SUMMARY OF THE REMEDIATION TECHNOLOGIES DEVELOPMENT FORUM SEDIMENTS REMEDIATION ACTION TEAM MEETING



Marriott Metro Center
Washington, D.C.
January 13-14, 1999

EXECUTIVE SUMMARY

About 65 participants attended the Sediments Remediation Action Team meeting in Washington, D.C. on January 13 and 14, 1999. Action Team Co-Chair Richard Jensen (DuPont Corporate Remediation) said that there is an important window of opportunity in the sediments field, which has been receiving increased attention recently. Technical activity and focus is accelerating, and many projects are approaching the construction phase. However, users manuals are needed, as is further study of a number of sediments remediation issues. The primary objectives of the meeting were to encourage participation by new members, narrow the focus of the Action Team, and identify two or more field projects for self-empowered subgroups to carry out. Several speakers were invited to the meeting to keep participants abreast of work being conducted by other organizations.

Project Briefings

The meeting featured presentations by a number of speakers about ongoing sediments remediation projects. The first presentation was given by Keith Jones of Brookhaven National Laboratory and included a description of the New York/New Jersey Harbor Decontamination Demonstration Project, the goal of which is to create self-sustaining facilities to decontaminate and remediate contaminated sediments from the harbor. The project plans to demonstrate several processes that could be employed at such facilities, including a manufactured soil process, sediment washing, rotary kilns, and a technology using a plasma torch.

Next, John Haggard described a contaminated sediments sites database being constructed by General Electric. The database presents a tremendous amount of information about 77 large-scale sediments remediation projects. It includes information on the type of contamination, cleanup objectives, cleanup costs, and cleanup results, as well as a list of references relating to each site. Next, Jim Keating of the U.S. Environmental Protection Agency (EPA) summarized parameters for sediment recovery demonstration projects outlined by the Clean Water Action Plan published in February 1998.

A review of research and development activities being conducted by the Environmental Protection Agency's (EPA's) Duluth Laboratory was provided by David Mount. The laboratory has been supporting EPA's effort to develop Equilibrium-Partitioning Sediment Guidelines (ESGs), formerly called Sediment Quality Criteria. It also has been developing sediment toxicity identification evaluation (TIE) methods, as well as studying a technology that reduces chemical availability in sediment, resulting in a reduction of pore water concentrations of the chemicals. Mount's presentation was followed by an account by Michael Coia (Phytoworks, Inc.) of potential applications of phytoremediation. The next morning, Walter Kovalick gave an overview of the Remediation Technologies Development Forum.

Discussions of Potential Projects

The balance of the meeting was devoted to discussing capping, treatment, natural attenuation, and assessment, and to identifying potential projects that subgroups of the Action Team might pursue in each area. An Alcoa site on the Grasse River in New York was the first site described. The site is contaminated with PCBs, and dredging has been conducted to help remediate the contamination. A particle broadcasting study has been proposed for the site. It would entail the placement of a 6-inch layer of clean sediment over the contaminated sediments, which is aimed at translocating the zone of bioturbation out of the PCB-tainted sediments. This could represent a permanent remedy for sequestering PCBs because there is no erosion in the river, little vertical migration of sediments, and the river is a net depositional area.

Pending approval from EPA and concurrence by the State of New York and the St. Regis Mohawk Tribe, Alcoa will proceed with the particle broadcasting study and would be willing to work with the Action Team to evaluate natural attenuation, phytoremediation, or other issues at the site. Participants supported the idea of the Action Team becoming involved with the site. An additional potential site for a treatment demonstration project is Puget Sound in Washington, where 500,000 cubic yards of PCB- and PAH-contaminated sediments will be removed and treated ex situ. Participants also expressed interest in the possibility of studying the use of a confined disposal facility (CDF) as a reactor, which might be tested in Puget Sound or elsewhere.

In connection with capping, Michael Palermo and Norman Francingues, Jr. gave presentations on the Army Corps of Engineers Waterways Experiment Station's Dredging Operations and Environmental Research (DOER) program. The DOER program has written a couple of guidance documents on capping, is evaluating a telescoping weir for confined disposal sites, and will soon complete an evaluation of dredging techniques utilized in Europe, among other endeavors. Representatives of the program encouraged the Action Team to pursue capping demonstration projects in the field to improve knowledge of capping and help convince decision-makers of the effectiveness of capping.

There was substantial subsequent discussion about the possibility of the Action Team conducting a capping demonstration project. Participants supported Jensen's suggestion to test different technologies adjacent to one another in a checkerboard arrangement at a field site containing homogenous sediments. A potential CDF site in the Port of Portland, Oregon, was suggested as a potential demonstration site. Participants agreed that a demonstration site would need to be tested 3 and 5 years after cap placement, although it would be desirable to test it 10 years later, as well. Some participants also suggested that a project could be conducted to assess the effectiveness of existing caps.

John Davis of Dow Chemical Company then outlined the use of natural attenuation as a remedial strategy. Natural attenuation is defined as relying on a series of physical, chemical, and biological processes to effectively reduce the toxicity, mobility, or volume of contaminants to levels that are protective of human health and the environment. Demonstration, documentation, and quantification of natural attenuation are required when it is selected as a remedy. It is often used in conjunction with other treatment technologies.

The application of natural attenuation to contaminated soils and groundwater has been fairly well-documented, but the principles and practices of applying natural attenuation to contaminated sediments have not been studied thoroughly. The Action Team might assist the development of these principles and practices by studying outstanding questions about natural attenuation such as the evidence required to evaluate it and the natural processes that affect it. The group decided to first develop a definition of natural attenuation in contaminated sediments and an outline of work that it might conduct to address the outstanding questions. These steps will be taken over the next 3 months or so. Then, the Action Team could consider potential opportunities to study natural attenuation at sites where it was planning to conduct other field studies.

The final discussion session was devoted to potential assessment projects that the Action Team might pursue. Based on work it conducts at an actual site or a virtual site (for which data would be made up), the Action Team might produce short white papers on a variety of assessment issues (for example, measures of project success, conceptual model development, and determining sediment fate and transport). Assessment is best discussed on a site-specific basis, participants agreed. Several individuals expressed support for the development of a game in which players decide how to proceed in response to scenarios involving contaminated sediments sites. Such a game would be similar to a game developed by industry that involved players in ecological risk assessment.

Samuel Kounaves of Tufts University described his research on in situ analysis of heavy metals in water using a microsensor. His laboratory has developed sensors to test concentrations of about 30 different metals in water. The sensors can effectively determine the order of magnitude of metals levels down to parts per billion or sub-parts per billion. Kounaves proposed designing a probe to use in sediments that the Action Team would test at several sites. How such a probe would be deployed would have to be determined, however, since it could not be deployed in the same way as groundwater probes. A number of participants expressed interest in further discussing a collaborative project.

Future Activities

For each of the potential demonstration projects discussed, a list was compiled of individuals wishing to be informed of upcoming conference calls and meetings. Participants discussed holding conference calls about the potential projects in the coming months. The Action Team agreed to meet as a whole in connection with the 1999 Conference on Hazardous Waste Research in St. Louis, Missouri. The conference is scheduled for May 25 through 27, 1999. The Action Team agreed to meet Tuesday, May 25. There will be sediments-related presentations given the following morning. It is possible that some of the subgroups may want to meet Monday evening, May 24. Participants will be advised of the schedule once it is finalized.

WEDNESDAY, JANUARY 13, 1999

WELCOME, OPENING REMARKS, AND DESCRIPTION OF MEETING OBJECTIVES
Richard Jensen, DuPont Corporate Remediation
Dennis Timberlake, U.S. Environmental Protection Agency (EPA)

Richard Jensen, co-chair of the Remediation Technologies Development Forum (RTDF) Sediments Remediation Action Team, welcomed approximately 65 participants (see Attachment A) to the second Sediments Remediation Action Team meeting within the past year. He began the meeting by giving a brief overview of previous Action Team meetings, including the three meetings in 1996 (in Cincinnati, Ohio, Wilmington, Delaware, and Vicksburg, Mississippi) and the September 1998 meeting in Cincinnati. Jensen said that it will be easy for new members of the group to catch up on its activities, since minutes from all the meetings and from several conference calls can be found on the Web at: http://www.rtdf.org/sediment.htm. The overheads that Jensen showed during his opening remarks are included as Attachment B.

The group has reconvened, Jensen said, because there is currently an important window of opportunity in the sediments area. Many projects are approaching the construction phase, after many years of study. Technical activity and focus is accelerating. Within the last year, the U.S. Environmental Protection Agency (EPA) prepared a report for Congress on contaminated sediments management strategy. The National Academy of Sciences (NAS) has completed a related study and held a conference over the summer. Shortly, it will begin a new study. Users manuals are needed; some are currently being drafted. Jensen said that the technical community has an opportunity to come together and contribute to these and other projects.

Jensen said that the goals of the meeting were to:

• Maintain the momentum that the group has been building over the last year.
• Gain active, new members, a goal which appeared to have been met, since at least twenty individuals who did not attend the 1998 Cincinnati meeting were in attendance.
• Find out about other organizations' activities through hearing presentations by invited speakers.
• Narrow the focus of the group.
• Identify two or more lab or field projects for self-empowered subgroups to carry out.
• Agree to meet again soon and establish a time and place to do so.

Jensen informed the group that RTDF working groups have a great deal of flexibility as to how they operate. Some groups have formed consortia, operating as legal entities on the basis of signed contracts. Others conduct informal field projects, in which representatives of government and industry work together without a contract. Other groups merely come together to exchange information. At a minimum, Jensen said, this group's goal is to convene regularly to exchange technical information. But the group is also hoping to embark on at least two joint field projects, either at the same site or different sites.

Jensen concluded by outlining the agenda for the meeting. The focus sessions on the second day, he explained, would concentrate on four topics: potential treatment projects, potential capping projects, potential natural attenuation projects, and potential assessment projects. The group's leadership decided not to divide into subgroups to discuss the four topics so that participants would not be limited to attending only one subgroup meeting. Although natural attenuation had been considered as either an assessment issue or a treatment issue at the previous meeting, Jensen said, it would be considered a separate issue at this meeting. There is precedent for doing so: at the first RTDF meeting, a natural attenuation subgroup was formed.

PRESENTATIONS

NEW YORK/NEW JERSEY HARBOR DECONTAMINATION DEMONSTRATION PROJECT

Keith Jones, Brookhaven National Laboratory

Jones began his presentation by saying that the Port of New York and New Jersey ranks first in volume of petroleum products handled every year. It handles over 28 billion gallons per year, which is more than three times the amount handled by the second largest handler of petroleum products, the Port of Houston. With this sort of volume moving in and out of the port, there is tremendous potential for contamination from oil spills, Jones explained. In addition, there is major industrial and commercial activity on the New York/New Jersey Harbor.

The most significant feature to the Decontamination Demonstration Project site is the fact that the border between New York and New Jersey runs through the middle of the Hudson River, which increases the bureaucracy involved. The Project's goal is to create facilities to decontaminate and remediate sediments dredged from the harbor. The harbor has become contaminated with petroleum hydrocarbons and other organic and inorganic contaminants, and most of the dredged material can no longer be easily disposed of.

The elements of the New York/New Jersey Contaminated Sediment Management Strategy for the harbor are prevention, assessment, abatement and control, remediation, dredged material management, research, and public outreach. The Project is funded under the Water Resources Development Act (WRDA). Funding began in 1990. Currently, the Project is mandated to develop, by 1999, one or more sediment decontamination technologies that can treat at least 500,000 cubic yards per year. This is a difficult mandate, Jones said, especially since the Project was only allocated $10 million over 3 years. Because there is not enough money to buy the requisite facilities outright, the Project is trying to develop partnerships with the private sector, on the premise that private companies could make a significant amount of money by building the facilities.

The Project team consists of EPA Region II, which is the lead agency, the Army Corps of Engineers, and the Department of Energy's Brookhaven National Laboratory, the overall technical program manager, with responsibilities that include overseeing the contracting and handling the money. Technical expertise has been provided by several area universities. There is also an Interagency Agreement supporting the work.

Jones outlined several companion projects to the one authorized by WRDA. The State of New Jersey is going to fund several demonstration projects, which will treat 30,000 to 35,000 cubic yards. Relatively small projects are also being run by the Port Authority of NY/NJ, as well as by EPA's Great Lakes National Program Office, in conjunction with the state of Michigan.

Jones described the contamination that has been found in the harbor. The Passaic River, which flows into the harbor via Newark Bay and contributes to the contamination, is the site of some of the worst dioxin-contaminated sediments in the country. While PAHs, PCBs, chlorinated pesticides and herbicides, dioxins, furans, metals, and other contaminants may be discharged into the harbor in large quantities, there is a large volume of sediments, so the average level of contamination is modest and usually below hazardous levels. Specifically, dioxin levels at various locations in the harbor sediment are about 50 to 100 parts per trillion (ppt), PCBs are a couple of parts per million (ppm), and PAHs are roughly 100 ppm. The sediment contains elevated levels of arsenic, cadmium, chromium, copper, mercury, lead, and zinc. Lead was measured from about 300 to 600 ppm. The sediments are primarily (80% to 95%) silts and clays, with a mean particle size of 15 to 20 microns. In some areas, there is a denser red clay that has to be disposed of, but usually it is not highly contaminated. Total Organic Carbon is 3% to 10%, modestly high. Estuarine salinity is 15 to 28 ppt, but reducing the elevated salt content is part of the treatment process. The solids content is 30% to 40%, and the raw sediment passes Toxic Characteristics Leaching Procedure (TCLP) tests.

According to Jones, roughly 200 miles of channels and 5 million cubic yards of sediment need to be dredged per year for the harbor area to continue to be navigable. Only 20% to 25% of the dredged sediment can be returned to the ocean. The rest has to be remediated in an economically feasible and environmentally responsible manner. The Project's objectives are to find and test the best ways to carry out the remediation.

Historic costs for disposal using dredge and dump were about $5 to $10 per cubic yard, Jones said. Most of the sediment cannot be dumped in the ocean, however, without causing adverse environmental impacts. Other disposal options, such as a Confined Disposal Facility operated by the Port Authority, range in cost from $32 to $53 per cubic yard. To beat the competition, therefore, the overall dredging, treatment, and disposal cost has to be about $30 to $35 per year to start and less in the future.

Jones stated that major element composition is important when considering some of the potential uses of the sediments. The harbor sediments contain silicon oxides and aluminum oxides, as well as other minor oxides. There is a reasonably high concentration of quartz silicon dioxide, which is useful for making cement, an aggregate, or glass. It is not necessary to separate the sediments by particle size, since the concentrations of metals and organics do not vary much by particle size.

Contaminant transport in the estuary is complex, Jones continued. Sediments are transported up and down the river by the tides, and contaminants originate from the atmosphere and from both upriver and downriver sources, such as a New York City sewage treatment plant. A participant asked if a major source of PAHs was runoff from asphalt roads, but Jones indicated that studies have found that the major contributors to Newark Bay were combined sewer overflows and water treatment plants. He said that the contaminants on the surface of the sediment have been mapped through EPA's Regional Environmental Monitoring and Assessment Program. The maps demonstrate a widespread distribution of the contaminants. Three-dimensional maps of contaminants, showing concentration versus depth, have also been created. Even though the contamination occurred 20 or 30 years ago, the maps show that the hot spots are still well localized.

Jones explained that the WRDA Project has been trying to put together a complete treatment train for the contaminated sediments, which requires integrating a large number of factors and problems. The treatment train begins with dredging and separating out the large material, which is then disposed of. The remaining materials, perhaps 2 mm or smaller, are subjected to either a low-temperature, intermediate-temperature, or high-temperature treatment technology. The low-temperature technologies are applied to the least contaminated material. Low-temperature technologies, which require an understanding of sediment characteristics and chemistry, include soil manufacturing and soil washing/chelation. Further research in the area of low-temperature technologies is needed, Jones said. Intermediate temperature approaches include thermal desorption and solvent extraction. High-temperature technologies, including a rotary kiln, a fluidized bed, or a plasma torch, are most effective at eliminating organic contaminants efficiently. They also have the potential to yield a high-value product for beneficial reuse.

Jones stated that, early in the Project, bench-scale testing (using 5-gallon samples of dredged materials) was carried out on 12 technologies. Next, five treatment technologies were tested on a pilot scale (2 to 25 cubic yards). Large-scale demonstrations (about 100,000 cubic yards), based on the analysis of results of the previous tests, will be conducted over the next year.

One low-temperature technology tested on a pilot scale by the U.S. Army Corps of Engineers' Waterways Experiment Station (WES) was the production of manufactured soil from the contaminated sediments. In the manufacturing process, the sediment was mixed with cellulose material, cow manure or sewage sludge, lime, and fertilizer to make topsoil. Jones described some of the problems with this treatment technology: the contaminants are not really dealt with, there might not be a beneficial reuse of the manufactured soil (it might only be usable in landfills, for example), and the sediments have to be diluted by about a factor of three. However, combining phytoremediation with manufactured soil production may have advantageous effects.

Testing of two intermediate-temperature technologies, solvent extraction and thermal desorption, was also conducted, Jones continued. Solvent extraction reduced semivolatiles and PCBs by 70% to 90% in bench-scale testing, but it did not reduce the metals content at all. Thermal desorption showed a 65% reduction in metals and a more than 99% reduction in semivolatiles and PCBs. Thermal destruction, a high-temperature process which operates at temperatures well above 1000 degrees Fahrenheit, reduced metals by between 58 and 83% and semivolatiles and PCBs by more than 99%. High-temperature technologies can reduce contaminant concentrations by several orders of magnitude more than low-temperature approaches.

One purpose of the bench and pilot testing was to determine an appropriate treatment train. The treatment train chosen includes both low-temperature and high-temperature technologies. In the coming months, Jones explained, there will be a demonstration of a manufactured soil process, which will treat up to 30,000 cubic yards of sediment. There will also be several demonstrations of sediment washing, a low-temperature technology. The first will be a 50-cubic-yard demonstration that will occur over the next few months, with the potential for the construction of a full-scale facility that could treat about 100,000 cubic yards per year, which might be ready to operate sometime in 1999. The sediment-washing process utilizes metals chelation to reduce the metals content of the sediments, followed by a high-pressure collision between water and the sediments to try to remove the organic materials, and some cavitation/oxidation before the solids and liquids are separated, hopefully resulting in a clean material. The product of the process could be used for construction fill or landfill cover.

Three high-temperature processes are also undergoing further investigation, Jones reported. A rotary kiln demonstration project run by the Institute for Gas Technology (IGT) will begin treating up to 10,000 cubic yards of sediment by July. If desired, IGT could utilize a rotary kiln that has the capacity to treat 100,000 cubic yards per year. Rotary kilns produce blended cement, which has a high value. A dredging company, J. Cashman, Inc. (JCI), is working with a company that operates a rotary kiln, Upcycle Associates, on a demonstration project, as well. The partnership results in a full treatment train for the contaminated sediment (incorporating removal of large objects from dredged material, etc.). The partnership has an existing facility near Albany, New York, for treating the sediments. Finally, Westinghouse has developed a plasma torch technology, which yields a black glass. This technology is not as far through the design process as the others. Further testing of whether the glass can be turned into glass tiles will be conducted over the next few weeks. Westinghouse's plans call for a facility to treat 10,000 cubic yards of sediments per year.

Jones reminded participants that the Project's goal is to create self-sustaining businesses that treat contaminated sediments, and individuals on the Project staff are working with the private groups running demonstration projects to that end. The Project is also conducting public outreach to try to increase public acceptance of the activities.

Jones summarized the Project's accomplishments to date:

• Several technologies have been successfully demonstrated at the bench and pilot scales.
• Many of the treatment train components have been designed.
• A public outreach program is underway.
• A plan for the commercialization of decontamination technologies is being developed and implemented.

Thus, assuming adequate funding, the necessary cooperation between public agencies, and the development of public-private partnerships, it appears that sediment processing facilities can be in operation in a relatively short time and at a reasonable cost to solve the problem of dredged material from the Port of New York and New Jersey. The slides that Jones showed during his presentation are not available electronically, but additional information on the New York/New Jersey Harbor Decontamination Demonstration Project is presented in Attachment C.

A participant asked whether Jones foresaw a large, regional processing center or multiple, small, disparate facilities to treat the contaminated sediments. Jones replied that he expected there would be multiple solutions at multiple locations, although there might be a central sediment storage facility, since the processing centers will all want a constant, even flow of sediments, while dredging can only occur within certain windows of time.

Ken Finkelstein asked where the money would come from to pay for the cost of the treatment technologies, insofar as they cost significantly more than conventional disposal methods. Jones said that the source of funding is unclear, but the hope is that it will come from the sale of the end product of the treatment plus a tipping fee of $20 to $30. Finkelstein suggested that parties responsible for the contamination have liability for remediation of the sediments. When sediment is removed, liability is eliminated, he said, and corporations could be forced to pay a price for the elimination of their liability. Jones said that the Army Corps of Engineers and EPA are discussing doing some work with that in mind.

A participant asked when it would be known whether the goal of treating the contaminated sediments at costs of less than $35 per cubic yard would be achievable. Jones replied that a strong business plan could probably be developed immediately. Some proposals have come in at $70 to $80 per cubic yard. That figure is driven up because it includes dredging, transport, processing, and disposal, as well as start-up and demobilization costs. It will probably be clear within the next year whether the Project will be able to meet its goals.

CONTAMINATED SEDIMENTS SITES DATABASE

John Haggard, General Electric

John Haggard described how General Electric (GE) is constructing a database of contaminated sediment sites around the country that have been or are being cleaned up. The overheads he showed during his presentation are available as Attachment D. The contaminated sediment sites database is being developed to shed light on: what drives decisions about the selection of remedial remedies, what types of remedies are selected, how technologies have performed (and what their costs and limitations have been), and what might be learned from past experiences. GE developed a methodical approach for gathering information, documenting information sources, and entering the information into an accessible database, Haggard stated. Much of the information is not published and comes from files or from people's heads; however, an effort is being made to rely on sources other than project proponents and opponents. Plans call for the database to continue growing, and Haggard asked for feedback about how the database might be used by other interested parties.

The database is limited to sites containing at least 3,000 cubic yards of contaminated sediment, Haggard said. GE chose this limitation because there are so many small-scale projects for which little information is available and because the developers of the database were interested in decision-making about larger-scale projects. The database also excludes navigational projects and sites outside of the United States. There are currently 77 projects in the database, representing 68 separate sites. For the 77 projects, there are about 1,500 individual references assembled and catalogued. The database is in Microsoft Access, Version 3.

Haggard explained that there are approximately 70 fields of data for each site, linked by a site identification number. Types of data in those fields include site status, types of contaminants originally present, planned remedy, cleanup objective, estimated vs. actual costs, post-remediation monitoring results, target cleanup levels from risk assessments (to enable a better understanding of the role of risk assessments in cleanup), and a list of available references.

According to Haggard, information in the database has come from such sources as Remedial Investigation/Feasibility Studies (RI/FSs) and equivalent reports, Records of Decision (RODs), consent orders, technical literature, progress reports, site visits, and phone conversations with agency personnel, Potentially Responsible Parties (PRPs), consultants, and contractors. GE tried to verify information learned from phone conversations through other sources, but that is not always possible.

Haggard displayed a list of the 77 projects and requested that participants advise him of any sites that are missing. Some very large sediment remediation projects are scheduled to begin in 1999, such as Saginaw Creek (Montana), Inland Steel (Indiana), Fox River (Wisconsin), and Cumberland Bay (New York). As an example, Haggard showed the database information on the New Bedford site.

Haggard said that the preliminary findings from the 77 projects indicate that the cost of remediation, which is often difficult to pinpoint, ranges from $83 to $1,670 per cubic yard, with a median of $350 to $400 per cubic yard. Specialty excavators and other innovative equipment were not employed very often. Additionally, very few sites treated their sediments, although information on how the sediments were disposed of was only available for 28 sites. GE also found that the amount of project documentation available varies widely and that there is little information-sharing occurring. On removal projects, GE found that the emphasis was placed on removing contaminant mass, rather than on decreasing risk.

The primary work remaining on the project is to assess what progress is being made through the cleanups, Haggard explained. Better post-project monitoring programs need to be designed and implemented and should include comparisons to pre-project conditions. GE has yet to determine what basic data to include (and where to get the data from) in analyses of the success of projects. Other work that remains on the project includes updating the data and continuing Quality Assurance (QA) efforts. The Sediment Management Workgroup is also going to look over the database, assess its utility, and make suggestions about its design. Finally, GE needs to decide how to present the database to the public.

A participant asked whether the rationale for choosing a particular cleanup level was included in the database. Haggard replied that the remedial action objective, the cleanup level goal, and the findings of the risk assessment were included, although those three pieces of information do not necessarily provide an understanding of why a particular cleanup level was chosen. Sabine Apitz commented that she sat on a National Research Council panel several years ago assessing sediment remediation. The panel concluded that risk assessment and rational management strategies almost never drove remedial decision-making. Instead, decisions were usually driven by economic or political factors. Finkelstein said that his view of the situation is that decisions are usually based on a rational framework, except when remediation to desired levels is not technically feasible.

Another participant inquired about the QA process for dealing with anecdotal information in the database that was not confirmed by other sources. Haggard answered that the GE team had tried to find written documentation or corroboration from other individuals. Also, another firm had attempted to independently verify the information in the database. Finally, data sheets about individual sites were sent out to the appropriate PRPs and industry contacts for confirmation. Some corrections have been received through that avenue, Haggard said, but they have been very minor.

Finkelstein said that the National Oceanic and Atmospheric Administration (NOAA) is working on a similar database effort, but is not as far along in the process. Finkelstein proposed that government and industry work together, if GE is willing to make its information available. Haggard replied that he intends to make all the facts available, so he would be happy to assist the NOAA effort.

SEDIMENT RECOVERY DEMONSTRATION PROJECTS UNDER THE CLEAN WATER ACTION PLAN

Jim Keating, EPA

Jim Keating began by displaying a copy of the Clean Water Action Plan (CWAP), which was signed by the Secretary of Agriculture and the EPA Administrator on February 14, 1998. The object of the CWAP is to restore and protect watersheds. Within the section of the plan called Actions to Strengthen Core Clean Water Programs is a Key Action that is described this way: "In 1998, EPA will initiate place-based contaminated sediment recovery demonstration projects in five watersheds selected from those identified in EPA's National Inventory of Sediment Quality as being of the greatest concern. Remediation efforts will be coordinated with federal natural resource trustees."

Keating said that, due to political pressure to adhere to the schedule set forth in the CWAP, discussions on planning the demonstration projects began quickly under a workgroup called the Sediment Network, which is made up of EPA personnel from some research labs, some program offices, and all of the regional offices. Keating heads the workgroup, which is devoted to sharing information on sediment-related issues of mutual concern. The workgroup decided that each regional or geographically focused program office could nominate candidate projects. Since it is not known how much funding will be available for EPA to contribute to projects, candidate projects would have to be planned and funded already. If funding becomes available from CWAP-targeted areas or other sources, the money will be used to expand or enhance the goals of existing projects or to conduct secondary analyses of the benefits of the approaches utilized at individual sites. Even if significant funding does not become available, EPA has the opportunity to demonstrate the array of remedies available for contaminated sediments and promote individual projects.

According to Keating, EPA is interested in selecting an array of projects, such as those that involve dredging and removal, demonstration of a new treatment technology, confined aquatic disposal, in situ or ex situ treatment, capping, promotion of natural attenuation, and/or source control. Since the goal is sediment recovery, rather than sediment remediation, there is a wider range of possibilities. The selection criteria are:

• Documented commitment to project completion
• Value of project expansion or additional analyses resulting from CWAP funds
• Relevance of project to reducing environmental and/or human health risk
• Demonstration of new and promising remediation/prevention technologies or approaches
• Demonstration of state/federal or other partnerships
• Demonstration of solutions to problems that are common to many areas of the country

Keating listed the eight candidate projects nominated: Lower Willamette River, Oregon; Detroit River, Michigan; Patrick Bayou, Texas; Capitol Lakes, Louisiana; Blackstone River, Massachusetts-Rhode Island; Woonasquatucket River, Rhode Island; Anacostia River, Washington, D.C.; and Baltimore Harbor, Maryland. After the projects were nominated, it became apparent that no further funding would be available for the projects immediately, so the list of eight has not yet been whittled to five.

Apitz commented that most of the candidate projects appeared to be freshwater sites and asked if EPA would distinguish between freshwater sites and marine sites when projects were selected. Keating replied in the negative. As it turns out, however, candidate projects represent a mix of sites, since several sites have tidal waters. The eight projects also represent a variety of contaminants, addressed by a variety of primary project elements.

Keating explained that EPA has not yet decided whether to narrow the list of eight projects to five, whether to work with all eight sites, or whether to add to the list of eight. He said that the Sediments Remediation Action Team is welcome to suggest additional projects or to pursue a more tangible partnership with EPA. For example, Keating suggested, it would be mutually beneficial if there were, in the future, a link on the EPA Sediment Recovery Demonstration Project Web page to the RTDF Web site and vice versa. (The Sediment Recovery Demonstration Project Web site is not yet available.) The overheads that Keating showed during his presentation are available as Attachment E.

UPCOMING MEETINGS

Danny Reible, Louisiana State University

Reible told the group about upcoming meetings and conferences of interest. The information he provided is available as Attachment F. One meeting that might be of particular interest, he said, is the 1999 Conference on Hazardous Waste Research in St. Louis, Missouri, from May 25 to 27, 1999. It will be hosted by the South/Southwest Hazardous Substance Research Center and the Rocky Mountain/Great Plains Hazardous Substance Research Center. The meeting will focus on contaminated sediments, brownfields, and emerging technologies, including phytoremediation. An open call for papers has been issued and is extended to the Sediment Remediation Action Team; abstracts are due February 1, 1999. The conference will include a workshop on assessment and remediation of contaminated sediments, with a special focus on in situ remediation. There will be 2 days of technical sessions, generated by the papers selected from those submitted.

Reible suggested that the Sediment Remediation Action Team could meet in conjunction with the conference. A meeting room would be available at the hotel for the Action Team to use. The conference requires a $200 registration fee, which includes lunches, but the fee could be waived for those who attend the Action Team meeting but do not attend the conference.

Other upcoming meetings include a proposed session on the bioavailability of contaminants in sediments at the Society of Environmental Toxicology and Chemistry (SETAC) meeting from November 14 to 18, 1999, in Philadelphia, Pennsylvania. In addition, there will be a symposium on environmental issues on the Gulf Coast at the American Chemical Society meeting from August 22 to 26, 1999, in New Orleans, Louisiana. There will also be a symposium on sediment contamination in coastal watersheds at the latter meeting.

Jensen told the group that another option would be to hold the next Action Team meeting in conjunction with the Battelle Memorial Institute bioremediation conference in San Diego, California, from April 18 to 23, 1999. The conference will include discussions of important issues, Jensen said, and is always very well attended. However, it would be difficult to work an Action Team meeting into the schedule of Battelle activities, which is full. The most likely possibility is Wednesday afternoon, which is a free period on the agenda for recreation and touring the area. The Action Team meeting would probably last all afternoon and into the evening. A catered dinner could be brought in. The sediments session of the Battelle conference is the following morning. The Battelle conference costs about $700, but there may be a smaller 1-day registration fee.

The Action Team voted on whether to meet during the Battelle conference in April or the Hazardous Substance Research Centers conference in May. A majority of the group voted for meeting in St. Louis in conjunction with the 1999 Conference on Hazardous Waste Research.

REVIEW OF RESEARCH AND DEVELOPMENT AT DULUTH LABORATORY

David Mount, EPA

Mount explained that the Duluth Laboratory falls within EPA's Office of Research and Development (ORD) and does research in support of programmatic and regulatory offices of EPA. He provided an overview of the projects that the lab is working on, and the overheads he used during his presentation are available as Attachment G. The lab does not design remediation strategies, but conducts research to provide regulatory and monitoring tools. The lab is also a part of the Effects Megalab within ORD and focuses on the effects of stressors on organisms. The premise of its research is that in order to study the biological effects of contaminated sediments, it is necessary to have a way to measure toxicity, to be able to predict the response to a given chemical composition, to have a way to find out what is toxic, and to consider bioaccumulative chemicals separately.

The lab assisted with recent revisions to the Freshwater Sediment Test Procedure Manual, originally published by EPA in 1994. According to Mount, there are minor changes to the solid phase tests, and there are two major additions: longer-term sub-lethal test methods (42-day survival, growth, and reproduction tests) for Hyalella and an 8-week life-cycle chronic test for a midge. The manual also outlines a 28-day bioaccumulation test with an aquatic earthworm to measure the uptake of chemicals from contaminated sediments into the organism, an attempt to represent the first step in the process that moves bioaccumulative chemicals from the sediment up through the food chain. The manual is currently being formatted by a contractor for printing; publication is expected this spring.

One way in which the Duluth lab has assisted with relating chemistry to biological effects is by assisting with the development of EPA's Equilibrium-Partitioning Sediment Guidelines (ESGs), formerly called Sediment Quality Criteria (SQCs). Mount explained that the name was changed because the word "criteria" led people to assume that the SQCs would be treated as standards that must be met. The Office of Water, however, has decided to implement the ESGs as guidance to assist with the determination of whether a particular sediment poses a risk, i.e., toxic materials in toxic amounts. The ESGs for endrin and dieldrin, which were selected because they were not controversial, have been written and will soon be printed. There are also Technical Basis Documents that describe the science behind the derivation of the ESGs. User's Guides for the endrin and dieldrin ESGs, which describe how the Office of Water suggests the guidelines be used by different regulatory programs under different legal authorities, are going out for public comment.

Mount said that an ESG for metals (copper, cadmium, zinc, nickel, lead, silver, and chromium) is being sent out for public comment and Science Advisory Board review. The ESG describes how to predict the absence of toxicity, rather than the presence of toxicity. Therefore, sediments cannot fail an ESG, they can only fail to pass an ESG. It is predicted that metal toxicity will be absent if either of two conditions is true: if the amount of acid volatile sulfide (AVS) is greater than the amount of simultaneously extracted metals (SEM) (i.e., there is excess sulfide present) or if the sum of dissolved metals in pore water, normalized to their water quality criteria (AWQC) concentrations, is less than 1. Mount is hoping the former model can be expanded to include other binding phases, such as organic carbon, to explain variability when SEM is greater than AVS.

Mount went on to say an ESG for PAHs is also being prepared. Originally, SQCs had been proposed for individuals PAHs--acenaphthene, phenanthrene, and flouranthene. The proposals were withdrawn because field samples from sites where there was strong evidence that PAHs were the source of toxicity showed high mortality of amphipods at concentrations well below the proposed criterion. The explanation is that PAHs do not occur individually, but in suites, so it is impractical to design guidelines for individual PAHs. Thus, PAHs will be dealt with as mixtures.

Apitz asked if the guidelines were being designed to recognize the two different modes of action of PAHs, narcosis and ultra violet-enhanced toxicity. Mount responded that the guidelines being developed are based on narcosis alone, because the implications of the ultra violet (UV) pathway are not well understood quantitatively. The implementation section of the document will talk about concerns related to high UV environments.

Mount explained that the approach to the PAH guideline is grounded in work by Rick Swartz, published in Environmental Toxicology and Chemistry beginning in 1995. Swartz developed a model called the Sigma-PAH Model, which predicts the mortality of estuarine amphipods as a function of PAH concentration. The model relies on the assumption that the toxicity of the mixture is equal to the sum of the individual fractional toxicities. Thus, the ESG for PAHs will call for summing the fractions generated by dividing the concentration of individual PAHs by a guideline for the particular PAH (which is derived from its threshold for toxicity). The desired result is for that sum to be less than 1. Mount stated that a guideline of just under 1,000 micrograms per gram of organic carbon does a pretty good job of separating the cases of increased mortality. One problem faced is that the toxicity caused by the multiple aromatics affecting a sediment is not necessarily represented by a measurement of the 16 priority pollutant PAHs.

Mount then touched on other work conducted by the Duluth lab. For example, the lab is doing research on methods of spiking chemicals onto clean sediments and ways of monitoring the equilibration of the chemicals over time. The lab is also active in the development of sediment toxicity identification evaluation (TIE) methods. This involves manipulating, in a test tube, sediments with an unknown source of toxicity. The manipulations are designed so that sediments respond differently to a given manipulation depending on the source of toxicity. One example is zero valent metal treatment, which adds zero valent magnesium to cationic cadmium, after which a redox reaction occurs that creates ionic magnesium and cadmium metal, which are nontoxic. A similar procedure calls for increasing the acid volatile sulfide (AVS) in sediment. Amorphous iron sulfide is a good medium for an exchange reaction in which cationic toxic metals precipitate as sulfides and free up relatively nontoxic iron. The understanding gained from this research about the way contaminants behave in sediments may have wider applicability.

For non-ionic organic toxicants, the lab has been running tests on a carbonaceous resin called Ambersorb. The tests call for incorporating 4% resin directly into the sediments. Chemicals are not removed from the sediment; instead, their availability in the sediment is being changed. The goal is to reduce the activity of chemicals in sediments by providing a high-affinity sorption site, thereby reducing concentrations in the pore water. Pore water concentrations are believed to be proportional to the potential toxicity of a chemical in sediment. Ambersorb seems to be extremely effective at removing exposure in pore water, preventing bioaccumulation in aquatic organisms, and preventing mortality under UV light. However, Ambersorb is not effective in DDT-contaminated sediments, for reasons that are still unclear, according to Mount.

The Duluth lab has analyzed chemistry of sediments after the addition of Ambersorb for periods of up to 60 days, but not for longer periods. The final condition of the sediment once mass transfer and equilibration have taken place has not been analyzed, though it would be of interest when considering Ambersorb as a remedial technology. It is believed that the efficiency of Ambersorb treatment is dependent on the K_OW (octanol water partition coefficient) of the chemical that was being sorbed. Tests indicate that the higher the K_OW, the lower the treatment efficiency. Chemicals with a high K_OW seem to react with Ambersorb as if it were ordinary organic carbon.

A participant asked whether the lab is actively considering residual questions mentioned related to PAHs. Mount replied affirmatively. For example, conventional wisdom holds that compounds with a high K_OW are not soluble enough to be toxic. The lab conducted an experiment in which it spiked clean sediment with only PAHs in that category, on the theory that spiking sediments with multiple high-K_OW chemicals would increase toxicity. The experiment supported the theory.

A participant commented that some of the technologies Mount discussed (e.g., adding resins and iron compounds to sequester contaminants) could be used as part of a cap, rather than added directly to contaminated sediments. Mount agreed. Another participant asked whether experiments have been conducted on metals other than cadmium. Mount explained that cadmium, copper, nickel, and zinc have been studied. With regard to sulfide spiking, he said, copper was found to be most responsive and nickel least responsive, with cadmium and zinc somewhere in the middle. For the base metals, cadmium, zinc, and nickel seem to work well, but copper behaves strangely.

A participant inquired when EPA might publish sediment TIE guidance. Mount answered that it is in part dependent on him. There is draft guidance on pore water TIE, and it is similar to effluent guidance. Solid phase guidance (for extremely high-K_OW chemicals for which pore water testing does not work well) is supposed to be drafted this year. Guidance on both types of testing will be issued together in one document.

POTENTIAL APPLICATIONS OF PHYTOREMEDIATION

Michael Coia, Phytoworks, Inc.

Coia began his presentation by explaining that the conventional use of phytoremediation involves in situ planting, although some aspects of applying phytoremediation to sediments are different from more conventional applications. He then compared phytoremediation with some of the conventional remedial approaches.

Coia stated that in situ capping of contaminated sediments has potential limitations, such as regulatory acceptance, the need for long-term containment assurance, and the degree of isolation required. In less-contaminated sediments, phytoremediation competes with excavation and incineration or landfill disposal (sometimes preceded by low-temperature thermal desorption). Often contamination levels which adversely affect the benthic community are lower than contamination levels which require treatment before land disposal. Phytoremediation also competes with dredging and sediment washing, but phytoremediation can be used in conjunction with them (for example to enhance the degradation that occurs in a sediment washing plant). Phytoremediation can compete with soil vapor extraction (SVE) or other bioventing techniques that remediate constituents of volatile organic compounds (VOCs), as long as root zone penetration can be achieved in an in situ planting. Some of the earliest research on phytoremediation was on nitroaromatics, such as TNT and DNT. This research demonstrated that nitroaromatics are phytodegradable in the same time frame and with the same kinetics necessary for conventional biotreatment technologies, such as bacteria-based land farming and composting.

The contaminants that can be remediated with phytoremediation are petroleum hydrocarbons (PAHs), chlorinated solvents (VOCs, which can be degraded or phytovolatilized), chlorinated pesticides (DDT, DDD, and Toxaphene), metals (so far there has been success with arsenic, mercury, lead, and cadmium), nitroaromatics (TNT, DNT, and RDX), agricultural nutrients (nitrates and phosphates), and mixed radionuclide wastes.

Understanding site conditions is key to utilizing phytoremediation, Coia said. First, the roots of the selected plants have to penetrate the area that requires remediation. Thus, before planting, it is necessary to know where the contamination is and whether plant growth can occur in that zone. If dredging occurs, dredged sediments and plantings can be placed in the desired relation to each other. Other site conditions that need to be understood are the relative proportions of different soil types in the sediment, the soil moisture, the partitioning coefficients, the agronomic parameters, and whether phytoremediation will occur in upland or wetland conditions. Some phytoremediation firms claim that phytoremediation will work if plants are growing on a site, but that is a simplistic assertion, in Coia's opinion. However, it is true that the presence of a tremendous hot spot where nothing is growing probably indicates that in situ phytoremediation will not work in that location.

Coia described three categories of phytoremediation techniques. The first, phytoextraction, relies on two types of plants to extract contaminants: hyperaccumulators (plants that harvest metals from sediments and bioaccumulate the metals in their tissues) and genetically altered plants. Coia said that his company, Phytoworks, has developed transgenic plants to take up methyl mercury from sediments utilizing bacteria-based genes that change the speciation of methyl mercury to ionic mercury and then to mercury gas, which is emitted through the plant tissues. At several universities across the country, researchers are studying how to genetically alter plants to be better bioaccumulators. If the metals are taken up more readily, the number of necessary harvesting cycles is reduced.

The second category of phytoremediation techniques is phytodegradation. The most prevalent type of phytodegradation is phytovolatilization, Coia explained. For example, hybrid poplars have been planted on a number of sites as hydraulic barriers that remediate chlorinated solvents. In some cases, the trees are not appropriate for degrading the organics, and VOCs are released from the plant tissue. Another type of phytodegradation is phytometabolism, which involves identifying the enzyme-based processes that are geared to addressing families of organics, such as nitroaromatics. The enzymes can be active extra-cellularly, so the plants containing the enzymes can be made into pulp, and the pulp can be introduced to sediments in an ex situ context. A third type of phytodegradation is rhizo-microbial phytostimulation, in which bacterial bioremediation and phytoremediation work together.

The third category of phytoremediation techniques is phytostabilization. One example is phytosequestration, the sequestration of metals, which is enhanced in the rhizosphere.

Coia went on to describe methods of applying phytoremediation to groundwater, surface water runoff, and industrial wastewater. These methods include phyto-hydraulic barriers, constructed wetlands (which include specially selected plant species), and vegetative caps and stormwater runoff swales planted with species that phytoremediate contaminants.

While phytoremediation is talked about a great deal and considered in feasibility studies, there are not a lot of projects actually utilizing the technology, Coia asserted. Last year, perhaps $20 million to $21 million was spent on phytoremediation, a very small fraction of what is spent on remediation overall. Phytoremediation is used more in laboratories and field demonstrations than in actual full-scale projects. This is largely due to the fact that there is not a lot of data demonstrating that phytoremediation will achieve ultimate cleanup criteria. The Action Team could help address this deficiency, Coia said.

Applications of phytoremediation that may develop in the next 3 or 4 years could save 20% to 70% of remedial costs, Coia predicted. He went on to describe some applications that may be used more frequently in the future:

• Reactive barrier walls. The use of reactive barrier walls is based on the fact that, in some plants, zero valent chemistry enhances certain plant enzyme activity that destroys specific organics. Slurry walls are now being installed that use frontal and gate technology, where the gate is a reactive barrier zone. The reactive barrier zone could involve plant enzymes.
• Urban stormwater management. In the future, phytoremediation could be used as a component of urban stormwater management plans.
• Engineered wetlands. Over time, Coia stated, engineered wetlands will probably become the most prevalent application of phytoremediation, especially in situations in which contaminants are discharged into a preexisting wetland.
• Sediment-washing reagents. Another potential scenario is one in which certain plant enzymes are concentrated and utilized as sediment-washing reagents to destroy organics.
• Dredge sediment admixing. In the future, dredge sediment admixing, the direct injection of plants into sediments that have been moved, might preclude the need for ex situ processing plants. If ex situ processing plants have to be constructed, it will be impossible for contaminated sediments treatment to cost as little as $35 per cubic yard. That price might be achievable, however, if the correct plant chemistry were injected into sluiced materials on their way to a confined disposal facility (CDF).
• 
  
• Biogenetically engineered plants/cultured species. Most biogenetically engineered plants and cultured species that might be used in phytoremediation are not agriculturally-oriented.
• Mixed radionuclide hazardous waste phytoremediation. Coia predicted that there may be more mixed radionuclide hazardous waste phytoremediation projects in the future. Currently, three Department of Energy facilities have active programs of that type.

To determine if phytoremediation can compete with conventional treatment technologies, it needs to be considered during the remedial remedy selection process and before the Record of Decision is issued, Coia suggested. Laboratory treatability work or demonstration work might occur during the site characterization or remedial investigation/feasibility study phase. Often at least a year's worth of design investigations are conducted, during which phytoremediation might be considered.

Coia presented the following matrix for comparing conventional technologies with their counterpart phytoremediation technologies:

In closing, Coia referred interested parties to the compendium of phytoremediation literature on the Phytoremediation of Organics Action Team Web site, at http://www.rtdf.org/public/phyto/phytobib/biba-b.html. Also, the overheads that Coia presented are available as Attachment H.

A participant asked what the biggest barriers to implementing phytoremediation were. Coia responded that the most significant barrier is the lack of a large body of data demonstrating definitively how the technologies work and outlining design parameters for utilizing them. Because of this, regulatory agencies are cautious about approving phytoremediation.

Finkelstein inquired whether there were difficulties using phytoremediation in sediments with very high concentrations of contaminants. Coia responded that there is not enough data on this topic. The Phytoremediation of Organics Action Team has evaluated some work showing that the very highly concentrated PAHs are not phytoremediable. However, it has not yet been determined what the achievable cleanup levels are for phytoremediation, and it is not necessarily the case that the cleanup levels that have been achieved in the laboratory can be achieved in the field.

FOCUS ON CAPPING

Michael Palermo, Waterways Experiment Station
Norman Francingues, Jr., Waterways Experiment Station

Francingues and Palermo shared some comments about capping projects. The overheads that each showed are included as Attachments I and J. Francingues explained that this presentation on capping was aimed at setting up the discussion of potential capping projects the following day. He said that more field information on capping is needed to fully evaluate its potential for remediating contaminated sediments.

Francingues said that the Army Corp of Engineers' Waterways Experiment Station (WES) is conducting research related to contaminated sediments. The station's Dredging Operations and Environmental Research (DOER) program supports engineer operations and maintenance in navigation programs in an effort to balance operational and environmental needs for dredging and disposal of dredged materials. Whether a project requires excavation of sediments to maintain a channel or to move contaminated sediments to a site where they can be remediated, it is necessary to consider the implementation of the project, such as how to utilize equipment most efficiently and in the most cost-effective manner.

Francingues described DOER's six focus areas as: contaminated sediments, near-shore placement, instrumentation, risk, innovative technologies, and environmental windows (i.e., at certain times of the year, such as when sensitive species are spawning in an adjacent habitat, dredging is not allowed). Instrumentation involves examining the economics of dredging by monitoring what is being dredged, how much is being dredged, and where the dredging is occurring in an effort to hold contractors to what they promise. In addition, DOER is evaluating the need for demonstrating instrumentation associated with better and more rapid site characterization. For example, DOER is investigating the research needs when it comes to distinguishing areas that require remediation from those that require dredging without remediation. The heterogeneity of sediments complicates delineation, Francingues said. DOER's work on risk supports its goal of making its risk assessment framework consistent with EPA's. Risk, in this case, is defined not only as environmental risk, but also as economic risk.

The objective of the innovative technologies demonstration area, which Francingues heads, is to identify and demonstrate emerging dredging and disposal technologies so that they can be used in cleanup projects. To this end, DOER cooperates with other agencies and field offices within Corps, as well as with regions and states. Innovation may occur not only in equipment or hardware, but also in bidding processes, instrumentation, or monitoring techniques. Innovative technologies are selected if they address an area in need of innovation, if they provide a cost savings, if there is a high probability of being able to implement them within a few years, and only if a demonstration co-sponsor (often within the Corps) is available.

Francingues anticipates that the innovative technologies focus team will be receiving, probably in May, an assessment of dredging techniques utilized in Europe and recommendations about those that might be appropriate for the Corps to use. The team also has a database of information about over 100 dredging techniques. The team has been identifying domestic and foreign commercial technologies, through literature reviews and some field visits, and evaluating the cost savings each might provide. The team will demonstrate the most appropriate innovative dredging equipment and techniques and recommend the use of the best-suited technologies. A participant asked if the focus of the DOER program is maintenance dredging or remediation dredging. Francingues indicated that the focus is on maintenance dredging for navigation, since that is the Corps' mission.

Francingues said that he was involved in a study by an international navigation association that produced a CD-ROM summarizing the handling and treatment of contaminated dredged materials on an international level. It includes 18 case studies and about 90 technology evaluations and is searchable. For more information, individuals should contact Maryjane Robertson at Robertson-Maryjane-CEWRC-IWR-XO@mail.wes.army.mil.

Francingues stated that the DOER team is evaluating a telescoping weir for confined disposal sites. In a confined disposal facility with ponded water, the effluent would conventionally be released over stop-log weirs to an outlet box. The telescoping weir is a series of concentric cylinders that are used to raise and lower the water level. An effluent of appropriate quality can be maintained, and dredged material can be dewatered so it dries out. Thus, the technology allows water to be managed at CDFs more effectively, more efficiently, and more economically than had been previously possible. Francingues mentioned that the Corps has applied for a patent on the technology. It is being used in one district, but it could be used anywhere. The weir is transportable, self-contained, and operates on solar power. DOER's goal is to promote this and other technologies to interested parties within the Corps and outside of it.

Palermo began by observing that the Corps is interested not only in capping, but also in in situ dredging and all the promising methods of treating and managing dredged contaminated sediments. The contaminated sediments focus area within DOER works primarily on CDFs, the mostly commonly considered disposal option for contaminated sediments, rather than capping projects. Contaminated sediments encountered when dredging for improved navigation are usually less contaminated than sediments dredged from sites that require remediation, but dredging for navigation usually results in a much larger volume of sediment to dispose of. Work related to CDFs covers such topics as separation, vegetation management, manufactured soil, the design and application of containment features, reclamation, site screening (which utilizes models to determine if groundwater leachate is a potential contaminant pathway), and assessment methods.

Palermo went on to discuss some of the capping work that has been completed by DOER. Two guidance documents have been written and are available on the Web. The first, entitled Guidance for Subaqueous Dredged Material Capping, focuses on dredged material capping, rather than in situ capping. The document addresses marine dredging more than dredging of riverine or near-shore locales. The second document, Guidance for In-Situ Subaqueous Capping of Contaminated Sediments, published by EPA's Assessment and Remediation of Contaminated Sediments (ARCS) Program, argues that capping can be very cost-effective and that effective caps can be designed and implemented. The document describes recommended technical procedures for evaluating an in-place capping project. It is probably the most comprehensive guidance available on capping as a remediation technique, Palermo said, although it would be impossible to cover every question related to capping, since some questions have not yet been conclusively answered (for example, questions relating to evaluating projects, construction techniques, design approaches, and models to use). Palermo stated that, by interacting with other groups, such as the Action Team, the Corps can help answer these questions.

The Corps has proposed establishing another work unit under DOER to study cap effectiveness. Palermo explained that the new unit is not funded this year, but may be funded next year. The goal of the unit would be to improve the available guidance on the design of caps, with a focus on investigating better ways to characterize sediments in order to define appropriate parameters for caps. DOER hopes that existing models can be improved and refined so that models can accurately predict if there is potential for contaminants to migrate slowly through a cap into the overlying water. Palermo said this is a particular concern for remediation projects.

Information about WES research is available on the Web at http://www.wes.army.mil/el/dots/. The Web site for the DOER program is located at http://www.wes.army.mil/el/dots/doer/.

WES has some ideas about what its would like the Action Team to consider, Palermo said, mostly related to ways to improve knowledge of capping as an option. Specifically, WES would like to improve strategies to support containment as a viable option. In order to address public concern about subaqueous containment, more data needs to be generated and compiled from models, tests, and monitoring to prove that it will be effective. WES would also like to develop procedures that would enhance its ability to design caps. (This is related to its goal of improving predictive models and design approaches.) Until now, cap thickness has been determined by adding together the cap thickness required to address five concerns: bioturbation by organisms, consolidation of sediments, potential erosion of sediments, chemical migration, and operational considerations. It might be possible, Palermo suggested, to look at how the five processes affect one another to determine whether a less thick (and therefore less expensive) cap would be effective. Finally, Palermo is hoping the Action Team will be able to assist with field verification of capping effectiveness, to validate models that predict capping effectiveness.

Palermo said that interested parties should also consider other nontraditional approaches to capping involving an inert layer of soil, rock, or geotextile. "Smart caps" can be designed to be self-maintaining or self-monitoring or to have other innovative features. For example, a contaminant adsorbing layer could be built into a portion of a cap. Caps can also be designed to enhance the aquatic environment, and it would also be useful to find a good way to predict how the placement of a cap will resuspend contaminated sediments. Tools for evaluating flux (both advective and diffusive) can also be improved, for example.

Palermo suggested that the Action Team could assist with field verification of capping effectiveness by conducting a demonstration project. He said that skeptics dubious about laboratory modeling are more likely to be convinced by data collected about contaminant migration at actual sites over time. He cautioned that the field site where a number of capping schemes are compared has to be controlled. The sediments need to be homogeneous, and the caps need to be isolated from storms and other variables. In addition, the physical integrity of caps has to be verified. For example, it would be useful to have proof that 3 feet of sand can be placed over fine, soft mud without much mixing occurring and without it turning over. Precise layering can be achieved in smaller, more controlled sites. Palermo said that field testing would utilize conventional data collection methods, such as coring and analysis of the cores for chemical concentrations and physical parameters. Smart cap features could also be employed at field sites. Palermo asserted that a multi-year time frame is critical for this type of project and that the field site would have to be monitored for at least a few years, though proof would be strongest after 5, 7, or even 10 years.

Palermo alerted participants to several other ongoing projects:

• A capping modeling project funded by EPA, which David Petrovski is working on.
• The Corps' work on model refinement related to contained aquatic disposal in pits that might be used in the New York Harbor.
• Active large-scale capping projects related to navigational dredging in Boston, New York, and Los Angeles, on which the Corps is collecting monitoring data.

One participant noted that it would be valuable to create a test site where effects of the cap on the environment as a whole were considered. Palermo agreed. However, he thinks that chemical isolation and other processes should not be studied at the same site because they might interfere with each other.

Coia asked whether the Action Team wanted to consider a site where existing contaminated sediment needs to be capped, rather than a site where contaminated sediment is dropped onto an area to be capped. Palermo replied that the latter would be desirable, in order to ensure homogeneity of the sediments capped. Coia stated that a contaminated sediment deposition project might encounter difficulties with resuspension during dredging and with dropping contaminated sediments through a significant depth of water. Palermo said that monitoring those two concerns would not be the focus of the study, though monitoring would have to occur to ensure that neither was a problem. A demonstration project would probably involve only a few thousand cubic yards of material, he said.

Jensen suggested using a checkerboard arrangement with different technologies adjacent to one another at the test site. He acknowledged that this could cause logistical problems: divers might have to place the caps, or a masking technique might have to be utilized. Palermo responded that he would prefer to see squares isolated from one another. He said that the focus of the project would be the effectiveness of the cap at sequestering the contamination after it is in place. There would have to be advection, in addition to diffusion, occurring in the water at the test site.

Finkelstein said that he had visited WES, which has wave tanks and other large basins. He asked if a test could be conducted in the lab in which waves and currents were mimicked to avoid some of the difficulties that would be encountered in the field. Palermo said he was less concerned about waves and currents in the field and more concerned about migration. If he receives funding, he will conduct further laboratory experiments in a test cell that is a meter square and about 10 feet high, which allows him to test caps and chemical migration. But he still believes that the results of full-scale tests in the field are necessary to convince people of the effectiveness of caps.

A participant observed that bioturbation, consolidation, erosion, chemical migration, and operational considerations -- the key factors in cap design, according to Palermo -- all vary from one site to another. He requested that Palermo comment on how important it is to take into consideration unique local environmental conditions when selecting test sites. Palermo responded that site selection would depend on the goals of the demonstration. If the focus is the isolation effectiveness of the cap and not its ability to resist wave action or currents, then wave action and currents should be eliminated for the purpose of the test.

A participant inquired whether the Corps might be able to go back to sites it has already capped and evaluate how successful the caps have been at isolating contaminants. Palermo responded that the Corps is doing that. The data collected by the Corps over time has been published and is cited in the Web reports that he mentioned.

Jensen reminded the group about an engineering solution called AquaBlok^TM, which was discussed at the last Action Team meeting. The product, designed by a company in Toledo, Ohio, has a layer of polymer and clay surrounding a piece of gravel. A cap consisting of a very large number of these gravel composites can be placed. The gravel weighs the material down, and after a few months, the clay and polymer become glued together, creating a sort of blanket. Jensen said that there is an opportunity for putting other chemistry, such as zero valent irons or enzymes, in the AquaBlok^TM. He inquired whether the material would qualify as a smart cap, and Palermo responded affirmatively. Jensen then asked about the potential use of something like the benthic flux chamber as an evaluation tool. Palermo responded that the chamber could be used as a monitoring tool at field sites. He went on to describe a self-monitoring technology called a peeper, which looks like a washboard. It is put in a cap, Palermo said, and it draws in contamination. It can be used as a monitoring tool that is removed and examined, in lieu of taking a sediment sample.

Reible commented that a problem with studying existing sites is that a lot of innovative technologies, like peepers and benthic flux chambers, have rarely been used. One purpose of demonstration projects would be to study problems with placing a cap, as well as the effectiveness of the capping approach.

Coia said that he believes it is a misnomer to call capping a much more cost-effective solution than dredging and treatment, if costs for the latter can be driven down to $35 per cubic yard or less. Capping costs can reach $25 to $28 per cubic yard, he said, and may no longer be the cheapest solution, especially if additives are used. Palermo agreed that, for some projects, capping will probably not be the least expensive solution. Also, he suggested that the $35 per cubic yard figure mentioned previously in connection with the New York/New Jersey Harbor Project may only include treatment and may not include dredging, dewatering, and other required pretreatment steps.

Jensen recounted that Francingues had explained to him that pulp floats. In cases where pulp is mixed with sediments for treatment, the pulp to treat them might end up being separated from them by the water. Jensen suggested that the Action Team might be able to develop an engineering solution to the problem of separation. Several participants said that a potential solution would be to use a mat.

One participant observed that capping using contaminated sediments creates a number of difficulties. For example, checking the thickness of the cap after a few years and after 10 years will entail punching a hole through it, which would release contaminated sediments. Palermo said he takes an opposing view about where the difficulties lies. Designing a cap to withstand waves and currents is a lot more straightforward than the question of long-term effectiveness and chemical migration. It is not hard for designers to create a cap that they feel is effective; the difficulty is in convincing skeptics that it will in fact isolate contaminants for hundreds of years. At some point, it is necessary to conduct a field verification of capping effectiveness under a controlled scenario to show people that models are accurate. Whether or not this is the most pressing step is a matter open to discussion. There are a lot of questions related to erosion resistance to investigate, for example, but they seem much more straightforward than demonstrating capping effectiveness.

Coia questioned whether the active projects Palermo described in Boston, New York, and Los Angeles could be monitored. Palermo answered that they are being monitored, but they are not controlled: there are not consistent conditions over a long period of time to analyze. Additionally, there is no opportunity to compare different approaches, including capping materials, designs, and thicknesses, under the same conditions. This information would probably never be furnished by any one full-scale site.

A participant suggested using a surrogate, rather than a contaminated site. One possibility is a bentonite blanket with a tracer--a well-defined contaminant--in it. Then the checkerboard approach could be applied, forestalling the need for homogenization. Palermo said that such a project would be informative and is a possibility, but that a real contaminated sediment in the field would be more convincing. The questioner said that trying to homogenize contaminated sediments that originally had hot spots would lead to questions about the success of the homogenizing process. Palermo agreed that such questions may arise, but thought that the concern might be defeated by the experimental design. It might be unnecessary to homogenize contaminated sediments if a site is located in a place where there is low variability in the characteristics of different samples. He added that Action Team members representing industry are most likely to know about potential demonstration sites.

Palermo asserted that sites must be significantly, rather than slightly, contaminated in order to test the efficacy of caps. There might have to be some negotiation with the regulatory agencies involved with a site to stress that the capping study would not be the final remedy and that an alternative remedial action could be chosen for the site later on. The Corps and industry have a substantial economic incentive to prove that capping will work because using it saves money.

Coia observed that a demonstration project could be conducted at a site while another remedial remedy is being designed, because the design process sometimes takes so long. The demonstration might produce enough data to convince those involved to change the remedial remedy selected, even if there had already been a ROD.

THURSDAY, JANUARY 14, 1999

RTDF OVERVIEW

Walter Kovalick, Jr., EPA

Kovalick, director of the Technology Innovation Office (TIO), began his presentation by providing information about the Year 2000 (or Y2K) problem. Kovalick explained that groundwater monitoring systems and other analytical tools often utilize expressions involving the date that will not function properly when the date changes from '99 to '00. For example, a power plant in the United Kingdom experimented with setting its clock to the year 2000, and the plant shut down because the system was trying to divide by zero. EPA has announced an enforcement policy that allows industry to announce, through the month of February, experiments of short duration to test whether systems will function in 2000, without being concerned about being out of compliance with any permits or regulations. Kovalick provided several Web addresses which describe the Y2K problem and sample tests of its effects: http://clu-in.org, http://www.epa.gov/year2000, http://www.millennia-bcs.com, http://www.year2000.com, and http://y2kjournal.com.

Kovalick explained that the primary purpose of his presentation was to outline what the Sediments Remediation Action Team could be. His presentation materials are included as Attachment K. The RTDF Action Teams are completely team-driven, Kovalick said, and the TIO does not have much money to contribute to their endeavors. The TIO is interested in helping deploy new technologies and addressing barriers encountered by the interested parties, including the responsible parties, federal and state project managers, consulting engineers, and technology vendors. The goal is to engage problems that do not have obvious, low-cost solutions.

Kovalick described one calculation of the future remediation market in the U.S., which estimates:

• 550 Superfund sites requiring a total expenditure of $7 billion
• 3,000 Resource Conservation and Recovery Act (RCRA) Corrective Actions requiring $39 billion
• 165,000 Underground Storage Tanks (USTs) requiring $21 billion
• 8,300 Department of Defense sites at 1,560 installations requiring $63 billion
• 10,500 Department of Energy sites at 137 installations requiring $63 billion
• 700 other federal agency installations requiring $15 billion
• 29,000 state sites requiring $13 billion

These figures are approximate and are in 1996 dollars, Kovalick said, so costs may be higher when cleanup occurs. The point is that a large amount of money is going to be spent by the federal government, especially by the Corrective Action program. Thus, there is an ongoing market for new technologies. More than 80% of new technologies have been related to soil and groundwater.

The TIO tracks all kinds of technologies and their deployment, Kovalick explained. There are about 700 uses of source control treatment technologies in Superfund remedial actions. About 48% are established technologies, such as incineration and solidification/stabilization. About 52% are classified as innovative technologies, which include soil vapor extraction, thermal desorption, bioremediation, in situ flushing, soil washing, solvent extraction, and dechlorination. A participant inquired whether the solidification/stabilization category included capping and containment projects, and Kovalick replied in the negative, stating that the category only included cases in which additives were combined with the contaminated sediments to treat them.

The TIO tries to raise awareness of innovative technologies and assist with their deployment, Kovalick said. Action Teams may become involved with a technology after it is beyond the proof of concept stage and the bench-scale testing stage, during the period in which it still needs to undergo pilot-scale testing and demonstration. Other EPA programs may work with innovative technologies at other stages of their testing and implementation. For example, the Bioremediation Action Committee becomes involved with pilot-scale testing, the Superfund Innovative Technology Evaluation (SITE) Program becomes involved in the demonstration stage, and the PRP Risk Sharing Initiative can be used in the commercialization stage. Through the Risk Sharing Initiative, EPA co-insures truly innovative technologies, so that EPA absorbs 50% of the cost of a failed innovative remedy.

The RTDF concept was established when industry representatives, representatives of the Corps and other agencies, researchers, and environmentalists resolved to create a mechanism under which all parties could come together to do joint work to address complex remediation problems. The purpose of the RTDF is to identify problems for which more effective, less costly solutions need to be developed and studied in the field. The government, industry, and academia then work together, through partnerships, to identify specific priorities and needs and address them through collaborative projects. Each problem set is addressed by an Action Team, and the name Action Team denotes the desire for projects involving field work produce results, not just discussion.

RTDF Action Teams are formed when there is strong interest in investigating potential solutions in problem area. Action Teams have governmental and non-governmental co-chairs, and they work best when multiple companies and multiple agencies take part. Participants do not necessarily have to contribute funding, but usually contribute sweat equity. Good ideas often attract partners and funding, Kovalick has seen. And lessons learned from projects are useful to all involved, which is one reason why Action Team meetings are open to the public. Information about the RTDF and its Action Teams is available on the Web at http://www.rtdf.org.

Of the seven Action Teams, the best-known is the Lasagna^TM Partnership, which formed to address the problem of trichloroethylene (TCE, which cannot be removed from clay-like soils by traditional methods). Kovalick explained that participants suggested using pneumatic fracturing to put "pancakes" above and below a region of contamination, placing a charge on the two "pancakes" to draw the ions one way and the cations the other way, then reversing the charge to create action similar to that of a washing machine. Placing a layer of biologically active media between the "pancakes" would aid with decontamination. The work, which resulted in a licensed technology, involved more than 3 years of research. The technology is already being used to clean up a DOE site in Kentucky.

The Permeable Reactive Barriers Team focuses on trying to manage plumes of chlorinated solvents and metals in groundwater. Many remediation projects are using permeable reactive barriers, and the Action Team is collecting data about them and considering questions about longevity of the barriers and optimizing their design. In addition, the Action Team will be conducting a training course that it created on the design of permeable reactive barriers.

The In Situ Flushing Action Team is still trying to determine its focus, Kovalick stated. While there are many researchers and surfactant suppliers on the team, few representatives of industry have become involved so far. The Action Team is trying to determine what seems most promising in the area of flushing NAPLs with solvents and surfactants.

The Bioremediation Consortium is focused on three technology areas, which include anaerobic biodegradation of chlorinated solvents and natural attenuation. Its work is being conducted at Dover Air Force Base, a national test site where solvents have been added to already-polluted groundwater to facilitate taking measurements. A study in the fall issue of the journal Environmental Science & Technology (ES&T), will describe how microbes from Florida degraded the chlorinated solvents at the air force base to the required levels.

The IINERT Soil-Metals Action Team (IINERT stands for In-Place Inactivation and Natural Ecological Restoration Technologies) is discussing using plants and other additives (primarily minerals) to try to achieve appropriate fixation of lead in soil.

Kovalick continued by informing participants that the Phytoremediation of Organics Action Team has already begun some demonstration projects on the effects of poplar trees on groundwater, surficial petroleum contamination in grasses, and alternative vegetative caps. The Action Team has three subgroups, devoted to TPH in Soil, TCE in Groundwater, and Alternative Cover.

All of the RTDF Action Teams provide a forum for information exchange through meetings, conference calls, and the availability of minutes. Some Action Teams have produced demonstration summaries or bibliographies, and others aim to produce technical reports or articles in peer-reviewed journals. Action Teams sometimes have advisory roles, Kovalick said. In some instances, partners work together through informal cooperative working relationships. Other times, formal partnership agreements are utilized. Within two Action Teams, industry representatives have signed agreements with one another and signed Cooperative Research and Development Agreements (CRADAs) with EPA. CRADAs are a formal mechanism to preserve intellectual property rights. The Lasagna^TM Partnership utilized a CRADA, for example.

The Web pages of Action Teams may contain member lists, meeting minutes, and technical documents. Also, if participants want to work together on a document, a password-protected portion of the Web page can be used. The Sediments Remediation Action Team has not made use of that capability yet, but if it does, participants should remember that individuals can only submit comments if they use Netscape. Those who have Internet Explorer can read but not input items. If they want to submit comments, they can e-mail them to Carolyn Perroni, who will enter them. Perroni also must be contacted to remove items from the password-protected area.

In sum, Kovalick concluded, the Action Teams are client-driven, collaborative, engaged in on projects of mutual interest, reliant on group dynamics, open to the public, driven by information and documentation, and directed at field work. In addition, the Action Team co-chairs have the final say when Teams are making decisions, and the Teams do not have the power of advisory committees to EPA.

Kovalick said that a lot of information about site characterization and monitoring, as well as remediation, can be found at the Clu-In Web site at http:/clu-in.org. The TIO also circulates a monthly e-mail message called TechDirect, which announces new publications and reaches 5,000 people, including federal employees, state employees, and consultants. A publication called RTDF Update is also circulated to interested parties, and a new issue has just been mailed.

Coia observed that many Remedial Project Managers may not be familiar with the PRP Risk Sharing Initiative. Kovalick said that only three parties have taken advantage of the initiative so far, which may demonstrate lack of interest in it, but EPA is continuing to advertise it.

TECHNOLOGY FOCUS SESSIONS

Jensen explained that the goal of each technology focus session would be to identify joint projects. Moderators would set the stage for each session and lead the discussions. Technologies proposed should have broad applicability. Once technologies were selected, potential sites should be suggested.

POTENTIAL TREATMENT PROJECTS

Karen Miller, Naval Facilities Engineering Services Center

John Smith, Alcoa

Smith presented information about a PCB-contaminated Alcoa site in upstate New York on the Grasse River, where there might be opportunities to conduct capping, treatment, or enhanced natural attenuation demonstration projects. However, work on the site has not yet been approved by EPA Region II, and the Action Team cannot become involved until approval has been obtained.

The Grasse River flows from the Adirondack Mountains to the St. Lawrence Seaway, Smith said. Historically, PCBs were discharged into various outfalls, so sediments have been contaminated from the Alcoa site up to the confluence with the St. Lawrence Seaway, about 6 miles north of the site. The site is located on an old power canal, and a nearby dam was built to generate electricity for the plant operations. The river was dredged in the early 1900's and moves very slowly. Often, the St. Lawrence flows into the river.

Particle broadcasting is being proposed for the site. In 1994, sediment dredging was evaluated. One acre, containing 3,500 cubic yards of PCB-contaminated sediments, was removed. There were a lot of boulders present at the site, and there were significant undulations in the bedrock. Before the dredging occurred, boulders had to be removed. Smith explained that during dredging, three layers of silt curtains were used. While only 1 acre was dredged, water in a 2- to 3-acre area was treated. Following every 8 hours of treatment, water treatment had to be conducted for 24 hours before dredging could resume.

Dredged sediments were disposed of at an on-site TSCA/RCRA landfill, which was double-lined. The dredging of 1 acre resulted in the removal of about 25% of all the PCBs in the river. The average PCB concentration at all depths was originally 1,200 ppm, Smith stated. After dredging, the average PCB concentration was 75 ppm. Buried sediments were removed, but contaminated sediments in contact with the water column, fish, and the environment remained, so contaminant levels before and after dredging in the water column and in fish were unchanged. The cost for removing the 3,500 cubic yards of contaminated sediments was about $4.9 million, a significant portion of which was spent on water treatment. Now that dredging is complete, another treatability study is being proposed in preparation for the selection of a final remedy.

Investigations of the Grasse River have been conducted since 1991, according to Smith. During 1994 and 1995, the non-time critical removal action occurred. From 1995 to 1998, activities were conducted at the site to further reduce ongoing discharges to the river. During that time, supplemental remedial studies on the source and transport of PCBs were also conducted. After the conclusion of those studies, an effective feasibility study can be conducted. In 1999, as part of treatability testing, Alcoa is submitting a proposal to EPA for the use of particle broadcasting. Particle broadcasting was originally termed "enhanced natural recovery" and involves placing a thin, 6-inch layer of sediment over the contaminated sediment to enhance natural processes that occur. Thus, particle broadcasting is a non-traditional capping approach, as well as a type of enhanced natural attenuation.

Smith said that the objective of the supplemental remedial studies was to develop an understanding of the fate and transport of PCBs in the river. With that information, Alcoa hoped to be able to effectively evaluate various remedial options so that the one selected would be truly effective at reducing PCB levels in fish and the water column . The river has been well characterized, Smith stated. A tremendous amount of testing has been conducted, including studies of high-flow events, die studies, sediment coring, fish sampling, PCB flux analysis, and PCB congener analysis. A talented team, consisting of individuals from a variety of labs, assisted with the tests.

Buried sediments do not affect the water column if they are not stirred up. In the Grasse River, the higher concentrations of contaminants are right on the hardpan, Smith explained. The average PCB concentration in the top 3 inches of sediment is 20 to 25 ppm. The contamination occurs over a wide area, though PCB concentrations gradually increase as one moves downriver. To reduce PCB levels in the water column, all of the surficial sediments will have to be addressed, rather than just a few hot spots. Given the circumstances, Alcoa is considering capping via particle broadcasting. Particle broadcasting seems appropriate because surface sediments (rather than buried sediments) are the source of contamination, external sources of contamination have been reduced, the water column is the source of PCBs affecting fish, resuspension is not important, and diffusive flux from pore water (as opposed to groundwater flux) is low. Even in the event of a 100-year storm, diffusive flux is the primary mechanism affecting the site.

Natural recovery of about 10% per year is occurring due to approximately 0.4 cm of natural sedimentation per year, Smith said. Since natural recovery is not occurring rapidly enough, Alcoa has proposed to enhance it via particle broadcasting, using a thin cap of clean sediment about 6 inches thick and 1 mile long. (Dechlorination and degradation tests are also being conducted, but those processes are not being considered for the final remedy.) The thin cap is proposed as a treatability study, not a final remedy, the effectiveness of which would be evaluated in terms of the resulting reduction in PCB flux through the water column and fish.

Smith also said that Alcoa has conducted a predesign baseline characterization over the past 2 years. Final design is currently being analyzed, and implementation has been proposed to begin in 1999. Post-monitoring and a final evaluation will occur after the cap is placed.

The ultimate goal of particle broadcasting, according to Smith, is to translocate the zone of bioturbation out of the PCB-containing sediments. During bioturbation, researchers have observed an increased flux of PCBs into the water column. Placing a layer of sediment above the bioturbation zone provides a sorptive barrier to the upward mobility of PCBs coming through the surface. This procedure could represent a permanent remedy for sequestering PCBs due to the site-specific conditions, because there is no erosion in the river, minimal vertical migration of sediments, and the river is a net depositional area. It is anticipated that once a thin cap is placed, no PCBs should be observed in the water column until the area beyond the cap.

Assuming approval is secured this year, Alcoa is willing to work with Action Team members who want to evaluate phytoremediation or actual natural degradation processes (as compared to predictions generated by models) at the site. Alcoa would also be interested in sharing the information generated with all interested parties.

Smith stated that Quantitative Environmental Analysis, LLC, did some modeling of mono-, di-, tri-, and tetra-PCBs at steady state conditions, assuming no deposition, although 0.4 cm of natural deposition actually occurs each year. According to the model, 200 years into the future, only the mono-PCBs would break through the 6-inch cap, and the flux would be lower than currently observed. If 0.4 cm of deposition per year is assumed, no PCBs will ever break through the cap, according to the model. Since the cap could provide an effective remedy for the contamination, Alcoa would like to conduct a treatability study on a large enough scale to gather data that can be used for evaluation.

Smith explained that there are many pre-design considerations for the particle broadcasting study, such as cap configuration, what technique to employ, and what materials to use. Alcoa would also like to determine how the cap materials will distribute and whether sorting or consolidation will occur. Various pre-design studies are being conducted to fully characterize the site conditions before beginning to lay the cap. The material to be used in the cap is being tested in the lab, through column settling tests, for example. Smith said that Alcoa was addressing all of the considerations mentioned the previous day by Palermo. There is also a particle broadcasting technical review team, comprised of some of the country's foremost experts on the technology. Careful attention is being paid to plans for the actual placement of the cap, so that fish habitat is not adversely affected, for example.

Smith concluded by reiterating that all the work proposed for 1999 can only occur if approval is received from EPA Region II and concurrence is received from the New York State Department of Environmental Conservation and the St. Regis Mohawk Tribe is received. If the approval is received, Alcoa would be glad to entertain proposals from the Sediments Remediation Action Team to work on projects at the site. Jensen collected a list of participants who would like to be contacted about the possibility of pursuing projects at the Grasse River site.

John Davis suggested there might be a problem at the Grasse River site with indemnification in the event that actions occurred during a project that exacerbated the contamination problem. John George of Alcoa responded that the issue would have to be worked out beforehand.

Miller described a site in Puget Sound, in the state of Washington, where dredging is being conducted to remove about 500,000 cubic yards of contaminated sediments. The sediments, which are contaminated mostly with PAHs and PCBs, will be treated ex situ. The Naval Facilities Engineering Services Center is working with the Corps to produce a treatability study. The site might also be a potential demonstration site, but it will be a few months before it is known whether the Action Team might be able to become involved with it. Miller will update the group at the May meeting or beforehand, if information becomes available sooner.

Smith inquired whether any other participants could suggest potential sites for testing treatment technologies. For example, he said, there might be an in situ project involving adding zero valent iron or an ex situ project related to stabilization.

Jensen inquired whether participants were interested in pursuing the idea of using a CDF as a reactor. He suggested that the Seattle site described by Miller might have potential in this regard. The group responded to the inquiry affirmatively. Smith asked whether the group would like to begin an initiative to track ongoing sediment remediation projects and examine the treatment technologies utilized at each site. A participant reminded the group that Haggard had described a database project the previous day. The participant indicated that interested individuals might be able to collaborate with GE on the database, and he suggested that the database should not be limited to intrusive remedies but should also include remedies such as capping, particle broadcasting, and natural recovery, which tend not to show up in collective databases. Haggard responded that the GE database covers all the large sediment remediation sites. He stated that Smith might be suggesting a database that would closely analyze the specific technologies being developed, even if they have not yet been applied to large-scale projects. Smith agreed with Haggard's characterization and added that he is interested in tracking and evaluating technologies collaboratively, as they are developed.

Smith also stated that he would like individuals involved with dredging operations to bring information about their projects to the treatment subgroup, which could evaluate the effectiveness of the dredging according to a group-developed plan. Monitoring of such variables as fate and transport is often not conducted after dredging operations, Smith pointed out. Jensen suggested this might be a possibility for the Puget Sound site.

Wendy Davis-Hoover stated that the National Risk Management Research Laboratory at EPA is looking for a site contaminated with lead or other heavy metals. The laboratory would use such a site to test an organism that concentrates lead and turns it into lead phosphate.

POTENTIAL CAPPING PROJECTS

Norman Francingues, Jr., Waterways Experiment Station
Danny Reible, Louisiana State University

Francingues reminded participants that a controlled field site is required for field verification of capping effectiveness. He said that components of a demonstration study would include examination of precise layering issues and verification of the physical integrity of caps, which would be protected from extreme events. Reible stated that a small-scale field demonstration project would allow the Action Team to study some issues that determine capping effectiveness, including cap placement, cap maintenance, and invective components. A field demonstration might also assess certain treatment issues, such as that of a reactive CDF. As he talked, Reible showed several overheads, which are included as Attachment L.

He went on to list some questions that could be addressed through a demonstration project for different types of caps and treatments. These include chemical containment effectiveness, the fate of contaminants beneath the cap, placement problems, cap integrity problems, and potential monitoring approaches and capabilities. The project would not address the hydraulic stability of the cap, once it was placed.

One possible demonstration site would be a near-shore confined disposal facility. There might be invective components in a near-shore environment, but the use of a CDF would allow a great deal of control over the system, including homogenization of the sediments, which is important when comparing different approaches. Reible stated that the Grasse River site would allow many important questions to be investigated, but would not allow different treatments to be compared equally. Comparisons could be made between uncapped areas, conventionally capped areas, and areas utilizing various amended caps. Other in situ treatment options and the resulting biotransformation could also be evaluated in the CDF. The test site might be similar to the Dover Air Force Base test cells being analyzed by the RTDF group devoted to bioremediation, Reible suggested.

Rich Landis asked whether some of the potential treatment technologies should undergo further laboratory testing before money is spent on field testing. Francingues responded that quite a bit of laboratory work has been conducted already, though not as much testing of amended caps has occurred as might be desired. In any event, Francingues said, there would certainly be laboratory work to design an appropriate field project, and there would have to be a technical design basis for selecting the design of caps. Reible added that many studies have been conducted on the potential capabilities of conventional caps. For some of the basic cap configurations, there would be no need for significant lab research before beginning a demonstration project. However, many of the possible amendments, such as potential mixes in reactive caps, have not been used, and further concept development in the laboratory would be required before testing in the CDF, Reible said.

Landis inquired what technique would be used to measure the level of goodness . Francingues answered that the design for a monitoring program would be solicited from the Action Team so that the definition of "success" could be agreed upon. A research plan that included such elements would be peer-reviewed by the group before embarking on any project.

Apitz suggested that the group discuss contaminants of concern, since different contaminants behave differently. Apitz also advocated the use of highly contaminated sediments by the group, which would make it easier to make compelling statements about a cap's effectiveness. She said that the group could proceed in one of two ways: either choose the site and then match the technology to it, or choose the technology and then select an appropriate site. Apitz warned that many of the known sites are not highly contaminated, and tests of them might result in fuzzy data. Reible stated it might not be necessary to conduct tests on very highly contaminated sediments, but using less contaminated sediments would drive up the cost of the testing because higher analytical precision would be required and indicators or trace components might have to be analyzed to determine the behavior of contaminants. On the other hand, with some approaches (such as biotransformation), test results from highly contaminated sediments might not accurately represent what would occur in low concentration scenarios.

Francingues commented that there are two parts of a potential project: what elements are desirable and what elements are available. The latter is the crucial factor, he said. The tests will have to be adapted to the characteristics of available sites. Objectives will not be compromised, they will only be modified to be consistent to what is available.

John Childs discussed the Port of Portland as an example of the type of site that is desired. Maintenance dredging is conducted at the port, he explained. The contained aquatic disposal unit that was used historically has been shut down and is not currently available. The port is considering using a slip and a terminal instead, although this has not yet been proposed. The slip would be blocked off and used as a CDF. An adjacent slip is contaminated with PAHs. The sediments which would be put into the CDF would probably be contaminated with TBT, PCBs, PAHs, and DDT. Childs said that the Portland site could work as a demonstration site, although some problems might be encountered. Francingues agreed that the site probably has the desired characteristics, particularly if near-shore elevated groundwater is fluxing through the site. Normally, a lot of flux is not desirable, Francingues said, but if the site can be isolated, the flux would not present a problem. Francingues stressed that there are many features of any given site that would have to be analyzed.

Timberlake questioned how a demonstration would be carried out in the field, since the major concern with capping is the long-term effectiveness of caps isolating contaminants. Laboratory studies evaluating capping or testing models use thin caps to enable measurements of breakthrough within a reasonable period of time. He asked how long monitoring would be required at a field site. If tests were conducted for 5 or 10 years and the cap seemed to be effective, some individuals would undoubtedly inquire whether the cap will be effective for another 10 years.

Reible responded that a very long period of time would not be required, and it would not be necessary to wait until contaminants broke through the cap. If the cap can be placed without stirring up sediments significantly and there is no invective component, he said, then the ability of the cap to contain the contaminants should not be a concern. It would not be necessary to wait until contaminants penetrated the cap completely, but monitoring could be conducted by coring the cap. The duration of a project would probably be 3 to 5 years, Reible stated, although it would be valuable to retest the site after 10 years to demonstrate preliminary results. It would not be effective to measure concentrations of contaminants in the water because a number of different treatment approaches would be tested in close proximity to one another. Flux chambers could be utilized in some cells, but coring would be necessary in capped cells. Coring is probably desirable in any event, to enable an analysis of chemistry under the cap, including the fate processes of the contaminants.

Timberlake inquired whether it would be valuable to analyze existing capped sites. For example, the group could evaluate the results of projects as compared to their design, how well caps were laid down, and contaminant migration. Reible said that there would be value to studying existing sites, but that assessments could only be conducted on the limited universe of caps which have already been used, and most of those caps have been relatively simple. Also, such a study would not allow for analysis of different capping technologies employed in sites with identical characteristics. In short, Reible said, the study would be useful, but would not address the same issues as could be addressed by designing demonstration sites. Francingues added that demonstration projects could be designed to utilize cost-saving techniques, such as reducing the thickness of a cap, thereby augmenting available knowledge on possible ways to reduce the cost of caps. Smith observed that owners of capped sites do not always want the effectiveness of their caps assessed, in case they are shown to be less effective that predicted.

Smith went on to say that groundwater flux through caps could be a problem. He asked whether groundwater advective flux could cause a cap to fail in 5 years and how that would be evaluated. Reible responded that advective flux varies widely, and, at some sites, advective flux can be the dominant factor (rather than bioturbation or diffusive flux). There are techniques to assess advective flux, such as seepage meters and measurements of vertical heads combined with assessments of permeability. Reible said that he has worked largely on large lakes, where seepage is minimal, but it is a concern at other sites in more active environments. It would have to be assessed at every site, he concluded, and perhaps assessment techniques could be tested, as well.

David Petrovski observed that advection not only is site-dependent, but can also be time-dependent. At some sites, advection can dominate after precipitation events and then be dormant while diffusion is the primary active agent. Advection during periods of short duration can have a larger effect on contaminant transport than diffusion over much larger time scales. Thus, one of the difficulties with predicting cap performance is that the models are based on a constant groundwater velocity (which is usually an average over time), and the unusual instances in which advection dominates causes significant contaminant transport not accounted for in the models. Usually, however advection is related to hydraulic conductivity.

Reible inquired whether participants are interested in conducting a field demonstration of capping effectiveness as outlined by the group thus far and whether there were additional ideas about how to conduct the demonstration. Haggard stated that GE has been evaluating capping in a non-traditional sense of the word--capping not for isolation, stabilization, containment, and reduction of flux, but capping as a restoration technique to reestablish ecological regimes after an invasive remedy has been completed. He asked whether there were any large-scale evaluations of the ability to restore sub-aquatic systems in near-shore and deep areas--not just benthic invertebrates, but also macrophytes and communities of fish. Francingues commented that there is a great deal of interest in environmental restoration projects using the beneficial aspects of navigation dredging projects. WES is conducting a lot of work in that area, Francingues said, and he promised to provide Haggard with a list of individuals at WES involved in that work. Francingues also stated that there is worldwide interest in restoration, particularly in riverine systems. Examples include stream bank restoration and bioengineering, not only for stabilization, but also for habitat enhancement.

A list was compiled of individuals interested in being notified of discussions about projects that would provide field verification of capping effectiveness. Jensen reminded the group that possible venues for testing different technologies in a CDF included the Grasse River site and the slip at the Port of Portland. He said that the types of issues to be addressed in by demonstration project had been well defined, and that the group would narrow down the list of technologies to test in the future.

Haggard inquired whether the possibility of designing a project at the Grasse River site would be discussed by the treatment subgroup or the capping subgroup. Smith responded that the Grasse River site was presented under treatment, but that it has elements of capping, assessment, and natural attenuation. Jensen suggested that the organization of the Action Team's subgroups might change from being technology-specific to being site-specific after available sites had been determined. Smith also commented that the Grasse River project is not a capping project in the traditional sense of the word, nor is it natural attenuation left to occur without any interference except for monitoring. Assessment issues, of course, are part of treatment, capping, and natural attenuation projects.

Timberlake indicated that he would like the subgroups to be better defined and inquired how other RTDF Action Teams handle subgroups. A participant explained that subgroups usually have a core group, which is the steering team, as well as a larger group comprised of other interested parties. The steering team designs agendas for meetings and fleshes out questions raised. The structure of the Action Team's subgroups may change over time. Francingues added that some RTDF Action Teams have subgroups that operate independently of the whole Action Team. But this Action Team will not be ready to operate that way until the subgroups and the projects they will work on are better defined.

Perroni explained how conference calls are scheduled and how notifications are distributed. She said that the proceedings are summarized and posted on the RTDF Web site. One meeting participant asked whether each Action Team member could be notified of upcoming calls. Jensen and Perroni explained that it is necessary for individuals to express interest beforehand so EPA knows how many telephone lines to reserve for each conference call.

POTENTIAL NATURAL ATTENUATION PROJECTS

John Davis, Dow Chemical Company

Davis began his presentation (see Attachment M) by stating that he would describe how natural attenuation can be used as a remedial strategy for contaminated sediments. Individuals tend to have different perspectives on natural attenuation, he said. Some see it as a "do nothing" or "risk away" approach. Davis went on to discuss the state of science related to natural attenuation in groundwater, a science that has been developed over the last 10 years. The application of natural attenuation to contaminated soils and groundwater has been fairly well documented, he stated. The science of natural attenuation in sediments is understood conceptually, but the principles, methods, and practices have not yet been well developed. However, the principles and practices developed for groundwater may be used as a starting point in developing those for sediments.

Natural attenuation as a risk management strategy for groundwater relies on natural mechanisms to reduce the mass of dissolved contaminants, according to Davis. Levels must be reduced to below regulatory standards before the contamination plume reaches potential receptors. Often, continued monitoring may be the only required remedial action at sites where natural attenuation is the selected remedy. At many sites, natural attenuation supplements other remediation techniques, as part of a whole treatment train, rather than being the only remedy selected for the site. For example, natural attenuation may occur under a cap.

Natural attenuation has also been called "passive remediation" and "intrinsic remediation," Davis explained. EPA has defined it as relying on a series of physical, chemical, and biological processes that effectively reduce the contaminants' toxicity, mobility, or volume to levels that are protective of human health and the ecosystem. The processes include biodegradation, dispersion, dilution, sorption, and volatilization.

Davis listed a series of advantages to using natural attenuation: contaminants are ultimately transformed to innocuous by-products; it is a non-intrusive technology that allows continuing use of infrastructure; other aggressive remedial technologies can sometimes pose greater risk to potential receptors; and it can be less costly than currently available technologies. Davis stressed that natural attenuation is not a "no action" alternative. In order to prevent false claims of natural attenuation, regulators require demonstration, documentation, and quantification of natural attenuation. This involves site characterization, computer modeling, risk evaluation, and long-term monitoring.

Natural attenuation should be considered for sites where natural attenuation processes have been observed or are strongly expected, Davis said. It should also be considered if no human or ecological receptors are adversely impacted by contaminants. Additionally, it may be selected when alternative remediation technologies are not cost-effective or are technically impractical, and it is a desirable alternative when other remedial technologies would pose added risk by transferring contaminants to other environment media, spreading contamination, or disrupting adjacent ecosystems. The key factor is that natural attenuation should not be considered when it will not be protective of human health and the environment.

Davis explained that the evidence required to evaluate, implement, and document natural attenuation is: documented loss of contaminants at the field scale; contaminant, biochemical, and geochemical analytical data; and direct microbiological evidence. Biochemical and geochemical data may relate to topography, soil type, climate, subsurface geology, or hydrogeology and may have been gathered during the site- characterization process. Hydrogeological data of interest include hydraulic conductivity, permeability, gradient, porosity, flow direction, and dispersion and sorption. However, Davis said, some biochemical and geochemical data required to support natural attenuation are not often collected during the course of site characterization, but could be collected at that time if site managers expect to consider natural attenuation as a remedy. (Collecting it at the outset is generally more cost-effective.) The list of such data includes: hydrogen and dissolved oxygen in groundwater, as well as sulfate and sulfide, ammonia, chloride, certain metals, nitrate and nitrite, VOCs and semi-VOCs, volatile fatty acids, methane, ethene, ethane, propane, propene, TOC/BOD/COD/TPH, pH, temperature, and redox potential.

Loss of contaminants can be documented by historical trends, contaminant concentration distribution, hydrogeologic evidence showing a reduction in the total mass of contaminants, or degradation products. Geochemical and biochemical indicators may show changes in the concentrations and distributions of contaminants over time. In general, Davis stated, a database developed over 2 or 3 years supports a case for using natural attenuation better than data collected once or twice. Microbiological evidence usually involves laboratory studies. When chlorinated solvents are involved, it is necessary to demonstrate that potential biodegradation is actually occurring, which also indicates the kinetics of the reactions.

A stepwise framework has been proposed to evaluate whether natural attenuation is occurring in groundwater. The framework involves identifying and collecting additional data that support the three lines of evidence for natural attenuation. It also involves integrating natural attenuation into a long-term site remediation/management strategy. Davis suggested that data collection can be conducted concurrent with other investigations and remediation planning activities. He added that the Bioremediation Consortium has developed a nine-stage process detailing required steps for demonstrating natural attenuation. Information about the Consortium's activities can be found on the web at http://www.rtdf.org/bioremed.htm.

Davis presented the following list of issues and questions that might be addressed by interested members of the Sediments Remediation Action Team:

• Is the definition of natural attenuation for sediments the same as that for groundwater?
• When should natural attenuation be considered for sediments?
• What are the dominant natural processes occurring in sediments that impact application of natural attenuation?
• What evidence is required to evaluate and implement natural attenuation in sediments?
• How are lines of evidence documented?
• What are the principles and practices for implementing natural attenuation in sediments?

Michael Montgomery, who has done significant work on quantifying natural attenuation of PAHs in sediments at the Naval Research Laboratory, showed several slides (see Attachment N) indicating some of the results of his research. His laboratory has looked at parameters used for examining carbon flows through marine systems. Two parameters of particular interest, Montgomery said, are bacterial production and an average of how fast the heterotrophic bacterial assemblage is growing, which is compared to total PAHs in the sediment. At low PAH concentrations, there is a wide variety of production values because a lot of factors affect how quickly bacteria metabolize organic carbon. At high PAH concentrations, there is low bacterial metabolism. PAH concentrations high enough to suppress total bacterial metabolism generate concern. When bacterial production was compared to PAH mineralization in tests, areas of low bacterial production showed a high rate of degradation of PAHs.

Apitz inquired whether PAH degraders were directly measured or whether they were inferred from PAH mineralization. Montgomery replied that only mineralization was examined. At some stations, naphthalene was added to sediment samples. Some stations showed a decrease in production proportional to the amount of naphthalene present, suggesting that bacteria were not often pre-exposed to naphthalene. At other stations where there was chronic exposure to naphthalene, an inhibition of bacterial production with increased naphthalene addition was not observed, suggesting that assemblages can adapt to naphthalene.

Davis stated that the journal Bioremediation is publishing a series of articles, beginning this month, on the state of the art in natural attenuation. In terms of work the Action Team might do on natural attenuation, he stated that it would probably be best to integrate assessments of natural attenuation into projects that have already been proposed. (For example, the Action Team could conduct a comparison of natural attenuation to natural processes occurring after the placement of a cap). EPA's Great Lakes National Program Office (GLNPO) is interested in analyzing how to deal with low contaminant levels in sediments in a wide area. Natural attenuation might be applied in some of those cases, Davis said.

Coia pointed out that one difference between natural attenuation in groundwater and natural attenuation in sediments is that in sediments, plants play a role, in addition to bacteria. He suggested that the Action Team might want to try to quantify the effects of the plants, in addition to those of the bacteria. Davis added that in groundwater, there are non-biological factors affecting natural attenuation, such as absorption. He said that he is also particularly interested in the role played by immobilization in the natural attenuation process.

Steve Mangion observed that an Office of Solid Waste and Emergency Response (OSWER) directive about natural attenuation in soils and groundwater can be found on EPA's Web site. He said that there is also a guidance document available (called something like Technical Protocol for Evaluating Natural Attenuation on Chlorinated Solvents), but that it is geared towards biological processes, rather than physical or chemical processes. The Agency is still wrestling with some non-biological aspects of natural attenuation, according to Mangion, and it is also working on guidance for site characterization requirements for natural attenuation. It is also trying to define the length required for long-term monitoring of natural attenuation.

Mangion said that natural attenuation of sediments is an especially difficult issue for EPA because the Agency has not yet thought a lot about how to characterize sediments to assess problems and how to conduct long-term monitoring of any remedial approaches. Also, the Agency expects that significant amounts of source control and source removal will occur when groundwater is contaminated. Davis commented that Mangion's important point about source control related to his own explanation that natural attenuation is usually not utilized in isolation. If there is still a source of contamination, it has to be removed, because the goal is to have no negative impact on human health or the environment. In addition, the concept of natural attenuation has become acceptable in a way that it was not 10 years ago.

Stahl said that the Society of Environmental Toxicology and Chemistry (SETAC) meeting in San Francisco, California, devoted an entire session to natural attenuation, which was very well attended. The meeting led to two articles, one of which was published in the SETAC newsletter and was about the need to consider natural attenuation at the beginning of the risk assessment process. In addition, Stahl said that he and Montgomery are editing a book for SETAC Press on the topic of natural attenuation, which is due out by November.

Hap Pritchard suggested that the group take advantage of Smith's concept of using a 6-inch cap, which would allow some fairly complicated estimates of natural attenuation. The 6-inch cap allows the bioturbating organisms to move to the surface and separate themselves from the contaminants, creating a distance through which contaminants must travel to reach potential receptors. If transport occurred much less quickly than expected, Pritchard said, it would be logical to study whether biodegradation was occurring. Placing a cap on contaminated sediments makes natural attenuation somewhat easier to study.

Pritchard said that, in riverine situations with PAH-contaminated sediments, there will be continual deposition of contaminated sediment on top of the cap after it is placed. In such situations, researchers must analyze whether the accumulation of new contaminated sediments over time is being remediated by existing microorganisms. A participant observed that, based on his understanding of Smith's presentation, such an analysis of natural attenuation would be appropriate at the Grasse River site.

Haggard stated that a fundamental understanding of the processes affecting long-term and short-term fate and transport and bioaccumulation issues allows sophisticated and innovative remedies to be designed, such as methods to enhance natural attenuation. He added that researchers should apply the same rigor to assessing all potential site remedies that they use to assess whether natural attenuation is an appropriate remedy at a site, either alone or coupled with other remedies. The point, Haggard said, is that people are often more aware of the limitations of natural attenuation than the limitations of other remedial approaches. Understanding the processes affecting sediments will facilitate the investigation of an informed answer to the question of whether any other remedial approach can bring better results than natural attenuation. Davis concurred and added that understanding the natural processes involved would facilitate the consideration of methods to potentially accelerate natural attenuation.

Jensen said that he recently read a paper describing work conducted in Seattle that involved a near-shore CDF and a berm. The berm had CDF material on one side of it and tidal water on the other side, and tidal action was causing cyclical movement of seawater in and out of the berm. As the seawater entered the berm, it brought oxygen, sulfates, and nitrates with it. A chemical-microbial reaction was occurring in the berm, making sulfide available, and the sulfide was immobilizing metals. The question posed in the paper was whether the CDF was seeping metals into the environment, and the answer given was that the cyclical tidal action was causing the berm to act as a precipitator for metals, taking them out of solution. Jensen commented that the situation is a serendipitous variation on natural attenuation. He stated that the paper made him wonder whether tidal action might also aid with the immobilization of metals such as arsenic or lead.

Finkelstein stated his belief that a site where fish have been contaminated for 20 years is not a candidate for natural attenuation. He said that he does not want to see the Action Team advocating a study of the potential application of natural attenuation in sediments at such a site. Finkelstein agreed that site conditions can be changed to enhance natural attenuation in some cases; however, he said that trying to come up with a natural attenuation remedy for a site that has been badly contaminated for years seems like an excuse to avoid the necessity of a more costly remedy. Davis emphasized that no one was trying to portray natural attenuation as a panacea, but only as one tool in a toolbox that ought to be better understood. More investigation is warranted of when and where natural attenuation can be applied and how it can be quantified.

Petrovski commented that although the groundwater paradigm for natural attenuation is useful when considering sediments, there is one significant difference. In the case of groundwater, it can safely be assumed that an aquifer will always be there and the setting will not change over time. With sediments, there is always a risk that a catastrophic storm event might flush all of the in situ contaminated materials out to a larger body of water and cause enhanced problems. Thus, in considering risk, scientists must consider not only the risk associated with current in situ conditions, but also the potential for contaminated materials to cause problems in any larger body of water where they might end up after a storm event.

George echoed Davis's response to Finkelstein -- that natural attenuation is one risk management strategy in an arsenal of potential strategies. Natural attenuation should be considered when modeling which contaminants are taken up by organisms. But the models, as well as the bioavailable components of contamination, need to be understood. With respect to the Grasse River, George said, evidence of both degradation and dechlorination has been observed. One of the interesting aspects of PCB migration through a cover material is that a situation might occur in which the lower chlorinated congeners actually emerge through the system because they are more amenable to biological action. The particle broadcasting approach being considered is not necessarily designed to address the biological component, as opposed to the physical component, of PCB flux from sediments to the water column.

Timberlake suggested that the Action Team should better specify its definition of natural attenuation with respect to sediments and outline some of the related work that needs to be done to answer questions about natural attenuation in sediments. Davis agreed and added that he thinks it would be desirable to integrate the work Timberlake described with a test project. Often, it is easier to define and address questions and answers when one is working with a specific example, Davis said. Another participant stated that the concept of natural attenuation was proposed in EPA's contaminant management strategy, issued in April 1998. The term was used, but it was not defined.

Jensen agreed that developing a document containing some definitions might be a good first step. He said that it might be drafted in about 3 months, and no research would be required to prepare it. A subset of the Action Team could get together to define what aspects of natural attenuation it would like to address. He explained that he hoped a draft could be ready by the next meeting of the Action Team, in May. As the Action Team considered other field projects, the natural attenuation opportunities available would also be discussed. Jensen said that participation and input of those skeptical about natural attenuation would strengthen any work the Action Team produces related to natural attenuation.

George stated that one of the benefits of the Action Team is that it brings together individuals involved in different aspects of sediment management strategy. He went on to mention an additional ad hoc industry group, called the Sediment Management Work Group (SMWG). The SMWG plans to develop some key technical papers on a number of issues, including natural attenuation. It would be willing to collaborate with the Action Team in any way that the Action Team would consider appropriate, George said. Davis supported Jensen's idea for the development of a document over the next 3 months. A list of individuals who wished to be advised about upcoming discussions of natural attenuation was compiled.

POTENTIAL ASSESSMENT PROJECTS

Ralph Stahl, DuPont Corporate Remediation

Stahl explained that, at the previous meeting in Cincinnati, the assessment subgroup decided to develop approaches for sediment assessment. The subgroup would like to address risk assessment and whether remedial actions are effective. It decided to begin a demonstration project and is anxious to settle on a site. In the meantime, the subgroup also discussed using a virtual site. About a year and a half ago, he explained, the EPA Superfund office and the American Industrial Health Council (AIHC) collaborated on a risk assessment project, working on a number of difficult issue areas through the development of white papers and, in some cases, the assessment of made-up data. The assessment group might base its virtual project on the EPA/AIHC project. If an actual site were available, individuals could collaborate on white papers about areas of interest. The subgroup hopes to continue sharing information about activities in which its members engage and might also be able to conduct workshops or design learning tools, Stahl said.

Stahl said that the Action Team should discuss approaches for a demonstration project. One method would be to utilize the EPA's Ecological Risk Assessment Framework to consider what issues are encountered when assessments of contaminated sediments are conducted. The Risk Assessment Framework involves problem formulation (including goal setting, establishing measures, and developing a conceptual model), analysis of effects and exposure, and risk characterization. Another possibility would be to develop a series of learning tools, such as assessment games based on actual or imagined data sets.

Stahl said that potential white paper topics that might come out the subgroup's work include:

• Evaluating the condition of a background or reference area
• Sediment sampling and analytical issues
• Measures of success, such as mass removal and ecological improvement
• Equilibrium partitioning-based sediment criteria
• Effects-based sediment criteria
• Applications of sediment toxicity testing
• Ecological assessment tools
• Bioavailability
• In situ test methods
• Conceptual model development
• Determining sediment fate and transport
• Food chain risk assessment techniques
• Monitoring remedial effectiveness

Stahl suggested that the white papers might be 1, 2, or 3 pages long and would enable individuals to share a common understanding of certain subjects. They could also be very useful to consideration of potential sites.

George stated that when the word "assessment" is used, people tend to think it relates only to the front end of a problem. He stressed that it is necessary to look at assessment in the broadest possible context and to consider not only the front-end characterization of a site and definition of remedial objectives, but also the back-end evaluation of whether remedial activities have accomplished the desired result. Stahl agreed that it is important to consider restoration, (e.g. to consider the success of restoration as measured by improvement to the benthos over time, as well as mass removal or changes in contaminant concentration).

Jim Clark stated that the subgroup had agreed on the importance of discussing assessment ideas on a site-specific basis. He said that any of the sites already discussed by the Action Team would serve the subgroup well. He added that Jim Keating has a list of eight or ten projects in which EPA is looking for participation, and the subgroup could become involved in some of those, independent of the use of a virtual demonstration project or becoming involved with any sites where the Action Team might conduct a demonstration project.

Jensen commented that there is opportunity for engineering new tools to aid with assessment, similar to a benthic flux chamber. He said that he has personally become interested in real-time pore water analysis, a technology developed by Sam Kounaves at Tufts University to test real-time concentrations of metals in water. Kounaves has developed a somewhat delicate probe, according to Jensen, but a more robust probe would be required to probe deep sediments.

Stahl then asked whether participants were interested in pursuing some of the projects he had suggested. About a dozen people concurred that the development of white papers would be a valuable activity for the subgroup. Stahl also asked for feedback about the idea of developing a game similar to the Ecochallenge game developed by industry a number of years ago. Ecochallenge was targeted at individuals in industry who suspected that ecological risk assessment was an arbitrary science. Another purpose of the game was to bring industry representatives together with the regulatory community. Stahl suggested that a new game could involve looking at a data set and deciding how to tackle it. An imaginary sum of funding would be available. The game would also serve to illuminate what people think is valuable and invaluable, as indicated by how they spend their imaginary money. Action Team members could use the game to enhance learning among their clients and within the organizations for which they work. Stahl asked the Action Team whether it would support trying to develop a game for sediments, and the group responded affirmatively. Stahl's overheads are included as Attachment O.

Kounaves then described his research on in situ analysis of heavy metals in water using a cone penetrometer-mounted microsensor (see Attachment P). Kounaves said that his research is devoted to ways to monitor chemistries in situ. He explained that the cone penetrometer-mounted sensors were originally developed with support from EPA through the Northeast Hazardous Substances Research Center. EPA's goal was to develop a rapid screening technology to use in the field to monitor heavy metals in groundwater. The sensors have the ability to monitor about 30 different metals, but each metal requires a different transduction scheme to be designed for the sensors. Arsenic, chromium, selenium, and zinc have been tested by the sensors in the lab, but not yet in the field. An article about a use of the technology to evaluate copper, cadmium, and lead was published in the January issue of the journal ES&T.

The sensor, which is a solid-state sensor, was developed in collaboration with IBM Watson Research Center and Orion Research, Inc. The sensors are only a couple of cubic millimeters and are very rugged, Kounaves said. The basis of the sensing mechanism is a preconcentration technique in which metals are plated out of solution at a strongly negative potential and then stripped out going towards a positive potential. As they are stripped out, a peak is obtained at the redox potential of the metal. A measurement can be taken in a few minutes.

Kounaves explained that his initial concept was for a 2 inch probe that could be dropped down wells. The finished system is precalibrated in the laboratory and then taken out in the field and run off a laptop computer. In the initial demonstration, which was conducted in lead-contaminated wells at Hanscom Air Force Base, there were two different peaks of redox potential of lead. However, all the samples were cross-checked by another laboratory, which demonstrated that the sensors were effective at indicating the order of magnitude of contamination. Kounaves said that the system is being used to analyze heavy metals and other electrochemical parameters in Martian soil and is being considered to test metal levels in blood.

Kounaves proposed that the Action Team charge his laboratory to design a probe to use in sediments, which the Action Team could then demonstrate at several sites. The data from the sensors could be compared to data tested by another means, and the metals level indicated by the probe could be compared to total metals and bioavailable metals at a site. The probe, which would have sensors protected by a porous membrane on the outside, would simply be placed in the ground to take measurements. A company called Fugro Geosciences already manufactures probes that might be used.

Potential partners in a project include Fugro Geosciences; Orion Research, which has the ability to add other sensors to the probe; Environmental Technology Group (ETG) in Baltimore, Maryland, which has developed a membrane technology which solutions can pass through; and IBM Watson, which has developed all the sensors for Kounaves's group and could develop sensors for additional metals, if desired. Kounaves's group will receive financial support from EPA later this year to conduct more field demonstrations and evaluations of the technology at a few sites in New England. In sum, the objective of any project the Action Team pursues would be to perform on-site rapid screenings of specific metals, which would extend the work conducted for EPA, and apply the technology to several sites. It would probably take 12 to 18 months to prepare for site demonstrations.

Apitz commented that the Navy has done a lot of work with site characterization and analysis penetrometer systems (SCAPS). These projects usually use the weight of a truck to push a probe 50 meters into soil, she said, but a truck usually cannot be used on marine sediments. Fortunately, probes only need to be pushed several meters into sediments. Probes might be deployed by divers in some circumstances. In short, people ought to remember that the deployment of probes becomes a difficult engineering problem for sediments. Kounaves responded that sediment density affects how hard it is to get the probes to the desired depth. Apitz added that sediment resistance and the depth in question determine the difficulty of the problem.

A participant inquired about the detection limits that might be possible for the secondary set of metals, such as arsenic and chromium. Kounaves replied that detection limits vary by metal and also depend on the length of preconcentration. Most levels can be tested to parts per billion or sub-parts per billion. Lead, cadmium, and copper can be tested on solid substrates. Zinc, as well as some other metals, requires mercury preactivation of the sensors. Analyzing for higher metal levels -- in the tens, hundreds, or thousands of parts per billion -- is easier than analyzing for lower levels. Kounaves said that any demonstration project involving the Action Team would probably analyze metals that are relatively simple to test for, such as cadmium and lead. Ionic mercury could also be tested. Analysis is best conducted by running a 100 second test in situ, then removing the probe and placing it in a buffer solution, and then conducting an analysis of the results.

Stahl inquired whether the Action Team is requesting a proposal from Kounaves that includes cost figures or is proposing another action item. Jensen suggested assembling a list of those interested in pursuing the problem, which involves mechanical engineering as well as analytical work, over the next year or so. Interested individuals could pursue the issue in upcoming conference calls. Such a list was compiled.

Finkelstein proposed the Industroplex site, which is about 2 miles from Tufts, as a potential demonstration site. Metals are a problem in the sediments at the site, he said, and studies of the site commence this spring. Bruce Yare, who was formerly involved with the RTDF, is a consultant at the site and could be contacted.

Stahl stated that the group interested in working on the probe will not be comprised of all of the same individuals who have expressed interest in participating in future discussions of assessment. Individuals who did not attend the assessment subgroup meeting in Cincinnati and would like to be involved in future assessment discussions should contact Stahl.

REFLECTIONS/PATH FORWARD

Richard Jensen, DuPont Corporate Remediation
Dennis Timberlake, EPA

Jensen invited comments on any issue or on potential improvements to the process utilized by the Action Team. Clark requested clarification of the date of the next Action Team meeting. Reible stated that scheduling had not been finalized; for example, subgroups may want to meet on Monday afternoon, May 24, before the full group meets on Tuesday, May 25. The final details of the schedule will be worked out well before the meeting, he said. Stahl stated that he would like to see more individuals from NOAA, the Fish and Wildlife Service, and EPA participating in the Action Team.

Timberlake concluded the meeting by stating that he was very pleased with the turnout, the organizations represented, and the comments offered by participants. Finally, he thanked participants for their support and assistance moving projects forward.

Attachments

Attachment A: Final Attendee List

Attachment B: Opening Remarks (Richard Jensen)

Attachment C: New York/New Jersey Harbor Decontamination Demonstration Project (Keith Jones)

Attachment D: Contaminated Sediment Sites Database (John Haggard)

Attachment E: Sediment Recovery Demonstration Project under the Clean Water Action Plan (Jim Keating)

Attachment F: Upcoming Meetings (Danny Reible)

Attachment G: Review of Research and Development at Duluth Lab (Dave Mount)

Attachment H: Potential Applications of Phytoremediation (Michael Coia)

Attachment I: Focus on Capping (Mike Palermo)

Attachment J: Focus on Capping (Norman Francingues, Jr.)

Attachment K: RTDF Overview (Walter Kovalick, Jr.)

Attachment L: Potential Capping Projects, (Danny Reible)

Attachment M: Potential Natural Attenuation Projects (John Davis)

Attachment N: Natural Attenuation of PAHs in Sediments (Michael Montgomery)

Attachment O: Potential Assessment Projects (Ralph Stahl)

Attachment P: In Situ Analysis of Heavy Metals in Pore Water Using a Cone Penetrometer-Mounted Microsensor (Samuel Kounaves)