National Academies Press: OpenBook

Sediment Dredging at Superfund Megasites: Assessing the Effectiveness (2007)

Chapter: 5 Monitoring for Effectiveness: Current Practices and Proposed Improvements

« Previous: 4 Evaluation of Dredging Effectiveness: What Has Experience Taught Us?
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 178
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 179
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 180
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 181
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 182
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 183
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 184
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 185
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 186
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 187
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 188
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 189
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 190
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 191
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 192
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 193
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 194
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 195
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 196
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 197
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 198
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 199
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 200
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 201
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 202
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 203
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 204
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 205
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 206
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 207
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 208
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 209
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 210
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 211
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 212
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 213
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 214
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 215
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 216
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 217
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 218
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 219
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 220
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 221
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 222
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 223
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 224
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 225
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 226
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 227
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 228
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 229
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 230
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 231
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 232
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 233
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 234
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 235
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 236
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 237
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 238
Suggested Citation:"5 Monitoring for Effectiveness: Current Practices and Proposed Improvements." National Research Council. 2007. Sediment Dredging at Superfund Megasites: Assessing the Effectiveness. Washington, DC: The National Academies Press. doi: 10.17226/11968.
×
Page 239

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5 Monitoring for Effectiveness: Current Practices and Proposed Improvements MONITORING FOR EFFECTIVENESS The effectiveness of environmental dredging in reducing risk, as predicted when the remedy was selected, can be verified only through monitoring. Monitoring includes • Monitoring of potential short-term risks due to dredging. • Verification that dredging has achieved its immediate target cleanup levels. • Long-term monitoring to determine whether remedial objectives have been or are likely to be achieved in the expected time frame. Monitoring of effectiveness is an essential part of the remedy and should be proportional to the size of the project. Through careful moni- toring it is possible to demonstrate whether environmental dredging minimizes risks to human and ecologic receptors during active opera- tions and to judge the success of contaminant cleanup in decreasing risk after the cessation of active remedial operations. Monitoring is the only way to determine short-term and long-term compliance with remedial- action objectives and evaluate net risk reduction of the remediation, and it forms the basis of the 5-year performance reviews after cleanup. Be- 178

Monitoring for Effectiveness 179 cause sediments typically pose long-term risks, monitoring often must span decades to assess risk reduction. The ultimate goal of monitoring is protection—that is, ensuring that short-term and long-term risks are minimized, by providing sufficient information to judge that the remedy is effective, or to adapt site man- agement to optimize the remedy’s performance to achieve risk-based objectives. Management adaptation may entail modification of dredging procedures—for example, if short-term exposures exceed expected mag- nitudes—or modification of the remedy itself by amendment or modifi- cation of the record of decision (ROD) if long-term risk reduction pro- ceeds more slowly or more rapidly than expected. An effective sediment-monitoring plan takes into account the suc- cessive stages of sediment cleanup: site characterization; selection, plan- ning, and implementation of the remedial action; effectiveness assess- ment; and adaptive management.1 Monitoring should build on the studies previously performed for the remedial investigation and feasibil- ity study (RI/FS), which should have • Determined the nature and extent of contamination and any trends in time (for example, due to natural recovery). • Supported or developed a conceptual site model. • Provided information to assess risks to the environment and people. • Evaluated remedial alternatives, including a quantitative com- parison of risks associated with implementation of each one. Once the remedy is selected and implementation begins, monitor- ing extends the record of site conditions into the future. MONITORING PRINCIPLES 1. Monitoring should be based on and inform the conceptual site model. 1 In general, adaptive management is the testing of hypotheses and conclu- sions and re-evaluation of site assumptions and decisions as new information is gathered (see Chapter 6 for further discussion). It is an important component of the updating of the conceptual site model (EPA 2005a).

180 Sediment Dredging at Superfund Megasites a. Appropriate metrics need to be chosen, measuring suc- cess against expectations based on the conceptual site model. b. Monitoring is an essential verification step, not an add- on activity or a second remedial investigation. 2. Effective monitoring of the remedy requires characterization of pre-remedial trends and reference conditions, in addition to post- remedial trends. a. Sufficient pre- and post-remedial sample sizes are needed, to allow for natural heterogeneity. b. The time span of pre- and post-remedial sampling needs to be sufficient to capture the time scale of recov- ery processes. c. Proper reference sites and conditions must be specified and monitored. Monitoring and the Conceptual Site Model Links between contamination, exposure, and risk can be highly complex, involving multiple physical, chemical, and biologic processes. A particular combination of these is present at each site. Monitoring pro- tocols and media to be monitored will vary accordingly, and should be closely linked to site conceptual models that link site conditions with biologic exposures and effects (EPA 1998). The expectations of the Super- fund ROD are a natural yardstick against which to judge effectiveness. Those expectations of short-term exposures and long-term risk reduction due to dredging should be based on the conceptual site model and its mathematical counterpart. Where site conceptual models are insufficiently developed, it is dif- ficult to develop an understanding of the factors driving trends in site- monitoring data. On major dredging sites, short-term and long-term ex- pectations based on site models will have been developed as part of the feasibility study supporting remedy selection. Collecting data to test whether expectations have been fulfilled is part of the process of concep- tual-model development, testing, and refinement that was begun with the initial site characterization. If the important cause-effect relationships between contaminant sources, transport mechanisms, exposure path-

Monitoring for Effectiveness 181 ways, and receptors have been well characterized by the time a remedy is selected, including bioavailability and food-web relationships as ap- plicable, and there has been sufficient pilot testing or other means of an- ticipating site-specific field conditions and implementation challenges, well-designed monitoring should indicate the remedy has performed as expected. If not, monitoring can help to identify important elements that are missing from the conceptual site model so that its predictions can be made more accurate and site management can be adapted accordingly, as recommended in EPA’s Contaminated Sediment Remediation Guidance (2005a). In monitoring of the effectiveness of a remedy, important transport mechanisms and exposure pathways to be monitored include not only the ones that control exposures and risks under normal conditions, but also the ones that may be triggered by dredging, such as releases that may occur during normal dredging operations or when debris or bed- rock is encountered. Therefore, before selection and implementation of a remedy, the site investigation should thoroughly examine factors that would complicate dredging and include them in the conceptual model. Complicating site conditions and operational limitations can also be identified through pilot studies to verify the performance of the selected technology under site-specific conditions. Data collection is one of the more expensive aspects of site man- agement (Box 5-1).2 Judicious use of the conceptual site model in design- ing the monitoring plan focuses data collection where it can best ensure protectiveness while conserving monitoring resources. Monitoring should target the key pathways and receptors necessary to determine whether remedial objectives have been met. If dredging is intended to reduce ecologic or human health risks, the conceptual site model can be used to focus sampling on locations and receptors that directly indicate risk related to the targeted sediments and contaminants and minimize spurious effects, such as increased body burdens in migratory species 2In addition to the example provided in Box 5-1, see the breakdown of costs of the Hylebos Waterway and 2004 dredging at New Bedford Harbor presented in Chapter 2 (Figures 2-4 and 2-5). However, it should be noted that these costs may not be directly comparable; it is not clear, for example, whether the costs include design costs and long-term monitoring.

182 Sediment Dredging at Superfund Megasites BOX 5-1 Estimated Monitoring Costs for Lower Fox River and Green Bay, Wisconsin, ROD Remedy The costs of construction monitoring (including verification sampling) and long-term monitoring (including an initial pre-dredging baseline survey of af- fected media and surveys of the same media continuing for decades after the remedy) for the ROD remedy for Operable Units 2-5 of the Lower Fox River and Green Bay, WI, have been estimated at $6 and $8 million, respectively (Shaw 2006). Together those costs exceed the estimated cost of engineering and con- struction support for the remedy, including development of design documenta- tion, plans, and specifications. and species with wide home ranges that are due to unrelated exposures at remote locations. If dredging is intended to minimize water-column contaminant transport, the site model can be used to control for the ef- fects of flow, temperature, seasonality, non-sediment-related stressors (such as point and nonpoint sources), and other ambient conditions to inform sampling plans and assist in interpreting the results. Monitoring decisions may be influenced by financial, jurisdictional, or political interests, even though they should be guided solely by the need to verify conceptual site models, inform remedy implementation, and to document when remedial objectives have been achieved. Cleanup negotiations between regulators and responsible parties can be conten- tious, and agreements on the scope of cleanups are often the results of a long and difficult process. The scope of post-remedial monitoring can also be established during those negotiations. The parties have few in- centives to seek actively to establish whether a chosen remedial action had its intended effect. This paradigm, wherein both regulators and re- sponsible parties may perceive that they have something to lose and nothing to gain in a robust post-remediation monitoring program, may be a reason for the lack of post-remediation confirmation sampling seen at some sites. Public-sector and private-sector designers of a monitoring plan may face strong pressures to demonstrate early success while con- trolling costs and may also feel pressure to divert remedial funding to support broader long-term natural-resource monitoring efforts. Those ancillary goals may be attractive to parties involved in designing a moni- toring program, but the fundamental objectives of monitoring are to per-

Monitoring for Effectiveness 183 form a fair and conclusive evaluation of remedy effectiveness and risk reduction, and resources and energy should be focused on this objective. Information developed from the monitoring program should be used to guide future decision-making in a manner which balances a realistic as- sessment of the projected environmental benefit relative to anticipated costs. Developing a body of well-designed site evaluations of dredging effectiveness will meet the broader programmatic objective of providing EPA and other lead agencies with invaluable information on strengths and weaknesses of dredging as a remedy—information that they can use in future remedial decision-making. Comparisons to Baseline Conditions To assess the effectiveness of the remedy, post-remedial monitoring should be compared with data trends and model forecasts developed before remedy selection. This requires that there be comparable datasets before (a “baseline”) and after dredging. As stated by EPA (2005a, page 8-2), During site characterization, the project manager should anticipate expected post-remedy monitoring needs to ensure that adequate baseline data are collected to allow comparison of future datasets. Monitoring plans should also be designed to allow comparison of results with model predictions that supported remedy selection. It is often difficult in practice for an effective monitoring plan to meet the above objectives. One important issue at Superfund megasites is that the time from initial site investigation to implementation of reme- dial measures can be 10 years or more; it is extremely difficult to ensure temporal and spatial consistency of baseline and post-remedial monitor- ing data, including data- quality assurance and control. Data collections that span many years can greatly complicate the selection of appropriate statistical tests for evaluating them. Those concerns are often manifested after the fact rather than being evident during the planning of the base- line and long-term monitoring programs.

184 Sediment Dredging at Superfund Megasites Consistent with its role in supporting hypothesis-testing, the moni- toring protocol should be rigorous enough to allow managers to evaluate critically the potential adverse effects of dredging on human and ecolo- gic receptors and potential risk reductions due to removal of contami- nated sediment. For example, proper reference sites or reference condi- tions should be established to allow comparison of affected media with pre-dredging or nondredged controls. Appropriate sample sizes should be determined from estimates of variability derived from pilot studies or other sources of data. In particular, the natural heterogeneity of biologic systems can be substantial and should be explicitly accounted for in de- fining sample sizes. CURRENT MONITORING PRACTICES According to Elzinga et al. (1998, as referenced in EPA 2004), moni- toring is “the collection and analysis of repeated observations or meas- urements to evaluate changes in condition and progress toward meeting a management objective.” Monitoring at Superfund sites is typically di- rected toward evaluation of the performance of a remedy and whatever environmental protections are in place during implementation of the remedy. Monitoring may include the collection of samples or real-time metered data • During implementation of the remedy to assess immediate hu- man health or environmental effects. • Soon after implementation to determine compliance with cleanup levels or other short-term objectives. • Over time to evaluate the achievement of the long-term reme- dial-action objectives, the need for maintenance or repair, and the con- tinued effectiveness of the remedy and associated source control. Ideally, the monitoring parameters measured are linked to site- specific risk factors so that success (or lack of success) of the remedy is evident and directly informs management of the site. There are no abso- lute requirements for monitoring elements or techniques, but a number of guidance documents have been published (Fredette et al. 1990; EPA/USACE 1998; EPA 2001a, 2004, 2005a) to identify relevant meas-

Monitoring for Effectiveness 185 urements and techniques and to guide the design of monitoring pro- grams for a contaminated sediment site undergoing remediation. Monitoring Parameters and Techniques Monitoring involves combinations of physical, chemical, and bio- logic methods. Three critical lines of evidence that increasingly define successful sediment remediation include sediment physical stability, sediment chemical stability (lack of movement of contaminants from the sediment to the water column), and biologic-ecologic integrity. These three concepts are integral components of remedy evaluation, and moni- toring should use techniques sufficient to measure progress toward these end points. A variety of techniques and measurement parameters exist for the characterization of the nature, extent, and potential effects of sediments. These techniques range from relatively simple and quick to elaborate and time consuming (e.g., EPA 2001a; Wenning et al. 2005). Several of the techniques are described below and summarized in Box 5-2. Physical Techniques Available physical techniques include direct sampling of sediment for laboratory analysis of geophysical properties, core sampling to iden- tify sediment layering or the presence of debris, side scan sonar to de- velop high resolution maps of bottom contours, acoustic sub-bottom pro- filers or magnetometers to map sub-bottom characteristics, remote sensing to document vegetative cover or other characteristics, videogra- phy or photography to document bottom features or shallow sediment profile characteristics, and instrumentation to measure environmental conditions (such as temperature and turbidity) or flow characteristics that may affect sediment and suspended solids transport. For example, sediment-profile imaging (a photographic technique) of surface (10-20 cm) characteristics can be conducted to establish various parameters in- cluding the depth of bioturbation, the depth of an oxygenated layer, general benthic community type and degree of recovery, or hydrogen sulfide gas production (see Figure 5-1 for an example). Other remote

186 Sediment Dredging at Superfund Megasites BOX 5-2 Common Physical, Chemical, and Biologic Measurements Used To Characterize Contaminated Sediments Common physical measurements include • Sediment geophysical properties, such as bulk density, particle size, and shear strength. • Pre-dredging and post-dredging bottom elevations, and sediment bed- forms. • Sediment layering, such as depth of disturbance or bioturbation, presence of gas bubbles, redox layers, and interfaces between sediment of different tex- tures. • Debris-field mapping (location, density, and size). • Conductivity, temperature, turbidity, and suspended particles under various flow conditions. • Stream velocities. Common chemical measurements include • Water-column parameters (such as dissolved oxygen and total and dis- solved chemicals under various flow conditions). • Surface- and subsurface -sediment chemistry, including magnitude, dis- tribution, and depth of contamination. • Pore water contaminant concentrations. • Bioavailable fractions of contaminants in sediment, on the basis of or- ganic-carbon normalization or acid volatile sulfide (AVS) analysis. • Tissue contaminant concentrations including tissues ingested by humans (in field collected or exposed aquatic organisms or plants) or tissue surrogates. • Air quality (including odor) during construction of remedy or handling of dredged material. Common biologic measurements include • Benthic invertebrate community structure (including abundance, diver- sity, and other structural or functional indexes). • Toxicity (acute and chronic effects measured in the laboratory or field). • Aquatic or wetland plant community structure (including species compo- sition and percentage of cover). • Fisheries status (including size, abundance, reproductive status, and inci- dence of lesions or parasites).

Monitoring for Effectiveness 187 FIGURE 5-1 Sediment profile imagery (SPI) equipment (two left photos) and sediment profile photograph (right) from New Bedford Harbor Superfund site (the outer harbor is the area of the site with the least contamination). This equipment is used as part of the long-term monitoring program at the site to assess benthic quality rapidly and augment traditional benthic survey techniques that entail sieving and enumeration. Source: W. Nelson, U.S. Environmental Pro- tection Agency. sensing techniques, such as lidar (light detection and ranging) can be used to map large-scale site characteristics, including the extent of eel grass beds or other vegetative cover. Assessments of physical stability of sediments (which translates into the likelihood for sediments to be dis- lodged and transported by erosive events) are based on site uses, hy- drology and geomorphology, sediment bed descriptions (radio dating deposits, stratigraphy, and physical characteristics), and measurement of sediment transport and sediment bed dynamics (erodability or bed ele- vation changes) (Bohlen and Erickson 2006). Chemical Monitoring Chemical monitoring can address multiple media⎯including air, sediment, water, biota, groundwater, and pore water⎯and can be de- signed to evaluate specific phases of chemicals of concern (for example, if they are dissolved or suspended in association with solids). It is impor-

188 Sediment Dredging at Superfund Megasites tant to monitor those parameters that affect chemical bioavailability, such as total and dissolved organic carbon, acid volatile sulfides (AVSs), grain size, and pore water fractions because organisms are exposed only to the bioavailable fraction (NRC 2003). The relationship between chemi- cal concentrations, the bioavailable fraction, and toxic effects is the foun- dation for establishing sediment quality guidelines (see next section). Chemical sampling may involve in situ instrumentation for water, sin- gle-point grab samples of water or sediment obtained with various de- vices, or use of samplers that integrate chemistry over time or space (such as sediment traps, composite water samplers, and peepers). Rapid chemical screening techniques that use immunoassay response (enzyme- linked immunosorbent assays [ELISA]) or chemical fluorescence to document relative exposures have also been developed, but these are generally single-contaminant or contaminant-class tests, and few rapid field screening techniques are available for measuring a broad array of contaminants. Some analytic methods for environmental samples can be time- consuming, labor-intensive, and expensive. For example, chemical measurements for persistent organic contaminants in sediments—such as polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs), and DDT—require extraction, cleanup, and instrument analyses with gas chromatography or mass spectrometry. None of those meas- urements is rapid or performed conveniently in the field. As with sedi- ments, the chemical analysis of biologic samples requires extraction, cleanup, and instrument techniques. Replicate measurements are neces- sary for both sediment samples and biologic tests because of inherent variability. Newer techniques have been developed for deployment of manu- factured materials in the form of passive sampling devices (such as semipermeable-membrane devices, solid-phase microextraction fibers, and Tenax) that can mimic biologic exposure to and tissue uptake of con- taminants from water, pore water, or sediments. For example, semiper- meable membrane devices (SPMDs) have been widely used in environ- mental applications since the early 1990s (Huckins et al. 1990; 1993) and applied at Superfund sites to monitor dissolved hydrophobic contami- nants and estimate water column concentrations of these contaminants (e.g., Hofelt and Shea 1997; Weston 2005). Polyethylene devices (PEDs) passively sample hydrophobic organic compounds in the aqueous phase.

Monitoring for Effectiveness 189 They are robust, simple, and inexpensive and have a short equilibration time (Booij et al. 1998; Adams 2003). The laboratory analysis of PEDs has fewer background-signal problems than the analysis of biologic samples. Another example is solid-phase microextraction (SPME) of sediment pore water, which may be quicker and more economical than conven- tional sampling. SPME uses fibers coated with a liquid polymer, a solid sorbent, or a combination. The fiber coating removes the compound from solution by sorption. The SPME fiber is then inserted directly into a gas chromatograph for desorption and analysis. Some of these passive sampling devices can reach equilibrium with environmental conditions much more rapidly than living organisms. They have been widely used in recent years for sampling metals and or- ganics in aquatic systems and found to be good indicators of fish and invertebrate bioaccumulation, that is, to be biomimetic (e.g., Arthur and Pawliszyn 1990; Huckins et al. 1993; Wells and Lanno 2001). Biomimetic samplers are easy to deploy and analyze, and they indicate exposure over time, but they are selective and do not indicate all chemical expo- sures or biologic effects (Table 5-1). Several of these techniques are still undergoing development and research is being conducted to better un- derstand the relationship between the sampler concentrations, environ- mental concentrations, and bioaccumulation in organisms (e.g., Leslie et al. 2002; Lohmann et al. 2004; Vinturella et al. 2004; Conder and LaPoint 2005; You et al. 2006). With time and refinement, these technologies will likely become more available for routine application at contaminated sediment sites. Biologic Monitoring Biologic monitoring looks at the sublethal to lethal responses of in- dividual organisms, populations, or communities in the environment or under controlled laboratory conditions. Biologic measurements and end points are usually more complex or difficult to obtain than physical or chemical measures, but biologic monitoring is the most definitive way to determine risk. Biologic monitoring typically provides a more integrated measurement of exposure (of both human and ecologic receptors) and is related more directly to ecosystem effects than is physical or chemical monitoring.

190 Sediment Dredging at Superfund Megasites TABLE 5-1 Strengths and Limitations of Methods for Assessing Biologic Effects in Aquatic Ecosystems Effect Assessment Method Advantages Limitations Criteria or Proven utility and ease of use Assume single chemical guidelines effect; based on laboratory exposures; causality link uncertain Biotic-ligand Proven utility and ease of use Insufficient research and model for accounting for metal validation for use with bioavailability in surface water sediment Empirically Proven utility and ease of use Bioavailability not accounted based guidelines for; may lead to incorrect conclusion of presence or absence of risk Equilibrium- Regulatory support; predictive Not applicable in dynamic based guidelines capability systems; does not consider all critical binding phases Species sensitivity Use of all available data for Lack of sufficiently large and distributions derivation of EQC or PNEC diverse sediment-toxicity datasets Indigenous biota Target receptors; lack of Habitat and other natural laboratory extrapolation; long- stressors or linkages term measure; proven utility; confound causality linkage; public interest; colonization inherent variability; loss of and transplant methods colonization units possible increase stressor diagnostic because of flow and power and experimental vandalism power Tissue residues and Documents exposure; use for Adaptation, acclimation, and biomarkers food web and risk models; metabolism confound widely used; very sensitive interpretations; uncertain and timely adverse- effect threshold levels Biomimetic Accumulates organics or Selectivity varies with devices: metals from waters and different chemicals; may not semipermeable sediments through diffusion mimic bioaccumulation of all membrane devices; and sorption; amounts organisms; some are subject

Monitoring for Effectiveness 191 TABLE 5-1 Continued Effect Assessment Method Advantages Limitations solid-phase accumulated on these inert to fouling, depending on microextraction; materials are similar to ecosystem; not standardized Tenax; diffusive amounts bioaccumulated in gradient transport fish tissues; can be placed in situ for short to long periods and then directly analyzed in laboratory Toxicity assays Bioavailability indicator; Causality link uncertain; (laboratory) proven utility; integrates laboratory-to-field effects of multiple chemicals; extrapolation; individual-to- does not measure natural community extrapolations; stressors does not measure natural stressors; cost of chronic assays Toxicity and More realistic exposure, which Most methods are not bioaccumulation reduces artifact potential; standardized; limited use; assays (field) measures many natural deployment can be difficult; stressors and interactions; possible caging effects with compartmentalizes exposures some organisms; causality to various media; exposure-to- link uncertain; loss of units effect linkage is strong possible because of predators and vandalism; acclimation stress possible because of temperature, salinity, or hardness differences Toxicity Better establishes specific Subject to manipulation fractionation chemical causality; standard artifacts; acute toxicity only; (laboratory) method for effluents limited use in sediments; large pore water volume requirements; limited sensitivity Toxicity More realistic exposure, which Very limited use; deployment fractionation (field) reduces artifact potential; can be difficult; shallow better establishes chemical environments only; acute causality toxicity only; loss of units possible because of high flow and vandalism; not standardized Source: Modified from Burton et al. 2005.

192 Sediment Dredging at Superfund Megasites Biologic testing often has both field and laboratory components— organisms collected from the field are identified and enumerated or, in marine systems, exposed to sediment-bound or water-borne chemicals in a laboratory (e.g., Barbour et al. 1999; EPA 2001a; Adams et al. 2005). In- digenous organisms can be collected with nets, hooks, traps, grab sam- plers, or other devices. Standardized laboratory sediment toxicity and bioaccumulation testing methods are commonly used in assessments of the potential hazard of dredged materials (e.g., Environment Canada 1992; EPA 1994, 2000a; EPA/USACE 1998; ASTM 2006). Standardized methods are available for freshwater and marine systems, in both short and long term (chronic) exposures. These tests often are one component of a “Sediment Quality Triad” and other weight-of-evidence based ap- proaches (Adams et al. 2005). A strong relationship has been docu- mented between the responses in these standardized laboratory test re- sponses (and indigenous benthic communities) and empirically-based sediment quality guidelines (discussed below) (Ingersoll et al. 2005). Field-collected or laboratory-reared organisms can be deployed in cages or nets for defined periods of exposure to water or sediment and retrieved for analysis (e.g., Ireland et al. 1996; Tucker and Burton 1999; Burton et al. 2000; Chappie and Burton 2000; Greenberg et al. 2002; Ad- ams et al. 2005; Crane et al. 2007). These caged-organism assays allow measurements of effects on growth and survival to be closely linked to environmentally-relevant chemical exposures (Table 5-1) (Solomon et al. 1997; Burton and Pitt 2002; Adams et al. 2005; Wharfe et al. 2007). Trans- plantation and recolonization of benthic macroinvertebrates on reference and site sediments have also been shown to be effective ways to measure site effects and risk, but they require exposures of up to a month (e.g., Clements and Newman 2002; Clark and Clements 2006). Biologic moni- toring can be cost-effective, relative to chemical monitoring (Karr 1993; Hart 1994). Transplantation, colonization, and caged-biota tests can be long and have deployment challenges. However, caged exposures often take only one to several days in freshwater systems (Ireland et al. 1996; Tucker and Burton 1999; Chappie and Burton 2000; Burton et al. 2002, 2005; Greenberg et al. 2002; Burton and Nordstrom 2004). In situ caged exposures of 2-4 days have been shown to provide uptake and toxicity information that is comparable with that of standardized laboratory tests that take 10-65 days at PCB-, chlorobenzene-, and metal-contaminated

Monitoring for Effectiveness 193 sediment sites (Greenberg et al. 2002; Burton et al. 2005). It is also useful to conduct laboratory-based exposures following standardized toxicity- test methods (EPA 2001a). Monitoring Human Exposures The biologic monitoring techniques listed and discussed above are related primarily to ecologic receptors and do not include monitoring of human subjects, which EPA and other Superfund lead agencies do not typically perform at contaminated sediment sites. However, biomonitor- ing of people who live near contaminated sediment sites is sometimes performed by other parties, such as local health authorities and academic scientists (Miller et al. 1991; Fitzgerald et al. 1996, 1999, 2004; MA DPH 1997; Korrick and Altshul 1998) and the Agency for Toxic Substances Disease Registry. Typical indicators of human exposure and risk reduction include contaminant concentrations in the subset of environmental media to which people might be directly or indirectly exposed in places where exposures might occur. Those media include surface sediment and sur- face water in areas accessible to people and aquatic biota used for food. Where there is interaction between contaminated sediments and flood- plain, the list may be expanded to include floodplain surface soil, terres- trial game species foraging in the floodplain, and agricultural products from the floodplain (see further discussion in sections below). Use of Sediment Quality Guidelines SQGs are numerical chemical concentrations intended to be either protective of biologic resources or predictive of adverse effects to those resources (Wenning and Ingersoll 2002). They are used to estimate the toxicity and risk from sediments. At a contaminated site, the SQGs can be used to establish contaminants of concern (COCs) from potentially long lists of contaminants of potential concern (COPCs), identify or rank problem reaches in a waterway, and classify hot spots (Long and Mac- Donald 1998). There are two basic categories of SQGs, empirical and determinis- tic. Empirical approaches use statistical methods to compare sediment

194 Sediment Dredging at Superfund Megasites chemistry to effects datasets to predict the probability of or the presence and absence of adverse or toxic effects (Word et al. 2005). A variety of those approaches have been used to develop toxic effects levels, thresh- olds, or concentrations (used as SQGs) (MacDonald et al. 2000; Burton 2002; Wenning and Ingersoll 2002). Deterministic approaches typically use equilibrium partitioning theory (Adams et al. 1985; Di Toro et al. 1991, 1992) to relate toxic concentrations found in water-only exposures to sediment exposures for the same organism. Effects are predicted to occur when toxic concentrations found in water occur in the pore water of the sediment (Word et al. 2005); complexing agents (organic carbon for hydrophobic non-ionic contaminants [Di Toro et al. 1991] and AVS for cationic metals [Di Toro et al. 1990, 1992]) are the basis of the equilib- rium calculations. There has been a lot of controversy and discussion on the use and viability of SQGs including their false positive and negative rates, their applicability to mixtures of chemicals, their ability to establish cause and effect relationships, and whether results can be extrapolated across spe- cies or biologic communities (Burton 2002; Wenning and Ingersoll 2002). Sediment quality guidelines are only indirect measures of effects and do not clearly establish whether risk or adverse biologic impacts are actually occurring (Table 5-1) (NRC 2003). A recent Pellston workshop summary, Use of Sediment Quality Guidelines and Related Tools for Assessments of Contaminated Sediments (Wenning et al. 2005), comprehensively reviews these approaches. A few conclusions reached from this workshop include that (1) although the scientific underpinnings of the different SQG approaches vary widely, none of the approaches appear to be intrinsically flawed; (2) chemically- based numeric SQGs can be effective for identifying concentration ranges where adverse biologic effects are unlikely, uncertain, and highly likely to occur, and; (3) in all cases, application of SQGs in a “toxic or nontoxic” context must be cognizant of the types and rates of errors as- sociated with each type of SQG (Wenning and Ingersoll 2002). EPA has supported the development of mechanistically based sediment quality guidelines (EPA 2000b, 2003a,b,c, 2005b) and the Na- tional Oceanographic and Atmospheric Administration (NOAA) has supported the development of empirical sediment guidelines (Long and Morgan 1990). It is expected that as the scientific issues continue to be

Monitoring for Effectiveness 195 resolved, they will see continued and greater use in toxicity evaluations, comparative risk analyses, and in remedial decision making. MONITORING-PROGRAM DESIGN Selection of the appropriate monitoring measures and design of a monitoring program depend on the development of clear hypotheses to be tested or questions to be answered that are directly linked to a de- tailed conceptual site model characterizing sources, pathways of expo- sure, and receptors that may be exposed during or after remediation. By the time a remedy is implemented, the understanding of site processes should be highly refined so that monitoring can be focused on the ex- pected beneficial and adverse effects of remediation. These effects in- clude releases to the water column or atmosphere, as monitored during dredging; residual sediment concentrations, as monitored by progress samples or post-dredging verification sampling; and reductions in expo- sures and risks, as observed through long-term monitoring. Monitoring During Dredging Dredging operations include material removal, transport, dewater- ing, final disposal, and onsite solids treatment, water treatment, and temporary storage. Therefore, monitoring during dredging operations may involve a variety of activities, including some that are not directly related to dredging operations or performance. An inherent difficulty is the need for rapid measurement techniques that can inform contingency actions and provide near-real-time feedback for executing corrective measures while the work is ongoing. Monitoring programs implemented during dredging are often based on the requirements of Clean Water Act Rule 401 water quality certification, typically administered by state environmental agencies. The focus of any required monitoring for water quality certification is effects on water quality, based on comparison with state or federal water qual- ity standards and criteria, taking upstream conditions into account. Sur- rogate or indicator parameters (such as turbidity or concentrations of a single chemical) are typically used to provide rapid information to the

196 Sediment Dredging at Superfund Megasites dredger and site manager and to develop a compliance history spanning the various phases of the project. With available technologies, some contaminant release and trans- port is inevitable during dredging (EPA 2005a). Depending on the vola- tility of the contaminant, there may be release to the atmosphere as well as the water column. On the basis of project data presented in this report, contaminant release to the water column might not depend on observ- able resuspension of solids (see Box 4-2 for an example). Nevertheless, monitoring of turbidity can provide real-time quality-assurance informa- tion to the dredge operator and allow adjustments in the field to reduce resuspension. Air monitoring can also identify potential exposures and facilitate needed operational adjustments to protect nearby populations. To quantify contaminant releases, however, upstream and down- stream water-column contaminant fluxes should also be monitored. This can be accomplished relatively quickly using immunoassay test kits or traditional grab sampling with subsequent analyses. Passive sampling devices or caged fish can also be placed at the site to indicate exposure over extended periods. These techniques make it possible to quantify the unintended contaminant loading to the water column, and this helps to explain increases in downstream exposures that are observed between the baseline and long-term monitoring and to distinguish short-term ef- fects due to dredging from continuing long-term releases attributable to uncontrolled sources. Monitoring Human Health Effects During Dredging During dredging and dredged-material handling, the surrounding community and remediation workers might experience higher exposures than before dredging. The increases could arise from chemical releases to the overlying surface water and ambient air, from uptake by biota con- sumed by people, and from creation of residual contamination in areas where people or edible biota come into contact with it. The surrounding community might also experience non-health-related effects, such as ac- cidents, noise, and residential or commercial disruption, which are po- tential ancillary consequences of dredging. An evaluation of net risk reduction should begin with sufficient datasets that permit comparison of exposure conditions before and dur-

Monitoring for Effectiveness 197 ing dredging operations. In some of the projects, increased exposure oc- curred during dredging in connection with the physical disruption of contaminated sediment. Monitoring during dredging should be de- signed so that data are sufficient to quantify changes in exposure result- ing from the dredging operation, specifically related to resuspension of sediment, release of chemicals from sediment, and creation of residuals. Changes in net risk resulting from transport, storage, treatment, and dis- posal of dredged sediments should be quantified, and this may include collection of monitoring data during dredging and dredging-related op- erations. Superfund remedial investigations often use concentrations of chemicals in environmental media—such as fish, sediment, surface wa- ter, and air—as surrogates for human exposure and do not study human subjects directly. Investigators need to monitor those media within the boundaries of the three-dimensional space in which people have direct or indirect contact. For example, people do not have direct contact with deep sediment. Unless that sediment becomes exposed in the future as a result of scouring, the dredging process itself, or some other process, sediment samples collected for evaluating direct contact should not ex- ceed the depth that a swimmer or wader might encounter. For indirect exposure to stable sediments through the food chain, the relevant sedi- ment sampling depth is limited to the biologically active zone. Sampling at greater depths is needed to assess potential exposures where sedi- ments are unstable. If sediment contamination has reached the terrestrial environment through atmospheric release and deposition or sediment deposition on floodplain soils, parallel monitoring and risk analyses should be performed for terrestrial exposure media. Direct studies of human exposure could help to quantify human exposure and risk but would be more invasive and expensive and would not necessarily yield a good measure of exposure reduction, given the difficulty in defining the exposed population and segregating site-related exposures from other exposures to chemicals of concern. Given that dredging remediation by definition involves an aquatic environment and that many of the most important sediment contami- nants are bioaccumulative, the consumption of fish and other aquatic organisms often contributes most to human health risk. However, until dredging is completed and cleanup goals have been met, EPA and state agencies with fisheries jurisdiction usually restrict fish consumption to

198 Sediment Dredging at Superfund Megasites protect human health. When members of the surrounding community comply with those restrictions, exposures of concern during dredging are limited to other pathways, such as releases to the atmosphere and surface water. Box 5-3 highlights examples of attempts to evaluate hu- man exposure during dredging. Ideally, dredging operations occur over a relatively short period that requires evaluation of acute and possibly subchronic risk, but not chronic risk, from these exposure pathways. To address those risks, EPA can establish acute and subchronic guidelines for air or other media for comparison with monitoring results. For example, at the New Bedford Harbor Superfund site, EPA selected air concentrations that if exceeded during dredging would require a change in the dredging operation. Also at that site, EPA detected increased hydrogen sulfide concentrations in a dredged sediment handling facility and changed the operation to reduce concentrations to safe levels.3 In such cases, monitoring results should be made available in a time frame that allows site managers to manage risks appropriately. The data should distinguish conditions upstream and downstream of dredging or upwind and downwind of dredging so that site managers can discern the effects of dredging relative to background exposure conditions including natural disturbances. Monitoring Ecologic Effects During Dredging Current practice often omits biologic monitoring during remedy implementation at sediment sites. Monitoring of bioaccumulation during dredging is typically not able to inform the project manager or operator in a timely fashion so that dredging protocols could be modified or addi- tional protections implemented, owing to the length of time that most organisms take to respond to environmental exposures. Other challenges in using bioaccumulation and tissue concentration monitoring data are in relating chemical concentration to ecologic relevance or adverse bio- logic effects and in the uncertainty of the relationship between exposure (such as to site sediments or resuspended materials) and tissue concen- trations in fish if they are able to move off site (these issues are described in greater detail in the next section). 3 The presence of hydrogen sulfide is related to the anaerobic environment of the sediments and not to chemical contaminants at the site.

Monitoring for Effectiveness 199 BOX 5-3 Monitoring of Conditions During Hot-Spot Dredging at the New Bedford Harbor Superfund Site for Effects on Human Exposure The New Bedford Harbor Superfund site has been the subject of extensive efforts to understand the effects of harbor contamination on aquatic species and people living near the harbor. In addition, EPA developed a plan to monitor the effects of dredging a contaminated hot spot on water quality, air quality, and bioaccumulation by benthic invertebrates. The hot-spot dredging occurred in 1994-1995. EPA (1997) compared results of monitoring conducted before, during, and after hot-spot dredging and concluded that the dredging resulted in few if any adverse effects on the marine ecosystem. EPA identified some air-quality issues that were remedied with changes in operation or engineering controls. Cullen et al. (1996) compared PCB concentrations in tomato samples col- lected downwind of the hot-spot dredging operation before and during dredg- ing and concluded that the average PCB concentration during dredging was about 6 times higher than the average PCB concentration before dredging. Choi et al. (2006) reported PCB concentrations in umbilical-cord blood samples among members of nearby communities that were collected before, dur- ing, and after hot-spot dredging. The authors reported that their results “support modest, transient increases in cord serum PCB levels during dredging, with sig- nificant declines in serum PCB levels observed after dredging, particularly for the more volatile PCBs and PCB-118.” They attributed the “significant declines,” in part, to the hot-spot dredging (see figure below). Figure shows covariate-adjusted smoothed plots of predicted ∑PCB (A), heavy PCB (B), light PCB (C), and PCB-118 (D) levels vs. infant’s date of birth. Vertical lines denote the (Continued on next page)

200 Sediment Dredging at Superfund Megasites BOX 5-3 Continued start and stop dates for dredging of contaminated New Bedford Harbor sediments. Plots are adjusted for child’s sex, maternal age, birthplace, smoking during pregnancy, previous lactation, household income, and diet (consumption of organ meat, red meat, local dairy, and dark fish). Source: Choi et al. 2006. EPA’s summary of tier 1 sediment remediation sites with pre-monitoring and post-monitoring data (S. Ells, EPA, unpublished information, March 22, 2006) presented during the committee’s first meeting indicates that the baseline average concentration of PCBs in surface sediment (no depth specified) at the hot spot was 25,000 mg/kg and lists a post-remedial average concentration the hot spot of 330 mg/kg. However, that information appears to be inconsistent with another EPA presentation (Nelson 2006) during which W. Nelson reported that the average concentration of PCBs in surface sediment (top 2 cm) of the up- per harbor, of which the hot spot makes up about 5% of total area, did not change significantly as a result of hot-spot dredging. It is possible that the aver- age was unchanged because the hot spot represents a small fraction of the upper harbor. However, if the hot spot represented the portion of the upper harbor with the highest PCB concentration, one would expect its removal to cause some decline in PCB concentration. Given that the PCB concentration apparently did not decline, it is not clear how hot-spot dredging might have led to reduced PCB concentrations in umbilical-cord serum after dredging. These collective efforts show how exposure might change during the pe- riod of dredging, but it is premature to use them to judge effectiveness, because EPA’s remediation is not yet complete. These studies illustrate the challenge of linking dredging in a large harbor with human exposure. Nevertheless, one of the main risks to ecologic receptors posed by release and transport of contaminants during dredging is increased con- taminant uptake and increased toxicity, and there are techniques for monitoring those effects, even if they are not in wide use. Subject to the timeliness limitations noted above for providing real-time feedback to dredging operations, studies can be designed to assess contaminant bio- accumulation and toxicity during dredging by using caged or sessile or- ganisms or using passive sampling devices such as SPMDs, as discussed above (Chappie and Burton 2000; Adams et al. 2005; Crane et al. 2007). The utility of caged-fish studies has been demonstrated, for example, in the 1995 Grasse River non-time-critical removal action (BBL 1995). Mus-

Monitoring for Effectiveness 201 sels deployed in mesh bags have been used in the long-term monitoring program at the New Bedford Harbor Superfund site to monitor trends in PCB bioaccumulation and evaluate the impact of dredging operations (Bergen et al. 2005). To quantify the spatial distribution of resuspended materials, the organisms can be placed at various distances from the sources of contamination. Comparisons with pre-dredging, reference, or upstream conditions allow managers to determine whether uptake of contaminants increases during dredging operations. Complex exposure dynamics cannot be mimicked in the laboratory. If standard test species are exposed in situ, exposures are more realistic. In situ testing with caged organisms has been shown to be an effective monitoring tool. Its primary advantages are the improved realism of ex- posure, the lack of sampling-induced artifacts, the ability to deploy and assess within days, and the ability to partition exposures of key envi- ronmental compartments and exposure time frames. (However, these techniques also have concerns regarding the modification of site condi- tions during exposure [Chappie and Burton 2000]). One can also link exposure with effects in that multiple end points can be assessed, such as tissue concentrations, growth, and reproductive status. Numerous stud- ies have demonstrated the approach in studies of runoff, base flow, and sediments (e.g., Ireland et al. 1996; Tucker and Burton 1999; Chappie and Burton 2000; Greenberg et al. 2002; Burton et al. 2005; Crane et al. 2007). Studies of marine systems have primarily used mussels (Salazar and Sa- lazar 1997; Bergen et al. 2005), and there has been less testing of amphi- pods (DeWitt et al. 1999). Freshwater studies have used a wide variety of organisms, such as fish, cladocerans, amphipods, midges, bivalves, may- flies, and oligochaetes (e.g., Chappie and Burton 2000). It is important to consider the likely response time when selecting test organisms; organ- isms that equilibrate with their environment more quickly would be more useful for evaluating releases during dredging. It is also advanta- geous, when possible, to use indigenous biota when conducting in situ caged testing. Standard toxicity test organisms (such as fathead min- nows) may have very different biologic responses than indigenous popu- lations that may have acclimated or adapted to toxics in the watershed, thus being less sensitive than the surrogate species. In that case, surro- gate species may be useful for detecting adverse effects, but they will not be a good indicator of effects to indigenous species.

202 Sediment Dredging at Superfund Megasites The use of natural resident populations collected during and im- mediately after dredging is an alternative to caging studies. For example, resident spottail shiners collected during the 2005 dredging operations at the Grasse River showed significant increases compared to sampling conducted during several years prior to dredging and the year following dredging (see Chapter 4, Figure 4-10 and associated text). Post-dredging Verification Sampling Verification sampling immediately after dredging allows site man- agers to determine whether cleanup levels or other short-term objectives have been met. The ability of the remedy to achieve short-term cleanup levels depends in part on how accurately the remedial investigation and additional pre-remedial sampling have characterized the extent and dis- tribution of contamination and on whether the dredging design based on that characterization encompasses the bounds of contamination encoun- tered by the dredger in the field. Dredging designs are based on interpo- lation of sediment core data, which are often sparse relative to the scale of sites. Even at major sites, the density of pre-design samples is typically less than one core per acre: for development of the Lower Fox River and Green Bay Operable Units 2-5 remedial design, the density was about one core per 1.6 acres (Shaw 2006). Depths of contamination in the wide expanses between core locations are therefore subject to uncertainty, and the dredging projects in this report provide evidence of that uncertainty in the form of sites where significant undisturbed residuals remained after dredging to design elevations. The probability of leaving consoli- dated sediments with elevated concentrations in place can be reduced by conducting more intensive pre-design sampling before dredging, by lowering the elevation of the dredge cut (that is, overdredging), or by verification sampling after dredging followed by redredging as needed (see examples in Box 5-4). Thus, there are tradeoffs between the volume of material removed and the intensities of pre-design and verification sampling. The greater the confidence in the methods used to develop the dredge prism, including sampling and interpolation, the less overdredg- ing and verification sampling may be needed to ensure protection. Those tradeoffs have been considered explicitly in pre-design studies for Lower Fox River Operable Units 2-5 (see Box 5-5).

Monitoring for Effectiveness 203 The dredging projects evaluated by the committee include numer- ous examples of sites where dredging generated substantial residual con- tamination. Verification sampling is needed to detect and quantify gen- erated residuals. Where possible, the samples should be collected in the form of cores long enough to penetrate and capture sediment underlying the generated residual layer, rather than grab samples (Palermo 2006). When cores cannot be obtained, grab samples should be taken. One promising technology for obtaining grab samples, using a hydraulic sampling device, is described in Box 5-6. It is also important, during col- lection and analysis of cores, to capture the unconsolidated “fluff” that may be generated from dredging activity. With core samples, the thick- ness and texture of the generated residual layer can be observed and dis- tinguished from underlying material to support planning of additional work that may be needed to minimize risk, such as backfilling with coarse-grained material. BOX 5-4 Verification Sampling at Harbor Island, Washington Dredging at Todd Shipyards (part of the Harbor Island Superfund site) re- lied on collection of shallow progress cores in each sediment management area (SMA) as dredging was completed. Results were compared with cleanup levels to determine whether additional dredging was needed. Dredging was sequenced in such a way that an SMA that had been remediated was not affected by dredg- ing in adjacent SMAs. Final verification samples were collected once all SMAs had been dredged at a relatively low density because of demonstrated compli- ance with cleanup levels based on the progress cores. The presence of extensive surface and subsurface debris in the areas to be dredged at Lockheed Shipyard (also part of the Harbor Island Superfund site) resulted in extensive residual contamination and undredged inventory at the end of the first dredging season. Shallow cores were collected throughout the dredged area after dredging to document the remaining contamination and dis- tinguish between a light unconsolidated sediment layer and more consolidated material. The latter material either had sloughed from the edge of the dredge cut or could not be removed by the dredger because of debris that remained on the site. On the basis of this sampling, it was decided to place a thin layer of clean material over the dredged area until it could be redredged in the following sea- son.

204 Sediment Dredging at Superfund Megasites BOX 5-5 Delineating the Dredge Prism in the Fox River, Wisconsin Sediment remediation areas and volumes were delineated for the remedial design of the lower Fox River, WI, between Operable Unit 2 and the mouth of Green Bay with an advanced interpolation method called full-indicator kriging. Full-indicator kriging provided a probability distribution of depth of contamina- tion to the ROD cleanup level at each sediment location. For areas where dredg- ing is the selected remedy, dredge-prism designs were developed at a range of significance levels (defined as the probability of exceeding the cleanup level at a given location) to inform risk management decisions. Those decisions involve balancing the risk of leaving contaminated sediment behind (a false-negative, or type 2, error) against the risk of unnecessarily dredging clean material (a false- positive, or type 1, error). Additional protection against false negatives will be provided in the remedy by post-dredging confirmation sampling. A significance level of 0.5 was chosen because it provided a reasonable Type 1 error and a low Type 2 error, reasonable accuracy, and the least bias in the dredge cut. This deci- sion was made acknowledging the importance of minimizing Type 2 errors be- cause remediation of clean sediments is cost that cannot be recovered. There was agreement that a robust verification sampling program would be developed and this would uncover significant Type 1 errors (that is, leaving behind sediments that should be remediated) which can subsequently be dealt with. In practice, more sediment will be removed than the selected significance level indicates because additional deepening of the dredge prism occurs during dredging-plan design and to account for contractor overdredging allowance (Anchor and Limno-Tech 2006a,b,c). In verification sampling, it is important that the spatial scale of remedy evaluation be consistent with the site’s remedial objectives. If the objectives require minimizing contaminant flux to the water column or minimizing sitewide exposures to widely ranging fish species, it is ap- propriate to compare area-weighted average concentrations with reme- dial goals. Protecting sensitive receptors that have more limited ranges would require verification that targets are achieved on a finer spatial scale. It should also be emphasized that although effectiveness of im- plementation, which is an intermediate goal, can be evaluated with veri- fication sampling, the ultimate goal is risk reduction through achieve- ment of remedial objectives, which is evaluated with long-term monitoring.

Monitoring for Effectiveness 205 BOX 5-6 Verification Sampling of Dredging Residuals at the Head of Hylebos Site, Commencement Bay, Washington Discrete sediment samples were collected on a daily basis immediately be- hind the operating dredge to provide immediate evaluation of the post-dredging residual layer for the Head of Hylebos project. Nearly 1,000 discrete samples of the residual layer were collected using a Marine Sampling Systems 0.3m2 Power Grab (Power Grab) generating measurements of residual layer thickness and sediment chemistry (24-hour chemistry turn around times). This program pro- vided immediate feedback on the nature of the residual layer generated during dredging, and allowed for ongoing adjustment of the dredging methods to fur- ther control the residual layer formation. Unlike typical surface grab samplers, the Power Grab is a hydraulically ac- tuated clamshell bucket that is capable of collecting 1-ft thick samples in many sediment types ranging from soft fine-grained sediment to more dense and com- pact silts and sands (not hardpan or glacial till), as well as through some debris. All of the sample contact surfaces on the Power Grab are stainless steel while the frame of the sampler is aluminum. The Power Grab features include (1) wide adjustable feet to control the depth of penetration to avoid over or under pene- tration of the sampler; (2) adjustable ballast (280-750 lb) to provide additional reaction weight for sampling in stiff material; (3) a semi-circular cutting profile to limit the disturbance of the sample; (4) and an enclosed bucket configuration to protect the sample from scour while being raised through the water column. These features allowed the sample team to consistently collect acceptable sam- ples (without over- or under-penetration) in all sediment types found on the site. Once the Power Grab sample was brought on board the sampling vessel, the overlying bucket covers were removed and overlying water decanted. The 0.3-m2 sample footprint (roughly 1 3/4 ft by 1 3/4 ft) was sufficient to allow for subsampling to measure the thickness and record the characteristics of the re- sidual layer, the characteristics of the underlying more compact native sediment, as well as the collection of sediment samples for chemical analysis. The Power Grab performed well throughout the two seasons of dredge confirmation sam- pling without any notable complications or problems. Source: Dalton, Olmsted & Fuglevand 2006. Long-Term Monitoring after Dredging Long-term monitoring is used to judge whether a remedy is reduc- ing risk at the expected rate and when remedial objectives have been

206 Sediment Dredging at Superfund Megasites achieved. Superfund remedies at sediment sites are typically subject to review at 5-year intervals when, following remediation, contamination exists that could limit potential uses of the site (EPA 2001b). At dredging sites this could occur for several reasons: residual contamination after the completion of the remedial action, the recontamination potential as- sociated with the dynamic nature of the aquatic environment, the fact that some sources may be undetected and that controls of known sources are not always implemented concurrently with the remedy (particularly at the watershed level), and the additional time required by remedies to achieve objectives when they rely in part on natural recovery processes and must counter past bioaccumulation of contaminant in the food chain. Ideally, reviews compare recovery at each 5-year interval with an expected trend of exposure and risk reduction under the recommended remedy as developed in the feasibility study. Remedy modification and additional data collection as needed to fill in gaps in understanding, would be triggered by a significant deviation from the expected trend. Otherwise, monitoring continues until remedial objectives are achieved. Although a rich set of data should already exist as a product of the remedial investigation, it is important that a complete baseline dataset be obtained before remedy implementation, observing the same pathways and exposures as planned for long-term monitoring, to support clear and definitive pre-remedial vs post-remedial comparisons and post-remedial trend estimates. The importance of establishing a baseline, especially to assess the effects of the remedy on fish, was emphasized by a previous National Research Council panel (NRC 2001). Because long-term moni- toring may continue for decades and trend estimates will be based on comparisons with data collected in the early years of monitoring, it is important that sampling and analytic methods selected for baseline and long-term monitoring be consistent with the technologic state of the art. It is also vital that the baseline data be collected before the commence- ment of remediation and encompass trends of sufficient duration for the effects of the remedy to be distinguished from ambient trends leading up to its implementation. To facilitate comparisons of data over a long pe- riod (decades), it is useful to store duplicate biologic samples (fish tis- sues, human tissues, blood) from the analyses (for example in a deep freezer or liquid nitrogen). These samples can be analyzed in later years to facilitate comparisons of analytical data.

Monitoring for Effectiveness 207 When a dredging remedy is implemented, surface sediment con- centrations can be affected by a combination of sediment removal, back- filling with clean material, and natural recovery processes. At larger sites, where remediation may proceed over a period of years, there is value in determining the relative importance of each of those processes in reducing surface concentrations. If burial under clean watershed sediment transported by riverine processes strongly reinforces dilution of exposures, this may create opportunities to adapt the remedy to re- duce cost without compromising effectiveness. For example, if burial by clean sediment is sufficiently rapid and uniform, it may be possible to achieve risk-reduction goals while tolerating higher generated residual concentrations or to reduce the thickness of post-dredging backfill (see Box 5-7). To be able to measure residuals, backfill, and long-term sedi- mentation as separate layers, baseline and long-term monitoring of sur- face sediment should include at least a subset of finely sectioned cores, which should be analyzed for geotechnical characteristics, including grain size, bulk density, and contaminant concentrations. It is important to stress that backfilling and burial by natural sedi- ment processes are not necessarily equivalent in protection even if they cover contamination to equal thicknesses. Sands, which are typically used as backfill material, may be more stable in the face of erosion dur- ing high-flow events than natural sediments but their lower sorptive ca- pacity may provide less effective attenuation of contaminant exposure than burial by natural sediment, especially when pore water is the path- way of potential exposure.4 Monitoring of surface concentrations should account for trends in bioavailability by normalizing contaminant concen- trations to sorbent material, such as organic carbon, in addition to meas- uring trends in bulk surface-sediment contaminant concentrations. Long-term Monitoring of Human Health Effects Results of the baseline human health risk assessment indicate which exposure pathways and chemicals warrant action, and this know- ledge is vital for making effective remedial decisions. As noted above, the consumption of fish and other aquatic organisms often contributes 4 It also is possible to specify backfill material with organic carbon.

208 Sediment Dredging at Superfund Megasites BOX 5-7 Dredging and Later Sedimentation at Manistique Harbor, Michigan After dredging ended at Manistique Harbor, 3-7 ft of sediment were de- posited from 1996 to 2005 (Weston 2005). With the deposition of the new surface sediment on post-dredging residuals, it was possible to meet revised dredging cleanup levels. The original cleanup level had been removal of all sediment con- taining PCBs at greater than 10 mg/kg anywhere in the sediment column, and it proved difficult to achieve. The cleanup level was revised to an average concen- tration of 10 mg/kg throughout the sediment column, with 95% removal of PCB mass also required (see Appendix C). most to human health risk. Ideally, the population consuming aquatic biota would have been studied in the baseline human health risk as- sessment, including quantification of variability in fish consumption rates among members of the population to ensure that the most highly exposed members of the population are evaluated in the risk assessment. During monitoring, any important changes in consumption patterns should be accounted for in the monitoring plans. However, other path- ways might be important, such as consumption of waterfowl, dermal contact with and ingestion of sediment, inhalation of volatilized con- taminants, and ingestion of surface water during swimming or other rec- reational activities or through use as drinking water. If sediment con- tamination reaches the floodplain, people could be exposed through consumption of game species, wild edible plants, and agricultural prod- ucts from the floodplain and through dermal contact with and ingestion of surface soil. Box 5-8 summarizes factors that one should consider in designing the aquatic sampling programs that are most often used to quantify human exposures. EPA’s goal at contaminated sediment sites is to protect human health, given that people could be exposed to sediment and other con- taminated media over long periods. Consequently, cleanup levels, if es- tablished to protect human health, usually represent long-term average concentrations that people can be exposed to without expectation of harm over long periods. Therefore, they should not be treated as abso- lute exposure limits that are not to be exceeded at any time during long- term monitoring. Cleanup levels and monitoring programs should be

Monitoring for Effectiveness 209 BOX 5-8 Collecting Aquatic Samples for Monitoring Human Exposure Fish and Shellfish • Sample the species commonly eaten by the local population and be sure to include species known to accumulate high concentrations of chemicals of con- cern. • Catch the size range of fish harvested by the local population, being sure to include the larger fish usually harvested because larger (older) fish in a popu- lation are generally the most contaminated with chemicals that bioaccumulate (such as PCBs, dioxins, and methylmercury). • Avoid sampling finfish species during their spawning period, because tissue concentrations of some chemicals (for example, such lipophilic chemicals as PCBs and dioxin but not methylmercury) may decrease during this time and because the spawning period is generally outside the legal harvest period. • Match assumed or known consumption patterns to sampled species. Fish-creel data (from data gathered by surveying anglers) from state fisheries departments constitute one justifiable basis for estimating types and amounts of fish consumed from a given body of water. It is important to account for the frac- tions that various trophic levels contribute to a fish consumer’s diet. • Composite samples of fish parts consumed by the local population. Peo- ple might eat skin-on fillets, skin-off fillets, or whole gutted fish (for example, in soups). Skin-off fillets will have the highest mercury concentrations, whereas whole-body fish samples will have the highest PCB and dioxin concentrations. PAHs do not tend to accumulate in finfish that metabolize them. Composites improve the chance of detecting chemicals and thus reduce the number of sam- ples without detectable concentrations in the resulting dataset and the need to determine how they will be factored into arithmetic averaging. • Use a probabilistic sampling design, randomly selecting sampling loca- tions to address spatial variability and to ensure that sufficient samples are col- lected to distinguish the site from reference areas. This approach allows statisti- cally valid inferences to be drawn on an area as a whole. Ideally, samples would be collected over a geographic area that represents the average exposure of those who eat fish from the body of water. If there are smaller areas where people are known to concentrate fishing, these areas should be intensively sampled. • Collect both weight and length data to control for the potential influence of fish nutritional state on chemical concentration, such as by normalizing fish concentrations to a standard body condition. (Continued on next page)

210 Sediment Dredging at Superfund Megasites BOX 5-8 Continued Sediment • Collect from accessible locations where people are likely to fish, swim, or engage in other activities (sediment samples in deep water, for example, may be relevant to the food-chain exposure pathway but not the direct-contact path- way). • To evaluate direct-contact exposures, collect sediment at depths that cor- respond to the depth to which a swimmer or wader might sink. • To evaluate indirect food-chain exposures, collect sediment from the bio- logically active zone. Surface Water • Collect from accessible locations where people are likely to fish, swim, or engage in other activities. • Collect from areas used as a drinking water source. • Measure total chemical concentrations if people ingest the water. Dis- solved-phase concentrations are more useful for some evaluations of dermal exposure. Source: Adapted from EPA 2000c. defined in the context of areas over which people average their exposure. For example, if the pathway of concern is direct contact with sediment, concentrations in sediment that are routinely beneath 10 ft of water are of less concern than concentrations in shallow, accessible waters at the shoreline. Cleanup levels should be compared with uncertainty bounds on average exposures, such as the 95% upper confidence limits of the mean concentration in each human exposure area, rather than the maxi- mum concentration detected in each exposure area.5 The conceptual site model and feasibility study results should be used to set expectations for the rate of risk reduction. To ensure that un- acceptable risks do not occur, site managers can track concentrations 5One caveat to the safety of long term averages and the relative unimportance of short term exceedances is that exposure can occur during a vulnerable period, such as fetal development or infancy.

Monitoring for Effectiveness 211 monitored over time to estimate expected cumulative cancer risk and noncancer hazards. If monitoring during or after dredging indicates that cleanup levels will not be met in the long run, site managers can use adaptive management to change this trend. Long-term Monitoring of Ecologic Effects A primary goal of long-term monitoring is to test the hypothesis that dredging has reduced injury to ecologic resources. Many of the techniques used to assess potential short-term adverse effects of dredg- ing on ecologic resources, described above, are also appropriate for as- sessing long-term effectiveness of sediment removal. For example, ben- thic toxicity testing has been successfully used as part of long-term monitoring at the former Ketchikan Pulp Company site in Alaska. In Waukegan Harbor, laboratory sediment toxicity assays conducted four years after dredging showed reduced toxicity compared to pre-dredging assessments, but nevertheless, toxicity still persisted (Ingersoll et al. 1996; EPA 1999; Kemble et al. 2000). Followup studies in the Black River showed that surficial sediments had reduced toxicity but PAHs were still causing toxicity in caged organisms (Burton and Rowland 1998). Con- taminant uptake and toxicity can also be quantified by using in situ ap- proaches with caged organisms or using passive sampling devices such as SPMDs, as discussed above. In addition to placing cages or SPMDs at different locations to quantify spatial distribution of contaminant con- centrations or effects, observations can take place over time to determine temporal changes in bioavailability and sediment toxicity. SPMDs were used in Manistique Harbor 4 years after dredging operations ceased. The SPMDs accumulated PCBs to detectable levels while PCBs were not de- tected in caged fish or surface water samples at the site (Weston 2005). Long-term monitoring of resident populations of fish and inverte- brates can also reveal changes in contaminant concentrations and ecolo- gic effects resulting from removal of contaminated sediment and natural processes. For example, the incidences of lesions and tumors in brown bullheads in the Black River showed initial increases and then marked reductions following dredging at the site (Baumann 2000) compared with pre-dredging conditions. Tissue data from fish whose habitat is lim- ited to the remediation site are valuable indicators because they integrate

212 Sediment Dredging at Superfund Megasites exposures over the remediation area. Several of the dredging projects (such as Waukegan Harbor, Grasse River, Black River, the Puget Sound Naval Shipyard, and GM Massena) that were evaluated by the commit- tee monitored fish tissue concentrations of contaminants to evaluate the effectiveness of sediment removal. Distinguishing the effects of remedia- tion from background trends on the basis of fish tissue data often proves problematic for the decision-making process because of the scarcity and variability of fish tissue data. In addition, the difficulty in quantifying the movements of fish in and out of a project area can make linkages be- tween exposure and effects problematic. It may be impossible to deter- mine how much time a fish has been exposed to study-site sediments compared with offsite sediments (which may also be contaminated). Bio- accumulation modeling approaches (e.g., Linkov et al. 2002) that include spatial and temporal characteristics of exposures based on a knowledge of the organism’s life history patterns can be useful in addressing that issue. Furthermore, pre-dredging data are often limited to a short time and very few fish (such as at Waukegan Harbor), so it might be impossi- ble to make statistically valid comparisons of trends (as discussed in the next section). In these situations, caged-fish studies can maximize expo- sure to test-site sediments and thereby reduce uncertainty. Fish may be the receptors of primary interest for both ecosystem and human health risk, but monitoring them supports effective decision-making only if suf- ficient samples are collected and their patterns of exposure are known. For determinations of ecosystem and human health risk, it is some- times more effective to monitor tissue concentrations in benthic inverte- brate organisms that reside at the test site (Adams et al. 2005; Burton et al. 2005; Solomon et al. 1997) because these organisms tend to be sessile or relatively immobile (for example, mussels). The organisms are ex- posed to the contaminated sediments through direct contact and are a food source for fish, birds, and mammals; therefore, food-web transfer and risk can be (and have been) modeled. The uncertainty of exposure is largely removed, and organisms are easier and less expensive to collect than fish and provide a convenient surrogate, as long as any assumed bioaccumulation link can be verified with site data. In addition, passive sampling devices that are biomimetic have recently been successfully used (see above discussion). The adsorption of organic and metal con- taminants on these devices has been shown to be similar to that of tissue concentrations in indigenous organisms, so they can be used as a surro-

Monitoring for Effectiveness 213 gate for fish (Arthur and Pawliszyn 1990; Huckins et al. 1990; Zhang et al. 1998; Wells and Lanno 2001; Lanno et al. 2004, 2005). Bioaccumulation and toxicity studies can also be conducted in the laboratory with sedi- ment and water collected from field sites after dredging operations. However, it is critical that laboratory studies consider abiotic factors that may influence contaminant bioavailability and degradation—such as ultraviolet light, suspended solids and colloids, and organic carbon— and the effect that removing sediments from the environment will have on bioavailability. Monitoring the structure and composition of benthic macroinverte- brate communities is a common approach to the assessment of effects of sediment contaminants and can be used when the benthic community is an important component of the conceptual site model of increased site risks. Such community characteristics as total abundance, species diver- sity, richness, and abundance of sensitive species can be compared with pre-dredging data and, when possible, with nearby reference sites. Again, it is critical that similar methods be used to collect and process pre-dredging and post-dredging samples. One of the greatest challenges associated with long-term monitoring of benthic communities is to sepa- rate effects of dredging from changes due to other environmental factors. The condition of benthic communities is generally expected to improve after the removal of contaminated sediment, as may be predicted by the conceptual site model. The failure of benthic communities to recover af- ter dredging could be a result of residual sediment contaminants, lack of colonizing organisms, conversion to an inhospitable or unsuitable habi- tat, or the presence of other stressors (Kelaher et al. 2003). Monitoring approaches using in situ cages that contain natural benthic communities offer an opportunity to demonstrate causal relationships between stress- ors and ecologically relevant responses. Demonstrating changes in the tolerance of populations or communities may also provide evidence of effectiveness of dredging in situations where traditional community metrics (such as abundance and species richness) do not show recovery. For example, increased tolerance to metals is often observed at metal- contaminated sites (Weis and Weis 1989; Clements 1999), so the loss of tolerance in a population or community after dredging is evidence that remediation was successful (Levinton et al. 2003). These experiments are a practical alternative to single-species toxicity tests and address the sta-

214 Sediment Dredging at Superfund Megasites tistical problems associated with field biomonitoring studies (Clark and Clements 2006). The rate of recovery of benthic communities after dredging will be determined by both biotic and abiotic factors (Yount and Neimi 1990). The rate of recovery will be influenced not only by the adverse effects of large-scale substrate disturbance and the presence of residual contami- nants, but also by proximity to reference areas and the availability of colonizing individuals. Ecosystems that have a direct connection to clean reference sites will probably recover faster than closed systems that have relatively little exchange. For example, the relatively fast recovery of stream ecosystems after remediation has been attributed to rapid coloni- zation by organisms from upstream reference sites (Clements and New- man 2002). DATA SUFFICIENCY AND STATISTICAL DESIGN Monitoring datasets should be rich enough to support testing of the hypothesis that dredging is effective in meeting its remedial goals and objectives. That requires that sampling targets the important exposure pathways and be designed to capture temporal and spatial variability and that sample sizes be sufficient for robust hypothesis-testing and sta- tistical modeling of dredging effectiveness goals. Standard statistical tests are often formulated as a null hypothesis representing no effect or no change vs an alternative hypothesis repre- senting an effect or change. When evaluating dredging effectiveness on the basis of pre- and post-dredging data, the null hypothesis represents no change due to dredging; the alternative hypothesis is that there was a change in environmental conditions because of dredging. Established formulas exist (EPA 2000d) for sample size determinations based on that traditional approach. With the required estimate of outcome variability (as can be obtained, for example, in a pilot study) and specification of the minimal effect size that should be detected, sample size determinations are based on optimizing the two types of statistical errors that can result. The probability of type I errors (incorrectly claiming dredging to be ef- fective when it was not) is fixed to be small, as is commonly done by set- ting the probability at 0.05. The probability of type II errors (failing to claim that dredging was effective when it really was) is minimized,

Monitoring for Effectiveness 215 maximizing statistical power (Mason et al. 1989). The approach is thus conservative; the burden of proof is on the monitoring data to provide enough data points to support dredging effectiveness with a high degree of confidence. One-sided alternative hypotheses can also be considered to test whether an effect or change was in a specific direction, such as a significant reduction in site conditions, and can consider varying mini- mal effect sizes, such as a reduction in site conditions of at least 90% from pre-dredging or related background values. The latter approach can be compared with hypothesis-testing techniques based on the bio- equivalence paradigm (McDonald et al. 2003). Sample size determination is crucial, but other components of sta- tistical experimental design should not be overlooked in developing monitoring plans to evaluate dredging effectiveness. A clear scientific definition of dredging effectiveness is needed so that appropriate statis- tical hypotheses can be formulated. Hypotheses to be tested should be based on and fully informed by the conceptual site model of exposures and risks, as developed in the remedial investigation and baseline risk assessment. Outcome variables need to be established, and their spatial and temporal support (where, when, and how much) should be deter- mined; all this should be consistent with and inform the statistical hy- potheses. Careful determination and measurement of potential sources of variation that may affect outcome variables are also important. Two sources of variation that deserve further focus are temporal and spatial variation in dredging effectiveness and their influence on monitoring and followup statistical analysis. It has been well established in this chapter and in the dredging pro- jects reviewed in the previous chapter that characterizing environmental conditions with monitoring before and after dredging is an important design consideration for evaluating dredging effectiveness. Less estab- lished are guidelines for determining when temporal characterizations should be assessed and whether assessment should follow a cross- sectional approach of one time before and one after dredging or be longi- tudinal and use multiple monitoring times before and after dredging. With just two time monitoring points before and after dredging (with multiple samples taken at each of these time points), one can determine whether a significant increase or decrease occurred between the two time points. When dredging (or another remedial action) takes place between these points, it is often assumed that the change results from the reme-

216 Sediment Dredging at Superfund Megasites diation, however, it is essential to consider trends that would occur re- gardless of dredging (for example, natural decreasing or increasing trends in contaminant concentrations in sediment or fish). The value of monitoring at several time points before and after dredging is that any trends not due to dredging can be determined and the effect of dredging more clearly established. Examples of fish tissue analyses that would have benefited from more complete time trend data are presented in Chapter 4 Boxes 4-1 and 4-6 on the Grasse River and Waukegan Harbor, respectively. Spatial or geographic variability can be an important component of overall variability to consider in designing monitoring plans to evaluate dredging effectiveness. There could be several reasons for spatial varia- tion in dredging effectiveness at a site. There could be naturally occur- ring variations in environmental factors, such as water flow, wind pat- terns, and sediment texture. There could be spatial variations in site conditions that affect the ability to dredge or dredge effectively, such as the presence of bedrock, harbor infrastructure, or debris. It is not only important to collect location information with monitoring data but to statistically inform the monitoring plan to determine appropriate loca- tions of monitoring samples. For example, in analysis of the Grasse River project (see Chapter 4), locations of sediment samples were too far apart to identify spatial variation in surficial PCBs. Analyses of the Lavaca Bay project (see Chapter 4) suggested that subarea variations in surface mer- cury were of interest, but samples were too small to test this hypothesis statistically while also considering temporal variation after dredge passes. The subfield of statistics known as geostatistics (for textbook treatments, see Cressie 1991; Goovaerts 1997; Diggle and Ribeiro 2007) deals with the design and analysis of spatially referenced data that commonly arise in monitoring of dredging applications and should in- form monitoring plans and data analysis. Even when spatial variation in dredging effectiveness is not of pri- mary interest, the data collected through monitoring may very well ex- hibit spatial dependence, that is, measurements of samples taken closer together are more similar than those of samples taken farther apart. Overlooking that property can result in hypothesis tests and statistical- model inference with biased levels of significance (Cressie 1991). Obtain- ing and including sample coordinates in monitoring databases will allow

Monitoring for Effectiveness 217 followup statistical analyses to include possible spatial dependence and make the appropriate adjustments when necessary. In the dredging projects reviewed, the committee found that the quantity and quality of available and accessible monitoring data varied considerably. Followup statistical analyses of monitoring data often were nonexistent or consisted of simple summaries and graphs lacking any formal notion of statistical uncertainty; this created a critical gap be- tween the large expenditures devoted to monitoring and the ability to provide scientifically defensible claims of dredging effectiveness based on monitoring data. It is imperative that rigorous statistical analysis of monitoring data be performed so that assessments of dredging effective- ness reflect the inherent uncertainties involved. APPROACHES TO IMPROVING MONITORING The dredging projects reviewed by the committee revealed limita- tions in the ability to make real-time adjustments in dredging operations to minimize contaminant releases; to connect remedial actions with their effects in space and time on exposure pathways, receptors, and ecosys- tems; to base monitoring on adequate conceptual site models of chemical fate and transport and of human health and ecologic risk; and to under- stand the roles of multiple processes in determining effectiveness of dredging. This section proposes approaches with promise to overcome those limitations. Each will require method development and evaluation before becoming part of the standard monitoring tool kit, and some may be appropriate only in particular cases. The methods can contribute to a weight-of-evidence basis of decision-making (Wenning et al. 2005) to reduce uncertainty in evaluation of risk reduction at specific sites. The topic of innovative monitoring methods is also reviewed by Viollier et al. (2003) and Apitz et al. (2005). Approaches with potential to improve site investigation and opera- tional and post-remedial monitoring include the following: • Measure sediment, pore water, and surface-water concentrations rapidly and accurately. • Monitor real-time contaminant releases during dredging. • Measure the bioavailable fraction of contaminants in the field.

218 Sediment Dredging at Superfund Megasites • Closely link exposure data (that is, chemical data) with biologic effect. • Understand and model biologic uptake. • Understand and model ecosystem response and recovery. • Understand and model reduction in human exposure. • Adequately and quickly identify generation, production, trans- port, and deposition of sediment residuals. • Understand and model processes responsible for recovery after dredging. • Quantitatively account for uncertainty in predictions of risk re- duction and in later monitoring. Measure Sediment Pore Water Concentrations Rapidly and Accurately A growing body of evidence suggests that sediment pore water concentrations are strong indicators of the effects of sediment- contaminant concentrations on benthic organisms (Adams et al. 1985; Di Toro et al. 1991; Jager et al. 2000; Kraaij et al. 2003; Wenning et al. 2005; Lu et al. 2006). Sediment pore water concentration is directly related to the amount of bioavailable contaminant and uptake by benthic organ- isms (McLeod et al. 2007). Current methods to measure sediment pore water involve the equilibration of sediment samples in the laboratory and extraction of equilibrated water or the use of biomimetic assays, as discussed above. Rapid techniques for measuring sediment pore water would provide more useful and timely information on the status of re- covery and resulting reduction in risk to humans and ecosystems. Method development and pilot testing are needed to determine how re- liably the available techniques can be adapted for laboratory and field conditions. Monitor Real-Time Contaminant Releases in the Field During Dredging Cost-effective methods are needed for real-time monitoring of con- taminant releases during dredging. At present, turbidity commonly is used as a surrogate for the release of persistent organic contaminants

Monitoring for Effectiveness 219 (such as PCBs) because the measurement is robust and quick and may be automated. It is often assumed that turbidity release is proportional to contaminant release. However, that may not be the case, as was shown in several of the Chapter 4 dredging project evaluations, because con- taminant fractionation between the aqueous phase and sediment parti- cles can result in releases of aqueous-phase (or colloid-associated) chemical contaminants, depending on the size and chemistry of the solid phase. Furthermore, if dredging exposed nonaqueous phases, such as liquid tar or hydraulic oil, contaminant release from such phases to over- lying water would not be related to the release of solids. In principle, some of the latest methods to measure contaminants in sediment pore water—such as ELISA, PEDs, and SPME—could be applied to monitor the release of contaminants to the aqueous phase if their detection limits prove adequate. The newer methods in conjunction with turbidity, to the extent that they are correlated, may facilitate more reliable and faster contaminant monitoring during dredging. Measure the Bioavailable Fraction of Contaminants in the Field A variety of potential methods are candidates for measuring the bioavailable fraction of organic and inorganic contaminants in sedi- ments. This information, for both baseline and post-remedial sampling, would complement chemical data to provide a more complete picture of changes in exposure due to the remedy. The National Research Council report on the bioavailabilty of contaminants in soils and sediments has a long and detailed chapter devoted to this topic (NRC 2003). However, most of the methods are not compound-specific or require detailed in- strumental methods of analysis. Field methods that are compound- specific are desired. One promising approach is immunoassay tech- niques for assessment of the bioavailable contaminant concentration. Contaminant-specific ELISAs (Johnson and Van Emon 1996) may be rapid, useful tools for measuring available contaminants. The immuno- assay uses the selectivity and sensitivity of antibody recognition coupled to an enzymatic reaction to rapidly determine chemical (such as PCB) concentrations in a variety of media, including wet sediment extracts and pore water. Ideally, the whole procedure, from extraction to colori-

220 Sediment Dredging at Superfund Megasites metric detection, could be carried out in the field with a test kit and port- able equipment [Ta 2001]. Understand and Model Biologic Uptake Site models that are used to support site investigations, remedy se- lection, and remedial monitoring should include a model of contaminant uptake by the affected biota. Biodynamic models describe the uptake of contaminants as a mass balance of uptake from water; uptake from food particles, including sediment; and loss rates. Such models would help to explain the relationship between level of sediment cleanup and concen- tration of contaminants in organisms. The typical bioenergetics-based toxicokinetic model (for example see Norstrom et al. 1976) assumes that uptake by each route is independent and additive. The model has been used to determine the uptake of contaminants by different routes with experimentally determined model parameter values (Boese et al. 1990; Weston et al. 2000; Lu et al. 2004). Luoma and Rainbow (2005) recently proposed biodynamics as a unifying concept in metal bioaccumulation, and similar formulations have been used in PCB food-web models (e.g., Connolly and Thomann 1992). It is proposed that a biodynamic model that integrates sediment, water, and organism data from field projects with the rapid assessment techniques described above be used to predict contaminant concentrations in several species of interest. For example, McLeod et al. (2007) showed that this biodynamic model successfully predicted PCB body burdens in the clam Macoma balthica exposed to un- treated and activated-carbon-amended Hunters Point sediment in labo- ratory experiments (see Box 5-9). Understand and Model Ecosystem Response and Recovery We lack rigorous modeling approaches to predict the ecologic characteristics of recovery after sediment cleanup. It is possible that an explanatory approach to ecologic recovery could build on the biody- namic modeling described above. If so, it could be incorporated into the conceptual site model and used to support the site investigation, remedy selection, and post-remedial monitoring. That belief is founded on the

Monitoring for Effectiveness 221 BOX 5-9 Biodynamic Modeling to Predict Organism PCB Concentrations If an organism is considered a single compartment for contaminant up- take, the following biodynamic equation describes its accumulation of a toxic contaminant (McLeod et al. 2007): dCorganism = FR ⋅ AE aq ⋅ Caq + IR ⋅ AE sed ⋅ Csed − k e ⋅ Corganism dt where, Corganism is the contaminant concentration in soft tissue (µg/g dry), FR is the water filtration rate (L of water per g dry per day), AEaq is the contaminant absorption efficiency from water, Caq is the aqueous contaminant concentration (µg/L), IR is the sediment-particle ingestion rate (g of sediment per g dry per day), AEsed is the contaminant absorption efficiency from sediment, Csed is the sediment contaminant concentration (µg/g dry), and ke is the proportional rate constant of loss (per day). Model parameters include organism and filtration and ingestion rates estimated from the literature. Sediment and aqueous contaminant concentrations would be measured in situ, and laboratory experiments would determine absorption-efficiency values and loss rates for the model organisms. The advantage of this approach is that once the organism parameters values are obtained, the conceptual model is transferable to other locations. fact that contaminants in sediments simplify community structure by eliminating some species but not others. Therefore, recovery should in- volve return of the contaminant-sensitive species to the community. In addition, benthic communities in estuaries are dynamic in space and time (Nichols and Thompson, 1985), so traditional ecologic observations should be frequent and detailed to resolve community recovery. Biody- namic modeling based on functional ecology may allow prediction of the species most sensitive to a contaminant, and this predictive capability will help biologists to identify which species from the available recruit- ment pool are likely to recolonize a site when the contaminant is re- moved or bioavailability is reduced. Recolonization of the contaminant- sensitive species in a recovering habitat reflects the success of remedia- tion. A hypothesis to be tested is that recolonization predictions can be built from basic information on taxon-specific functional ecology and

222 Sediment Dredging at Superfund Megasites biodynamics and contaminant metabolism, combined with data on spe- cies availability for community recruitment. Understand and Model Reduction in Human Exposure Monitoring programs should include measurement of surface sediment, surface water, edible aquatic species, and other environmental media found in the baseline risk assessment to present unacceptable human health risks through either direct or indirect exposure. Monitor- ing determines whether exposure concentrations have declined as pre- dicted. In addition, some systematic studies of the U.S. population, such as the National Health and Nutrition Examination Survey (NHANES), include biomonitoring data on some of the contaminants commonly de- tected at Superfund megasites. At the New Bedford Harbor Superfund site, members of the surrounding community have been studied, includ- ing collection of umbilical-cord serum and breast-milk samples, as part of an epidemiologic study of PCB effects on young children. Such human biomonitoring studies can be expensive and invasive and are not entirely without risk to those being monitored. Therefore, the committee does not recommend implementing human biomonitoring sampling for all dredg- ing projects. However, if relevant human biomonitoring data exist, they can be reviewed for evidence that dredging resulted in reduced human exposure and risk. Noninvasive biomonitoring might also improve fu- ture assessments of human exposure. For example, Fitzgerald et al. (2005) reported a significant correlation between a noninvasive test of enzyme activity related to PCBs and serum concentrations of PCBs in members of a Mohawk tribe living near the General Motors-Central Foundry Division Superfund site along the St. Lawrence River. Serum PCB concentrations in this population had previously been correlated with consumption of fish from the river (Fitzgerald et al. 1996, 1999, 2004). Adequately and Quickly Identify Generation, Production, Transport, and Deposition of Sediment Residuals The purpose of sediment verification sampling is to ascertain whether additional dredging passes, backfilling, or other remedial fol-

Monitoring for Effectiveness 223 lowup is needed to meet risk-based cleanup levels. Downtime for dredg- ing equipment and operators is expensive, but verification sampling and laboratory analysis can be slow and laborious and require dredgers to move on to other locations and return when results are available and have been reviewed. Methods include grab sampling, coring, and visual inspection by diver or with an underwater camera. Operator response to verification sampling is limited by best achievable laboratory- turnaround times and would be improved by de- velopment of more reliable field methods of analysis and greater use of mobile laboratories. In combination with the most rapid methods of analysis, the development of correlations between target chemical con- centrations and sediment physical properties, as may be reflected in sediment layering and other geomorphologic features, has the potential to streamline sampling and analysis and to provide cost efficiencies and much more rapid feedback to operators (Dow 2006). Understand and Model Processes Responsible for Recovery After Dredging As discussed above, a dredging remedy can affect surface sediment concentrations through a combination of sediment removal, backfilling, and enhancement of natural recovery processes by creating areas of preferential settling and deposition. Backfilling in particular was a com- ponent of the remedy at many of the sites considered in Chapter 4, but its risk-reduction efficacy is uncertain and probably depends on the na- ture and thickness of the backfill material. Although backfilling provides a separation layer between the water column and the contaminant, the effective attenuation of exposure by backfill material may be minimal if it is in a thin layer and has low adsorptive capacity. To understand the long-term effects of dredging remedies on risk, those issues should be evaluated as part of the monitoring program. Be- cause undredged residuals, generated residuals, and backfill material would be expected to differ in grain size and bulk density, these layers should be delineated, after dredging, through physical and chemical analysis of finely segmented cores. A time series of similar followup cor- ing data, as part of the long-term monitoring program, would suffice to distinguish the effects of post-dredging burial from those of backfilling.

224 Sediment Dredging at Superfund Megasites To estimate the combined effects of backfilling and burial on bioavail- ability, organic-chemical concentrations in surface layers should be or- ganic-carbon normalized, and acid-volatile sulfide analyses of metals should be conducted. Emerging field methods of pore water and bio- availability analysis should also be applied as they become more reliable and widely available. Account for Uncertainty Quantitatively in Predictions of Risk Reduction and Later Monitoring All risk assessments have inherent uncertainty in fate and transport modeling and quantification of exposure and toxicity. The goal of moni- toring should be to measure a given level of net risk reduction with a reasonable degree of confidence. By acknowledging uncertainty, one is better equipped to design an effective monitoring program and to an- swer questions from affected communities. For example, quantification of uncertainty may enable site managers to inform community members that the vapor-phase concentration of a chemical that will be released during dredging operations and reach the nearest neighborhood, on the basis of the best available modeling, is well below levels of concern at a specific high level of confidence. CONCLUSIONS The committee draws the following conclusions concerning moni- toring of dredging effectiveness in reducing risk: • Monitoring is the only way to evaluate the success of a remedy in reducing risk and is therefore an essential part of the remedy. • Trends that occur at these sites are subject to biologic, chemical, and physical processes that often operate on long time scales. The trends and processes may be best described and understood with long-term modeling and monitoring in pre-remedial and post-remedial time frames. • In the absence of sufficient baseline data, it is impossible to evaluate effectiveness. Where pre-dredging conditions are not static, a

Monitoring for Effectiveness 225 pre-remedial time-trend analysis is needed to judge remedial effective- ness. • In most cases reviewed by the committee, monitoring has not been adequately designed or implemented. Specifically, o The design of the monitoring has often not been linked sufficiently to the conceptual site model. o Tools developed for the remedial investigation, includ- ing numerical models and baseline risk assessments, are often neglected in formulating monitoring plans. o Baseline datasets have not always been consistent with long-term monitoring data. o Contaminant exposure and effects have not always been adequately linked in time and space. o In many cases, the quality and quantity of monitoring have been insufficient to support rigorous statistical analyses. • Some of the currently used monitoring techniques have proved useful in determination of short-term and long-term effects of remedia- tion. These include: o Monitoring during dredging, such as measurement of mass flux through upstream and downstream chemical monitoring and biologic monitoring, including caged- fish studies. o Long-term monitoring of fish tissue, where appropriate, and other pathways that contribute substantially to hu- man health risks. o Long-term monitoring of affected benthic communities, including tissue concentrations and health of benthic communities. o Laboratory toxicity testing using benthic organisms in sediment to monitor long-term changes following dredging. • If fish are exposed to offsite conditions, there is uncertainty as to their exposure and the relationship to risk. In those cases, benthic organ- isms may be better indicators of exposure, provided that their use is con- sistent with the conceptual site model of exposure pathways. If biologic testing is not possible, passive sampling biomimetic devices provide in- dications of contaminant exposures.

226 Sediment Dredging at Superfund Megasites RECOMMENDATIONS The committee offers the following recommendations for improv- ing monitoring of dredging effectiveness: • EPA should ensure that monitoring is conducted at all contami- nated sediment megasites to evaluate remedy effectiveness. That will require a commitment of resources commensurate with the scale and complexity of the site. • Monitoring plans should focus on elements required to judge ef- fectiveness and inform management decisions for the site. Care should be taken to select the correct indicators of ecologic or human risk care- fully. All aspects of monitoring—including planning, evaluation, and adaptive management based on monitoring findings—should be closely linked to the to conceptual site model so that the hypotheses and as- sumptions that led to the selected remedy can be tested and refined. • The breadth and richness of monitoring datasets should be suffi- cient to support the testing and full evaluation of effectiveness goals. Sta- tistical expertise should be included to inform well-designed monitoring programs, guide database development, and perform rigorous statistical analysis of monitoring data aimed at evaluating effectiveness. • EPA should ensure that monitoring information on all Super- fund megasites is systematically collected, organized, analyzed to assess the effectiveness of remediation, and made available to the public in such a form that effectiveness evaluations can be independently verified. • Numerical models that are used in the remedial investigation and feasibility study to design the remedy should be revisited during the remediation phase to help in the evaluation of the effectiveness of reme- diation. • If possible where combination remedies have been used, the relative contributions of dredging, capping and backfilling, and natural recovery should be measured through sediment monitoring, and the re- sults of monitoring should be used to adapt and optimize remedies. • Remediation decision makers should examine the expected net risk reduction associated with each remedial alternative before selecting a remedy that will be implemented and link the monitoring program to the assessment of net risk reduction for the selected remedy.

Monitoring for Effectiveness 227 • Pre-remediation baseline monitoring methods and strategies should be developed to allow statistically valid comparisons with future monitoring datasets that rely on time-series data. The ultimate goal is to assemble a consistent long-term dataset for conducting evaluations. Dur- ing preliminary and final remedy design, monitoring should be initiated to help to establish a time trend integrating earlier characterization data as technically appropriate. • Monitoring of the benthic community, as surrogate species re- flecting food-web transfer, can provide valuable information about eco- system health and integrate short-term and long-term exposures and multiple life stages. When consistent with the conceptual site model of exposure and risk development of site-specific approaches is recom- mended. • Faster and less expensive monitoring methods that are deploy- able in the field are needed to better inform dredging operations in real time and to improve predictive capability. REFERENCES Adams, R.G. 2003. Polyethylene Devices and the Effects of Sediment Resuspen- sion on the Cycling of PAHs and PCBs in the Lower Hudson Estuary. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA. Adams, W.J., R.A. Kimerle, and R.G. Mosher. 1985. Aquatic safety assessment of chemicals sorbed to sediments. Pp. 429-453 in Aquatic Toxicology and Hazard Assessment: Seventh Symposium, R.D. Cardwell, R. Purdy, and R.C. Bahner, eds. STP 854. Philadelphia, PA: American Society of Testing Materials. Adams, W.J., A.S. Green, W. Ahlf, S.S. Brown, G.A. Burton, B. Chadwick, M. Crane, R. Gouguet R, K.T. Ho, C. Hogstrand, T.B. Reynoldson, A.H. Ring- wood, J.D. Savitz, and P.K. Sibly. 2005. Using sediment assessment tools and a weight of evidence approach. Pp. 163-226 in Use of Sediment Quality Guidelines and Related Tools for the Assessment of Contaminated Sedi- ments, R.J. Wenning, G.E. Batley, C.G. Ingersoll, and D.W. Moore, eds. Pensacola, FL: SETAC Press. Anchor and LimnoTech. 2006a. Evaluation of Geostatistical Methods for Deline- ating Remediation Boundaries in Operable Unit 3, Lower Fox River Reme- dial Design, OU 2-5. February 28, 2006. Anchor and LimnoTech. 2006b. Evaluation of Geostatistical Methods for Deline- ating Remediation Boundaries in Operable Unit 4, Lower Fox River Reme- dial Design, OU 2-5. February 28, 2006.

228 Sediment Dredging at Superfund Megasites Anchor and LimnoTech. 2006c. Supplemental Information on Geostatistical Per- formance Metrics, Operable Units 3 and 4, Lower Fox River Remedial De- sign, OU 2-5. February 28, 2006. Apitz, S.E., J.W. Davis, K. Finkelstein, D.W. Hohreiter, R. Hoke, R.H. Jensen, J. Jersak, V.J. Kirtay, E.E. Mack, V.S. Magar, D. Moore, D. Reible, and R.G. Stahl, Jr. 2005. Assessing and managing contaminated sediments: Part 2. Evaluating risk and monitoring sediment remedy effectiveness. Integr. En- viron. Assess. Manage. 1(1):e1-e14. Arthur, C.L., and J. Pawliszyn. 1990. Solid phase microextraction with thermal desorption using fused silica optical fibers. Anal. Chem. 62(19)2145-2148. ASTM (American Society of Testing and Materials). 2006. Standard Test Method for Measuring the Toxicity of Sediment-Associated Contaminants with Es- tuarine and Marine Invertebrates. ASTM E1367-03. ASTM Annual Book of ASTM Standards, Vol. 11.06. American Society of Testing and Materials, West Conshohocken, PA. Barbour, M.T., J. Gerritsen, B.D. Snyder, and J.B. Stribling. 1999. Rapid Bio- assessment Protocols for Use in Streams and Wadeable Rivers: Periphyton, Benthic Macroinvertebrates and Fish, 2nd Ed. EPA 841-B-99-002. Office of Water, U.S. Environmental Protection Agency, Washington, DC [online]. Available: http://www.epa.gov/owow/monitoring/rbp/ [accessed Jan. 16, 2007]. Baumann, P.C. 2000. Health of Bullhead in an Urban Fishery after Remedial Dredging Final Report - January 31, 2000. Prepared for U.S. Environmental Protection Agency, Great Lakes National Program Office, Chicago IL, by U.S. Geological Survey Field Research Station, Ohio State University, Co- lumbus, OH [online]. Available: http://www.epa.gov/glnpo/sediment/ Bullhead/index.html [accessed Jan. 9, 2006]. BBL (Blasland, Bouck & Lee, Inc.). 1995. Non-Time-Critical Removal Action: Documentation Report, Vol. 1. Grasse River Study Area, Massena, New York, Aluminum Company of America, Massena, New York. Draft. Blasland, Bouck & Lee, Inc., Syracuse, New York. December 1995. Bergen, B.J., W.G. Nelson, J. Mackay, D. Dickerson, and S. Jayaraman. 2005. Envi- ronmental monitoring of remedial dredging at the New Bedford Harbor, MA, Superfund site. J. Environ. Monit. Assess. 111(1-3):257-275. Boese, B.L., H. Lee II, D.T. Specht, R.C. Randall, and M.H. Winsor. 1990. Com- parison of aqueous and solid-phase uptake for hexachlorobenzene in the tellinid clam, Macoma nasuta (Conrad): A mass balance approach. Environ. Toxicol. Chem. 9(2):221-231. Bohlen, W.F., and M.J. Erickson. 2006. Incorporating sediment stability within the management of contaminated sediment sites: A synthesis approach. Integr. Environ. Assess. Manage. 2(1):24–28.

Monitoring for Effectiveness 229 Booij, K., H.M. Sleiderink, and F. Smedes. 1998. Calibrating the uptake kinetics of semipermeable membrane devices using exposure standards. Environ. Toxicol. Chem. 17(7):1236-1245. Burton, G.A., Jr. 2002. Sediment quality criteria in use around the world. Limnology 3(2):65-76. Burton, G.A., Jr., and J.F. Nordstrom. 2004. An in situ toxicity identification evaluation method. Part II. Field validation. Environ. Toxicol. Chem. 23(12):2851-2855. Burton, G.A., Jr., and R. Pitt. 2002. Stormwater Effects Handbook: A Tool Box for Watershed Managers, Scientists and Engineers. Boca Raton, FL: Lewis [online]. Available: http://www.epa.gov/ednnrmrl/publications/books/ handbook/index.htm [accessed Jan. 16, 2007]. Burton, G.A., Jr. and C. Rowland. 1998. Assessment of Sediment Toxicity in the Black River Watershed. Final Report. Great Lakes National Program Office, U.S. Environmental Protection Agency, Chicago, IL [online]. Available: http://epa.gov/greatlakes/sediment/BlackRiver/Index.htm [accessed Jan. 16, 2007]. Burton, G.A., Jr., R. Pitt, and S. Clark. 2000. The role of traditional and novel tox- icity test methods in assessing stormwater and sediment contamination. CRC Crit. Rev. Environ. Sci. Technol. 30(4):413-447. Burton, G.A., Jr., P.M. Chapman, and E.P. Smith. 2002. Weight-of-evidence ap- proaches for assessing ecosystem impairment. Hum. Ecol. Risk Assess. 8(7):1675-1696. Burton, G.A., Jr., M.S. Greenberg, C.D. Rowland, C.A. Irvine, D.R. Lavoie, J.A. Brooker, L. Moore, D.F.N. Raymer, and R.A. McWilliam. 2005. In situ expo- sures using caged organisms: A multi-compartment approach to detect aquatic toxicity and bioaccumulation. Environ. Pollut.134(1):133-144. Chappie, D.J., and G.A. Burton, Jr. 2000. Applications of aquatic and sediment toxicity testing in situ. J. Soil Sediment. Contam. 9(3):219-246. Choi, A.L., J.I. Levy, D.W. Dockery, L.M. Ryan, P.E. Tolbert, L.M. Altshul, and S.A. Korrick. 2006. Does living near a Superfund site contribute to higher polychlorinated biphenyl (PCB) exposure? Environ. Health Perspect. 114(7):1092-1098. Clark, J.L., and W.H. Clements. 2006. The use of in situ and stream microcosm experiments to assess population and community-level responses to met- als. Environ. Toxicol. Chem. 25(9):2306-2312. Clements, W.H. 1999. Metal tolerance and predator-prey interactions in benthic macroinvertebrate stream communities. Ecol. Appl. 9(3):1073-1084. Clements, W.H., and M.C. Newman. 2002. Community Ecotoxicology. Chiches- ter, UK: Wiley.

230 Sediment Dredging at Superfund Megasites Conder, J.M., and T.W. La Point. 2005. Solid phase microextraction for predicting the bioavailability of 2.4,6-trinitrotoluene and its primary transformation products in sediment and water. Environ. Toxicol. Chem. 24(5): 1059-1066. Connolly, J.P., and R.V. Thomann. 1992. Modeling the accumulation of organic chemicals in aquatic food chains. Pp. 385-406 in Fate of Pesticides and Chemicals in the Environment, J.L. Schnoor, ed. New York: Wiley. Crane, M., G.A. Burton, J.M. Culp, M.S. Greenberg, K.R. Munkittrick, R. Ribeiro, M.H. Salazar, and S.D. St-Jean 2007. Review of aquatic in situ approaches for stressor and effect diagnosis. Integr. Environ. Assess. Manage. 3(2):234- 245. Cressie, N. 1991. Statistics for Spatial Data. New York: Wiley. Cullen, A.C., D.J. Vorhees, and L.M. Altshul. 1996. Influence of harbor contami- nation on the level and composition of polychlorinated biphenyls in pro- duce in Greater New Bedford, Massachusetts. Environ. Sci. Technol. 30(5):1581-1588. Dalton, Olmsted & Fuglevand, Inc. 2006. Remediation Action Construction Re- port, Part 1: Head of Hylebos Waterway Problem Area Commencement Bay Nearshore/Tideflats Superfund Site Tacoma, Washington, Review Draft, July 21, 2006. Prepared for Head of Hylebos Cleanup Group, Arkema, Inc, General Metals of Tacoma, Inc, by Dalton, Olmsted & Fugle- vand, Inc., Kirkland, WA. July 21, 2006. DeWitt, T.H., C.W. Hickey, D.J. Morrisey, M.G. Nipper, D.S. Roper, R.B. Wil- liamson, L. Van Dam, and E.K. Williams. 1999. Do amphipods have the same concentration-response to contaminated sediment in situ as in vitro? Environ. Toxicol. Chem. 18(5):1026-1037. Diggle, P.J., and P.J. Ribeiro, Jr. 2007. Model-based Geostatistics. New York: Springer. Di Toro, D.M., J.D. Mahony, D.J. Hansen, K.J. Scott, M.B. Hicks, S.M. Mayr, and M.S. Redmond. 1990. Toxicity of cadmium in sediments: The role of acid volatile sulfide. Environ. Toxicol. Chem. 9(12):1487-1502. Di Toro, D.M., C.S. Zarba, D.J. Hansen, W.J. Berry, R.C. Swartz, C.E. Cowan, S.P. Pavlou, H.E. Allen, N.A. Thomas, and P.R. Paquin. 1991. Technical basis for establishing sediment quality criteria for nonionic organic chemicals us- ing equilibrium partitioning. Environ. Toxicol. Chem. 10(12):1541-1584. Di Toro, D.M., J.D. Mahony, D.J. Hansen, K.J. Scott, A.R. Carlson, and G.T. Ankley. 1992. Acid volatile sulfide predicts the acute toxicity of cadmium and nickel in sediments. Environ. Sci. Technol. 26(1): 96-101. Dow (Dow Chemical Company). 2006. GeoMorph Sampling and Analysis Plan: Upper Tittabawassee River, The Dow Chemical Company, Michigan Op- erations, Midland, MI [online]. Available: http://www.deq.state. mius/documents/deq-whm-hw-dow-Text-Final%20GeoMorphSAP- 612006.pdf [accessed Jan. 17, 2007].

Monitoring for Effectiveness 231 Elzinga, C.L., D.W. Salzer, and J.W. Willoughby. 1998. Measuring and Monitor- ing Plant Populations. BLM Technical Reference 1730-1. BLM/RS/ST- 98/005+1730. Bureau of Land Management [online]. Available: http:// www.blm.gov/nstc/library/pdf/MeasAndMon.pdf [accessed Jan. 17, 2007]. Environment Canada. 1992. Biological Test Method: Acute Test for Sediment Toxicity Using Marine or Estuarine Amphipods. Report EPS 1/RM/26. Ottawa (ON): Environment Canada. Conservation and Protection [on-line]. Available: http://www.etc-cte.ec.gc.ca/organization/bmd/pubs/pubs_en/ 1RM26Englishfinal.pdf [accessed May 11, 2007]. EPA (U.S. Environmental Protection Agency). 1994. Methods for Assessing the Toxicity of Sediment-Associated Contaminants with Estuarine and Marine Invertebrates. EPA 600/R-94/025. Office of Research and Development, U.S. Environmental Protection Agency, Washington, DC. June 1994. EPA (U.S. Environmental Protection Agency). 1997. Report on the Effects of the Hot Spot Dredging Operations: New Bedford Harbor Superfund Site, New Bedford, MA. U.S. Environmental Protection Agency, Region 1. October 1997 [online]. Available: http://www.epa.gov/region01/superfund/sites/ newbedford/47203.pdf [accessed Jan. 18, 2007]. EPA (U.S. Environmental Protection Agency). 1998. Guidelines for Ecological Risk Assessment. EPA/630/R095/002F. Risk Assessment Forum, U.S. Envi- ronmental Protection, Washington, DC. April 1998 [online]. Available: http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=12460 [accessed Jan. 18, 2007]. EPA (U.S. Environmental Protection Agency). 1999. Evaluation of Toxicity and Bioaccumulation of Contaminants in Sediments Samples from Waukegan Harbor, Illinois. EPA-905-R-99-009. Great Lakes National Program Office, U.S. Environmental Protection Agency, Chicago, IL. October 1999 [online]. Available: http://www.epa.gov/glnpo/sediment/waukegan/whrpt.pdf [ac- cessed Jan. 9, 2007]. EPA (U.S. Environmental Protection Agency). 2000a. Methods for Measuring the Toxicity and Bioaccumulation of Sediment-Associated Contaminants with Freshwater Invertebrates, 2nd Ed. EPA/600/R-99/064. Office of Research and Development, U.S. Environmental Protection Agency, Duluth, MI, and Office of Science and Technology, U.S. Environmental Protection Agency, Washington, DC. March 2000 [online]. Available: http://www.epa.gov/ waterscience/cs/freshmanual.pdf [accessed May 11, 2007]. EPA (U.S. Environmental Protection Agency). 2000b. Equilibrium Partitioning Sediment Guidelines (ESGs) for the Protection of Benthic Organisms: Metal Mixtures (Cadmium, Copper, Lead, Nickel, Silver and Zinc). EPA 822R02045. Office of Science and Technology, U.S Environmental Protec- tion Agency, Washington, DC.

232 Sediment Dredging at Superfund Megasites EPA (U.S. Environmental Protection Agency). 2000c. Guidance for Assessing Chemical Contaminant Data for Use in Fish Advisories, Vols. 1 and 2. EPA 823-B-00-007. EPA 823-B-00-008. Office of Water, U.S. Environmental Pro- tection Agency, Washington, DC [online]. Available: http://www.epa. gov/waterscience/fish/guidance.html [accessed Jan. 18, 2007]. EPA (U.S. Environmental Protection Agency). 2000d. Guidance for Data Quality Assessment: Practical Methods for Data Analysis, EPA QA/G9, QA 00/Update. EPA/600/R-96/084. Office of Environmental Information, U.S. Environmental Protection Agency, Washington, DC [online]. Available: http://www.epa.gov/quality/qs-docs/g9-final.pdf [accessed Jan. 18, 2007]. EPA (U.S. Environmental Protection Agency). 2001a. Methods for Collection, Storage and Manipulation of Sediments for Chemical and Toxicological Analyses: Technical Manual. EPA-823-B-01-002. Office of Water, U.S. Envi- ronmental Protection Agency, Washington DC. October 2001 [online]. Available: http://www.epa.gov/waterscience/cs/toc.pdf [accessed Jan. 18, 2007]. EPA (U.S. Environmental Protection Agency). 2001b. Comprehensive Five-Year Review Guidance. EPA 540-R-01-007. Office of Emergency and Remedial Response, U.S. Environmental Protection Agency, Washington, DC. June 2001 [online]. Available: http://www.epa.gov/superfund/resources/5year/ guidance.pdf [accessed Aug. 3, 2006]. EPA (U.S. Environmental Protection Agency). 2003a. Procedures for the Deriva- tion of Equilibrium Partitioning Sediment Benchmarks (ESBs) for the Pro- tection of Benthic Organisms: Dieldrin. EPA/600/R-02/010. Office of Re- search and Development, U.S Environmental Protection Agency, Washington, DC [online]. Available: http://www.epa.gov/nheerl/public ations/files/dieldrin.pdf [accessed May 11, 2007]. EPA (U.S. Environmental Protection Agency). 2003b. Procedures for the Deriva- tion of Equilibrium Partitioning Sediment Benchmarks (ESBs) for the Pro- tection of Benthic Organisms: Endrin. EPA/600/R-02/009. Office of Research and Development,U.S Environmental Protection Agency, Washington, DC [online]. Available: http://www.epa.gov/nheerl/publications/files/endrin. pdf [accessed May 11, 2007]. August 2003. EPA (U.S. Environmental Protection Agency). 2003c. Procedures for the Deriva- tion of Equilibrium Partitioning Sediment Benchmarks (ESBs) for the Pro- tection of Benthic Organisms: PAH Mixtures EPA/600/R-02/013. Office of Research and Development, U.S Environmental Protection Agency, Wash- ington, DC [online]. Available: http://www.epa.gov/nheerl/publications/ files/PAHESB.pdf [accessed May 11, 2007]. EPA (U.S. Environmental protection Agency). 2004. Guidance for Monitoring at Hazardous Waste Sites: Framework for Monitoring Plan Development and Implementation. OSWER Directive 9355.4-28. Office of Solid Waste and

Monitoring for Effectiveness 233 Emergency Response, U.S. Environmental Protection Agency. January 2004 [online]. Available: http://www.epa.gov/superfund/action/guidance/dir 9355.pdf [accessed Jan. 18, 2007]. EPA (U.S. Environmental Protection Agency). 2005a. Contaminated Sediment Remediation Guidance for Hazardous Waste Sites. EPA-540-R-05-012. OSWER 9355.0-85. Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency. December 2005 [online]. Available: http://www.epa.gov/superfund/resources/sediment/pdfs/guidance.pdf [ac- cessed Dec. 26, 2006]. EPA (U.S. Environmental Protection Agency). 2005b. Procedures for the Deriva- tion of Equilibrium Partitioning Sediment Benchmarks (ESBs) for the Pro- tection of Benthic Organisms: Metal Mixtures (Cadmium, Copper, Lead, Nickel, Silver and Zinc). EPA 600/R-02/011. Office of Research and Devel- opment, U.S Environmental Protection Agency, Washington, DC. January 2005 [online]. Available: http://www.epa.gov/nheerl/publications/files/ metalsESB_022405.pdf [accessed May 11, 2007]. EPA/USACE (U.S. Environmental Protection Agency and U.S. Army Corps of Engineers). 1998. Evaluation of Dredged Material Proposed for Discharge in Waters of the U.S.—Testing Manual. EPA-823 -B-98 -004. U.S. Environ- mental Protection Agency, and Department of Army, U.S. Army Corps of Engineers. February 1998 [online]. Available: http://www.epa.gov/water science/itm/ITM/ [accessed Jan. 18, 2007]. Fitzgerald, E.F., K.A. Brix, D.A. Deres, S.A. Hwang, B. Bush, G. Lambert, and A. Tarbell. 1996. Polychlorinated biphenyl (PCB) and dichlorodiphenyl di- chloroethylene (DDE) exposure among Native American men from con- taminated Great Lakes fish and wildlife. Toxicol. Ind. Health 12(3-4):361– 368. Fitzgerald, E.F., D.A. Deres, S.A. Hwang, B. Bush, B.Z. Yang, A. Tarbell, and A. Jacobs. 1999. Local fish consumption and serum PCB concentrations among Mohawk men at Akwesasne. Environ. Res. 80(2Pt.2):97–103. Fitzgerald, E.F., S.A. Hwang, K. Langguth, M. Cayo, B.Z. Yang, B. Bush, P. Wor- swick, and T. Lauzon. 2004. Fish consumption and other environmental exposures and their associations with serum PCB concentrations among Mohawk women at Akwesasne. Environ. Res. 94(2):160–170. Fitzgerald, E.F., S.A. Hwang, G. Lambert, M. Gomez, and A. Tarbell. 2005. PCB Exposure and in vivo CYP1A2 activity among Native Americans. Environ. Health Perspect. 113(3):272-277. Fredette, T.J., D.A. Nelson, T. Miller-Way, J.A. Adair, V.A. Sotler, J.E. Clausner, E.B. Hands, and F.J. Anders. 1990. Selected Tools and Techniques for Physical and Biological Monitoring of Aquatic Dredged Material Disposal Sites. Technical Report D-90-11. U.S. Army Corps of Engineers, Waterways Experiment Station, Vicksburg, MS [online]. Available: http://stinet.dtic.

234 Sediment Dredging at Superfund Megasites mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADA22944 2 [accessed Jan. 17, 2007]. Goovaerts, P. 1997. Geostatistics for Natural Resources Evaluation. New York: Oxford University Press. Greenberg, M.S., G.A. Burton, Jr., and C.D. Rowland. 2002. Optimizing interpre- tation of in situ effects of riverine pollutants: Impact of upwelling and downwelling. Environ. Toxicol. Chem. 21(2):289–297. Hart, D.D. 1994. Building a stronger partnership between ecological research and biological monitoring. J. N. Am. Benthol. Soc. 13(1):110-116. Hofelt, C.S., and D. Shea. 1997. Accumulation of organochlorine pesticides and PCBs by semipermeable membrane devices and Mytilus edulis in New Bed- ford Harbor. Environ. Sci. Technol. 31(1):154 – 159. Huckins, J.N., M.W. Tubergen, and G.K. Manuweera. 1990. Semipermeable membrane devices containing model lipid: A new approach to monitoring the bioavaiiability of lipophilic contaminants and estimating their biocon- centration potential. Chemosphere 20(5):533-552. Huckins, J.N., G.K. Manuweera, J.D. Petty, D. Mackay, and J.A. Lebo. 1993. Lipid-containing semipermeable membrane devices for monitoring organic contaminants in water. Environ. Sci. Technol. 27(12):2489-2496. Ingersoll, C.G., P.S. Haverland, E.L. Brunson, T.J. Canfield, F.J. Dwyer, C.E. Henke, N.E. Kemble, D.R. Mount and R.G. Fox. 1996. Calculations and evaluation of sediment effect concentrations for the amphipod Hyalella az- teca and the midge Chironomus riparius. J. Great Lakes Res. 22(3):602-623. Ingersoll, C.G., S.M. Bay, J.L. Crane, L.J. Field, T.H. Gries, J.L. Hyland, E.R. Long, D.D. MacDonald, and T.P. O’Connor. 2005. Ability of SQGs to estimate ef- fects of sediment-associated contaminants in laboratory toxicity tests or in benthic community assessments. Pp. 497-556 in Use of Sediment Quality Guidelines and Related Tools for the Assessment of Contaminated Sedi- ments, R.J. Wenning, G.E. Batley, C.G. Ingersoll, and D.W. Moore, eds. Pensacola, FL: SETAC Press. Ireland, D.S., G.A. Burton, Jr., and G.G. Hess. 1996. In Situ toxicity evaluations of turbidity and photoinduction of polycyclic aromatic hydrocarbons. Envi- ron. Toxicol. Chem. 15(4):574-581. Jager, T., F.A. Antón Sánchez, B. Muijs, E.G. van der Velde, and L. Posthuma. 2000. Toxicokinetics of polycyclic aromatic hydrocarbons in Eisenia andrei (Oligochaeta) using spiked soil. Environ. Toxicol. Chem. 19(4):953-961. Johnson, J.C., and J.M. van Emon. 1996. Quantitative enzyme-linked immu- nosorbent assay for determination of polychlorinated biphenyls in envi- ronmental soil and sediment samples. Anal. Chem. 68 (1):162-169. Karr, J.R. 1993. Defining and assessing ecological integrity: Beyond water quality. Environ. Toxicol. Chem. 12(9):1521-1531.

Monitoring for Effectiveness 235 Kelaher, B.P., J.S. Levinton, J. Oomen, B.J. Allen, and W.H. Wong. 2003. Changes in benthos following the clean-up of a severely metal-polluted cove in the Hudson River Estuary: Environmental restoration or ecological distur- bance? Estuaries 26(6):1505-1516. Kemble, N.E., D.G. Hardesty, C.G. Ingersoll, B.T. Johnson, F.J. Dwyer, and D.D. MacDonald. 2000. An evaluation of the toxicity of contaminated sediments from Waukegan Harbor, Illinois, following remediation. Arch. Environ. Contam. Toxicol. 39(4):452-461. Korrick, S.A., and L. Altshul. 1998. High breast milk levels of polychlorinated biphenyls (PCBs) among four women living adjacent to a PCB- contaminated waste site. Environ. Health Perspect. 106(8):513-518. Kraaij, R., P. Mayer, F.J. Busser, M. van het Bolscher, W. Seinen, J. Tolls, and A.C. Belfroid. 2003. Measured pore-water concentrations make equilibrium par- titioning work: A data analysis. Environ. Sci. Technol. 37(2):268-274. Lanno, R.P., J. Wells, J. Conder, K. Bradham, and N. Basta. 2004. The bioavailabil- ity of chemicals in soil for earthworms. Ecotox. Environ. Safe. 57(1):39-47. Lanno, R.P., T.W. La Point, J.M. Conder, and J.B. Wells. 2005. Applications of solid-phase microextraction fibers as biomimetic sampling devices in ecotoxicology. Pp. 511-523 in Techniques in Aquatic Toxicology, Vol. 2, G. Ostrander, ed. Boca Raton: CRC Press. Leslie, H.A., T.L. ter Laak, F.J.M. Busser, M.H.S. Kraak, and J.L.M. Hermens. 2002. Bioconcentration of organic chemicals: Is a solid-phase microextrac- tion fiber a good surrogate for biota? Environ. Sci. Technol. 36(24):5399- 5404. Levinton, J.S., E. Suatoni, W. Wallace, R. Junkins, B. Kelaher, and B.J. Allen. 2003. Rapid loss of genetically based resistance to metals after the cleanup of a Superfund site. Proc. Natl. Acad. Sci. U.S.A 100(17):9889-9891. Linkov, I., D. Burmistrov, J. Cura, and T.S. Bridges. 2002. Risk-based manage- ment of contaminated sediments: Consideration of spatial and temporal patterns in exposure modeling. Environ. Sci. Technol. 36(2):238-246. Lohmann, R., R.M. Burgess, M.G. Cantwell, S.A. Ryba, J.K. MacFarlane, P.M. Gschwend. 2004. Dependency of polychlorinated biphenyl and polycyclic aromatic hydrocarbon bioaccumulation in Mya Arenaria on both water column and sediment bed chemical activities. Environ. Toxicol. Chem. 23(11):2551-2562. Long, E.R., and L.G. Morgan. 1990. The Potential for Biological Effects of Sedi- ment-Sorbed Contaminants Tested in the National Status and Trends Pro- gram. NOAA Technical Memorandum NOS OMA 52. National Oceano- graphic and Atmospheric Administration, Seattle, WA. March 1990 [online]. Available: http://www.ccma.nos.noaa.gov/publications/tm52.pdf [accessed May 11, 2007].

236 Sediment Dredging at Superfund Megasites Long, E.R., and D.D. MacDonald. 1998. Recommended uses of empirically de- rived sediment quality guidelines for marine and estuarine ecosystems. Hum Ecol Risk Assess. 4:1019-1039. Lu, X., D.D. Reible, and J.W. Fleeger. 2004. Relative importance of ingested sedi- ment versus pore water as uptake routes for PAHs to the deposit-feeding oligochaete Ilyodrilus templetoni. Arch. Environ. Contam. Toxicol. 47(2):207- 214. Lu, X., D.D. Reible, and J.W. Fleeger. 2006. Bioavailability of polycyclic aromatic hydrocarbons in field-contaminated Anacostia River (Washington, DC) sediment. Environ. Toxicol. Chem. 25(11):2869-2874. Luoma, S.N., and P.S. Rainbow. 2005. Why is metal bioaccumulation so variable? Biodynamics as a unifying concept. Environ. Sci. Technol. 39(7):1921-1931. MacDonald, D.D., C.G. Ingersoll, and T.A. Berger. 2000. Development and evaluation of consensus-based sediment quality guidelines for freshwater ecosystems. Arch. Environ. Contam. Toxicol. 39(1):20–31. Mason, R.L., R.F. Gunst, and J.L. Hess. 1989. Statistical Design and Analysis of Experiments with Applications to Engineering and Science. New York: Wiley. McDonald, L.L., S. Howlin, J. Polyakova, and C.J. Bilbrough. 2003. Evaluation and Comparison of Hypothesis Testing Techniques for Bond Release Ap- plications. Prepared for Abandoned Coal Mine Lands Research Program, University of Wyoming, Laramie, WY, by Western EcoSystems Technol- ogy, Inc., Cheyenne, WY, and Land Quality Division, Department of Envi- ronmental Quality, Cheyenne, WY. May 28, 2003 [online]. Available: http://www.west-inc.com/reports/aml.pdf [accessed Jan.18, 2007]. MA DPH (Massachusetts Department of Public Health). 1997. Housatonic River Area PCB Exposure Assessment Study, Final Report. September 1997. Bu- reau of Environmental Health Assessment, Environmental Toxicology Unit, Massachusetts Department of Public Health. McLeod, P.B., M.J. van den Heuvel-Grave, S.N. Luoma, and R.G. Luthy. 2007. Biological uptake of polychlorinated biphenyls by Macoma balthica from sediment amended with activated carbon. Environ. Toxicol. Chem. 26(5): 980-987. Miller, D.T., S.K. Condon, S. Kutzner, D.L. Phillips, E. Krueger, R. Timperi, V.W. Burse, J. Cutler, and D.M. Gute. 1991. Human exposure to polychlorinated biphenyls in greater New Bedford, Massachusetts: A prevalence study. Arch. Environ. Contamin. Toxicol. 20(3):410-416. Nelson, W. 2006. Long-term chemical and biological monitoring efforts at the New Bedford Harbor (NBH) Superfund site. Presentation at the Second Meeting on Sediment Dredging at Superfund Megasites, July 12, 2006, New Bedford, MA.

Monitoring for Effectiveness 237 Nichols, F.H., and J.K. Thompson. 1985. Time scale of change in the San Fran- cisco Bay benthos. Hydrobiologia 129(1):121-138. Norstrom, R.J., A.E. McKinnon, and A.S.W. DeFreitas. 1976. A bioenergetics- based model for pollutant accumulation by fish. Simulation of PCB and methylmercury residue levels in Ottawa River yellow perch (Perca flaves- cens). J. Fish. Res. Board Can. 33(2):248-267. NRC (National Research Council). 2001. A Risk-Management Strategy for PCB- Contaminated Sediments. Washington, DC: National Academy Press. NRC (National Research Council). 2003. Bioavailability of Contaminants in Soils and Sediments. Washington, DC: National Academies Press. Palermo, M. 2006. Overview of Dredging Equipment, Processes: Operating Methods and Strategies. Presentation at the First Meeting on Sediment Dredging at Superfund Megasites, March 22, 2006, Washington DC. Salazar, M.H., and S.M. Salazar. 1997. Using caged bivalves to characterize expo- sure and effects associated with pulp and paper mill effluents. Water Sci. Technol. 35(2-3):213-220. Shaw Environmental, Inc. 2006. Final Basis of Design Report: Lower Fox River and Green Bay Site, Brown, Outagamie, and Winnebago Counties, Wiscon- sin. Prepared for Fort James Operating Company, Inc., NCR Corporation, by Shaw Environmental, Inc. June 16, 2006 [online]. Available: http://www.dnr.state.wi.us/org/water/wm/foxriver/documents/BODR/Fina l_BODR_VolumeI.pdf [accessed Jan 18, 2007]. Solomon, K.R., G.T. Ankley, R. Baudo, G.A. Burton, Jr., C.G. Ingersoll, W. Lick, S.N Luoma, D.D. MacDonald, T.B. Reynoldson, R.C. Swartz, and W. War- ren-Hicks. 1997. Workgroup summary report on methodological uncer- tainty in conducting sediment ecological risk assessments with contami- nated sediments. Pp. 271-296 in Ecological Risk Assessments of Contaminated Sediments, C.G. Ingersoll, T. Dillon, and G. R. Biddinger, eds. Pensacola, FL: SETAC Press. Ta, N. 2001. Rapid Sediment Characterization of PCBs with ELISA: An Immuno- assay Technique a Rapid Sediment Characterization (RSC) Tool. TechData Sheet. TDS-2086-ENV. Naval Facilities Engineering Service Center, Port Hueneme, CA. June 2001 [online]. Available: http://costperformance.org/ monitoring/pdf/elisa_2.pdf [accessed Jan. 18, 2007]. Tucker, K.A., and G.A. Burton, Jr. 1999. Assessment of nonpoint source runoff in a stream using in situ and laboratory approaches. Environ. Toxicol. Chem. 18(12):2797-2803. Vinturella, A.E., R.M. Burgess, B.A. Coull, K.M. Thompson, and J.P. Shine. 2004. Use of passive samplers to mimic uptake of polycyclic aromatic hydrocar- bons by benthic polychaetes. Environ. Sci. Technol. 38(4):1154-1160. Viollier, E., C Rabouille, S.E. Apitz, E. Breuer, G. Chaillou, K. Dedieu, Y. Fura- kawa, C. Grentz, P. Hall, F. Janssen, J.L. Morford, J-C. Poggiale, S. Roberts,

238 Sediment Dredging at Superfund Megasites G.T. Shimmield, M. Taillefert, A. Tengberg, F. Wenzhoefer, and U. Witte. 2003. Benthic biogeochemistry: State of the art technologies and guideline for the future of in situ survey. J. Exp. Mar. Biol. Ecol. 285:5-31. Weis, J.S., and P. Weis. 1989. Tolerance and stress in a polluted environment: The case of mummichog. BioScience 39(2):89-95. Wells, J.B., and R.P. Lanno. 2001. Passive Sampling Devices (PSDs) as biological surrogates for estimating the bioavailability of organic chemicals in soil. Pp. 253-270 in Environmental Toxicology and Risk Assessment: Science, Policy, and Standardization – Implications for Environmental Decisions: Tenth Volume, B.M. Greenberg, R.N. Hull, M.H. Roberts, Jr., and R.W. Gensemer, eds. ASTM STP 1403. West Conshohocken, PA: American Soci- ety for Testing and Materials. Wenning, R.J., and C.G. Ingersoll. 2002. Summary of the SETAC Pellston Work- shop on Use of Sediment Quality Guidelines and Related Tools for the As- sessment of Contaminated Sediments, 17-22 August 2002; Fairmont, MT. Society of Environmental Toxicology and Chemistry (SETAC), Pensacola FL [online]. Available: http://www.cerc.usgs.gov/pubs/sedtox/WG0_SET AC_SQG_Summary.pdf. [accessed May 11, 2007]. Wenning, R.J., G.E., Batley, C.G. Ingersoll, and D.W. Moore, eds. 2005. Use of Sediment Quality Guidelines and Related Tool for the Assessment of Con- taminated Sediments. Pensacola. FL: Society of Environmental Toxicology & Chemistry Press. 783 pp. Weston, D.P., D.L. Penry, and L.K. Gulmann. 2000. The role of ingestion as a route of contaminant bioaccumulation in a deposit-feeding polychaete. Arch. Environ. Contam. Toxicol. 38(4):446-454 Weston (Weston Solutions Inc.). 2005. Final Manistique Harbor and River Site, Manistique, Michigan, Data Evaluation Report: Revision 1-19 May 2005. Document No. RWW236-2A-ATGN. Prepared for U.S. Environmental Pro- tection Agency, Chicago, IL. Wharfe, J., W. Adams, S.E. Apitz, R. Barra, T.S. Bridges, C. Hickey, and S. Ire- land. 2007. In situ methods of measurement-an important line of evidence in the environmental risk framework. Integr Environ. Assess. Manage. 3(2):268-274. Word, J.Q., B.B. Albrecht, M.L.Anghera, R. Baudo, S.M. Bay, D.M. Di Toro, J.L. Hyland, C.G. Ingersoll, P.F. Landrum, E.R. Long, J.P. Meador, D.W. Moore, T.P. O’Connor, and J.P. Shine. 2005. Predictive ability of sediment quality guidelines. Pp. 121-161 in Use of Sediment Quality Guidelines and Related Tools for the Assessment of Contaminated Sediments, R.J. Wenning, G.E. Batley, C.G. Ingersoll, and D.W. Moore, eds. Pensacola, FL: SETAC Press). You, J., P.F. Landrum, and M.J. Lydy. 2006. Comparison of chemical approaches for assessing bioavailability of sediment-associated contaminants. Environ. Sci. Technol. 40(20):6348-6353.

Monitoring for Effectiveness 239 Yount, J.D., and G.J. Niemi. 1990. Recovery of lotic communities and ecosystems from disturbance: A narrative review of case studies. Environ. Manage. 14(5):547-569. Zhang, H., W. Davison, B. Knight, and S. McGrath. 1998. In situ measurements of solution concentrations and fluxes of trace metals in soils using DGT. Envi- ron. Sci. Technol. 32(5):704-710.

Next: 6 Dredging at Superfund Megasites: Improving Future Decision-Making »
Sediment Dredging at Superfund Megasites: Assessing the Effectiveness Get This Book
×
Buy Paperback | $63.00 Buy Ebook | $49.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Some of the nation's estuaries, lakes and other water bodies contain contaminated sediments that can adversely affect fish and wildlife and may then find their way into people's diets. Dredging is one of the few options available for attempting to clean up contaminated sediments, but it can uncover and re-suspend buried contaminants, creating additional exposures for wildlife and people. At the request of Congress, EPA asked the National Research Council (NRC) to evaluate dredging as a cleanup technique. The book finds that, based on a review of available evidence, dredging's ability to decrease environmental and health risks is still an open question. Analysis of pre-dredging and post-dredging at about 20 sites found a wide range of outcomes in terms of surface sediment concentrations of contaminants: some sites showed increases, some no change, and some decreases in concentrations. Evaluating the potential long-term benefits of dredging will require that the U.S. Environmental Protection Agency step up monitoring activities before, during and after individual cleanups to determine whether it is working there and what combinations of techniques are most effective.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!