Journal of Environmental Engineering

This paper attempts to contribute to the fluid dynamics knowledge on the migration of chloroform, and other related volatile organic constituents, in ground‐water systems, and to complement the existing qualitative information with the development and analysis of quantitative models. Two exploratory models are developed and tested as to their ability to reproduce the limited historical information on the evolution of the contaminant concentration in the unsaturated and saturated zones of the alluvial aquifer in Louisville, Kentucky. The model's development and an uncertainty analysis originating from measurement errors, model assumptions, nonpoint source distribution, and computational approximations, are conducted using recent contributions to the theory of stochastic partial differential equations.

Results from a field program to collect data for the evaluation of alternative collection‐well design parameters for the hydraulic containment and removal of dense non‐aqueous‐phase liquids (DNAPL) and aqueous‐phase liquids (APL) are described. The Cooper‐Jacob method is utilized to estimate the aquifer transmissivity and storage coefficient. Estimates of hydraulic conductivity and permeability from drawdown data for DNAPL and APL are derived to indicate how the relationships from fluids principles may be successfully employed for the design of recovery systems. The diameter of the pumping‐well casing or borehole is shown to not influence the long‐term volume of DNAPL that can be pumped. Large‐diameter wells are found to take longer to equilibrate and allow intermittent but high‐volume pumping for short durations. With small‐diameter wells, a larger drawdown is obtained for the same pumping rate, and this collects more APL; but the small‐diameter well creates a larger gradient and thus more incentive to move the DNAPL for recovery.

A mathematical model is developed to evaluate applying municipal wastewater sludges to forest land. The model is used to estimate sludge loading rates that produce steady state nitrate concentrations in percolation that meet water quality standards. Annual loading rates for aerobically and anaerobically digested sludges consistent with a 10‐mg/l standard are computed for conifer and hardwood forests in the eight forest regions of New York. These rates vary from 1.7 to 2.7 Mg/ha of dry mass for conifers and from 1.4 to 2.3 Mg/ha for hardwoods among the various forest areas, or in terms of sludge nitrogen, 85 to 135 kg/ha for conifers and 70 to 115 kg/ha for hardwoods. Sludge application rates are sensitive to annual nitrogen uptakes by vegetation, and estimates of these quantities are very uncertain. A geographic information system is used to identify the amount of forest land within each region that meets state regulatory restrictions for sludge disposal. It is determined that sufficient forest land is available to accept large portions of the state's sludge production.

Sampling in an incompletely mixed effluent was studied with the aid of a hydraulic model. The effluent was the discharge from a major wastewater treatment facility. Residual chemicals from the chlorination/dechlorination process were of concern. Concentration time‐series data obtained in a 1:12 scale model were used to determine the best sampling location and duration and to predict expected sampling errors relative to spatially and temporally averaged concentration values. Some general concepts for sampler placement in not fully homogeneous streams were derived. Spatial concentration gradients have to be determined to select number and location of multiple sampling points. When concentrations fluctuate in time due to process controls, the flow patterns and velocity distributions have to be known to select representative sampling points. For example, representative sampling within a separated flow region requires carefully selected probe placement and long sampling durations. Physical model simulations can provide the information on flow and concentration time series that are necessary to select the sampling strategy.

A simple modification of the conventional countercurrent air‐stripping process is proposed as “cascade air stripping” for cost‐effective removal of volatile organic contaminants from water. The feasibility of cascade air stripping in removing organic contaminants of very low volatility (e.g., 1,2 dibromo‐3‐chloropropane, or DBCP) from water was investigated in a prototype scale study. The results of this study snowed that the cascade system could remove more than 95% of DBCP by allowing more air flow than the conventional air‐stripping system at lower pressure drops and equal power input. This study also enabled the process model to be verified under three different packing configurations and several air‐to‐water ratios. From the results, it can be seen that the proposed system extends the viability of air stripping to contaminants of low volatility. These prototype scale results were found to be in agreement with the findings of a previous pilot‐scale study done on common volatile and semivolatile contaminants.

Adsorption onto activated carbon in a fixed‐bed reactor has been found to be an attractive process for removing hazardous organics from water and wastewater. Current fixed‐bed reactor design procedures can be both expensive and time‐consuming. Mathematical process modeling can reduce the cost and time by decreasing laboratory and pilot‐scale experimentation. A model will provide the most utility if it is computationally efficient, and both stable and accurate for a wide variety of system conditions. A global method of orthogonal collocation is currently favored over finite differences. This method, however, results in an unstable algorithm as a result of spatial oscillation. The use of the semidiscrete Galerkin finite‐element method with an asymmetric interpolating function was found to eliminate this spatial oscillation, resulting in a computationally efficient algorithm. A relationship was developed that allows the optimum value of the asymmetric parameter to be determined. This results in a stable algorithm and minimizes numerical dispersion. Additionally, the model predictions were in excellent agreement with experimental results.

A drift‐flux formulation that accounts for multiple bubble interactions is applied to describe the mean water velocity, air fraction, and plume growth for a vertical air diffuser plume that is bound on one side. A coarse‐bubble line‐diffuser source in a rectangular mixing tank is considered. Physical velocity and plume width measurements are used to calibrate the numerical model. The model is shown to be sensitive to the initial plume width and the entrainment coefficient. A dimensionless equation for the ambient water velocity at the outer edge of the plume is developed from physical measurements. The water velocity and air fraction are assumed to be uniform over the plume width. Four ordinary differential equations are developed and solved simultaneously to obtain the changes in significant plume variables with fluid depth. To reproduce the observed plume characteristics, the entrainment coefficient was 30‐60% higher than the single‐phase jet entrainment coefficient. The effect of the wall friction was minor for depths less than 1 m.

The U.S. Environmental Protection Agency has recently promulgated drinking‐water regulations, known as the surface water treatment rule (SWTR), to control viruses, Giardia cysts, Legionella, and heterotrophic bacteria. The SWTR specifies overall minimum removal/inactivation efficiencies by filtration and disinfection for Giardia lamblia and viruses and uses the CT concept to predict in‐activation efficiencies. CT stands for the product of characteristic exposure time (T) and the characteristic concentration (C) in the disinfection chamber. This paper evaluates possible methods to determine the characteristic exposure time for ozone disinfection. The extent to which different measures of T (such as the hydraulic retention time, $T_{10}$ and $T_{50}$) actually ensure compliance with the required level of inactivation is evaluated by using segregated flow analysis and classical residence time distribution models. The analysis indicates that $T_{10}\text{,}$ the minimum exposure time of 90% of the water in the disinfection unit, is a good measure to estimate disinfection performance in most cases. $T_{50}$ does not guarantee compliance with the SWTR in most cases.

It is claimed that waste‐stabilization ponds provide an efficient, inexpensive, and appropriate form of wastewater treatment for developing countries. Based on negative experiences in Tanzania, this paper concludes that waste‐stabilization ponds designed by standard western design practice cannot always be expected to perform well without at least minimal operator training and funds for operation and maintenance. Moreover, in this study the performance of a series of waste‐stabilization ponds in Dar es Salaam, Tanzania, is investigated. The removal efficiencies of permanganate value (PV), biochemical oxygen demand $\left({\mathrm{BOD}}_5\right)\text{,}$ suspended solids, and ammonia nitrogen are good. ${\mathrm{BOD}}_5$ has a good correlation with PV. Though the fecal coliform removal efficiency is impressive, the mean survival is high. While dissolved oxygen has a marked variation with pond depth, temperature has no significant variation with depth. It is suggested that there is the potential for improving the mechanistic and analytic approaches characterizing the present waste‐stabilization pond research efforts.

This paper presents results of a study undertaken to rationalize the incubation time and temperature in biochemical oxygen demand (BOD) tests with respect to higher ambient temperature conditions in tropical countries. The experimental protocol comprised two phases. In the first phase studies were concurrently conducted at the laboratories of the National Environmental Engineering Research Institute (NEERI) at Nagpur, Bombay, Hyderabad, Kanpur, and Madras to estimate progressive BOD values. They also estimated the reaction kinetics for each of the 110 sets of samples at 20, 25, 30, and 35°C with acclimated seed. In phase 2, further studies were carried out at 27°C incubation temperature in five NEERI laboratories, along with laboratories of Jadhavpur University, Central Pollution Control Board (CPCB), and pollution control boards from 10 Indian states. The protocol included 17 replicates of six pure chemicals, sewage samples from respective cities, wastewaters from six types of industries, and surface water samples. Statistical analyses of data reveals that the BOD_5,20 is equivalent to the BOD_3,27 irrespective of the composition of test sample.