$ < 73 \ r" pro^ 111 CD «T U.S. ENVIRONMENTAL PROTECTION AGENCY OFFICE OF INSPECTOR GENERAL Catalyst for Improving the Environment Evaluation Report EPA Should Improve Its Oversight of Federal Agency Superfund Reviews Report No. 10-P-0133 June 2, 2010 ------- Report Contributors: Carolyn Copper Steve Hanna Tapati Bhattacharyya Anne Emory Chad Kincheloe Kate Kimmel Brooke Shull Abbreviations CERCLA Comprehensive Environmental Response, Compensation, and Liability Act CERCLIS Comprehensive Environmental Response, Compensation, and Liability Information System DOD U.S. Department of Defense DOE U.S. Department of Energy EPA U.S. Environmental Protection Agency FFA Federal facility agreement FFRRO Federal Facilities Restoration and Reuse Office GAO U.S. Government Accountability Office NPL National Priority List OECA Office of Enforcement and Compliance Assurance OIG Office of Inspector General OSWER Office of Solid Waste and Emergency Response OU Operating unit RPM Remedial Project Manager SMP Site management plan TCE Trichloroethene ------- *. U.S. Environmental Protection Agency 10-P-0133 f \ Office of Inspector General June 2 2010 \ w ° At a Glance Why We Did This Review Prior studies have identified weaknesses in the Superfund 5-year review process. We evaluated how the U.S. Environmental Protection Agency (EPA) identifies and monitors issues and recommendations in Reviews conducted at federal facility Superfund sites. We also examined how EPA achieves compliance with unimplemented recommendations and nonconcurrence issues. Background EPA's Superfund 5-year review is a required process that examines the clean-up remedies at Superfund sites where hazardous substances remain at levels that may pose unacceptable risks. The Reviews are required every 5 years. They determine whether remedies adequately protect human health and the environment. Federal facility Superfund sites (e.g., military sites) complete their own Reviews and submit them to EPA. EPA is required to report to Congress on these Reviews. For further information, contact our Office of Congressional, Public Affairs and Management at (202) 566-2391. To view the full report, click on the following link: www.epa.qov/oiq/reports/2010/ 20100602-10-P-0133.pdf. Catalyst for Improving the Environment\ EPA Should Improve Its Oversight of Federal Agency Superfund Reviews What We Found EPA does not have effective management controls to monitor the completion of Review recommendations at federal government Superfund sites. For Reviews signed since 2006, 84 percent of Review recommendations were overdue as of April 28, 2009. EPA regional staff do not consistently follow Superfund 5-year review process guidance and policies for updating the status of Review issues and recommendations in the Comprehensive Environmental Response, Compensation, and Liability Information System (CERCLIS). Overdue or unimplemented recommendations to improve underperforming or nonperforming clean-up remedies may increase the risk to human health and the environment. Federal facilities are responsible for their own Reviews, and EPA states in a letter its concurrence or nonconcurrence with these Reviews. However, EPA's management of the concurrence process has resulted in some Reviews being conducted more than 5 years apart and some issues not being addressed. EPA has no management controls or policy for communicating, following up on, or resolving issues it does not agree with or it believes need improvement. Enforcement options are not clearly documented in EPA guidance or policy statements, and are not consistently understood or followed by staff. We also identified data quality problems. Discrepancies in the presentation of issues and recommendations exist between the Reviews and CERCLIS, some data in CERCLIS are logically inconsistent, and recommendations from prior Reviews are not always closed out. What We Recommend We recommend that EPA implement improved management controls to monitor the completion of federal facility Review recommendations, ensure Reviews are submitted every 5 years, improve the management of the nonconcurrence process, clarify and describe enforcement options to achieve completion of recommendations, and improve data quality. EPA initially agreed with all recommendations, except for recommendation 3-1. After additional discussions, EPA agreed with all recommendations and proposed actions to address them. All recommendations are open with agreed-to actions pending. In its final response to this report, EPA should provide estimated or actual completion dates for all recommendations. ------- ^t0 | \ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY 3 S WASHINGTON, D.C. 20460 \ c/ PRO^ OFFICE OF INSPECTOR GENERAL June 2, 2010 MEMORANDUM SUBJECT: FROM: TO: EPA Should Improve Its Oversight of Federal Agency Superfund Reviews Report No. 10-P-0133 Wade T. Najjum ~ J Assistant Inspector General J-! Office of Program Evaluation Cynthia Giles Assistant Administrator Office of Enforcement and Compliance Assurance Mathy Stanislaus Assistant Administrator Office of Solid Waste and Emergency Response This is our report on the subject evaluation conducted by the Office of Inspector General (OIG) of the U.S. Environmental Protection Agency (EPA). This report contains findings that describe the problems the OIG has identified and corrective actions the OIG recommends. This report represents the opinion of the OIG and does not necessarily represent the final EPA position. Final determinations on matters in this report will be made by EPA managers in accordance with established resolution procedures. The estimated cost of this report - calculated by multiplying the project's staff days by the applicable daily full cost billing rates in effect at the time - is $971,193. Action Required In accordance with EPA Manual 2750, you are required to provide a written response to this report within 90 calendar days. Your response should include a corrective action plan for ------- agreed-upon actions, including actual or estimated milestone completion dates. We have no objections to the further release of this report to the public. If you or your staff have any questions regarding this report, please contact Carolyn Copper, Director for Program Evaluation, Hazardous Waste Issues, at (202) 566-0829 or copper.carolyn@epa.gov; or Chad Kincheloe, Project Manager, at (312) 886-6530 or kincheloe.chad@epa.gov. ------- EPA Should Improve Its Oversight of Federal Agency Superfund Reviews 10-P-0133 Table of C Chapters 1 Introduction 1 Purpose 1 Background 1 Noteworthy Achievements 2 Scope and Methodology 3 Limitations 4 Prior Evaluation Coverage 5 2 EPA Needs a Management Control for Monitoring the Completion of Review Recommendations 6 Entry of Review Data in CERCLIS is Required 6 Recommendations Are Not Monitored for Completion 7 Conclusions 9 Recommendations 9 Agency Comments and OIG Evaluation 10 3 EPA Needs Additional Oversight and Enforcement Guidance for Reviews. .. 11 Lack of Controls and Guidance Impedes Compliance with Legal Requirements and Affects the Quality of Reviews 11 Guidance on Enforcement of Review Recommendations Needed 13 Conclusions 15 Recommendations 15 Agency Comments and OIG Evaluation 16 4 CERCLIS Data Quality Problems Impede Accurate Documentation of Issues and Recommendations 17 Discrepancies Exist Between Information in Published Reviews and in CERCLIS 17 CERCLIS Data Quality Issues Affect EPA's Ability to Monitor Reviews 18 Conclusions 19 Recommendations 19 Agency Comments and OIG Evaluation 19 Status of Recommendations and Potential Monetary Benefits 21 ------- EPA Should Improve Its Oversight of 10-P-0133 Federal Agency Superfund Reviews Appendices A OSWER Response to OIG Draft Report and OIG Response 23 B Distribution 32 ------- 10-P-0133 Chapter 1 Introduction Purpose The purpose of this project was to evaluate the U.S. Environmental Protection Agency's (EPA's) oversight of 5-year reviews conducted at Superfund National Priority List (NPL) sites that are also federal facility sites. These Reviews assess the performance of clean-up remedies to ensure the remedies adequately protect human health and the environment. We addressed the following questions: 1. How does EPA (Headquarters and regions) ensure that all federal facility issues and recommendations are identified and tracked? 2. Are EPA's actions to address unimplemented federal facility Review recommendations and nonconcurrence issues achieving federal facility compliance? Background The Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA, or Superfund) requires a review every 5 years at Superfund sites in which contaminants remain above levels that could pose unacceptable risks to human health and the environment. These reviews assess the performance of clean-up remedies to ensure the remedies protect human health and the environment. Reviews identify issues that affect current or future remedy protectiveness and may include recommendations to address issues for each operating unit (OU) at a site. Based on the issues identified, the protectiveness of each OU is characterized by one of five categories defined by EPA, listed in decreasing order of protectiveness: • Protective: Protective of human health and the environment. • Will be protective: Will be protective once the remedy is completed. • Protective in the short term: Protective in the short term; however, for the remedy to be protective in the long term, follow-up actions need to be taken. • Protectiveness deferred: Protectiveness cannot be determined until further information is obtained. • Not protective: Not protective unless specified actions are taken. EPA is required to report to Congress on the status of nonfederal and federal NPL site Reviews. In these reports, EPA defines the protectiveness of the site as not more protective than the least protective OU. 1 ------- 10-P-0133 Federal facilities are defined as the "buildings, installations, structures, land, public works, equipment, aircraft, vessels, and other vehicles and property, owned by, or constructed or manufactured for the purpose of leasing to, the Federal government." Federal facilities have the lead role and responsibility in the clean- up of their Superfund sites. In addition, federal facilities conduct their Superfund Reviews, prepare the Review reports, and submit their reports to the appropriate regional EPA office for review and comment. The federal facility is also responsible for ensuring that recommendations and follow-up actions identified during Reviews are completed. Federal agencies or departments fund Reviews at federal facilities. One of EPA's primary roles in federal facility Reviews is providing oversight to determine whether federal facility decisions protect human health and the environment. Through a concurrence process, EPA regions determine whether Review decisions are adequately supported. In this process, EPA regions examine federal facility Reviews and prepare a letter that either indicates concurrence with the Review or provides independent findings. EPA regions should enter the issues and recommendations identified by the Review into the Comprehensive Environmental Response, Compensation, and Liability Information System (CERCLIS). As of April 2009, there were 172 federal facility NPL sites (Table 1-1). The U.S. Department of Energy (DOE) and U.S. Department of Defense (DOD) are the two largest owners of federal facility Superfund sites, which include nuclear weapons plants, military bases, and fuel distribution stations. Examples of these sites are the DOE Hanford Nuclear Reservation and the DOD Rocky Mountain Arsenal. Federal facility Superfund sites may contain hazardous waste, unexploded ordnance, radioactive waste, or other toxic substances. According to EPA, federal facilities account for approximately one-half of the liability for Superfund clean-ups across the United States, including the largest single sites and sites with the widest varieties of contamination. Federal facility sites pose significant clean-up challenges. According to an EPA document, it may take up to 75 years and $400 billion to clean up federal facility Superfund sites. Noteworthy Achievements In 2006: • EPA implemented CERCLIS modifications to facilitate the input and monitoring of Review data. • The Federal Facilities Restoration and Reuse Office (FFRRO) of the Office of Solid Waste and Emergency Response (OSWER) issued a memorandum to the Regional Superfund National Policy Managers. This Table 1 -1: Federal Facility NPL Sites Federal NPL Sites Percent Agency DOD 140 81.4 DOE 21 12.2 Other 11 6.4 Total 172 100 Source: EPA. 2 ------- 10-P-0133 memorandum emphasized completion of the Reviews and follow-up of recommendations as an office priority. In 2007: • OSWER issued a memorandum that identified Review program priorities, including: > Improving the quality and consistency of Reviews. > Continuing to improve the timeliness of Reviews. > Tracking and implementing Review issues and recommendations. > Ensuring CERCLIS accurately reflects Review planning information, conclusions, and current progress on implementing recommendations. > Continuing to improve coordination between EPA Headquarters and regions. • EPA conducted Superfund Program Reviews designed to increase understanding of challenges with respect to 5-year reviews. These reviews continued into 2008. In 2009: • In September 2009, OSWER issued a document titled Five-Year Reviews, Frequently Asked Questions and Answers, to clarify EPA's 2001 Comprehensive Five-Year Review Guidance. • During the course of our evaluation, EPA regions updated CERCLIS Review recommendations, reducing overdue recommendations from 84 percent to 39 percent.1 Scope and Methodology We conducted our work from April 2009 to February 2010 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the evaluation to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our objectives. We assessed EPA's management controls for oversight of the completion of Review recommendations for NPL federal facility sites. We believe the evidence obtained provides a reasonable basis for our findings and conclusions based upon our objectives. We obtained and analyzed Review data for federal facility NPL sites from EPA's CERCLIS. Using CERCLIS, we identified the total number of sites and overdue recommendations in 2006-2008 Reviews and selected the 5 EPA regions (Regions 3, 4, 8, 9, and 10) with the most federal facility NPL sites and overdue recommendations. 1 We did not verify the accuracy of this reduction in overdue recommendations. 3 ------- 10-P-0133 To address Question 1, we requested information from and interviewed Remedial Project Managers (RPMs) and other staff involved in the Review process in the selected regions. We used this information and these interviews to identify the process by which recommendations are monitored, the reasons for overdue recommendations, and the reasons for discrepancies between Review and CERCLIS data. We interviewed regions based on the number of Review recommendations reported after 2006, focusing on recommendations that were overdue or that affected current or future protectiveness. We also conducted interviews with OSWER, FFRRO, and the Office of Enforcement and Compliance Assurance (OECA), Federal Facilities Enforcement Office to obtain information on EPA's oversight role. We reviewed sites that had determinations of "not protective," "protectiveness deferred," and "protective in the short term," to determine why recommendations or actions at these sites were overdue. We compared the protectiveness statements listed in CERCLIS with those in the published Reviews and in reports to Congress. We also reviewed the EPA response letters to federal facility Reviews to determine the Agency's concurrence. This analysis included the most recent concurrence letter for all 137 sites that filed a Review. To address Question 2, we reviewed EPA management controls for ensuring federal facility completion of Reviews and completion of the recommendations. Our working definition of a completed recommendation was one with a status of "complete," "addressed in the next review," or "considered but not implemented" in CERCLIS. We defined a recommendation as overdue if it had not been completed and the current planned completion date had passed. We requested information and interviewed Federal Facilities Enforcement Office to determine EPA's enforcement actions related to Reviews and unimplemented recommendations. We interviewed FFRRO to identify any management controls for monitoring Review recommendations. We reviewed 15 federal facility agreements (FFAs) signed since 2001 to determine whether the Reviews were defined as a primary document and therefore subject to dispute resolution and stipulated penalties. Limitations We relied on CERCLIS data for much of our analysis. We identify CERCLIS data quality issues germane to the objectives of this evaluation in Chapter 4 of this report. We did not independently verify that recommendations EPA regional staff designated as complete in CERCLIS were in fact complete. 4 ------- 10-P-0133 Prior Evaluation Coverage The following recent EPA Office of Inspector General (OIG) and U.S. Government Accountability Office (GAO) reports addressed issues related to the scope of our review. • EPA OIG, EPA Has Improved Five-Year Review Process for Superfund Remedies, But Further Steps Needed, Report No. 2007-P-00006, December 5, 2006. • GAO, Greater EPA Enforcement and Reporting Are Needed to Enhance Cleanup atDOD Sites, GAO-09-278, March 13, 2009. 5 ------- 10-P-0133 Chapter 2 EPA Needs a Management Control for Monitoring the Completion of Review Recommendations EPA does not have a management control for monitoring the completion of Review recommendations. EPA is not consistently updating the status of, or monitoring the completion of, Review recommendations as communicated in EPA guidance and emphasized by EPA management. When we began our review, CERCLIS data showed that 84 percent of the Review recommendations made since 2006 were overdue. Further, regions are not submitting annual status reports on Reviews as specified in guidance. Consequently, review recommendations may not be implemented in a timely fashion, or at all, and underperforming or nonperforming clean-up remedies may not be improved, potentially increasing the risk to human health and the environment. Entry of Review Data in CERCLIS is Required Both EPA policy and management communications require that progress on Review recommendations be monitored and entered into CERCLIS. EPA's 2001 Review Guidance states, "Regions should track the progress and completion of recommendations and/or follow-up actions with documentation in the site file, and upon completion update the administrative record in the site information repository." In addition, a 2007 OSWER memorandum states that EPA regions should maintain timely and accurate Review data to ensure CERCLIS accurately reflects Review planning information, conclusions, and current progress on implementing recommendations. Each Review issue and its associated recommendation that affects current or future protectiveness of remedies should be entered into CERCLIS by regional staff. EPA's Five-Year Review CERCLIS Manual discusses entering the following issue and recommendation data: • Issue and recommendation details. This information includes the category of the issue, such as remedy performance or changed site conditions, and the recommendation to address the issue. • Protectiveness indicators. For each issue, CERCLIS captures a separate entry indicating whether the issue affects both current and future protectiveness. 6 ------- 10-P-0133 • Milestone dates. Each recommendation contains three milestone dates: > The initial planned completion date, which should not be modified after entry. > The current planned completion date, which reflects any changes in the planned date. > The actual completion date. • Status. This is an indicator of the completion of the recommendation and includes values of "complete," "considered but not implemented," "ongoing," "under discussion," and "blank." Recommendations are considered complete when they are designated as complete, addressed in the next Review,2 or considered but not implemented. A recommendation is considered overdue if it is past its current planned completion date and the status is not complete. EPA staff have developed a number of internal CERCLIS reports that allow the data to be extracted and displayed in a variety of ways. These reports include the identification of overdue recommendations (i.e., recommendations not yet completed, with a past-due planned completion date). Recommendations Are Not Monitored for Completion EPA guidance affirms that Review recommendations should be monitored. However, EPA does not have a management control for this. Section 3.8 of EPA's 2001 Comprehensive Five Year Review Guidance states that each EPA region should report annually to EPA Headquarters on the progress of the Reviews for each of its sites. The guidance states the reports are to include, among other things, the following specific information: • For each completed Review, a summary of the protectiveness determination(s), issues that impact protectiveness, follow-up actions, and the schedule and entity responsible for implementing such actions; • Status of protectiveness when Review reports from previous fiscal years made a "not protective" determination or "needed further information" before making a protectiveness determination, or deferred protectiveness; and • Status of follow-up actions identified in Review reports from previous fiscal years. EPA Headquarters staff stated these reports have never been submitted. In response to a previous OIG report, EPA agreed that the annual reporting requirement in Section 3.8 of EPA's 2001 Review Guidance could be replaced by Review data in CERCLIS. EPA agreed in this previous report that regions would 2 While conducting this review, we used EPA's definition of a complete recommendation, which includes the status of "addressed in the next Review." This status means the recommendation is carried over and will be addressed during the next Review. No "addressed in the next Review" or "considered and not implemented" recommendations were found in the April 28, 2009, data. 7 ------- 10-P-0133 be required to keep CERCLIS up to date with Review data. However, this recommendation has not been implemented as agreed. As of April 28, 2009, 84 percent of 222 recommendations made in Reviews during the 2006-2008 period were overdue and not designated as "complete" in CERCLIS. During the course of our review, EPA regional staff updated CERCLIS data, reducing the percent of overdue recommendations to 39 percent (as of August 26, 2009).3 According to FFRRO staff, twice each year Headquarters informally distributes to EPA regions a CERCLIS Review report that identifies overdue recommendations. This report could serve as part of an effective management control to monitor the completion of recommendations. However, this report was not distributed to any of the regional federal facility branch chiefs. EPA Headquarters staff advised that the distribution procedure for the report would be updated to include regional federal facility contacts. The lack of consistent updates to CERCLIS Review information may also be affected by regional variations in monitoring and data input. Methods of regional monitoring included status reports received from the federal facility and spreadsheets, tables, or action item lists maintained by the RPM. According to Region 9 staff, they use an "outstanding recommendation" report that is managed at the section chief level. In contrast, an RPM in another region was unaware that Review recommendations and issues were monitored, and still another RPM stated that monitoring Review recommendations was not a priority because of competing demands for resources. However, this RPM did state that regional staff meet frequently with the federal facility to discuss recommendations. Irrespective of the method used, recommendations were overdue for sites in all regions we selected for review. The issues identified in Reviews can be significant to human health and the environment, underscoring the need to monitor the completion of Review recommendations. Specific examples of issues from recent Reviews are: • Ogden Defense Depot, Utah > After 11 years of operation, the groundwater clean-up remedy has not been effective in reducing vinyl chloride concentrations to public safety standards. > Workers at the Site could be exposed to chemical vapors from contaminated groundwater underneath buildings if concrete floors are not intact and impervious. 3 We did not verify the accuracy of this reduction in overdue recommendations. 8 ------- 10-P-0133 • Fort Ord, California > The long-term protectiveness of an area at the site that has a history of trespassing has been deferred because all explosives have not yet been remediated. • Eielson Air Force Base, Alaska > Ingestion of fish with high levels of polychlorinated biphenyls, or PCBs, may pose unacceptable health risks. > The extent of trichloroethene (TCE) groundwater contamination is unknown. • Fairchild Air Force Base, Washington > TCE has been detected at increasing levels in an offsite groundwater monitoring well. This condition may not be related to the remedial action, but could indicate another source of TCE contamination. This condition has been determined to affect future site protectiveness. Conclusions EPA lacks a management control for monitoring the progress and completion of Review recommendations. Regions are not updating CERCLIS information with the status of Review recommendations as required. EPA is not using CERCLIS consistently to monitor recommendations. Regions are not providing annual updates on Reviews as recommended in guidance, and Headquarters is not providing oversight to ensure submittal of these updates. Methods used to monitor recommendations vary by region and RPM. Improvements are needed to ensure that recommendations are completed in a timely fashion to assure remedy performance and overall protectiveness of federal facility NPL sites. Recommendations We recommend the Assistant Administrator for Solid Waste and Emergency Response: 2-1 Develop and implement a management control for monitoring completion of federal facility Review recommendations. 2-2 Implement policies and procedures to address Section 3.8 of EPA's 2001 Review Guidance, including (1) a summary for each Review completed during the year, (2) an update of the status of OUs from sites designated as "not protective" or "protectiveness deferred" in Reviews from prior years, and (3) the status of follow-up actions identified in Reviews from prior years. To the extent that this can be accomplished through the use of CERCLIS, specific reports should be implemented and monitored. 9 ------- 10-P-0133 Agency Comments and OIG Evaluation OSWER agreed with Recommendations 2-1 and 2-2 and it proposed acceptable corrective action plans to address them. Recommendations 2-1 and 2-2 are open with agreed-to actions pending. In its final response to this report, OSWER should provide estimated or actual completion dates for these recommendations. Although our review was limited to NPL federal facility sites, in its response OSWER said it would work to ensure that its other offices are aware of the recommendations and, if needed, will implement processes to improve on monitoring, tracking, and completing recommendations listed in all Reviews. Appendix A provides the complete Agency comments, including follow-up responses, and the OIG response. 10 ------- 10-P-0133 Chapter 3 EPA Needs Additional Oversight and Enforcement Guidance for Reviews EPA's management of the concurrence process has resulted in some Reviews being conducted more than 5 years apart and issues potentially not being addressed. EPA uses its concurrence date as the Review completion date instead of the date that the federal facility submits its Review. This practice can cause delays in identifying potential problems with site clean-ups. EPA has no management controls or policy for communicating, following up on, or resolving issues it does not agree with or it believes need improvement. Enforcement options are not clearly documented in EPA guidance or policy statements and are not consistently understood or followed by staff. These conditions can impede the effectiveness of the Review process to protect human health and the environment, demonstrating the need for guidance on enforcement and oversight. Lack of Controls and Guidance Impedes Compliance with Legal Requirements and Affects the Quality of Reviews CERCLA requires Reviews every 5 years. However, current EPA guidance establishes the due date for federal facility Reviews as 5 years from the date of EPA's concurrence letter for the preceding Review. Because EPA has not implemented any controls over the concurrence process, EPA regions may issue concurrence letters several months, or years, after a Review is submitted. This delay in the issuance of concurrence letters results in some Reviews being issued beyond the 5-year legal requirement. Late Reviews Occur Because of Delays in EPA Concurrence Twenty-nine percent of the Reviews completed4 since 2006 were late by more than 6 months, measured from the date of the prior review (Figure 3-1). Seven of the Reviews (13 percent) were more than 1 year overdue, and one Review was overdue by almost 3 years. Reviews were more than 1 year overdue in part because of EPA delays in signing the concurrence letter for the Review. According to EPA Regional staff, signed concurrence letters were delayed because Reviews were submitted late, Reviews were incomplete or in draft form and had to be resubmitted, or EPA disagreed with the Reviews and time was needed to resolve disagreements. 4 A Review is "completed" as of the Review date entered into CERCLIS, which is the date EPA signs its concurrence letter. 11 ------- 10-P-0133 EPA has no policy or guidance to define when federal facility Reviews should be submitted to EPA to allow for timely concurrence review within the 5-year timeframe. The site management plan (SMP) could potentially be used to minimize delays in EPA concurrence with the Review. An SMP provides a list of steps and a summary of CERCLA response actions and associated work products required at each federal facility. The SMP provides milestones that reflect the schedule of completing CERCLA response actions that have been agreed to by the lead agency and the regulatory agencies. The SMP could include a date for submittal of a draft Review by the federal facility, which allows EPA time to review the document and resolve disagreements prior to the Review's statutory deadline. While the target date could serve as guidance, EPA could not enforce the SMP's date. However, EPA could enforce the statutory deadline if the federal agency failed to submit the Review within 5 years of initiation of the remedial action. EPA's Concurrence Process Lacks Other Management Controls Agency guidance states that EPA has final authority over whether Reviews adequately address the protectiveness of remedies. Agency guidance also states that EPA will concur with federal agency Reviews or EPA may provide independent findings. The concurrence process provides a control for monitoring the quality of the Reviews and associated recommendations to ensure overall remedy protectiveness. However, EPA lacks policy and procedures for its concurrence process. Specific issues are: • No guidance on timeframes and process for concurrence review. EPA has no guidance on when federal agencies should submit their Reviews to EPA regions, how long EPA's review should take, or what the process for review is and whether and what internal EPA coordination is needed. Lack of clear policies in this matter may compromise the validity of the reviews. According to EPA's SuperfimdProgram Reviews 2007-2008 Cycle, some regions stated they will only provide feedback on a federal facility Review if they receive it with enough time to do so before the final due date. Figure 3-1: Timeliness of 5-Year Reviews Signed After 2006 On time (19 sites, 35%) Late more than 6 Late within 6 months months (16 sites, 29%) (20 sites, 36%) Source: OIG analysis of CERCLIS data. 12 ------- 10-P-0133 • Lack of standards for concurrence letters. > Inconsistent content: Letters take various forms. Letters for 12 Reviews consisted only of EPA signatures with no statement of concurrence. > Unclear nonconcurrence: Some letters did not include EPA statements that the Agency agreed or disagreed with the federal facility's protectiveness statement. EPA did not concur with 14 of 137 federal facility Reviews. However, EPA stated its nonconcurrence in only 3 letters. The other 11 letters included an EPA revised protectiveness statement, but not an EPA nonconcurrence statement. • No follow-up and resolution of nonconcurrence issues. EPA has no controls to prioritize and monitor the resolution of issues included in nonconcurrence letters. The lack of controls over the concurrence process can result in inconsistency in how disagreements are communicated and resolved, or whether disagreements and concerns are disclosed by EPA. • Some letters missing from the online record. EPA provides access to Reviews through the Internet. However, EPA's concurrence letter is not always included in the online federal facility Review record. As a result, EPA staff and the general public may not be able to ascertain the Agency's concurrence or nonconcurrence. Guidance on Enforcement of Review Recommendations Needed Guidance can support EPA's ability to enforce the completion of recommendations at federal facilities, which may be necessary to maintain site protectiveness. According to EPA, its enforcement authority over federal facility NPL sites is limited by CERCLA. EPA's ability to take CERCLA enforcement actions at a federal facility is determined by the wording of the FFA.5 FFAs are negotiated agreements developed separately for each federal facility. FFAs vary in the wording that provides EPA the ability to enforce the submittal of a Review or the completion of recommendations. EPA Regions Have Differing Views of Their Enforcement Options Implementation of Review recommendations can be necessary to ensure site protectiveness. However, EPA regions have differing views on their ability to enforce Review recommendations at federal facilities. In addition, EPA Headquarters is uncertain of the enforceability of the recommendation milestones. Two of the five regions interviewed stated they can use the Superfund Record of 5 An exception is EPA's authority related to imminent and substantial endangerment, which is a worst-case scenario. U.S. Department of Justice concurrence is necessary for EPA to issue a CERCLA order to another federal agency in cases of imminent and substantial endangerment. 13 ------- 10-P-0133 Decision to enforce the completion of Review recommendations that affect site protectiveness. However, two other regions reported they have little enforcement authority because the federal facility is the lead agency. Staff in some regions are not aware of the tools they can use to enforce the completion of recommendations. For instance, staff in one region believed that they could enforce Review recommendations through the FFA as long as the issue related to protectiveness. Staff in another region believed that Review dates could not be enforced unless they were listed as primary documents in the FFA. Staff in two of the five regions interviewed believed that EPA could not enforce the implementation of Review recommendations because EPA is not the lead agency. These differing opinions among regions may be due to a lack of enforcement guidance - EPA has not clearly identified the set of enforcement tools available to the regions to assist in the timely completion of Review recommendations. Reviews Should Be but Are Not Consistently Enforceable under Existing Federal Agreements A component of EPA's Review enforcement capability is based on the definition of the Review as a "primary document" in the FFA. Inclusion as a primary document is dependent upon the consent of the federal agency, as EPA lacks the authority to compel DOD to include language in a consensual agreement. However, Reviews are not consistently listed as primary documents in FFAs. Designation as a primary document in the FFA generally provides authority for formal dispute resolution as well as the assessment of stipulated penalties. Enforcement options vary by site, and the specifics of the process are defined in each site's FFA. If an issue is not resolved informally, then the formal process is decided by the EPA Administrator. According to OECA staff, completion of Reviews may be enforced by EPA under at least one of three conditions: 1. Reviews are defined as primary documents in FFAs. EPA's2001 Review Guidance states, "For Federal facilities only, EPA considers Review reports to be stand-alone primary documents or part of another related primary document that should have an enforceable schedule within the framework of the FFA." 2. Reviews are defined as a "periodic assessment" in DOD FFAs. DOD began referring to a Review as a "periodic assessment" in the FFA in 2003 and does not classify these as primary documents. However, OECA stated that periodic assessments should be enforceable as a term or condition of the FFA. 3. FFAs should act in accord with CERCLA. In the FFA, the federal facility commits to implementing a remedy in accordance with CERCLA. 14 ------- 10-P-0133 Since the Review is required under CERCLA, failure to complete the Review is a violation of the FFA. Because of the importance that OECA staff placed on the wording of FFAs, we examined the 15 FFAs signed since 2001. Both of the non-DOD FFAs identified the Review as a primary document consistent with EPA's guidance. Of the 13 DOD FFAs, almost all (12) identified the Review as a periodic assessment. Therefore, all of the FFAs with the Review identified as a periodic assessment should be enforceable. Conclusions EPA lacks a management control to ensure EPA's concurrence letters are timely, consistent, and transparent. EPA's policy of basing future Reviews on the date of the concurrence letter has resulted in submissions that violate the statutory requirement for a Review every 5 years. EPA's ability to enforce the submission of Reviews and the completion of recommendations is unclear. EPA lacks guidance on Review enforcement tools. Improvement in EPA's ability to enforce Review submission and the completion of recommendations would enhance the effectiveness of the Review process, leading to improved remedy performance and protection of human health and the environment. Recommendations We recommend the Assistant Administrator for Solid Waste and Emergency Response: 3-1 Establish a policy to set the due date for Reviews every 5 years, consistent with law. Current guidance sets the due date as 5 years from EPA's concurrence date for the prior review. 3-2 Develop and consistently implement a transparent and public mechanism to identify, monitor, and resolve EPA nonconcurrence with federal facility Reviews. This should include policies and procedures for conducting concurrence reviews, addressing the lead time needed for EPA regional review prior to the due date, and defining a standard format for concurrence letters. 3-3 Develop a management control to monitor overdue Reviews. 15 ------- 10-P-0133 We recommend the Assistant Administrator for Enforcement and Compliance Assurance: 3-4 Develop enforcement policy that describes EPA enforcement tools, processes, and authorities to achieve completion of Review recommendations, as well as the timely submittal of complete Reviews. As appropriate, the guidance should include the desired wording in the FFA and SMP to support this enforcement. Agency Comments and OIG Evaluation The OIG made changes to the report based on the Agency's comments where appropriate. OSWER provided the official Agency response, which incorporated OECA's comments. OSWER agreed with recommendations 3-1, 3-2, and 3-3. However, in its official response to the draft report, OSWER disagreed with recommendation 3-1. OSWER stated that it believed its current policy can achieve the intent of this recommendation, and proposed a joint memorandum to the regions to emphasize the policy. The OIG did not agree with this proposed corrective action and met with OSWER to discuss the recommendation and alternatives. OSWER subsequently agreed with recommendation 3-1 and provided an acceptable corrective action plan. In its final response to this report, OSWER should provide estimated or actual completion dates for recommendations 3-1,3-2, and 3-3. These recommendations are open with agreed-to actions pending. OECA agreed with Recommendation 3-4. However, the draft report response did not provide a clear corrective action plan to address the recommendation. After follow-up discussion, OECA provided an updated response to recommendation 3-4 that presented an acceptable corrective action plan. Appendix A provides the full text of the Agency comments, including follow-up responses, and the OIG's response. 16 ------- 10-P-0133 Chapter 4 CERCLIS Data Quality Problems Impede Accurate Documentation of Issues and Recommendations EPA lacks management controls to ensure that CERCLIS information accurately reflects the issues and recommendations reported in Reviews, which can impact EPA's efforts to monitor progress. Discrepancies exist between the Review issues and recommendations and those entered in CERCLIS. In addition, other inconsistencies are introduced into CERCLIS Review data after initial entry. CERCLIS lacks a method to confirm that recommendations from prior Reviews have been addressed. EPA should have controls for accurate entry of issues and recommendations into CERCLIS, and controls for timely updates to Review information to ensure that issues and recommendations are accurately recorded, documented, and understood by EPA staff that rely on CERCLIS data. Discrepancies Exist Between Information in Published Reviews and in CERCLIS In the regions where we conducted interviews, the RPMs or other designated regional staff enter data on issues, recommendations, and protectiveness determinations from the Review into CERCLIS. Recommendation data, such as completion status, Figure 4-1: Comparison of Recommendations in Reviews milestone dates, and and CERCLIS comments, can be updated to reflect changes. In a sample of 36 Reviews completed after 2006 that we selected, we identified a total of 495 recommendations that appeared in the Reviews, in CERCLIS, or in both (Figure 4-1). Only 238 recommendations were present in both CERCLIS and the Reviews. We could not find in CERCLIS 136 (36 percent) of the 374 recommendations that appeared in our sample of Reviews. In addition, we could not identify in our sample of Reviews 121 (34 percent) of the 359 recommendations in CERCLIS. For 14 of the 36 Reviews (39 percent) we selected, all recommendations that were in the published Reviews were also in CERCLIS. 5-Year Review Recommendations CERCLIS Recommendations Source: OIG analysis of CERCLIS Reviews after 2006. 17 ------- 10-P-0133 In some cases, the descriptions of the recommendations in CERCLIS were so different from the descriptions in the Review that it was difficult to determine whether all recommendations were entered in CERCLIS. This analysis was further complicated for some sites by the use of different OU naming conventions in the Review and CERCLIS. CERCLIS Data Quality Issues Affect EPA's Ability to Monitor Reviews We encountered multiple data quality problems that can affect EPA's ability to use CERCLIS to monitor the completion of Review recommendations. These include: • Review data modified after entry. A May 2007 memorandum on Five- Year Review Program Priorities states OSWER's intent to ensure that CERCLIS accurately reflects Review planning information, report conclusions, and current progress on implementing recommendations. The memorandum further states that CERCLIS should reflect the original planned completion date, while allowing the update of the recommendation status, current planned date, and comments if the planned date changes. We observed that changes were made to the original planned completion date and other Review data to reflect changes in the recommendation status, rather than to the status, current planned date, and comments. This misunderstanding of how Review data in CERCLIS should be maintained should be corrected. • Some CERCLIS data logically inconsistent. We observed several types of logical inconsistencies regarding the entry of Review data into CERCLIS. These issues include: > Protectiveness designation inconsistent with other data. Each Review issue includes a Yes/No (Y or N in CERCLIS) determination as to whether the issue affects current or future protectiveness. Of the 48 sites filing Reviews after 2006, 11 are designated in CERCLIS as "protective" but have issues that affect current and/or future protectiveness. Conversely, 7 sites have OUs listed in CERCLIS as "less than protective" but have no issues listed that affect either current or future protectiveness. The 2001 EPA guidance does not address the use of current and future protectiveness in determining OU protectiveness. > Sitewide protectiveness defined incorrectly. Sites with more than one OU that have completed construction of the remedy are required to identify the sitewide protectiveness in CERCLIS. According to EPA's 2001 guidance, a site is considered protective if the remedies at all OUs are protective, but if at least one OU is not protective, the site is not protective. We found seven federal 18 ------- 10-P-0133 facility sites with a sitewide protectiveness that is inconsistent with the protectiveness of the OUs.6 • Data from prior years not always closed out. Although the 2001 EPA guidance states that the Review should address the status of recommendations from the prior Review, regional staff we interviewed did not understand how these recommendations should be closed out in CERCLIS. Conclusions Significant inconsistencies between issues and recommendations in the Review submitted by the federal facility and data entered in CERCLIS affect the ability of EPA to effectively monitor the completion of recommendations that address site protectiveness. CERCLIS data quality problems compromise its potential as a tool for monitoring Reviews. Recommendations We recommend the Assistant Administrator for Solid Waste and Emergency Response: 4-1 Develop a management control to ensure consistency between the Review and CERCLIS. At a minimum, this control should ensure that all Review recommendations that affect current or future protectiveness are entered into CERCLIS for monitoring, consistent with EPA's 2001 guidance. 4-2 Develop guidance and/or CERCLIS controls to address data quality issues, including the modification of data after entry, logical inconsistencies between CERCLIS data elements, and the close-out of recommendations from prior reviews. Agency Comments and OIG Evaluation OSWER agreed with Recommendations 4-1 and 4-2. However, for Recommendation 4-1, the OIG requested additional information from OSWER on its proposed corrective plan, which was provided. OSWER's proposed corrective actions for recommendations 4-1 and 4-2 are acceptable. These recommendations are open with agreed-to actions pending. In its final response to this report, EPA should provide estimated or actual completion dates for these recommendations. 6 The seven sites and their Review year are Aberdeen Proving Ground (Michaelsville Landfill) 2003; Aberdeen Proving Ground (Michaelsville Landfill) 2009; Castle Air Force Base 2004; Fort Richardson 2008; Joliet Army Ammunition Plant (Load Assembly Area) 2004; Naval Air Station, Whidbey Island (Ault Field) 2004; Naval Air Station, Whidbey Island (Seaplane Base) 2004. 19 ------- 10-P-0133 Appendix A provides the complete Agency comments, including follow-up responses, and the OIG response. 20 ------- 10-P-0133 Status of Recommendations and Potential Monetary Benefits RECOMMENDATIONS POTENTIAL MONETARY BENEFITS (In $000s) Rec. No. Page No. Subject Status1 Action Official Planned Completion Date Claimed Amount Agreed To Amount 2-1 9 Develop and implement a management control for monitoring completion of federal facility Review recommendations. 2-2 9 Implement policies and procedures to address Section 3.8 of EPA's 2001 Review Guidance, including (1) a summary for each Review completed during the year, (2) an update of the status of OUs from sites designated as "not protective" or "protectiveness deferred" in Reviews from prior years, and (3) the status of follow-up actions identified in Reviews from prior years. To the extent that this can be accomplished through the use of CERCLIS, specific reports should be implemented and monitored. 15 Establish a policy to set the due date for Reviews every 5 years, consistent with law. Current guidance sets the due date as 5 years from EPA's concurrence date for the prior review. 15 Develop and consistently implement a transparent and public mechanism to identify, monitor, and resolve EPA nonconcurrence with federal facility Reviews. This should include policies and procedures for conducting concurrence reviews, addressing the lead time needed for EPA regional review prior to the due date, and defining a standard format for concurrence letters. 3-1 3-2 Assistant Administrator for Solid Waste and Emergency Response Assistant Administrator for Solid Waste and Emergency Response Assistant Administrator for Solid Waste and Emergency Response Assistant Administrator for Solid Waste and Emergency Response 3-3 15 Develop a management control to monitor overdue Reviews. 3-4 16 Develop enforcement policy that describes EPA enforcement tools, processes, and authorities to achieve completion of Review recommendations, as well as the timely submittal of complete Reviews. As appropriate, the guidance should include the desired wording in the FFA and SMP to support this enforcement. 4-1 19 Develop a management control to ensure consistency between the Review and CERCLIS. At a minimum, this control should ensure that all Review recommendations that affect current or future protectiveness are entered into CERCLIS for monitoring, consistent with EPA's 2001 guidance. Assistant Administrator for Solid Waste and Emergency Response Assistant Administrator for Enforcement and Compliance Assurance Assistant Administrator for Solid Waste and Emergency Response 21 ------- 10-P-0133 RECOMMENDATIONS POTENTIAL MONETARY BENEFITS (In $000s) Rec. No. Page No. Subject Status1 Action Official Planned Completion Date Claimed Agreed To Amount Amount 4-2 19 Develop guidance and/or CERCLIS controls to address data quality issues, including the modification of data after entry, logical inconsistencies between CERCLIS data elements, and the close-out of recommendations from prior reviews. Assistant Administrator for Solid Waste and Emergency Response 1 0 = recommendation is open with agreed-to corrective actions pending C = recommendation is closed with all agreed-to actions completed U = recommendation is undecided with resolution efforts in progress 22 ------- 10-P-0133 Appendix A OSWER Response to OIG Draft Report and OIG Response OIG Note: Page numbers referenced in the Agency response refer to the Draft Report. Page numbers may not be the same in this final report. April 29, 20107 MEMORANDUM SUBJECT: OSWER's Response to the Office of the Inspector General Draft report, "EPA should Improve Its Oversight of Federal Agency Superfund Reviews" FROM: Mathy Stanislaus/s/ Assistant Administrator TO: Carolyn Copper Director for Program Evaluation Hazardous Waste Issues Thank you for the opportunity to review and respond to the Office of the Inspector General's (OIG) Draft Evaluation Report entitled "EPA Should Improve Its Oversight of Federal Agency Superfund Reviews." We also appreciate the meetings with your staff to discuss your findings prior to issuance of the draft report. Our response incorporates the Office of Enforcement and Compliance Assurance (OECA) and Regional comments. An important part of the context for this work is the Executive Order and applicable appropriations. Under the Executive Order, most federal departments have served as lead agency for Superfund remediation at their own facilities, and appropriations are structured accordingly. Thus, EPA's role is oversight of work led by the responsible federal department. We agree that EPA needs to implement improved management controls to monitor the completion of federal facility five-year review (Review) recommendations. We believe that the majority of the OIG recommendations can be implemented in CERCLIS and through a joint OSWER/OECA policy memorandum to the Regions. The Agency does not agree with the implication raised at several points in the report that the lack of full management controls may result in increased risk to human health and the environment. We suggest that it would be more appropriate to say that "Improvements are needed to ensure that recommendations are completed in a timely fashion so that remedy performance and protectiveness are documented." 7 Additional information provided by OECA on May 18, 2010, and by OSWER on May 19, 2010. 23 ------- 10-P-0133 As a follow-up to the OIG evaluation, OSWER and OECA will issue a joint memorandum to the Regions that will address the recommendations in the draft report. This memorandum will explain changes in how Review recommendations are tracked, monitored, and implemented. The memorandum will provide clarification to the Regions on issues that affect present and future protectiveness of the remedy. We will also explain the enforcements options and tools available to the Regions to help ensure that five-year recommendations are being implemented. We also understand that the universe of sites for your review was limited to NPL federal facility sites. Therefore, we also understand that the recommendations submitted in your draft report only pertain to federal facility sites. I will work to ensure, however, that other OSWER offices are aware of the recommendations and, if needed, will implement processes to improve on monitoring, tracking, and completing recommendations listed in all Reviews. Our attached response includes comments on the findings, agreement/disagreements with the recommendations, planned corrective action, and an action plan for implementing the recommendations. Our second attachment includes technical comments that should be considered. If you have any questions regarding this response, please contact John Reeder at (703) 603-9089. Attachments 24 ------- 10-P-0133 Attachment 1 EPA Response to OIG Recommendations Recommendation 2-1 Develop and implement a management control for monitoring completion offederal facility Review recommendations. EPA Response We agree with this recommendation. FFRRO will begin to address the Review recommendations during our mid-year and end-of-the-year planning meetings with the Regions. Twice a year, we will send e-mails to the Regional Federal Facility Managers with a list of recommendations that have not been completed or are overdue. Our workplanning meetings occur in March and August. We will continue to monitor the implementation of the recommendation. OIG Response: The OIG agrees with the Agency's proposed actions in response to Recommendation 2-1. The recommendation is open with agreed-to actions pending. In its final response to this report, EPA should provide estimated or actual completion dates for this recommendation. Recommendation 2-2 Implement policies and procedures to address Section 3.8 of EPA 's 2001 Review Guidance, including (1) summary for each Review completed during the year, (2) an update on the status of OUs from sites designated as "not protective " or "protectiveness deferred" in Reviews from prior years, and (3) the status offollow-up actions identified in Reviews from prior years. To the extent this can be accomplished through the use of CERCLIS, specific reports should be implemented and monitored. EPA Response We agree with this recommendation. OSWER believes that there are existing reports within CERCLIS that can implement this recommendation. FFRRO will use the mid-year and end-of- the-year planning meetings as an opportunity for the Regions to provide updates on the Reviews. In the OSWER and OECA policy memorandum, FFRRO will include expectations for reporting requirements. OIG Response: The OIG agrees with the Agency's proposed actions in response to Recommendation 2-2. The recommendation is open with agreed-to actions pending. In its final response to this report, EPA should provide estimated or actual completion dates for this recommendation. 25 ------- 10-P-0133 Recommendation 3-1 Establish a policy to set the due date for Reviews every five years, consistent with the law. Current guidance sets the due date as 5 years from EPA 's concurrence date for the prior Review. EPA Response We do not agree with this recommendation as drafted. We believe that our current policy, if properly implemented, will achieve the intent of the OIG recommendation and is consistent with the law. We believe that dates are missed because of the late reports submitted by the federal agency or the continuous negotiations with federal agencies on the protectiveness determination for a site. Our joint OSWER/OECA memorandum to the Regions will emphasize that the review time should be based on what is indicated in the Federal Facility Agreement (FFA) for documents and that if agreement can not be reached, then the Regions should make an independent protectiveness determination statement. EPA Revised Response (5/19/2010) We agree with the intent of the recommendation. Starting in FY 2011, all Federal Facility five- year reviews will be due on the date five years from the remedial action start date. We will work with the Regions to ensure that both the Federal Agencies and the Regions know that the completion date for a five-year review is five years from the remedial action start date. The OSWER/OECA policy memorandum will reiterate to the Regions that they should communicate to the other federal agencies that review time is needed and should be added to the schedule for the completion of the report. FFRRO will continue to emphasize that five-year reviews must be completed on time and due dates will not be adjusted if reports are late. The issue of due dates will be addressed in the policy memorandum. ()l(» Response: In follow-up communication, the Agency agreed with the recommendation, and stated that it will require that all federal facility 5-year re\ iews he due on the date 5 years from the remedial action date The Agency stated it will issue an OSWI-R ()!¦( A policy memorandum to the regions to reiterate this policy The OICi accepts this conecti\e action plan The recommendation is open with agreed-to actions pending In its final response to this report. I-PA should pro\ ide estimated or actual completion dales for this recommendation. Recommendation 3-2 Develop and consistently implement a transparent and public mechanism to identify, monitor, and resolve EPA nonconcurrence with federal facility Reviews. This should include policies and procedures for conducting concurrence reviews, addressing the lead time neededfor EPA regional review prior to the due date, and define a standardformat for concurrence letters. 26 ------- 10-P-0133 EPA Response We agree with this recommendation. As stated before, OSWER and OECA will develop a joint memorandum that will address procedures for conducting the federal facility Reviews and how to resolve nonconcurrence with the Reviews. As an attachment to the policy memorandum, OSWER and OECA will provide examples of concurrence and nonconcurrence letters. During our quarterly meetings with the other Federal Agencies, OSWER will add a standing agenda topic of Reviews and tracking and implementing the recommendations stated in the Reviews. Once the OSWER and OECA joint policy memorandum is finalized, we will post it on our web site. Also, the EPA concurrence and nonconcurrence letters and the final five-year review report are posted on the Regional web site and are part of the administrative record for the site. OK; Response: The ()l(i agrees with the Agency's proposed actions in response to Recommendation 3-2 The recommendation is open with aureed-to actions pending In its final response to this report. I-PA should pro\ ide estimated or actual completion dates for this recommendation Recommendation 3-3 Develop a management control to monitor overdue Reviews. EPA Response We agree with this recommendation. Currently, we send letter to all of the Federal Agencies identifying Reviews which are due or overdue. In addition, we report to Congress every year those Reviews that have been completed or are overdue. During our mid-year and end-of-the-year planning meetings and our quarterly meetings with the other federal agencies, we will continue to address the due dates for Reviews. OIG Response: The OIG agrees with the Agency's proposed actions in response to Recommendation 3-3. The recommendation is open with agreed-to actions pending. In its final response to this report, EPA should provide estimated or actual completion dates for this recommendation. Recommendation 3-4 Develop enforcement policy that describes EPA enforcement tools, processes, and authorities to achieve completion of Review recommendations, as well as the timely submittal of complete Reviews. As appropriate, the guidance should include the desired wording in the FFA and the SMP to support this enforcement. EPA Response We agree with the recommendation and we will ensure that the EPA Regions are more aware of the possible enforcement options for Reviews. In fact, OECA has already provided 27 ------- 10-P-0133 new guidance on the use of stipulated penalties and will issue additional new enforcement policy which describes the Agency's enforcement tools, processes and authorities for these Reviews. OIG's report should make clearer this critical point: that EPA's ability to have the five- year review report as a primary document (and clearly enforceable) is dependent on a federal agency's agreement that such document should be primary. The Defense Department, over at least the last 10 years, has not agreed to EPA requests that the Review report be a primary document in the FFA. Further, the Site Management Plan (SMP) has been included in FFAs only since 1999 so a significant number of earlier FFAs - approximately 100 FFAs, more than half the total - do not include the SMP and would not be affected by the above recommendation. Given that the vast majority of EPA's FFAs are with the Defense Department, EPA does not expect the Defense Department to change its well-established position regarding the content of FFAs. Thus, while we support the OIG's recommendations in concept, it is not within EPA's control to effectuate the change OIG seeks in its recommendation to OECA regarding the wording in the FFA and SMP. EPA Revised Response (5/18/2010): We agree with the recommendation and we will ensure that the EPA Regions are more aware of the possible enforcement options for Reviews. In fact, OECA has already provided the Regions with a new enforcement tool consisting of a concise summary of EPA procedures for assessing stipulated penalties which includes practical examples of how this authority may be used. OECA will issue new enforcement policy which describes the Agency's enforcement tools, processes and authorities for these Reviews. EPA agrees to develop a guide informing EPA regions of the various enforcement options when a Federal agency is late submitting the 5-year review or fails to develop an adequate Review. EPA will ensure regions are aware of FFA model language and how it can be used to support enforcement. When actions are taken to enforce 5-year review requirements, OECA will inform the regions of those actions. OIG's report should make clearer this critical point: that EPA's ability to have the five-year review report as a primary document (and clearly enforceable) is dependent on a federal agency's agreement that such document should be primary. The Defense Department, over at least the last 10 years, has not agreed to EPA requests that the Review report be a primary document in the FFA. Further, the Site Management Plan (SMP) has been included in FFAs only since 1999 so a significant number of earlier FFAs - approximately 100 FFAs, more than half the total - do not include the SMP and would not be affected by the above recommendation. Given that the vast majority of EPA's FFAs are with the Defense Department, EPA does not expect the Defense Department to change its well-established position regarding the content of FFAs. Thus, while we support the OIG's recommendations in concept, it is not within EPA's control to effectuate the change OIG seeks in its recommendation to OECA regarding the wording in the FFA and SMP. 28 ------- 10-P-0133 OIG Response: In a follow-up communication, OECA provided specific information stating that it will issue a new enforcement policy that describes the Agency's enforcement tools, processes, and authorities for these Reviews. The OIG agrees with the Agency's revised proposed actions in response to Recommendation 3-4. The recommendation is open with agreed-to actions pending. In its final response to this report, EPA should provide estimated or actual completion date for this recommendation. A sentence has been added to the final report to indicate that inclusion as a primary document is dependent upon the consent of the federal agency. Recommendation 4-1 Develop a management control to ensure consistency between the Review and CERCLIS. At a minimum, this control should ensure that all Review recommendations that affect current or future protectiveness are entered into CERCLIS for monitoring, consistent with EPA's 2001 guidance. EPA Response We agree with this recommendation. We will reiterate to our Regional offices that only recommendations that affect present or future protectiveness should be tracked and monitored in CERCLIS. Question 28 in our recent FAQ on Reviews provides guidance to the Regions on the types of issues and recommendations that should be monitored and tracked. EPA Revised Response (5/19/2010): We agree with this recommendation. FFRRO will reiterate to the Regions in the joint OSWER/OECA policy memorandum that only the recommendations that affect current and future protectiveness should be tracked in CERCLIS. FFRRO will also remind the Regions that recommendations in the five-year review should be tracked in CERCLIS and that Regional managers should periodically check to ensure that the information is inputted in CERCLIS correctly. FFRRO will ask Regions to report on the implementation of this recommendation during the mid-year and end-of -the year work planning meeting. OIG Response: In a follow-up communication, OSWER provided information stating that it would address the recommendation in the OSWER/OECA memorandum and would ask regions to report on implementation of the recommendation during midyear and end-of-year work planning meetings. The OIG agrees with the Agency's revised proposed actions in response to Recommendation 4-1. The recommendation is open with agreed-to actions pending. In its final response to this report, EPA should provide estimated or actual completion date for this recommendation. 29 ------- 10-P-0133 Recommendation 4-2 Develop guidance and/or CERCLIS controls to address data quality issues, including the modifications of data after entry, logical inconsistencies between CERCLIS data elements, and the close-out of recommendations from prior Reviews. EPA Response We agree with this recommendation. There are audit reports available which flag data errors. FFRRO will continue monitoring and developing reports, as necessary, to ensure that data quality issues are being addressed. Also, our recent FAQ on Reviews provides guidance to the Regions on how to close-out recommendations which were not implemented from prior Reviews. The FAQ states that the Region must change the status of the recommendation to "Considered and Not Implemented" and enter text of why the recommendation was not implemented. We will continue to monitor those fields to ensure that the Regions are closing-out recommendations. OKi Response: The ()l(i agrees with the Agency's proposed actions in response lo recommendation 4-2 The recommendation is open with auieed-to actions pending. In its final response to this report. I-IW should pro\ ide estimated or actual completion date lor this recommendation 30 ------- 10-P-0133 Attachment 2 Technical Comments by OECA Specific Comments 1. Under "Late Reviews Occur Because of Delays in EPA Concurrence," on page 11, second full paragraph, OIG's draft report discusses the SMP as a tool that could be used to avoid delays in EPA concurrence. The SMP is a document typically used only in FFAs signed since 1999. For the FFAs signed since EPA and DoD resolved the FFA dispute (February, 2009), the terms of the FFA state that the SMP's deadline for submission of the Periodic Assessment Report is merely a target date, unenforceable under the terms of the FFA. While the target date could serve as a guideline, EPA could not enforce the SMP's date. However, EPA could enforce the statutory deadline if the Federal agency failed to submit the report within 5 years of initiation of the remedial action as that would be a term or condition of the FFA. OIG Response: Language was added to the final report to reflect that EPA could enforce the statutory deadline if the federal agency failed to submit the report within 5 years of initiation of the remedial action. 2. Under "Reviews Should be but are Not Consistently Enforceable under Existing Federal Agreements," on page 13, first full paragraph, the OIG's draft report could more clearly state that EPA lacks the authority to compel the Defense Department to include language in a consensual agreement. Whether a Review report is a primary document is dependent upon the consent of the Federal agency. The reader of the OIG report could likely get the impression that EPA has, within its control, the power to add the Review report as a primary document. EPA has no such control. As the OIG is aware, EPA recently resolved a several year-long FFA dispute with the Defense Department where the department refused to enter into FFAs altogether despite the requirement to enter one with EPA in CERCLA. The Defense Department refused to enter into FFAs where the requirement is clearly enunciated in statute. There is no analogous requirement that a Federal agency submit the Review report to EPA for review, let alone concurrence. OIG Response: Language was added to the final report stating EPA's lack of authority. 3. On page 14, first full paragraph above "Conclusions," the OIG reviewed recent FFAs (signed since 2001) to reach a conclusion that nearly all FFAs should be enforceable. It seems that the OIG meant to say that the periodic assessment is enforceable, rather than "nearly all FFAs should be enforceable." EPA views that all FFAs are enforceable; however, not all FFAs have the "Periodic Assessment" provision, and, in fact, most do not have the language reviewed by the OIG. This means that EPA could not generalize over the hundred other FFAs that the Review provisions are enforceable under the FFA to the same degree as the recent Periodic Assessment provision. OIG Response: Language was modified in the final report to be consistent with EPA's statement. 31 ------- 10-P-0133 Appendix B Distribution Office of the Administrator Assistant Administrator, Office of Solid Waste and Emergency Response Assistant Administrator, Office of Enforcement and Compliance Assurance Regional Administrators, Regions 1-10 Director, Office of Regional Operations Agency Follow-up Official (the CFO) Agency Follow-up Coordinator General Counsel Principal Deputy Assistant Administrator, Office of Solid Waste and Emergency Response Principal Deputy Assistant Administrator, Office of Enforcement and Compliance Assurance Associate Administrator for Congressional and Intergovernmental Relations Associate Administrator for Public Affairs Director, Office of Federal Facilities Restoration and Reuse, Office of Solid Waste and Emergency Response Deputy Director, Office of Federal Facilities Restoration and Reuse, Office of Solid Waste and Emergency Response Director, Office of Federal Facilities Enforcement, Office of Enforcement and Compliance Assurance Deputy Director, Office of Federal Facilities Enforcement, Office of Enforcement and Compliance Assurance Audit Follow-up Coordinator, Office of Solid Waste and Emergency Response Audit Follow-up Coordinator, Office of Enforcement and Compliance Assurance Audit Follow-up Coordinators, Regions 1-10 Acting Inspector General 32 ------- |