Sediment Dredging at Superfund Megasites ... - Emsus.com

Sediment Dredging atSuperfund Megasites:Assessing the EffectivenessKarl GustavsonArmy Engineer Researchand Development Center (ERDC)Federal Remediation Technologies RoundtableWashington DC - June 5, 2008

Committee Membership• Charles R. O’Melia (Chair),Johns Hopkins University• G. Allen Burton, Wright StateUniversity• William H. Clements,Colorado State University• Frank C. Curriero, JohnsHopkins Bloomberg School ofPublic Health• Dominic Di Toro, University ofDelaware• Norman R. Francingues, OASystems Corporation• Richard G. Luthy, StanfordUniversity• Perry L. McCarty, StanfordUniversity• Nancy Musgrove,Management of EnvironmentalResources, Inc.• Katherine N. Probst,Resources for the Future• Danny D. Reible, University ofTexas• Louis J. Thibodeaux,Louisiana State University• Donna J. Vorhees, TheScience Collaborative• John R. Wolfe, Limno-Tech,Inc.

Reviewers of the Final ReportReview Referees• George M. HornbergerUniversity of Virginia• C. Herb Ward, Rice UniversityReviewers• Todd S. Bridges, U.S. ArmyEngineer Research andDevelopment Center• Edwin H. Clark II, Earth PolicyInstitute• J. Paul Doody, BBL-Arcadis• Rick Fox, Natural ResourceTechnology, Inc.• Paul Fuglevand Dalton,Olmsted & Fuglevand, Inc.• John C. HenningsonHenningson EnvironmentalServices• Chris Ingersoll, U.S. GeologicalSurvey• Stephen U. Lester, Center forHealth & Environmental Justice• Jeffrey S. Levinton, StonyBrook University• Victor S. Magar, ENVIRON• Larry McShea, Alcoa• Steven C. Nadeau, HonigmanMiller Schwartz and Cohn, LLP.• Clay Patmont, AnchorEnvironmental, LLC.• Harlee S. Strauss, H. StraussAssociates, Inc.

Statement of TaskEvaluate the expected effectiveness of sediment dredgingin achieving risk reduction at Superfund megasites.The committee was charged to consider:• Short and long-term effects on ecologiccommunities• Site specific factors that contribute to or hindereffectiveness• Whether current monitoring technologies aresufficient and how they can be improved• How to improve sediment megasite decisionmakingin the future.

Outline• Introduction• Evaluation of Dredging Effectiveness• Monitoring at Sediment Megasites• Improving Future Decision-Making

Risk-based Decision-Making•Risk relates to exposures tocontaminants above acceptablelevels.•Remediation should be designed toeliminate or reduce those exposures.•Exposures are dictated by surfacesediments – contaminants buriedbelow the biologically-active zone arenot available to organisms.•Disrupting sediment beds to removeburied contaminants can exposeorganisms to otherwise inaccessiblecontaminants.•However, removal of deeply buriedcontaminants may thwart futureexposure and transport during severeevents.Historical changes in sediment core profilesof mercury concentrations in BellinghamBay, WA. Source: Patmont et al. 2004.

Committee’s Evaluation ofDredging Sites• The committee examined experiences at 26 dredgingprojects and evaluated whether, after dredging, thecleanup goals had been met.-Had a site achieved its quantitative cleanup levels (typically aspecified contaminant concentration in sediment).-Had a site achieved its remedial-action objectives (what thecleanup is expected to accomplish in the long term).-Factors that contributed to the success or adversely affecteddredging operations.• Various sites were examined, including full-scaledredging projects, pilot studies at sites, and dredgingprojects within a large-scale remediation effort.

26 Environmental-Dredging ProjectsSelected for Evaluation• Bayou Bonfouca LA• Lavaca Bay, TX• Black River, OH• Outboard Marine Corporation– Waukegan Harbor, IL• Commencement Bay–Head ofHylebos, Tacoma, WA• Commencement Bay–Sitcum,Tacoma, WA• Duwamish Diagonal, Seattle,WA• Puget Sound Naval Shipyard,Bremerton, WA• Harbor Island–Lockheed,Seattle, WA• Harbor Island–Todd, Seattle,WA• Cumberland Bay, NY• Dupont–Christina River, DE• Lower Fox River (OU-1), WI• Lower Fox River (Deposit N), WI• Lower Fox River (SMU 56/57), WI• Ketchikan Pulp Company, WardCove, AK• Newport Naval Complex–McCallister Landfill, RI• GM Central Foundry, St. LawrenceRiver, NY• Grasse River, NY (non-time-criticalremoval action)• Grasse River, NY remedial optionspilot study (ROPS)• Lake Jarnsjon, SwedenManistique Harbor, MI• Reynolds Metals, St. LawrenceRiver, NY• Marathon Battery, Hudson River,Cold Spring, NY• New Bedford Harbor, MA (hotspot)• United Heckathorn, Richmond, CA

Evaluation of Dredging Effectiveness• Operational goals (mass removal or dredging toelevation) were achieved at many sites.• Dredging is effective at removing large volumes ofcontaminated sediments from the environment.– This doesn’t necessarily equate to risk reduction, but can beuseful if those sediments are likely to be transported to lesscontaminated areas.• Dredging alone achieved desired contaminant-specificcleanup levels at only a few of the reviewed sites.– Capping after dredging was often necessary to achieve cleanuplevels.• At some sites, dredging achieved cleanup levels andpresumably contributed to declines in biota contaminantconcentrations.

Evaluation of Dredging EffectivenessDifficulties achieving low postdredgingsurface sedimentcontaminant concentrationsarise primarily from:• dredging’s limitations in fullyremoving contaminatedsediments.Courtesy of Marc Mills, US EPA - ORD

Evaluation of Dredging EffectivenessDifficulties achieving low postdredgingsurface sedimentcontaminant concentrationsarise primarily from:• site conditions that limitcomplete removal ofcontaminated sediments.The result is residualcontamination that hindersthe ability to achieve riskreduction.(Dalton, Olmsted & Fuglevand, Inc. 2006)

Evaluation of Dredging Effectiveness• Dredging can cause releases of contaminants duringoperations that increase water column and fish tissuecontaminant concentrations in the short term.(NRC 2007; data from Alcoa 2006)

Evaluation of Dredging Effectiveness• The longer-term benefits to human health and the environment ofdredging contaminated sediments are not well understood ordocumented.– Monitoring at most sites does not include the full array ofmeasures necessary to evaluate risk.– Dredging may have occurred in conjunction with otherremedies or natural processes.– Insufficient time may have passed to evaluate long-term riskreduction.• Due to these deficiencies, the committee was generally unable toestablish whether dredging alone is capable of achieving longtermrisk reduction.

So, Does Dredging Work or Not?• There is not a single answer to thisquestion.• Results will be site specific and depend onsite conditions, the adequacy of predredgingcharacterization of thoseconditions, and the implementation of theremedial action.• For example…14

Range of Experiences at Dredging ProjectsGrasse River, NY• Removed 24,600 cy in 2005.• Average surficial sediment PCBconcentrations increased afterdredging from 4.1 mg/kg to 150mg/kg. No cleanup level.• Intended to remove 64,000 cy; Ableto dredge 40% of the targeted area.Capped entire dredged areafollowing remediation.• Residuals: Average of 16 in. ofcontaminated sediments remainedafter dredging (range from 3 to 32in.)• Increased contaminantconcentrations in fish and waterduring dredgingHead of Hylebos, WA• Removed 404,000 cy from 2004 to2006.• Average surficial sediment PCBconcentrations decreased afterdredging from 0.69 mg/kg to 0.07mg/kg. Cleanup level: 0.3 mg/kg.• Sediments from targeted areas weresuccessfully dredged. Had to cap onearea due to remaining contamination.• Residuals: No summary statistics;Ranged from 0 to >12 in. Residualcontaminated sediments were belowcleanup level.• No sampling for contaminantconcentrations in water and in fish.15

Differences in Site Conditions Influence ResultsGrasse River, NY• Contaminated sedimentsunderlain by bedrock and hardpan(not “dredgeable”).• River channel deepened byblasting contains unevensubstrate (ridges, gulleys,scattered boulders) beneathsediments that make dredgeaccess difficult.• Difficult to characterize the depthof contaminated sediment usingstandard characterizationtechniques due to heterogeneity ofunderlying material.• Debris present; removed prior todredging operations.Head of Hylebos, WA• Contaminated sediments underlain bydense, sandy material (“dredgeable”).• Former tidal flat has relativelyfeatureless and homogeneoussubstrate underlying contaminatedsediments.• Relatively easy to characterize thedepth and extent of contaminationbased on historic channel depth andvisual differences betweencontaminated and clean sediment.• Debris present; removed prior todredging operations.16

Manistique River and Harbor•Removal was initially expected totake 3 seasons and remove 104,000cy – project took 6 seasons andremoved approximately 190,000 cy.•Issues encountered in characterizingthe extent of contaminated sedimentand completely removing sedimentresiding over fractured bedrock.•Cleanup level: average PCBconcentration of 10 mg/kg throughoutthe sediment column within the AOC.•Post-dredging average PCBconcentrations (top foot) throughoutthe site were 9.0 mg/kg (2000) and7.3 mg/kg (2001).

Manistique River and HarborPost-dredging Sedimentation•Following cessation of dredging in2000, there has been substantialsedimentation in the harbor.•Upwards of 83,000 cubic yardswere deposited in the harbor; insome places up to 5 meters thick.•This has buried residual PCBs; in2004 surface sediment averageconcentration was 0.88 mg/kg.•Indicates the potential importanceof other processes in combatingresidual contamination andachieving risk reduction.

Evaluation of Dredging EffectivenessRECOMMENDATIONS:• Remedies should be designed to meet long-term riskreductiongoals (as opposed to metrics not strictly related torisk, such as mass-removal targets).• Environmental conditions that limit or favor the effectivenessof dredging should be given major consideration in decidingwhether to dredge at a site.• Resuspension, release, and residuals will occur if dredging isperformed. Decision-making should explicitly consider thoseprocesses in expectations of risk reduction.• To reduce adverse effects, best-management practices thatlimit resuspension and residual contamination should be usedduring dredging. The ability of combination remedies tolessen the adverse effect of residuals should be considered.

Monitoring for EffectivenessCurrent monitoring techniques are generally adequate for describingrisk. However, they have not been adequately applied at manySuperfund sites to describe whether long-term risk reductionobjectives have been achieved.– Sparse or incomplete monitoring data were collected.– Selected organisms were not linked to remediation at the site.– Preremediation monitoring approaches were not consistentwith those used for long-term postremediation monitoring.– Pre-remediation trends were not of sufficient duration to enablejudging the effect of the remedial action.– Monitoring was of insufficient quality or quantity to supportrigorous statistical analyses.

Monitoring at Sediment MegasitesRECOMMENDATIONS:We need to do a better job evaluating the effectiveness of remedialactions at Superfund sites.• EPA should ensure that monitoring is conducted at all contaminatedsediment megasites to evaluate remedy effectiveness.• Descriptions of risk reduction require:– A pre-remedial baseline data set for comparison to post-remediationdata– Appropriate controls and reference sites– Consistent sampling and appropriate statistical power.• Monitoring data should be compiled and made publicly availablesuch that it’s possible to verify evaluations of remedial efficacyindependently.

Improving Future Decision-Making• The large spatial scale and long remedial timeframes ofcontaminated sediment megasites make it difficult topredict and quantify the human health and ecosystemrisk-reduction benefits achieved by isolated remediationsin a large-scale watershed.• The complexity and heterogeneity of large-scalemegasites suggest that a variety and combination ofremedial approaches will often be appropriate.• The committee’s retrospective analysis on ‘what worksand why’ in sediment remediation is an essential part ofa shared experience among regulators, practitioners,and the public. This type of effort needs to be ongoing.

Improving Future Decision-Making• An adaptive-management approach is essential to the selectionand implementation of remedies at contaminated-sedimentmegasites where there is a high degree of uncertainty about theeffectiveness of dredging.This approach:– replaces the paradigm of establishing all remedial actionsprior to implementation.– focuses on achieving objectives during a specified timeframe.– is based on the use of monitoring data from pilot studies andongoing remediation.– uses data collected during remedial operations to assess andimprove remedies to optimize progress towards remedialgoals.