frequency‐domain analyses were carried out, and the dependent stochastic component was represented by various stochastic models. Synthetic water quality data were generated by using the probability distribution of the independent residuals, and forecasting of future water quality data was done using a Box‐Jenkins‐type difference model. Eighteen years of data were used for model development, while the model performance was compared with the data for the remaining three years. The comparisons were found to be satisfactory.

Sodium and chloride contamination of shallow groundwater can be frequently attributed to three major sources: (1) Individual sewage disposal systems (ISDS); (2) road salt application during winter deicing activities; and/or (3) road salt storage. Effluent streams in a large reservoir watershed were sampled during baseflow conditions. Sodium and chloride concentrations were regressed against different land use types including roadway, residential, commercial, industrial, agricultural, and forest areas. Results indicate that roadway areas are the largest single influence of sodium and chloride concentrations. Data from other watersheds are compared to the developed relationships and agreement is favorable. Additionally, a mass balance is performed and indicates that, typically, roadway deicing activity in the test watershed can potentially introduce far greater amounts of sodium and chloride in the local groundwater than the next greatest influence, residential septic systems.

Krypton‐85 was used to measure reaeration coefficients in numerous streams and rivers in Texas with various hydraulic characteristics. Flow rates ranged from 0.027 to 16.0 cms, velocities from 0.009 to 0.87 m/s, and depths from 0.17 to 1.83 m. Using a linear multiple regression technique, the Texas Reaeration Equation was developed for estimating reaeration coefficients in Texas streams and rivers. Statistical and graphical tools were used to evaluate this equation for accuracy and reliability and to determine the limits of its application. The methods of analysis of variance and analysis of residuals were used to evaluate the accuracy of the Texas Reaeration Equation. The 95% confidence intervals of the regression coefficients were also determined. The Texas Reaeration Equation adequately predicted reaeration coefficients in Texas streams and can be reliably applied to streams with depths from 0.2 to 1.0 m and velocities from 0.01 to 0.30 m/s.

Extensive land use, nonpoint source (NPS), and lake water quality data are used to determine the interaction of lake water quality and watershed development in the Lake Austin watershed. Historical trends in total suspended solids and total phosphorus are examined over the 1979–84 period, during which impervious cover increased by 79% in the lake watershed. A PC‐based continuous stirred tank reactor (CSTR) model is used to depict the response of lake water quality to loadings associated with a variety of development options ranging from impervious cover of 10–40%. The reliability of the model is supported by the unusually simple lake hydraulics, exclusive NPS loading, and extensive data base. Model results are used to create a set of recommendations that should protect the water quality of the lake.

This paper describes work toward development of a convenient decision support system to improve everyday operation and control of the wastewater treatment process. The goal is to help the operator detect problems in the process and select appropriate control actions. The system comprises subsystems for database management, data analysis including several statistical methods, and expert advice. The statistical methods investigated include construction of external reference distributions, computation of moving averages, graphical methods, discriminant functions for early detection of upsets and several kinds of control charts. The latter two methods are not built into the expert system. Reference distributions are used to help define fuzzy terms such as high and low for the conditions of the variables involved in the expert system control rules. The system is written in dBase‐III and Lotus 1‐2‐3 and has a user‐friendly interface for communication between the user and the system.

A post-audit study was made of a predictive water quality model. An analytical model using annually averaged parameters is in good agreement with recent data. The model uses an analytical loading function, first-order kinetics, and constant coefficients. The model is accurate in predicting both whole-body dieldrin residue and oil-normalized dieldrin residue in carp for an impoundment of the Iowa River. The oil-normalization of the field data reduce its variability of dieldrin in mg/kg oil and emphasize the importance of lipid bioconcentration of hydrophobic pollutants. Dieldrin has a low water solubility, and partitions to sediments. Carp were chosen as the worst case for pesticide accumulation because of their high fat content and bottom-feeding behavior. Dieldrin accumulation data for carp were available for the period 1968–1988. Dieldrin concentrations are expected to be 30 μg/kg in carp in 1990.