Journal of Environmental Engineering

Concerns about environmental data quality and the need for data documentation are increasing due to increased accessibility to and demand for environmental data. Local, state, and federal agencies as well as consultants, researchers, and the public are interested in current, reliable, accessible, and understandable data. It is often necessary to rely on data obtained from secondary sources, originally collected for purposes other than present applications. Information describing form, specificity, and processing are needed, in addition to assessment of data quality, to make informed decisions about the usefulness of secondhand data. This paper merges the thinking of environmental sciences and spatial information systems, and presents an initial proposal for an environmental data lineage report. The proposal describes two kinds of information needed in a lineage report—an overview of the data collection process, and an overview of relevant legal and organizational factors driving the data collection. The process description includes information about sample design, site conditions, instrument calibration, data recording and processing, and so forth. Description of legal and organizational factors such as laws, decision criteria, and existing standards explain the original purpose for data collection.

An electrical conductivity method has been developed to determine the effective diffusivity of contaminants through porous solids, such as hazardous waste solidified with cementitious binders. The electrical conductivity method calculates the MacMullin number, which is the ratio of the effective diffusivity of a compound in a porous solid to the molecular diffusivity of that compound. The effects of water‐to‐cement ratio, curing time, and sample thickness on the MacMullin number were determined. The validity of this method was established through comparison with effective diffusivities measured by conventional techniques. These values were measured during the study with a two‐chamber diffusion device; other values were obtained from the literature. Correlations of the MacMullin number with porosity and bulk density were developed. The electrical conductivity method was found to be a rapid and reliable method for measuring effective diffusivities of contaminants in porous solids such as solidified hazardous wastes and offers the advantages of much more rapid analysis than conventional methods.

An investigation is conducted to determine the potential for and extent of ground‐water contamination by organic matter arising from ferrous foundry solid‐waste landfills. In the laboratory, the Environmental Protection Agency's (EPA's) toxicity characteristic leaching procedure (TCLP) is used on waste samples representing nine common binder systems to identify organic compounds that may be released from these materials in landfills. Leachates are analyzed by gas chromatography/mass spectrometry (GC/MS) for qualitative analysis against a data base of 45,000 compounds, and by gas chromatography/flame ionization detector (GC/FID) for quantitative analysis. A field study at four ferrous foundry waste landfills is conducted to determine presence of organic compounds leached from the solid wastes. Laboratory results indicate that although a wide variety of organic compounds are detected, most are present at low concentrations. Core oil and phenolic urethane binder systems leached the largest number of organic compounds and at the highest concentrations. No samples produced concentrations above the TCLP regulatory levels. In field investigations, trace concentrations of some organic compounds, all below quantitative limits, were found in ground‐waters adjacent to three landfills.

Two microcosm experiments were conducted to study the effects of low pH on leaf decomposition in low‐alkalinity lakes. In the first experiment, a mix of oak and birch litter was incubated for six months in water from Little Rock Lake, Wisc., at pH 5.0 and 6.5. In the second experiment, maple litter was incubated for three months in water from Round Pond, Mass., at pH 4.0, and in water from Walden Pond, Mass., at pH 6.0. In both experiments, litter weight loss was less at the lower pH. However, the initial carbon loss from maple litter at pH 4.0 was faster than at pH 6.0 in the second experiment. Calcium and magnesium leaching were greater at the lower pH in both experiments, suggesting that low pH may also be associated with increased leaching of individual elements. Data on ionic inputs to Little Rock Lake from leaves demonstrate the potential importance of leaves as a nutrient source to seepage lakes, or any lakes in which leaf inputs may be significant.

A simple laboratory test, referred to here as the dredging elutriate test (DRET), was sought to predict the concentration of contaminants in the water column at the point of dredging. The DRET is procedurally similar to the modified elutriate test developed by the U.S. Army Corps of Engineers to predict concentrations of contaminants at the point of disposal of dredge material. The objective of experimentation was to investigate the release of dissolved and suspended polychlorinated biphenyls (PCBs) as a function of initial concentration of sediment, aeration time, and settling time in the DRET. Settling time was found most important. Field data from a pilot dredging project in New Bedford Harbor, Mass. compared well to results of the DRET. Most of the PCB in the field and DRET were sorbed to small particles rather than dissolved. A good correlation was obtained between total PCB and total suspended solids. A sorption equilibrium model based on available partition coefficients described the sorbed‐dissolved PCB distribution. Particle size distribution analysis showed that the mean diameter was less than 10 μm after just 0.5 h of settling and decreased very little with further settling. These particles and their PCB burden have the potential to be transported away from a dredging site.

In this paper, the sediment‐side abiotic mechanisms for contaminant transport across the sediment‐water interface of rivers are compared. The transport processes considered include molecular diffusion, colloidal diffusion, pore‐water advection between the stream and its associated aquifer, localized advection due to sediment surface roughness, and sediment movement. A simple model for contaminant transport by active bed‐load transport of noncohesive sediments in triangular dunes is developed and compared to both linearly sorbing and nonsorbing contaminant transport via passive pore‐water processes. The model is limited to homogeneous noncohesive sediment beds and specific flow conditions that lead to steady dune‐migration rates. The results suggest, however, that advection should normally be the dominant mechanism in stable sediments and for nonsorbing contaminants in unstable sediments near the threshold for sediment movement. Sorbing contaminant transport should normally be controlled by the rate of sediment movement.

The maximum capacity chemical oxygen demand (COD) loading in upflow anaerobic sludge blanket (UASB) reactors is evaluated using three 8.5 L reactors and high‐strength synthetic wastewaters composed of milk and sucrose at 37°C. The study was conducted over a wide‐range COD loading rate (18–260 g·L⁻¹·day⁻¹), by varying hydraulic retention time (HRT) (1.8–10 hr) and COD levels in wastewater (6,000–20,000 mg/L). At COD loading rates of up to 160 g·L⁻¹·day⁻¹, which corresponded to HRT of 1.8 h and COD of 12,000 mg/L in the wastewater, the process removed 94–98% of soluble COD and 75–90% of total COD. At rates higher than 160 g·L⁻¹·day⁻¹, the efficiency of COD removal deteriorated because of sludge disintegration and washout resulting from vigorous mixing by the biogas produced. The capability of volatile suspended solids (VSS) to convert COD to methane increased with COD loading; it reached the maximum of 1.7 g methane COD per gram VSS per day at the COD loading of 100 g·L⁻¹·day⁻¹, corresponding to the food‐to‐microorganism (F/M) ratio of 3 g COD per gram VSS per day. Characteristics of the sludge granules, effluent, and biogas are also discussed.

This research documents the use of oxidation‐reduction potential (ORP) to control two lab‐scale aerobic‐anoxic sludge digesters. The reactors were operated in sequencing batch reactor fashion, and one reactor had the ratio of air‐on/air‐off arbitrarily “fixed” at 3 h for each portion of the cycle. The other reactor operated in a real‐time manner with 3 h of air‐on and the air‐off time determined by computer detection of the nitrate breakpoint in the orp‐time profile. Reactor performances were evaluated from a chemical (mass balance) removal perspective [total and volatile suspended solids (TSS and VSS), nitrogen, and phosphorus] coupled with a consideration of the control stability achieved when the reactors were subjected to spikes of sodium nitrate, ammonium chloride, and hydrogen peroxide. The reactors showed comparable chemical removals, however the real‐time control reactor displayed a greater ability to accommodate the disturbances investigated.

A mathematical model (CRAS) has been developed to evaluate the behavior of trivalent chromium [Cr(III)] in the activated‐sludge process. The validity of the model was tested by comparing its prediction with experimental data. Laboratory‐scale continuous‐flow activated‐sludge reactors receiving a continuous Cr(III) dosage provided the data. The independent experimental parameters were sludge age $\left({\theta }_x\right)$ and $\mathrm{pH}:{\theta }_x=5\text{,}10\text{,}\mathrm{and}15\mathrm{days}$; $\mathrm{pH}=6\text{,}7\text{,}\mathrm{and}8$. Comparison of the experimental results with the predictions revealed that: (1) More Cr(III) was intracellularly transported into the solid phase than Cr(III) adsorbed on activated sludge; and (2) the solid‐bound Cr(III) per unit weight of suspended solids in the effluent was considerably greater and more varied than that in the mixed liquor. By incorporating these two findings, the CRAS model was found capable of predicting quantitatively the percentage removal of Cr(III) (the average relative error was 4.5%) and the total Cr(III) concentration in the mixed liquor (14.9%) as a function of pH and ${\theta }_x$.

The mechanisms responsible for overdosing are studied using organic polymers as well as inorganic chemicals as sludge conditioners. Anaerobically stored sludges were used to minimize changes of sludge characteristics due to microbial activity within the period of experimentation. Conditioned sludge is considered as a two‐phase system consisting of a liquid phase and a solid phase. Sludge dewaterability is characterized by capillary suction time (CST) and specific resistance to filtration (SRF), and the sludge liquid phase is characterized by its viscosity. Based on the experimental results, it is concluded that: overdosing is only associated with polymer conditioning; the traditional use of CST and SRF does not explicitly account for the mechanisms causing overdosing behavior; increasing viscosity of the liquid phase, not reduced flocculation performance, is primarily responsible for overdosing as measured by CST and SRF: and CST is incapable of detecting measurable changes in sludge dewaterability within a fairly wide range around the optimal polymer dosage.

Experimental work on the role of feed suspension polydispersivity on the performance of granular media filters is presented. Suspensions of latex particles of sizes 0.09 μm, 0.944 μm, 7.04 μm, and 7.6 μm, and their paired combinations were filtered through a porous bed of glass spheres. During the ripening stage of the filtration, the presence of small particles was observed to improve the removal of large particles, while the removal of small particles themselves was not significantly affected by the presence of larger particles. These results are explained by the relative contributions of particles deposited in the filter bed to the total available surface area for subsequent particle deposition. Filter ripening evolves faster when particles in the filter influent have a smaller mass‐averaged diameter. Deposit morphology and relative distribution of the mass in the bed appear to depend on suspension polydispersivity and play an important role in head‐loss development. When particles smaller than 1 μm are present in the feed suspension, relatively short periods of filtering to waste should improve the removal of the bacteria‐sized particles that may be poorly removed by clean filter beds.