«This page left intentionally blank. United States Environmental Protection Agency EPA-540-R-05-012 Office of Solid Waste and Emergency Response OSWER ...»
Highly chlorinated congeners of PCBs may gradually partially dechlorinate naturally in anaerobic sediment (Brown et al. 1987, Abramowicz and Olsen 1995, Bedard and May 1996). In general, lesschlorinated PCBs bioaccumulate less than the highly chlorinated congeners, but are more soluble and, therefore, more readily transported into and within the water column than highly chlorinated PCBs. The less chlorinated PCBs exhibit significantly less potential human carcinogenic and dioxin-like (coplanar structure) toxicity (Abramowicz and Olsen 1995, Safe 1992), but may be transformed in humans into forms with potential for other toxicity (Bolger 1993).
Aerobic processes may then biodegrade the less chlorinated PCB congeners (Flanagan and May 1993, Harkness et al. 1993). The sediment concentrations of other chemicals and the total organic content tend to control these processes. However, little evidence exists that lower chlorinated congeners under the anaerobic or anoxic conditions found in most sediment are significantly transformed. Therefore, these partially dechlorinated organics tend to accumulate and persist (U.S. EPA 1996d, Harkness et al. 1993).
Although desirable, it is unclear whether biologically mediated dechlorination of PCBs would be effective in achieving remedial objectives in a reasonable time frame and may result in the production of more toxic byproducts.
4-8 Chapter 4: Monitored Natural Recovery
4.4 EVALUATION OF NATURAL RECOVERYAn evaluation of MNR as a potential remedy or remedy component should generally focus on
considering, at a minimum, the following questions:
• Is there evidence that the system is recovering?
• Why is the system recovering or not recovering?
• What is the pattern of recovery or non-recovery expected in the future?
This evaluation should be supported with a variety of types of site-specific characterization data and, often, modeling. The lines of evidence approach for evaluation of natural attenuation of contaminants in soil and ground water can provide a general framework for evaluating MNR in sediment (e.g., U.S. EPA 1999d). Swindoll and his colleagues include a chapter on natural remediation of sediment that presents a useful summary discussion (Swindoll et al. 2000). EPA’s Office of Research and Development (ORD) is in the process of drafting a technical resource document specifically for MNR in sediments and may also include suggested protocols. In addition, members of the joint industry–EPA Sediments Action Team of the Remedial Technologies Development Forum (RTDF) has developed a series of working papers on MNR that can be found at http://www.rtdf.org/public/sediment/mnrpapers.htm (Davis et al. 2003, Dekker et al. 2003, Erickson et al. 2003, Magar et al. 2003, Patmont et al. 2003).
As with the evaluation of any sediment alternative, an evaluation of MNR should be generally based on a thorough conceptual site model that includes current and future pathways of human and ecological exposure to the contaminants. This conceptual understanding should be based on site-specific data collected over a number of years and, for factors known to fluctuate seasonally, data collected during different seasons. Lines of evidence that can be used to construct a plausible case for the use of MNR include those listed in Highlight 4-4. It is important to note that not all lines of evidence or types of information are appropriate at every site, but, generally, multiple lines of evidence are needed. Project managers should be aware that a substantial spacial and temporal record may be useful to establish a reliable trend, especially for surface sediment data, which typically vary widely.
Highlight 4-4: Potential Lines of Evidence of Monitored Natural Recovery • Long-term decreasing trend of contaminant levels in higher trophic level biota (e.g., piscivorous fish) • Long-term decreasing trend of water column contaminant concentrations averaged over a typical low-flow period of high biological activity (e.g., trend of summer low flow concentrations) • Sediment core data demonstrating a decreasing trend in historical surface contaminant concentrations through time • Long-term decreasing trends of surface sediment contaminant concentration, sediment toxicity, or contaminant mass within the sediment
Examples of types of site-specific information that could be collected to support the lines of evidence
listed in Highlight 4-4 include the following:
• Identification and characterization of ongoing sources of contamination;
• Evaluation of historical and current contaminant levels in biota and surface water;
• Evaluation of geomorphology, long-term accretion, and erosion;
• Evaluation of sequestration mechanisms (e.g., sorption, precipitation) and rates of degradation or transformation;
• Development of a tool to allow prediction of future recovery and risk reduction (e.g., sediment and contaminant fate and transport modeling).
The amount of physical, biological, and chemical process information needed to assess the applicability of MNR adequately is site specific. An important step in documenting the potential for MNR as a management alternative normally is to show observed reductions in exposure and risk can be reasonably expected to continue into the future. In systems where the mechanisms causing the recovery are uncertain, or where the fate and transport processes driving recovery may be complex and changing with time, simple extrapolation of historical trends may not be appropriate. In such cases, a wellconstructed model can be a useful tool for predicting future behavior of the system. The use of models is discussed further in Chapter 2, Section 2.9 Modeling.
Integration of the data quality objective (DQO) process with risk evaluation can help identify which natural processes are most critical to the evaluation of MNR at a site. Generally, the identification of MNR data needs and preparation of study design can be structured similarly to the DQO process (U.S.
EPA 2000a) that is normally integrated within the remedial investigation and feasibility study (RI/FS).
The DQO process is discussed in greater detail in Chapter 2, Section 2.1.1.
4-10 Chapter 4: Monitored Natural Recovery
4.5 ENHANCED NATURAL RECOVERYIn some areas, natural recovery may appear to be the most appropriate remedy, yet the rate of sedimentation or other natural processes is insufficient to reduce risks within an acceptable time frame.
Where this is the case, project managers may consider accelerating the recovery process by engineering means, for example by the addition of a thin layer of clean sediment. This approach is sometimes referred to as “thin-layer placement” or “particle broadcasting.” Thin-layer placement normally accelerates natural recovery by adding a layer of clean sediment over contaminated sediment. The acceleration can occur through several processes, including increased dilution through bioturbation of clean sediment mixed with underlying contaminants. Thin-layer placement is typically different than the isolation caps discussed in Chapter 5, In-situ Capping, because it is not designed to provide long-term isolation of contaminants from benthic organisms. While thickness of an isolation cap can range up to several feet, the thickness of the material used in thin layer placement could be as little as a few inches. The grain size and organic carbon content of the clean sediment to be used for thin-layer placement should be carefully considered in consultation with aquatic biologists. In most cases, natural materials (as opposed to manufactured materials) approximating common substrates found in the area should be used. Clean sediment can be placed in a uniform thin layer over the contaminated area or it can be placed in berms or windrows, allowing natural sediment transport processes to distribute the clean sediment to the desired areas.
Project managers might also consider the addition of flow control structures to enhance deposition in certain areas of a site. Enhancement or inception of contaminant degradation through additives might also be considered to speed up natural recovery. However, when evaluating the feasibility of these approaches, project managers should consult state and federal water programs regarding the introduction of clean sediment or additives to the water body. For example, in some areas, potentially erodible clean sediment already is a major nonpoint source pollution problem, especially in areas near sensitive environments such as those with significant subaquatic vegetation or shellfish beds.
4.6 ADDITIONAL CONSIDERATIONS
MNR is likely to be effective most quickly in depositional environments after source control actions and active remediation of any high risk sediment have been completed. Where external sources were controlled many years previously and no discernable high risk sediment areas can be identified, yet site risks remain unacceptable, it may be questionable whether natural processes alone will reduce risks satisfactorily in the future. At these sites, it can be especially important to evaluate the effectiveness of previous source control actions and to evaluate potential additional active sediment source control or remediation methods for selected areas. For MNR, as for other sediment remedies, effective source control is often critical to reaching remedial objectives in a reasonable time frame and to preventing recontamination.
As discussed in Chapter 7, Remedy Selection Considerations, when evaluating MNR, the shortterm effects on human health and the environment during the recovery period (i.e., the baseline risks for the site) should be compared to the short-term effects of other approaches such as effects of resuspension of contaminants due to dredging and habitat changes caused by capping. Section 7.3, Considering Remedies, discusses the process of comparing short-term and long-term risks associated with various approaches in a net comparative risk analysis.
4-11 Chapter 4: Monitored Natural Recovery In most cases, the long-term effectiveness of MNR is dependent on the dynamic processes of mixing and burial over time remaining dominant over sediment resuspension or contaminant movement via advective flow or other mechanisms. Assessment of sediment and contaminant fate and transport are, therefore, very important at most sites. Some potential mechanisms for physical disruption of overlying cleaner sediment, such as keel drag or pipeline construction, may be amenable to human management controls. Others mechanisms for physical disruption, such as ice scour or flooding, may be only partly manageable or not manageable. The importance of contaminant movement through overlying sediment to surficial sediment and the overlying water can depend on several factors, including the chemical characteristics of the contaminant, physical characteristics of the sediment, and patterns of ground water flow. These issues can also be of concern for in-situ capping and are discussed further in Chapter 2, Section 2.8, Sediment and Contaminant Fate and Transport, in Chapter 5, In-Situ Capping, and in the U.S.
Army Corps of Engineers (USACE) Technical Note, Subaqueous Capping and Natural Recovery:
Understanding the Hydrogeologic Setting at Contaminated Sediment Sites (Winter 2002). In general, the presence of processes, such as erosion or ground water flow, that cause release of contamination to the water column should not eliminate consideration of MNR as a remedy; instead, they should lead to evaluation of the consequences of those processes on exposure and risk.
Generally, regions should consider using MNR either in conjunction with source control or active sediment remediation or as a follow-up measure to an active remedy. For example, MNR may be an appropriate approach for some sediment sites after control of floodplain soils and NAPL seeps. At other sites, MNR may be an appropriate approach to control risk from areas of wide-spread, low-level sediment contamination, following dredging or capping of more highly-contaminated areas. MNR may also be an appropriate measure to reduce residual risk from dredging or excavation in cases where the active cleanup is not expected to achieve risk-based measures alone.
When considering the use of MNR as a follow-up measure, project managers should consider the change in conditions caused by the active remedy. As noted by the SAB (U.S. EPA 2001j): “If MNA [or, as used in this guidance, MNR] is to be considered after a remedial action (e.g., the removal of heavily contaminated portions or capping), the effects of the remedial action on the chemistry, biology, and physics of contaminated sediments should be evaluated. The effects include: 1) potential disturbances on reaction conditions and aquatic life when dredging is used, and 2) changes on reaction conditions and mass transfer in the sediment and at the sediment/water interface when capping is used.” MNR should be considered when it would meet remedial objectives within a time frame that is reasonable compared to active remedies. However, the Agency recognizes that MNR may take longer to reach cleanup levels in sediment than dredging or in-situ capping and, therefore, may take longer to reach all remedial action objectives, such as contaminant reductions in fish. It is important to compare time frames on as accurate a basis as possible, including for example, accurate assessments of time for design and implementation of dredging or capping and realistic assumptions concerning dredging residuals.
Where possible, estimates of the uncertainty in the recovery time frame associated with each alternative should also be made. Factors that the project manager should consider in determining whether the time
frame for MNR is “reasonable” include the following:
• The extent and likelihood of human exposure to contaminants during the recovery period, and if controlled by institutional controls, the effectiveness of those controls;
4-12 Chapter 4: Monitored Natural Recovery
As with any remedy, project managers should carefully evaluate the uncertainties involved and consider the need for contingency measures, contingency remedies, or interim decisions where there is significant uncertainty about effectiveness. For MNR, as for other approaches which take a period of time to reduce risk, project managers should carefully consider how risks can be controlled during the recovery period. For sites with bioaccumulative contaminants, institutional controls such as fish consumption advisories are frequently needed to reduce human exposures during this period. In most cases, no institutional controls are possible for reducing ecological exposure during the recovery period.
See Chapter 3, Section 3.6, Institutional Controls, and Chapter 7, Section 7.5, Considering Institutional Controls, for more information concerning institutional controls at sediment sites. Highlight 4-5 lists some important points to remember from this chapter.
Highlight 4-5: Some Key Points to Remember When Considering Monitored Natural Recovery • Source control should be generally implemented to prevent recontamination • MNR frequently includes multiple physical, biological, and chemical mechanisms that act together to reduce risk • Evaluation of MNR should be usually based on site-specific data collected over a number of years. At some sites, this may include an assessment of seasonal variation for some factors • Project managers should evaluate the long-term stability of the sediment bed, the mobility of contaminants within it, and the likely ecological and human health impacts of disruption • Multiple lines of evidence are frequently needed to evaluate MNR (e.g., time-series data, core data, modeling) • Thin-layer placement of clean sediment may accelerate natural recovery in some cases • Contingency measures should be included as part of an MNR remedy when there is significant uncertainty that the remedial action objectives will be achieved within the predicted time frame • Generally, MNR should be used either in conjunction with source control or active sediment remediation