A Preliminary Laboratory Study of Initial Copper Release from Dredge Residuals

Authors


Abstract

A preliminary laboratory study was conducted to investigate the impact of different residual types and sediment surface roughness on copper contaminant fluxes to the water column. Sediments from Torch Lake, Michigan served as the test samples. These sediments are mining by-products with elevated Cu levels. Six experiments were run during which the sediments were conditioned to simulate different forms of residuals. During these experiments, the water column above the sediments was circulated via peristaltic pumping or orbital shaking and the total and dissolved Cu levels were monitored periodically for 15 days. Dissolved Cu levels indicated that during the first 48 hr the water column concentrations approached equilibrium for all six cases. Total Cu levels increased with time and did reach equilibrium but were more susceptible to fluctuations in water column suspended solids levels. Analysis of the resulting dissolved Cu data indicated that the resulting water column Cu concentrations differed with sediment surface and residual type. The highest dissolved Cu water column concentrations were observed for a roughened surface with a larger surface area. The lowest water column dissolved Cu levels were observed for the case with sediment slurry placed over clean sand. The dissolved Cu levels in the water column for all six simulated conditions were several orders higher than the USEPA ambient water quality criteria for protection of aquatic life. © 2014 Wiley Periodicals, Inc.*

INTRODUCTION

Dredging is the excavation of bottom sediments from water bodies for disposal elsewhere. This technique is often used to keep waterways navigable as well as replenish sand on public beaches where erosion has occurred. The technology has been adapted to remediate contaminated sediment and reduce risks to aquatic life and humans.

A dredger (or “dredge” as is the general usage in the Americas) is any device, machine, or vessel that is used to excavate and remove material from the bottom of a water body. For example, a scoop attached to the end of a rope or pole by which a person can draw sediments up from the bottom of a pond is a dredger. Developing this idea further, a motorized crane equipped with a drag bucket or clamshell (grabber) that is used to scoop material from the bottom of a water body is also a dredger. The crane could be located on the bank or mounted on a barge. If the crane is mounted on a barge, the entire vessel is referred to as a dredger. The process of dredging creates dredged material, which is transported from the dredged area. Dredging can produce materials for land reclamation or other purposes (usually construction-related), and has also historically played a significant role in gold mining. Dredging can disturb aquatic ecosystems, often with adverse impacts.

Dredging as mentioned is one of the three major sediment remediation/management technologies. Dredging disturbs the sediment bed and, in the process, resuspends sediment that is disturbed and can release the contaminants as sorbed to the sediment or in the interstitial water. Release of contaminants is broadly defined as transfer from the sediment bed to the water column due to resuspension when the sediment bed is disturbed. Sediment resuspension caused by dredging is defined as the sediment particles suspended into the water column during the dredging operation that do not rapidly settle out of the water column following resuspension. Sediment resuspension during dredging is unavoidable and occurs whenever materials are dredged, regardless of the dredge type or precautions that may be taken during dredging operations. However, the degree of sediment resuspension from dredging depends on many site and operation-specific variables as described by Herbich and Brahme (1991), Collins (1995), Johnson and Parchure (2000), and Hayes and Wu (2001).

When dredging sediments that are more contaminated at depth than at the surface of the sediment bed, the dredging residuals may be more contaminated and release more contaminants to the water column than the original, undisturbed sediment bed. The contaminant release from residuals may be greater than the release from sediment resuspension by dredging. Patmont and Palermo (2007) measured 2 percent to 9 percent of the contaminated sediments remaining in the residuals after environmental dredging, while Hayes and Wu (2001) reported that losses of contaminated sediment by resuspension were generally less than one percent. In addition, dredge material may contain hazardous substances that may adversely affect the sediment disposal area; furthermore, the process of dredging often dislodges chemicals residing in benthic substrates, thereby releasing them into the water column. Some resuspended contaminants may dissolve into the water column and become available for uptake by biota. Release of hazardous substances into the water column represents a potential short-term as well as long-term dredging-related environmental exposure.

The contaminants released from dredge resuspension and residuals are important information in making decisions to ensure that dredging is a viable option for the sediment remediation at a particular site. The primary objective of this study was to develop a preliminary understanding of the effect of sediment residual type and surface roughness on water column contaminant levels.

MATERIALS AND METHODS

The approach used in this study was to structure a series of experiments using the same sediment as a contaminant source. The sediment characteristics were altered in an effort to mimic different sediment surfaces and residual conditions after dredging. Identical experiments were run on these sediments and the level of dissolved and total contaminant fluxes to the water column recorded.

Sediments

Sediments from Torch Lake in Houghton, Michigan were used for this work. This area had been a historic copper mining area in the 19th and 20th centuries. Milling byproducts and waste materials discharged to Torch Lake resulted in an accumulation of sediments with high copper concentrations. Exhibit 1 lists the pertinent characteristics and the contaminants present in the Torch Lake sediments. As shown in Exhibit 1, the Cu concentrations appeared to be extremely high (9,130 mg/kg) compared to other inorganic contaminants, which was a motivation of selecting copper for the study.

Exhibit 1. Typical characteristics of Torch Lake sediments

Characteristics (Unit)ValueCharacteristics (Unit)Value
pH6.86Dissolved copper (mg/L)0.939
ColorReddishArsenic (mg/kg)74.9
Specific gravity2.60Cadmium (mg/kg)17.2
D10 (µm)0.73Chromium (mg/kg)109
D50 (µm)2.74Nickel (mg/kg)109
D90 (µm)9.19Lead (mg/kg)1430
Porewater DO (mg/L)4.1Zinc (mg/kg)1130
Porewater DOC (mg/L)50.5Mercury (mg/kg)0.732
Total copper (mg/kg)9,130  

Experimental Scenarios

An approach was developed which enables the isolation and observation of different contaminant water column flux mechanisms. These experiments were constructed such that analysis of individual scenario results or in combination with other experimental results would identify the role that specific characteristics play in contaminant flux to the water column. The scenarios are as follows:

Scenario 1: Undisturbed sediment

  1. Sediment thickness in the tank/aquarium: 1 inch.
  2. Sediment surface condition: Smooth.
  3. Running condition: Continuous recycled flow with 3-inch depth of initially deionized (DI) water above the sediment surface with a flow rate of 1,560 mL/min.

Scenario 2: Roughened sediment

  1. Sediment thickness in the tank/aquarium: 1 inch.
  2. Sediment surface condition: Rough (top 3/8-inch was roughened with scouring using a 3/8-inch triangular notch trowel).
  3. Running condition: Continuous recycled flow with 3-inch depth of DI water above the sediment surface with a flow rate of 1,560 mL/min.

Scenario 3: Externally generated residuals with clean bottom

  1. Sediment thickness in the tank/aquarium: ½ inch sediment with ½ inch sand below the sediment.
  2. Sediment surface condition: Smooth.
  3. Running condition: Continuous recycled flow with 3-inch depth of DI water above the sediment surface with a flow rate of 1,560 mL/min.

Scenario 4: Externally generated residuals with scored bottom

  1. Sediment thickness in the tank/aquarium: ½ inch fluff generated from the settlement of sediment slurry on top of ½ inch roughened sediment at the bottom. The supernatant was removed after slurry solids settled and replaced with 3-inch depth of DI water above the fluff.
  2. Sediment surface condition: Sediment fluff.
  3. Running condition: Continuous recycled flow with 3-inch depth of DI water above the sediment surface with a flow rate of 1,560 mL/min.

Scenario 5: Externally generated residuals with clean bottom/clean water

  1. Sediment thickness in the tank/aquarium: ½ inch sediment with ½ inch sand below the sediment (same setup as Scenario 3)
  2. Sediment surface condition: Smooth.
  3. Running condition: Continuously shaken on shaker table at 25 rpm with 3-inch depth of DI water above the sediment surface.

Scenario 6: Externally generated residuals with scored bottom/clean water

  1. Sediment thickness in the tank/aquarium: ½ inch fluff generated from the settlement of sediment slurry on top of ½ inch roughened sediment at the bottom. The supernatant was removed after slurry solids had settled and replaced with 3-inch depth of DI water above the fluff (same setup as Scenario 4).
  2. Sediment surface condition: Sediment fluff.
  3. Running condition: Continuously shaken on shaker table at 25 rpm with 3-inch depth of DI water above the fluff surface.

Experimental Setup

The experimental scenarios were conducted in Plexi-glass tanks. These tanks tapered from top to bottom and had mid dimensions of 14-inch × 9-inch × 9-inch (depth). These tanks were selected for two reasons. The rectangular shape would facilitate steady flow over the entire sediment surface. In experimental Scenarios 1 to 4, water was injected into one end of the tank through stainless steel diffusers located ½ inch above the sediment. An identical diffuser located on the opposite end of the tank withdrew water. This arrangement ensured that all portions of the water column received equal exposure to the sediments. The second reason for selecting these tanks is that their size allowed for exchange rates to be high or residence times low for the pumps used. A schematic of the experimental setups are illustrated in Exhibit 2. The details for the experimental set up are described in Karim (2013).

Exhibit 2.

A schematic of the experimental setup (NTS)

Sampling and Measurements

Water samples were collected from each tank immediately after the start of the experiment (within 5 min), at 1 hr, 2 hr, 4 hr, 8 hr, 12 hr, and 24 hr the first day and at every day thereafter for an additional 14 days. Water samples were collected with a plastic sample syringe. The pH, turbidity, dissolved Cu, and total Cu were measured for all the samples. The main reason for measuring the pH and turbidity was to observe changes in the system during the experimental run as well as to find any effect of pH and turbidity changes in the release of total and dissolved Cu in the water column. It was expected that lower pH in the system would release more dissolved Cu and higher turbidity in the column water would contribute to an increase in the release of total Cu from the sediment to the water column.

Methods of Analysis

The pH was measured with a Denver Instrument (Bohemia, NY) UltraBasic digital pH meter and standard bulb probe. The turbidity was measured with a Hach (Loveland, CO) 2100N Turbidimeter. Dissolved and total Cu concentrations were measured according to standard method of examination of water and wastewater using a Perkin Elmer (Waltham, MA) Optima 4300 flame emission atomic absorption spectrophotometer (AAS). Experiments for Scenarios 5 and 6 were run with a BigGER Bill Thermolyne Shaker at 25 rpm that appeared to create enough turbulence/flow to release the contaminants from the sediment. Cole-Parmer (Vernon Hills, IL) Masterflex L/S 4-channel peristaltic pumps were used for the experiments for Scenarios 1, 2, 3, and 4 to recirculate the water and the pumps were operated at 600 rpm that appeared to create enough flow rate (1,560 mL/min ≈ 0.412 gpm ≈ 0.055 ft3/min) to be comparable with the shaker as well as to simulate the field conditions. A 3-inch deep water column with 6.5 L of water was used for all six scenarios. By normalizing the depth of water (3 inches) and tank width (9 inches), the flow speed of recirculation can be estimated to be 0.293 ft/min.

RESULTS AND DISCUSSIONS

The experimental results obtained from the study are discussed below. No mass balance analysis for the contaminants was performed as the main intent of the study was to investigate the effect of residual type and surface roughness on water column contaminant levels. As mentioned earlier, the turbidity was measured, but the results are as total suspended solids (TSS) instead of typical turbidity units (nephelometric turbidity units). A relationship between the turbidity and TSS was developed for Torch Lake sediment by measuring turbidity and TSS for six samples of the sediment slurry. The data were plotted and a best fit regression line was calculated (y = 0.4219x + 2.6422, R2 = 0.9868) as shown in Exhibit 3. Approximately 98.68 percent of the variation in the values of TSS was accounted for by the linear relationship with turbidity. Several data values of turbidity were assumed and the corresponding values of TSS were estimated using the best fit regression equation.

Exhibit 3.

Relationship between TSS and turbidity

Using the relationship in Exhibit 3, TSS for all the samples under the six scenarios were estimated and plotted as a function of time. The plots are presented in Exhibit 4. The concentrations of total Cu released in the water column as a function of time for the six simulated scenarios are also plotted in Exhibit 4 to assess the relationship, if any, between TSS and total Cu concentrations. The TSS values varied widely with a range of 0.40 mg/L to 2,278 mg/L over the different experimental scenarios. The hike in TSS for Scenario 6 for the last four sampling events clearly appeared to be the experimental disturbances as the other scenarios showed consistent results. Without the hike, the TSS values ranged from 0.40 mg/L to 98 mg/L. The highest TSS was observed for Scenario 6 followed by Scenarios 1, 2, 5, 4, and 3. It was expected that Scenarios 4 and 6 would exhibit the highest TSS as the sediment samples were prepared with fluffs generated from the settlement of sediment slurry (approximately 100 g/L) on top of ½ inch roughened sediment at the bottom. The sediment concentrations for the samples for Scenarios 1, 2, 3, and 5 were approximately 833.71 g/L. As indicated in Exhibit 4, the total Cu released showed very good correlations with the TSS. Under most of the scenarios and sampling events total Cu concentrations released in the water column increased with increased suspended solids in the water column. It was obvious as the Cu could bond with the suspended solids, thereby showing higher concentrations of total Cu in the AAS analysis.

Exhibit 4.

Correlation of total Cu released with TSS

The variations of pH, total Cu, and dissolved Cu with time are illustrated in Exhibit 5. As seen in the exhibit, pH remained around 7.0 for the duration of the experiment varying between 6.72 and 7.21, meaning no pH changes occurred during the experimental period. This was a good indication that the release of dissolved Cu should not change significantly with time under the different scenarios which was discussed, verified, and confirmed later in this section.

Exhibit 5.

Variation of pH and Cu released with time

Exhibit 6 highlights the variation of total Cu and dissolved Cu during the first 48 hr. Initially, Cu levels were close to 0.0 mg/L. During the ensuing 48 hr, dissolved Cu levels increased but appeared to be reaching equilibrium. Total Cu levels also were changing during this time, some of which was likely attributable to fluctuations in TSS. Of interest during the first 48 hr was that changes in total Cu did not necessarily result in changes in dissolved Cu which was a further indication that a portion of the changes in total Cu were associated with Cu adsorbed to the suspended sediments.

Exhibit 6.

Variation of Cu released with time for initial 48 hr

Individual R2 values based upon the first 48 hr indicated that the power equation provided a suitable method for capturing sediment water column flux behavior. The individual equations for each experiment were different. Based on Exhibit 6, the initial concentrations released in the water column and R2 values for both the total Cu and dissolved Cu are summarized in Exhibit 7. The observed initial flux for the total Cu ranged from 0.0504 mg/L to 0.3148 mg/L and for the dissolved Cu 0.0087 mg/L to 0.0125 mg/L. The initial flux for total Cu appeared to be 6 to 25 times of the initial flux of the dissolved Cu, which appeared to be consistent to some extent with the literature, laboratory, and field studies conducted by several researchers (Brannon et al., 1976; Hirst & Aston, 1983; Lee et al., 1975; Wright, 1978).

Exhibit 7. Regression analysis summary

ScenarioTotal CuDissolved Cu
 Initial Release (mg/L)Release Rate Constant (per hr)R2Initial Release (mg/L)Release Rate Constant (per hr)R2
Scenario 10.11300.29370.69780.01010.47340.9265
Scenario 20.13290.46320.73440.02800.45280.7329
Scenario 30.05040.45200.67760.00870.40200.8163
Scenario 40.31480.08900.03770.01250.45130.7723
Scenario 50.17190.09530.24650.01180.41590.9094
Scenario 60.26650.09340.10450.01170.50790.9193

Dissolved Cu appeared to be more bioavailable compared to total Cu. As a result, dissolved Cu levels for the first 24 hr and 48 hr of each experiment were modeled using the power functions from Exhibit 6 and presented in Exhibit 8.

Exhibit 8.

Variation of predicted dissolved Cu released with time for initial 48 hr

The experimental approaches behind Scenarios 2 to 6 were designed to illuminate the role residual characteristics had on Cu flux from the sediments to the water column. Scenario 1 served as a base representing a smooth, undisturbed baseline condition. Various hypotheses were developed during experimental preparation as to what the effects of the different experiments would be on the Cu flux to the water column. Experimental duration was selected to be 15 days. This duration was expected to be adequate for the Cu to be transferred from the residuals to the water column to reach equilibrium. Since the largest gradients between the water column and the residuals/sediments were expected to occur immediately after mixing, the first 24 hr was monitored more frequently. Following that, the remainder of the experiment was monitored on a 24-hr basis.

The results indicate that much of the changes in water column dissolved Cu concentrations occurred during initial phases of the experiments. Following that, dissolved Cu levels remained somewhat constant during each experiment for the duration of the simulation. Total Cu levels did not exhibit similar behavior. Total Cu levels exhibited spikes which appeared to be related to increases in TSS resulting from accidental sediment resuspension during experimental operations. The resuspended solids would settle again and total Cu would return to pre-event levels. At the same time, dissolved Cu levels did not increase during these events and remained constant throughout the remainder of the experiment near the level it had reached during the first 48 hr. On this basis a decision was made to focus on the analysis of the first 48 hr of the experiments.

Dissolved Cu levels for the individual experiments all had similar patterns. The magnitude of the dissolved Cu levels varied but did not appear to be greatly different from one experiment to the next. Clear discernible behavior characteristics were observed only when the various regressions were developed. Scenario 2, which used a scored sediment surface representing a scoured sediment bottom as a residual source, appeared to predict the highest levels of dissolved Cu. The regression for Scenario 2 predicted a 48-hr water column dissolved Cu concentration of 0.16 mg/L (Exhibit 6), twice that of the next highest experiment (Scenario 6). Scenarios 4 and 6, both of which used a settled fluff from thin sediment slurry (100 g/L) as a residual source, were next highest as 0.07 mg/L and 0.08 mg/L, respectively. The base condition, Scenario 1, representing an unaltered sediment surface, predicted a 48-hr dissolved Cu level of 0.06 mg/L. Regressions for Scenarios 3 and 5, both of which used thick slurry (833.71 g/L) as the residual source placed on a sand bed, predicted the lowest dissolved Cu levels of 0.04 mg/L and 0.056 mg/L, respectively. The dissolved Cu concentrations in the water column for all scenarios were higher than USEPA water quality criteria concentrations for copper (CMC = 13.0 µg/L, CCC = 9.0 µg/L for freshwater; CMC = 4.8 µg/L, CCC = 3.1 µg/L for saltwater; where CMC stands for criteria maximum concentrations, and CCC for criteria continuous concentrations). The CMC is an estimate of the highest concentration of a material in surface water to which an aquatic community can be exposed briefly without resulting in an unacceptable effect (often called the criteria for acute toxicity to aquatic organisms). The CCC is an estimate of the highest concentration of a material in surface water to which an aquatic community can be exposed indefinitely without resulting in an unacceptable effect (often called the criteria for chronic toxicity to aquatic organisms).

In Scenario 2 the working hypothesis was that increasing the surface area by roughening the sediment surface would increase the dissolved Cu flux to the water column. The sediment in Scenario 2 had the same physical characteristics as the sediments in Scenario 1 so there should be the same amount of Cu in the sediments. Any difference between Scenarios 1 and 2 would be attributable to the increased surface area in Scenario 2 allowing a greater flux in the water column. Results seemed to indicate that this did occur. The increase in surface area per unit area of the sediment surface was about 2.4 when compared to the base condition of Scenario 1. In addition, creating ridges in Scenario 2 resulting in the groove faces having a rougher surface than the smooth surface of Scenario 1. The end result was due to the fact that in Scenario 2 there was an increase in surface area due to the grooves and also the roughness of the grooves.

As mentioned earlier, Scenarios 3 and 5 both used thick slurry (833.71 g/L) as their residual source and the slurry was placed over a bed of clean sand. Scenario 3 had water flowing over the sediment surface in the same manner as Scenarios 1, 2, and 4 while circulation in Scenario 5 was provided by placing the tank on an orbital shaking table. These two experimental scenarios generated the lowest water column dissolved Cu levels. The hypothesis was that these scenarios would generate higher levels of dissolved Cu than that of the base condition (Scenario 1) since the slurry nature of the residuals would facilitate more dissolved Cu exchange with the water column. This hypothesis was not supported by the results which indicated that the water column dissolved Cu levels were lower than that of the base. Instead, it appeared that placing the slurry on the clean sand provided a second flux pathway for Cu. Instead of just diffusing into the overlying water column, Cu diffused into the interstitial water of the sand bed and the overlying water column. No water samples were collected from the sand bed so this cannot be verified. Another possibility is that the iron content in the sand may have facilitated the removal of the Cu from the interstitial water in the sand bed, thereby further increasing the Cu flux gradient in favor of the sand bed over the water column. The sand used for the bed was a graded clean sand that had been washed and dried and had no obvious impurities. Iron in the sand was not thought to be a reason for the results generated by Scenarios 3 and 5 but cannot be ruled out without further investigation.

Scenarios 3 and 5 differed only in the means used to circulate the water column. Flow rates in Scenarios 1 to 4 were maintained at sufficiently high levels (1,560 mL/min ≈ 0.412 gpm ≈ 0.055 ft3/min) to generate residence times of about 4.17 min in the tanks. The orbital speed on experimental Scenarios 5 and 6 (25 rpm) was selected to provide the highest possible level of circulation in tank without purposely inducing sediment resuspension. Since both Scenarios 3 and 5 had similar results, it is reasonable to deduce that the level of circulation in both tanks were similar. Further, the orbital motion in Scenario 5 would promote movement in the interstitial waters of the sand bed more so than that of pumping water above the sediments in Scenario 3.

Scenarios 4 and 6 were performed using a thin sediment slurry (100 g/L) as a residuals source. In these cases, the thin slurry was allowed to settle and the overlying water removed and replaced with clean water. This resulted in the residual source being a thin “fluff” layer immediately on top of the scored sediment base. The hypothesis underlying this experimental setup was that the thin fluff would provide the maximum opportunity for leachable Cu to transfer to the overlying water column. Using scored sediment as a base of the fluff decreased the gradient from the fluff to sediments and increased the likelihood that leachable Cu would migrate to the overlying water column.

Comparisons of the results among the scenarios showed that copper release is a function of the effective surface area of the residuals. Scenario 2 had a serrated surface with an effective surface area that was nearly 100 percent greater than the uniform surfaces of the other scenarios. The overall results in Exhibit 8 showed 100 percent greater or more copper release in Scenario 2 than in the other scenarios. Comparisons between Scenarios 6 and 5 as well as between Scenarios 4 and 1 or 3 showed greater releases from sediment surfaces that had greater porosity. Scenarios 1, 3, and 5 have a porosity of about 0.65 while Scenarios 4 and 6 had porosity of about 0.95, nearly 50 percent greater. The overall results in Exhibit 8 showed that copper releases for higher porosity residuals were on average about 40 percent greater than copper releases from low porosity residuals.

CONCLUSIONS

During the first 48 hr of the various scenarios tested, dissolved Cu concentrations increased from nearly 0.0 mg/L to equilibrium levels and ranged between 0.04 mg/L and 0.16 mg/L. Results indicated that the experimental scenario with an increased sediment surface area had the highest dissolved Cu concentration. Total Cu levels increased in all experimental scenarios but, in general, indicated fluctuations with changes in TSS levels indicating that part of the total Cu observations represented suspended particulate material containing Cu. None of the scenarios indicated dissolved Cu levels with significant variability with TSS changes. Regressions for dissolved Cu levels generated R2 values that ranged from 0.7329 to 0.9265.

The study results supported the simulation of Cu contributions from sediment in water bodies with flowing water and showed that the approach could be suitable to other sediments and contaminants. The experiments confirmed that releases increase with effective surface area and sediment porosity, characteristic of diffusional losses. Because Torch Lake sediments originate from mining and metal producing by-products, the sediments have relatively little leachable Cu. This may have impacted some experiments where the available leachable Cu at the sediment surface was exhausted. The highest dissolved Cu water column concentrations (0.16 mg/L) were observed for a roughened surface with larger surface area (Scenario 2) followed by Scenario 6 (0.08 mg/L), Scenario 4 (0.07 mg/L), Scenario 1 (0.06 mg/L), Scenario 5 (0.056 mg/L), and Scenario 3 (0.04 mg/L). The dissolved Cu concentrations in the water column for all scenarios were several orders of magnitude higher than the USEPA ambient water quality criteria for protection of aquatic life from copper in both freshwater and saltwater environments.

ACKNOWLEDGMENTS

This work was supported by the Environmental Laboratory, U.S. Army Corps of Engineers under the auspices of the U.S. Army Research Office Scientific Services Program administered by Battelle (Delivery Order 0139, Contract No. W911NF-11-D-0001). The authors thank Mr. Richard Hudson and Dr. Catherine Thomas for their laboratory assistance.

Biographies

  • M. A. Karim, PhD, P.E., is an assistant professor of civil and construction engineering at Southern Polytechnic State University, Marietta, Georgia, where he is responsible for environmental engineering courses for civil and construction engineering programs. He received his PhD in civil and environmental engineering from Cleveland State University and has been a certified professional engineer in the state of Ohio (inactive since January 2010) and Virginia since 2003. He is involved in research in the area of soil and sediment contamination and remediation, environmental management, solid waste management, and statistical hydrology. He has more than 12 journal publications and 3 professional reports.

  • Barry W. Bunch, DEng, P.E., is a research civil engineer for Environmental Laboratory at the U.S. Army Corps of Engineers, Engineer Research and Development Center, Waterways Experiment Station, in Vicksburg, Mississippi, where he has been a team member of the Water Quality and Contaminant Modeling Branch since 1989. He received his doctorate in environmental engineering from Louisiana Tech University and has been a certified professional engineer for the state of Louisiana since 1993. He is involved in research in the area of civil and environmental engineering. He has more than 12 journal and proceedings publications and over 25 professional reports.

  • Paul R. Schroeder, PhD, P.E., is a research civil engineer for the Environmental Laboratory at the U.S. Army Corps of Engineers, Engineer Research and Development Center, Waterways Experiment Station, in Vicksburg, Mississippi, where he has been a leader in assessment of environmental effects of engineering activities for over 30 years. He received his PhD in civil and environmental engineering from The Ohio State University and has been a registered professional engineer in the state of Mississippi since 1985. He is involved in research in the areas of dredged material management and disposal, sediment-associated contaminant risk, and engineering software development. He has more than 50 journal and conference publications and over 150 professional reports.

Ancillary