§ < 30 \ z o ^ c>v° P R 0^ OFFICE OF INSPECTOR GENERAL Catalyst for Improving the Environment Evaluation Report Better Enforcement Oversight Needed for Major Facilities with Water Discharge Permits in Long-Term Significant Noncompliance Report No. 2007-P-00023 May 14, 2007 ------- Report Contributors: George Golliday Renee McGhee-Lenart Laura Tam Andrew Creath Dan Engelberg Abbreviations DMR Discharge Monitoring Report ECHO Enforcement and Compliance History Online EMS Enforcement Management System EPA U.S. Environmental Protection Agency NPDES National Pollutant Discharge Elimination System OECA Office of Enforcement and Compliance Assurance OIG Office of Inspector General OTIS Online Tracking Information System PCS Permit Compliance System QNCR Quarterly Noncompliance Report SNC Significant Noncompliance Cover Photo: Sloss Industries wastewater treatment facility. (Black Warrior RIVERKEEPER Website photo) ------- tf£D sr^ / \ U.S. Environmental Protection Agency 2007-P-00023 ? ^ Hffiro rkf Inenortnr (Perioral May 2007 0* U ¦ O • L. I I V11 Ul IIIICI I Lul a I UlCvll Office of Inspector General At a Glance PRO"*^ Catalyst for Improving the Environment Why We Did This Review We undertook this review to assess oversight of major facilities in long-term significant noncompliance with water discharge permit requirements. We sought to determine if the U.S. Environmental Protection Agency (EPA) and States are taking timely and appropriate enforcement actions against facilities in long-term significant noncompliance. Background EPA has authorized 45 States to administer the National Pollutant Discharge Elimination System program, including enforcement of discharge permits. EPA still maintains responsibility for oversight and ensuring that Clean Water Act regulations are enforced. According to EPA's current guidance, several basic oversight criteria define a "well-performing" compliance and enforcement program, including (1) timely and appropriate enforcement response, and (2) accurate recordkeeping and reporting. For further information, contact our Office of Congressional and Public Liaison at (202) 566-2391. To view the full report, click on the following link: www.epa.qov/oiq/reports/2007/ 20070514-2007-P-00023.pdf Better Enforcement Oversight Needed for Major Facilities with Water Discharge Permits in Long-Term Significant Noncompiiance What We Found EPA did not provide effective enforcement oversight of major facilities with National Pollutant Discharge Elimination System permits in long-term significant noncompliance. While flexibility is required in a national program, EPA inconsistently applied guidance defining timely formal actions. Also, EPA guidance did not provide meaningful direction on what constitutes "appropriate" actions. Moreover, for 21 of 56 facilities reviewed, EPA and States did not take suitable formal enforcement actions to address all instances of significant noncompliance. At the remaining 35 facilities, none of the actions we could assess were timely based on criteria in EPA's Enforcement Management System. EPA and States also did not maintain complete and accurate records of National Pollutant Discharge Elimination System compliance and enforcement activities. Many region and State files were incomplete, and data in EPA's information systems were incomplete and inaccurate. Further, regions and States did not report inspection-related violations in EPA's Permit Compliance System. We also noted that bacteria exceedances are not required to be reported as significant noncompliances. Timely actions could help minimize the millions of pounds of excess pollutants released by these facilities. We estimate that up to 51 million pounds of excess pollutant loads were discharged from July 2002 through June 2005 by 44 facilities reviewed, representing loads that could be minimized. What We Recommend We recommend that the Assistant Administrator for the Office of Enforcement and Compliance Assurance clarify and implement guidance regarding facilities in significant noncompliance, implement a quality assurance program, and establish controls allowing EPA leadership to identify significant noncompliance by bacteria-only violators. EPA disputed many of our findings, but stated general concurrence with our recommendations and identified planned actions. However, the Office of Enforcement and Compliance Assurance's planned actions generally do not address the intent of our recommendations, and the issues are considered unresolved. ------- ^tDsx ^ J WASHINGTON, D.C. 20460 < JML, \ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY *l PRO"*^ OFFICE OF INSPECTOR GENERAL May 14, 2007 MEMORANDUM SUBJECT: Better Enforcement Oversight Needed for Major Facilities with Water Discharge Permits in Long-Term Significant Noncompliance Report No. 2007-P-00023 FROM: Wade T. Najjum ^ / Assistant Inspector General, Office of Program Evaluation TO: Granta Y. Nakayama Assistant Administrator, Office of Enforcement and Compliance Assurance This is our report on the subject review 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. Accordingly, the findings described in the report are not binding upon EPA in any enforcement proceeding brought by EPA or the Department of Justice. 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 $932,305. Action Required In accordance with EPA Manual 2750, you are required to provide a written response to this report within 90 calendar days. You should include a corrective action plan for agreed upon actions, including milestone dates. We have no objections to the further release of this report to the public. This report will be available at http://www.epa.gov/oig. If you or your staff have any questions regarding this report, please contact me at (202) 566-0832 or naiiurn. wade@epa. gov; or Dan Engelberg, Director for Program Evaluation, Water Issues, at (202) 566-0830 or engelberg.dan@epa.gov. ------- Better Enforcement Oversight Needed for Major Facilities with Water Discharge Permits in Long-Term Significant Noncompliance Table of C Chapters 1 Introduction 1 Purpose 1 Background 1 Noteworthy Achievements 5 Scope and Methodology 5 2 Oversight Limited at NPDES Major Facilities in Long-Term SNC 6 EPA Guidance Misinterpreted and Vague 6 Lack of Suitable Enforcement at 21 Facilities 8 Untimely Enforcement at 30 Facilities 8 Timely Enforcement Helps Minimize Excess Pollutant Discharges 9 Conclusion 9 Recommendation 10 Agency Comments and OIG Evaluation 10 3 Incomplete and Inaccurate Records and Reporting Inhibit EPA Oversight of NPDES Major Facilities 11 Compliance and Enforcement Files Incomplete 11 EPA's Enforcement Data Systems Inaccurate 12 Many Violations Found During Inspections Not Reported in PCS 13 Bacteria Not Reported as SNC 13 Conclusion 14 Recommendations 14 Agency Comments and OIG Evaluation 14 Status of Recommendations and Potential Monetary Benefits 15 Appendices A Details on Scope and Methodology 16 B Overall Agency Comments 20 C OIG Evaluation of Overall Agency Comments 23 D Detailed Agency Comments and OIG Evaluation 25 E Distribution 43 ------- Chapter 1 Introduction Purpose We initiated this review to assess oversight of major facilities in long-term significant noncompliance (SNC) with National Pollutant Discharge Elimination System (NPDES) permits. The U.S. Environmental Protection Agency (EPA) has overall responsibility for implementing the NPDES program and ensuring adherence to Clean Water Act regulatory requirements. We sought to determine how well EPA is ensuring that timely and appropriate enforcement actions are taken against NPDES facilities in long-term SNC. We also sought to determine what excess pollutant loads could be minimized if facilities in long-term SNC for effluent violations achieved compliance. Background The Clean Water Act authorizes the NPDES permit program. The program is designed to control water pollution by regulating point sources that discharge pollutants into surface waters of the United States. Point sources are discrete conveyances, such as pipes from municipal and industrial facilities or man-made ditches. These facilities are generally classified as major or minor, depending on size and nature of the discharges. For example, a major municipal treatment plant typically discharges one million gallons or more per day of wastewater. Facilities that intend to discharge pollutants into the Nation's waters are required to obtain an NPDES permit. The permit sets limits on the amount of pollution a facility can discharge into waterbodies, to protect human health and aquatic resources. Federal regulations require permitted facilities to monitor discharged pollutants and periodically report those monitoring results to EPA or States using Discharge Monitoring Reports (DMRs). Noncompliance with permit requirements can pose risks to human health and the environment. Violations of pollutant permit limits can expose aquatic life and humans to excess levels of toxic or harmful pollutants. Similarly, failure to submit DMRs means risks to aquatic life and humans are unknown. EPA Oversight of NPDES Compliance and Enforcement Programs The EPA Office of Enforcement and Compliance Assurance (OECA) is responsible for ensuring that the regulated community complies with Federal environmental statutes, including the Clean Water Act. The Clean Water Act gives EPA authority to allow States to administer the NPDES program. 1 ------- Currently, 45 States administer the program, including compliance and enforcement. Where the State is the administering agency, EPA has continuing overall responsibility for implementing and overseeing the NPDES program. In the remaining States, EPA is still the administering authority. For certain States, including Oklahoma and Texas, EPA has compliance and enforcement authority for some facilities while the States have that authority over others. According to EPA's current guidance, 1986 Revised Policy Framework for State/EPA Enforcement Agreements, several basic elements define a "quality" compliance and enforcement program, including: • Timely and appropriate enforcement response • Accurate recordkeeping and reporting Compliance with the nation's environmental laws is the goal of enforcement. Deterrence of noncompliance is achieved through: (1) a credible likelihood of detecting a violation, (2) the speed of the enforcement response, and (3) the likelihood and severity of the sanction. Enforcement is a vital part of encouraging regulated entities to meet environmental obligations. Additionally, enforcement levels the playing field with environmentally compliant companies and deters those who might otherwise profit from violating the law. Types of Noncompliance Title 40, Code of Federal Regulations, Section 123.45(a)(2) requires EPA and States to report when NPDES major dischargers are in noncompliance with their permit conditions. Reportable noncompliance can include a violation of effluent limits in NPDES permits, violations of enforcement orders or schedules for compliance with permit conditions, failure to provide compliance schedule reports, unauthorized bypasses or discharges, and failure to submit DMRs on time. SNC is a subset of reportable noncompliance. SNC involves priority violations of NPDES permit conditions that EPA believes merit special attention, including: • Violations of monthly and non-monthly effluent limits for two or more months during two consecutive quarterly review periods, by (a) 20 percent for toxic pollutants, such as metals; and (b) 40 percent for conventional pollutants, such as total suspended solids. • Non-effluent violations, such as bypasses or unpermitted discharges, which cause or have the potential to cause a water quality problem, such as beach closings. • Permit compliance schedule violations. • Reporting violations, including failure to submit timely DMRs (filing a DMR more than 30 days late or not at all). • Violations of existing enforcement orders, including judicial or administrative orders. 2 ------- There are approximately 6,400 active NPDES major facilities. About 4,800 facilities reported being in noncompliance for at least one quarter during our 12-quarter review period (July 2002 through June 2005). About 2,650 of those facilities were reported to be in SNC for at least one quarter during the same time. Enforcement Management System Criteria EPA's national guidance for developing and improving NPDES compliance tracking and enforcement systems is called the Enforcement Management System (EMS). According to the EMS, there are three levels of response to all violations. For some violations, a response may not be necessary. The second level response is an informal response. An informal enforcement response can be an inspection, phone call, or a violation letter. Informal responses inform the permittee that the Agency is aware that a violation occurred. The third level of response is a formal action. A formal enforcement action: • Requires a facility to take action to achieve compliance. • Specifies a timetable. • Contains consequences for noncompliance that are independently enforceable without having to prove the original violations. • Subjects the facility to adverse legal consequences for noncompliance. According to the EMS, facilities with SNC violations must receive a formal enforcement action, from the administering authority, that is timely and appropriate, or return to compliance within the quarter following the SNC violation. If formal action is not taken, the State or EPA is expected to produce a written record clearly justifying why an alternative action (other than a formal action) was more appropriate. If the State, as the NPDES program authority, does not take timely and appropriate formal enforcement action, EPA is expected to take a formal enforcement action. EMS states a timely formal enforcement action must be taken by the end of the quarter following the SNC violation. For instance, if a facility experiences an SNC violation in February of the first quarter, a formal enforcement must be taken by the administering agency by June 30 of the second quarter. EMS states an appropriate response for SNC is a formal action. The guidance states an appropriate formal enforcement action is determined by its effectiveness and whether the noncomplying facility returns to compliance as quickly as possible. An appropriate action also establishes a deterrent to future violations and promotes fairness of government treatment. Compliance and Enforcement Data Systems and Reports EPA and States use the Permit Compliance System (PCS) to manage and assess compliance and enforcement program information. This includes such 3 ------- information as permit issuance, permit limits, facility monitoring data, and enforcement and inspection activity for facilities regulated under the Clean Water Act. EPA and States enter compliance and enforcement information directly into the PCS database. For example, States enter DMR data and information submitted by NPDES facilities into PCS. A few States submit DMR information electronically. The Online Tracking Information System (OTIS) enables EPA staff, State/local/tribal governments, and Federal agencies to access enforcement and compliance data from systems like PCS. The public can access the Enforcement and Compliance History Online Website (ECHO),1 which provides similar data. Each quarter, the States and/or EPA produce a Quarterly Noncompliance Report (QNCR) from PCS. The report flags NPDES major facilities that were in reportable noncompliance with permit requirements during the previous 6 months. States and EPA regions use the QNCR to determine what actions to take at these facilities. The QNCR encompasses a rolling 6-month period such that the QNCR for the second quarter (April through June) captures reportable effluent violations from January through June. The Watch List is an EPA management tool, not for public access, that tracks facilities with serious or chronic noncompliance that have not received formal enforcement action. PCS reports effluent violations on the QNCR by using a computer comparison of permitted effluent limits to a facility's actual self-reported monitoring data from DMRs. If those effluent violations are severe enough, the facilities should be flagged as SNC for effluent violations. The QNCR is submitted to EPA headquarters on the last working day of the second month following the QNCR period. Compliance Inspections Inspections are another critical source of compliance and enforcement information. During inspections, information is gathered to determine compliance. Inspections include direct observations of facility operations and/or conditions. Inspections allow EPA and States to determine reliability of a permittee's self-reported data. They also allow EPA to evaluate compliance with permit conditions, applicable regulations, and other requirements. The Agency goal is to have EPA and/or States annually inspect 100 percent of all NPDES major facilities or equivalent coverage of a combination of major and priority minor facilities. 1 http://www.epa.gov/echo/. 4 ------- Noteworthy Achievements OECA piloted the State Review Framework in 2004 to foster dialogue on enforcement and compliance performance that should lead to improved program management and environmental results. This framework involves a process for conducting performance reviews of enforcement and compliance activities in the States (as well as for non-delegated programs implemented by EPA regions). In January 2004, OECA initiated the "Watch List," which tracks facilities with serious or chronic violations of environmental laws that have received no formal enforcement response. The Watch List for NPDES facilities uses existing data reported by EPA regions and States to PCS. EPA, in cooperation with State governments and the Environmental Council of the States, developed ECHO, which provides integrated compliance and enforcement information for approximately 800,000 regulated facilities nationwide. The site allows the public to find inspection, violation, enforcement action, informal enforcement action, and penalty information about facilities for the past 3 years. Scope and Methodology We conducted our review from January 2005 through December 2006 in accordance with Government Auditing Standards, issued by the Comptroller General of the United States. We reviewed EPA's oversight of NPDES major facilities in long-term SNC using the following quality elements in EPA's 1986 Revised Policy Framework for State/EPA Enforcement Agreements: (1) timely and appropriate enforcement response, and (2) accurate recordkeeping and reporting. We interviewed staff in OECA and Office of Water's Office of Wastewater Management. We interviewed staff in selected EPA regions, as well as officials and staff in selected States. We selected 84 of the 132 facilities in long-term SNC, according to EPA data systems, in EPA Regions 4, 5, and 6 for our review. We reviewed available compliance and enforcement files to determine their accuracy and completeness for the 84 NPDES facilities in long-term SNC. We utilized data from PCS and OTIS to determine the excess pollutant loads released from facilities in long-term SNC for effluent violations. We evaluated the effectiveness of management controls and considered findings in prior EPA Office of Inspector General (OIG) reports. Details on our scope and methodology are in Appendix A. 5 ------- Chapter 2 Oversight Limited at NPDES Major Facilities in Long-Term SNC EPA did not provide effective enforcement oversight of major facilities in long- term SNC with NPDES permits. EPA inconsistently interpreted guidance defining timely formal actions. Also, EPA's guidance does not provide a means for determining the appropriateness of actions. Problems with EMS guidance prevented us from determining whether enforcement actions taken were appropriate. However, we found that for the 56 NPDES major facilities in long- term SNC reviewed:2 • At 21 facilities, EPA and States did not take suitable formal enforcement actions to address SNC violations during our review period. • At the remaining 35 facilities, none of the actions that we could assess were timely based on the criteria in EPA's EMS. Timely and appropriate formal enforcement actions are important to minimize additional pollutants from being discharged into the Nation's waters to ensure protection of human health and the environment. We estimate that up to 51 million pounds of excess pollutant loads were discharged during our review period by 44 facilities reviewed, and that represents loads that can be minimized. EPA Guidance Misinterpreted and Vague EPA did not consistently interpret its EMS guidance for timely actions, even though EMS guidance is clear with regard to timeliness. Further, EMS guidance is unclear regarding appropriateness of enforcement actions and is thus subject to interpretation. This ambiguity prevented us from evaluating the appropriateness of the EPA and State formal enforcement actions taken. However, we concluded that EPA's ambiguous guidance may result in inconsistency in interpretations between regions and States. Flexibility is needed in any national enforcement program. Nonetheless, definitions for program fundamentals such as the timeliness and appropriateness of action are necessary to establish minimum acceptable performance and oversight. 2 Due to significant EPA data system errors for 27 facilities in Alabama, Michigan, and Texas, as discussed in Chapter 3, we could only evaluate 57 of the 84 facilities selected for review for timely formal enforcement action. We also eliminated one additional facility in Tennessee from our sample that was shown to be a minor facility by OECA. 6 ------- EPA Inconsistently Interprets Guidance for Timely Actions During our review, we were given inconsistent interpretations of what EPA considers timely. In a 1999 EMS memo, OECA's water enforcement division director wrote that for a formal enforcement action to be considered timely, it must be taken by the end of the quarter following the SNC violation. In other words, if a facility experiences an SNC violation in February of the first quarter, then a formal enforcement action must be taken by the administering agency by June 30 of the second quarter - the last day of the quarter - to be timely. Although we believe the policy in the EMS memo is clear on the definition of "timely," and the memo has not been rescinded, EPA allows other interpretations. For example: • In response to an earlier version of our report, EPA asserted that timely enforcement actions could be 60 days later than the timely definition above. • In subsequent data submissions to the OIG, EPA noted that a timely response must occur within 6 months of the facility appearing and remaining on the QNCR. This implies EPA has up to a full year from the time of an SNC violation to provide a response. EPA's stated that "not only is there flexibility in the EMS language itself regarding timeliness, but the fact that the EMS is a guidance document and not a regulation infers flexibility." However, that is inconsistent with written guidance in EMS, which provides a clear definition of timely. EPA Guidance Does Not Address Appropriateness of Actions EMS guidance states that the appropriate response to an SNC violation is a formal action that should ensure that the facility returns to compliance expeditiously, creates a deterrent effect, and is fair. However, the guidance does not state how expeditiously facilities should come back into compliance after an action is taken. In fact, EPA staff stated that "appropriate" is defined flexibly and a formal action can address the same violations for many years. Since EPA has no clear criteria, we could not determine if the formal actions taken were appropriate. We questioned whether the outcome of some formal actions was appropriate. For example, EPA took a formal enforcement action in 1988 against the East Chicago Sewage Treatment Plant.3 According to OECA, that action "renders all effluent exceedances (sic) resolved pending since that time." East Chicago has been in SNC for effluent violations - including pollutants such as phosphorus, chlorides, and sulfates - in every quarter from 1996 to June 2005. During the 3 years of our 3 EPA issued Pretreatment Administrative Orders on January 25, 1993, and September 20, 2004. The September 2004 order addressed pretreatment issues found during an audit. 7 ------- review period, East Chicago discharged between 76,000 and 230,000 pounds of excess phosphorus and between 12,000,000 and 38,000,000 pounds of excess chlorides. We could not determine the excess loadings of sulfates. Seventeen years after formal action, as of June 2005, the facility still had excess discharges to waters that were currently impaired. Lack of Suitable Enforcement at 21 Facilities Of the 56 NPDES major facilities in long-term SNC that we were able to assess, EPA and States did not take suitable formal enforcement actions to address all SNC violations against 21 of them during the 3-year period July 2002 through June 2005. Some facilities received informal actions, such as letters of violation. According to the EMS, facilities with SNC violations must receive a formal enforcement action, from the administering authority, that is timely and appropriate, or return to compliance within the quarter following the SNC violation. Specifically: • Eight of the 21 facilities in SNC during our review had not received a formal action as of June 2005, according to OECA. • The remaining 13 facilities were operating under enforcement orders issued prior to the time period, but these actions did not address all violations for which the 13 facilities were in SNC. For example, the Lafayette Sewage Treatment Plant in Tennessee received formal action regarding only two of three pollutants for which it was in SNC. During our review period, this facility was in SNC for effluent violations for six quarters for ammonia-nitrogen, seven quarters for settleable solids, and two quarters for total suspended solids. The Tennessee Department of Environment and Conservation issued an Administrative Order in April 2002 to address ammonia-nitrogen and total suspended solids violations. However, the Order did not address the settleable solids violations. This facility discharged between 15,000 and 182,000 pounds of excess settleable solids from July 2002 to June 2005. The facility discharges to a stream that is impaired for organic enrichment and low dissolved oxygen, and Lafayette's excess settleable solids discharges could be further worsening the existing water quality impairments. Untimely Enforcement at 30 Facilities EPA and States took 57 formal enforcement actions to address SNC violations between July 2002 and June 2005 at 35 facilities.4 We were able to assess 45 of 4 Some facilities received more than one enforcement action. Thus, a facility may have had one action judged untimely and another in which timeliness could not be determined. This explains the apparent inconsistency in the total number of facilities associated with untimely actions and actions whose timeliness could not be determined. A total of 35 facilities were included in the assessment. 8 ------- the 57 actions at 30 facilities for timeliness and, based on EMS criteria, none of the 45 was timely. For example, EPA issued an Administrative Order against ConocoPhillips Company on August 14, 2003, to address effluent violations for selenium beginning July 2002. The action was untimely because it was not taken by December 2002 (the end of the next quarter), as required by EMS. According to PCS, the facility continued to exceed selenium permit limits through June 2005. Our calculations show that this facility discharged about 950 pounds of selenium above its allowable permit level between July 2002 and June 2005. Selenium is toxic to aquatic life and some wildlife and harmful to humans. We could not determine the timeliness of the remaining 12 formal actions at 10 facilities because available information was flawed. Specifically, the enforcement actions did not explain what SNC violation was being addressed or when the SNC violations occurred. Without this information, we would not determine whether the action was timely. For example, the Consent Decree issued against the City of Columbiana Wastewater Treatment Plant in Alabama on December 10, 2003, simply accelerates an existing compliance schedule for a Consent Order issued on March 21, 2002. This action does not directly link to an SNC violation. Therefore, we could not address timeliness. Timely Enforcement Helps Minimize Excess Pollutant Discharges Timely actions are important because they can help deter or minimize environmental impact. The sooner SNC is identified and formal enforcement actions taken, the sooner the excess pollution can be controlled. We estimate that up to 51 million pounds of excess pollutant loads were discharged from July 2002 through June 2005 by 44 of the facilities we reviewed.5 Seventeen of those facilities discharged excess pollutants into already impaired waters. Conclusion EPA has not provided its regions and States with clear and consistent guidance for taking timely and appropriate enforcement actions. Incomplete and misleading guidance reduces the effectiveness of regions' and States' oversight of NPDES facilities. At the Agency level, we concluded that EPA does not provide effective oversight over the regions and States. Noncompliance with permit requirements can pose risks to human health and the environment through exposure to excess levels of toxic or harmful pollutants. 5 Due to a lack of data in EPA's data system, we could only calculate a range of pollutant discharges for some facilities. See Appendix A for more details. 9 ------- Recommendation We recommend that the Assistant Administrator, Office of Enforcement and Compliance Assurance: 2-1 Clarify and implement the EMS guidance to ensure that timely and appropriate formal enforcement actions are taken against NPDES facilities in SNC. Agency Comments and OIG Evaluation EPA agreed to clarify the EMS guidance by rescinding the 1999 OECA memorandum in conflict with the EMS. EPA's planned action is to eliminate the guidance that clarifies when timely enforcement action should be taken. While reducing the ability to measure the effectiveness of enforcement actions increases the flexibility to define all actions as timely, it does not meet the intent of the recommendation. EPA's detailed response and OIG's evaluation are in Appendices B, C, and D. The OIG has incorporated technical corrections and clarifications from EPA's response into the final report as appropriate. 10 ------- Chapter 3 Incomplete and Inaccurate Records and Reporting Inhibit EPA Oversight of NPDES Major Facilities Records of compliance and enforcement activities were often incomplete and inaccurate. Specifically: • Many region and State files were incomplete. • Data in EPA's information systems were incomplete and inaccurate. • Regions and States did not report inspection-related violations in PCS, although EPA required such reporting. An accurate history of the compliance and enforcement activities at a facility is important for oversight and making future enforcement decisions. The lack of accurate information inhibits EPA's ability to provide effective oversight to NPDES major facilities and thus protect human health and the environment from excess levels of toxic or harmful pollutants. During our review, we also found that bacteria exceedances are not required to be reported as SNC, but we think such reporting should be considered. Compliance and Enforcement Files Incomplete Many EPA and State compliance and enforcement files were incomplete for NPDES major facilities. We reviewed EPA and State files to determine if they were complete by comparing enforcement information from EPA and State file reviews to formal enforcement actions listed in PCS/OTIS. Some of the information in the compliance and enforcement files did not match the information recorded in EPA's data systems. Table 3-1 on the following page shows the results of comparing enforcement files with EPA's data system information. Additionally, 12 formal enforcement actions in the EPA and State files were not listed in EPA's data systems. We also found one action listed in OTIS for which we could find no other evidence in either PCS or EPA and State files. Incomplete files and data systems prevent EPA from having the needed information to provide effective oversight. 11 ------- Table 3-1: Completeness of EPA and State Enforcement Files Region State Actions reported by PCS/OTIS No. of those actions found in EPA files Percent complete No. of those actions found in State files Percent complete Region 4 Alabama 8 5 63% 8 100% Tennessee 7 4 57% 4 57% Total 15 9 60% 12 80% Region 5 Indiana 4 2 50% 1 25% Michigan 4 2 50% 1 25% Total 8 4 50% 2 25% Region 6 Oklahoma 12 10 83% 5 (of 9)* 55%* Texas 22 12 55% 5 (of 10)* 50%* Total 34 22 65% 10 (of 19)* 53%* *When calculating completeness of State files, we did not consider actions taken against facilities for which EPA retains primary enforcement authority. Source: EPA OIG review EPA's Enforcement Data Systems Inaccurate EPA's data systems used to track NPDES enforcement program are inaccurate. EPA's data systems do not accurately reflect the SNC status of facilities. Once a formal enforcement action is issued for an SNC violation, a facility should be designated as "resolved pending" for that violation, according to Title 40, Code of Federal Regulations, Section 123.45. However, OECA states that Regions 4, 5, and 6 (the three regions reviewed) do not designate facilities in "resolved pending." Thus, facilities in these regions may appear to remain in SNC status in EPA's data systems for a long time despite operating with an existing enforcement action. EPA's data system incorrectly classified Michigan facilities in SNC for DMR violations due to data system errors in Michigan. These errors have existed since 2003. EPA's PCS data system showed that 30 facilities from Michigan were in long-term SNC. However, OECA reviewed the facilities' status and determined that 27 of the facilities were not actually in SNC.6 OECA stated that Michigan was having problems entering DMR information into PCS in a timely manner. Staff from Michigan Department of Environmental Quality also noted that their data system was not capable of correctly communicating with PCS and that SNC data was not available through their system at the time we interviewed them. However, the data entry and transmission problems continue, which seriously compromises EPA's oversight of Michigan's NPDES facilities. 6 We excluded 25 Michigan facilities with significant data errors from our analysis. Two other facilities, one from Alabama and one from Texas, were also excluded for similar reasons. See Appendix A for more details. 12 ------- In a related matter, EPA staff stated that once a facility receives a violation for submitting DMRs late, that violation will continually show as a violation in EPA's data systems until regions or States manually correct the problem. EPA needs to ensure its data systems are accurate so it can provide proper oversight to regions and States. Many Violations Found During Inspections Not Reported in PCS Despite EPA requirements, most EPA regions and States do not enter into PCS violations found during inspections. OECA reported that no States in EPA Regions 1, 2, 5, 7, 9, and 10 reported these violations in Fiscal Year 2005. Inspections allow EPA and States to determine the reliability of the permittee's self-reported data and evaluate compliance with permit conditions and applicable regulations. Entering these violations into PCS allows EPA and States to track these violations and report them. EPA currently requires data entry of violations found during inspections (called single-event violations) for NPDES major permittees in PCS in the single event category. However, most States are not performing this procedure. Staff from EPA Region 4 stated that EPA does not ensure that States enter single event violation data from inspections. Staff from EPA Regions 5 and 6 acknowledged that entry of single event violations is required. Failing to enter inspection-related violations in PCS may prevent these types of violations from being identified and acted upon. Bacteria Not Reported as SNC We noted numerous and significant exceedances of bacteria permit limits at 37 of 84 facilities we reviewed. Almost 75 percent of the exceedances would be classified as SNC using the criteria for conventional pollutants. The violations ranged up to 59,000 percent of the permit limit. None of these violations was reported as SNC in PCS or OTIS because bacteria are not a pollutant subject to SNC criteria. Since these violations are not classified as SNC, they would likely not be reported on the QNCR that regions and State use to address facilities with significant violations. Although SNC reporting is not required, bacteria are a leading cause of water quality impairments across the nation. When present at levels exceeding water quality standards, bacteria can make swimmers ill and contaminate shellfish for human consumption. Our review only discovered the significant bacteria violations because those facilities had SNC violations for other pollutants. In the absence of other SNC violations, EPA and States may not address significant bacteria violations. Because this issue was outside the scope of our review of long-term SNC, we did not conduct a separate review focused on the extent of bacteria-only violations 13 ------- and the adequacy of the response to them. However, it is enough of a concern that it warrants further review by OECA. Conclusion While EPA has overall responsibility for implementing and overseeing the NPDES program, EPA generally has to rely on the States for information. EPA cannot attest to the reliability of records and data for those facilities in long-term SNC. Incomplete and inaccurate records and information represent a management control weakness. EPA cannot rely on NPDES enforcement and compliance data from its internal systems being complete or accurate. Also, enforcement and compliance records maintained by regions and States are incomplete and inaccurate. OECA needs to implement a quality assurance program that, at a minimum: • Assesses accuracy of region and State compliance and enforcement files. • Verifies status of facilities in SNC. • Verifies that formal actions taken are correctly entered into EPA systems. • Assures reporting issues (with Michigan and others) are resolved or reports annotated to disclose known inaccuracies. • Verifies that all EPA regions and States are entering violations found as a result of inspections in PCS. Recommendations We recommend that the Assistant Administrator, Office of Enforcement and Compliance Assurance: 3-1 Implement a quality assurance program addressing the completeness of compliance and enforcement files, accuracy of EPA data systems, and reporting in PCS the violations found during inspections. 3-2 Establish controls allowing EPA leadership to identify significant noncompliance by bacteria-only violators for enforcement action. Agency Comments and OIG Evaluation EPA generally concurred with our recommendations. However, OECA's response that its existing actions and processes provide a reasonable amount of quality assurance is not responsive to Recommendation 3-1. We consider the issue unresolved. For Recommendation 3-2, we accept OECA's response. OECA will need to provide us with an action plan and milestone dates for when the proposed actions will occur. EPA's detailed response and OIG's evaluation are in Appendices B, C, and D. The OIG has incorporated technical corrections and clarifications from EPA's response into the final report as appropriate. 14 ------- 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 10 Clarify and implement EMS guidance to ensure that timely and appropriate formal enforcement actions are taken against NPDES facilities in SNC. 3-1 14 Implement a quality assurance program addressing the completeness of compliance and enforcement files, accuracy of EPA data systems, and reporting in PCS the violations found during inspections. 3-2 14 Establish controls allowing EPA leadership to identify significant noncompliance by bacteria-only violators for enforcement action. Assistant Administrator, Office of Enforcement and Compliance Assurance Assistant Administrator, Office of Enforcement and Compliance Assurance Assistant Administrator, Office of Enforcement and Compliance Assurance 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 15 ------- Appendix A Details on Scope and Methodology We conducted our review from January 2005 through December 2006 in accordance with the Government Auditing Standards, issued by the Comptroller General of the United States. We interviewed staff in OECA and the Office of Water's Office of Wastewater Management to determine applicable compliance and enforcement guidance and criteria for EPA regions and States. We reviewed NPDES major facilities in long-term SNC. We classified facilities as being in long-term SNC if they had been in SNC for 8 of 12 quarters from the third quarter of 2002 to the second quarter of 2005 in PCS. However, due to continual updating of EPA data systems, 7 of the facilities selected were classified in SNC for only 7 of 12 quarters by the end of our field work. We selected EPA Regions 4, 5, and 6 for review because they have the highest concentration of NPDES major permits with 3,506 facilities (55 percent of the universe). Further, of EPA's 10 regions, they are ranked high in the number of long-term SNC facilities (see Table A-l). Table A-1: Number of SNC Facilities/Ranking of Regions Regions No. of Long-Term SNC Facilities Ranking Region 4 38 4th Region 5 51 2nd Region 6 43 3rd Source: EPA OIG Review Although EPA Region 2 had the highest number of long-term SNC facilities, many were in the territory of Puerto Rico, and we chose not to review that region due to travel and time constraints. We interviewed staff in all three regions selected for review. We also interviewed EPA Region 7 staff during the preliminary research phase of our review. We reviewed 84 of the 132 facilities in long-term SNC in EPA Regions 4, 5, and 6 according to PCS. These 84 facilities represent about 1.3 percent of the total number of active NPDES major facilities from 2002 through 2005. We selected facilities by choosing two States from each of the three EPA regions reviewed (see Table A-2). We interviewed staff in those States to gather information about compliance and enforcement programs. Table A-2: Facilities Reviewed by State Regions States No. of Facilities Reviewed Region 4 Alabama 8 Tennessee 12 Region 5 Indiana 8 Michigan 30 Region 6 Oklahoma 4 Texas 22 Source: EPA OIG Review 16 ------- We reviewed compliance and enforcement files for 84 NPDES facilities in long-term SNC. We reviewed files to confirm enforcement actions taken by EPA and States as reported in PCS and OTIS, determined whether timely formal enforcement actions were taken at facilities, and determined the completeness of the compliance and enforcement files. To do this, we verified the SNC status and formal actions taken at each facility through OTIS. We relied on the classification system in PCS and OTIS to indicate whether an action was formal or informal. During file reviews, we documented only those actions classified as formal - such as administrative orders and consent orders - and compared them to information from PCS and OTIS to confirm that our data were complete. Finally, we compared the formal actions taken against each facility with the facility's SNC status in PCS and OTIS to determine if timely formal enforcement actions were taken consistent with criteria from EPA's EMS document. We could not assess the appropriateness of enforcement actions because the EMS does not state how expeditiously facilities should come back into compliance after an action is taken. This ambiguity prevented us from evaluating the appropriateness of the EPA and State formal enforcement actions taken. Of the 84 initial facilities we evaluated, 25 of the 30 facilities in Michigan were excluded from our timeliness analysis due to EPA data system errors. Additionally, one facility each in Texas and Alabama were excluded for similar issues. We also excluded one facility because it was a minor facility even though OECA's data systems showed it as a major facility in SNC. We analyzed the remaining 56 facilities to determine if timely formal enforcement actions were taken to address SNC. Due to EPA data system inaccuracies, 2 of the 56 facilities that were classified as major facilities in PCS were actually minor facilities for part of our 3-year evaluation period. As such, we evaluated these facilities only for that period of time when the facilities were considered major facilities. Because 21 facilities did not have actions taken during our review period, we only identified and analyzed formal enforcement actions at the remaining 35 facilities. We reviewed 69 actions from these 35 facilities. We excluded 12 of those 69 actions from our analysis for timeliness (8 addressed sludge or sludge reporting violations that were not directly related to the SNC violations, 3 addressed citizen complaints at specific facilities, and 1 represented a parallel action that allowed the State to assume primary enforcement authority from EPA). Thus, we evaluated 57 formal actions. To evaluate timeliness for the 57 formal actions, we reviewed each individual action found during the file reviews and from OTIS to determine the following information: (1) the date the formal action was taken, (2) the specific SNC violations the action addressed, and (3) the dates the violations occurred. We then determined the length of time between the dates the actions were taken and the dates the violations occurred. We compared the violations and formal actions to the 3-year compliance status history as shown in OTIS for each facility on a quarterly basis. We compared the length of time between violation and action with EPA's EMS criteria for timeliness. If the length of time exceeded the time period allowed by the criteria, the action was found to be untimely. Although we conducted an extensive file review on each facility, we did not have enough information on 12 formal actions to determine whether they were timely. 17 ------- We analyzed data from OTIS and PCS to determine the excess pollutants released from facilities in long-term SNC for effluent violations. We calculated the excess pollutant load by determining the pollutant load each facility is permitted to discharge and subtracting that load from the pollutant load each facility actually discharged when it violated its permit limits. We determined the pollutant load actually discharged from facility DMR information by multiplying the actual effluent flow times the actual effluent pollutant concentration times a constant of 8.34. EPA uses this formula to determine mass-based permit limitations and was reviewed by staff from EPA's Office of Wastewater Management. We calculated the permitted pollutant load for each facility using the same formula and each individual facility's pollutant permit limits. We summed the daily excess pollutant loads for each facility over the 3-year time period from the third quarter of 2002 through the second quarter of 2005. In certain cases, OTIS did not report actual effluent pollutant concentrations. However, the percentage by which the permit limit was exceeded was reported in OTIS. In this case, we back- calculated the actual effluent pollutant concentration using the pollutant permit limit and the percentage by which the permit limit was exceeded as reported by PCS. Since the percentage exceedance is calculated by PCS automatically using the data as reported by each facility on its DMRs, this method did not compromise the accuracy of the actual effluent pollutant concentration. Five of the 84 evaluated facilities were listed as SNC for effluent violations, but OTIS had no information indicating which effluent limits were violated or by what percentage. For this reason, we did not include those five facilities in our analysis. Due to varying availability of information for each facility, we calculated both minimum and maximum excess pollutant loads. If the required information for each facility - including design and actual reported flows, permit effluent limits, and actual effluent concentrations - was available in the detailed facility reports from PCS, then it was not necessary to determine minimums and maximums. However, if the only information available was the design flow, permit limit, and the percentage by which the facility exceeded its permit limit from the detailed facility report, then we calculated both minimum and maximum excess pollutant loads. This is because the 3-year compliance status by quarter only lists the highest monthly percentage by which the facility violated its permit limits in that quarter. The magnitude of any violations in other months for that quarter was unknown. Both minimum and maximum excess pollutant load were calculated to account for this uncertainty. While interviewing EPA region and State staff, we also sought to obtain information on what tools and best practices are available to help address facilities in long-term SNC. Management Controls We identified the following management controls as applicable to our objectives: Effectiveness of Program Operations: We evaluated the effectiveness of the compliance and enforcement program by determining if EPA and States were providing timely and appropriate enforcement actions to facilities in long-term SNC, and if EPA provided effective oversight to selected regions and States. Details on what we found are in Chapter 2. 18 ------- Validity and Reliability of Data: We evaluated the validity and reliability of PCS/OTIS data by comparing it with the data found in EPA regions and State files. We compared the number and types of formal enforcement actions in PCS/OTIS for the 84 facilities we reviewed with the information found in EPA and State files. Details on what we found are in the Chapter 3. Compliance with Applicable Laws and Regulations: In our evaluation, we determined whether regions and States were following EMS guidance in addressing facilities in SNC with timely enforcement actions. We found problems in the guidance provided. Details on what we found are in Chapter 2. Prior Coverage We reviewed the following prior EPA OIG audits as they pertained to our review: • Water Enforcement: State Enforcement of Clean Water Act Dischargers Can Be More Effective, 2001-P-00013, August 14, 2001. • North Carolina NPDES Enforcement and EPA Region 4 Oversight, 2000-P-00025, September 28, 2000. • Kansas National Pollutant Discharge Elimination System Program, E1HWF7-07- 0022-8100089, March 31, 1998. • Region 10's National Pollutant Discharge Elimination System Permit Program, E1HWF7-10-0012-8100076, March 13, 1998. 19 ------- Appendix B Overall Agency Comments March 30, 2007 MEMORANDUM SUBJECT: OECA Comments on Draft Report, "Better Oversight Needed for Maj or Facilities with Water Discharge Permits in Long-Term Significant Noncompliance" FROM: Granta Y. Nakayama Assistant Administrator TO: Dan Engelberg Director, Water Issues, Office of Program Evaluation Office of Inspector General The Office of Enforcement and Compliance Assurance (OEC A) appreciates the opportunity to review and comment on the February 28, 2007 Office of Inspector General (OIG) draft report titled, "Better Oversight Needed for Major Facilities with Water Discharge Permits in Long- Term Significant Noncompliance." This draft report was revised after OIG received OECA's extensive comments in October 2006 on its initial August 2006 draft report. The revised draft is an improvement over the initial draft. However, despite exhaustive efforts by OECA and Regional staff to provide well-documented comprehensive comments and to be available for follow-up consultation, significant factual inaccuracies and misinterpretations remain. These undermine the credibility of OIG's analysis and diminish the value of its recommendations. These include: • Inaccurate Interpretation of National EPA Enforcement Guidance and Policy. Despite OECA's attempts to clarify the operation of its guidance and policy, OIG fails to recognize the difference between regulatory requirements and EPA guidance and, as a result, applies EPA's timely and appropriate enforcement response guidelines as a rigid requirement. OIG's draft report fails to consider the deliberate flexibility provided to Regions and states in EPA's guidance resulting in inaccurate OIG findings such as, "at 22 of the 57 facilities reviewed, EPA and the states did not take sufficient formal enforcement action," and "at the remaining 35 facilities, none of the actions were timely." In stark contrast, OECA's assessment of the 57 facilities indicated that the vast majority (approximately 53 or 93%) of facilities included in the OIG review received an adequate response to address the SNC violations. • Inaccurate Data Analysis. At least seven facilities (three in Region 4 and four in Region 6) should not have been included in OIG's analysis because they were not classified as 20 ------- major facilities during all or part of the evaluation period and, therefore, would not have been screened for SNC violations by EPA. A request for change in a facility's status is submitted in writing from a Region to OECA's Office of Compliance. Status change dates are not archived in EPA's Permit Compliance System (PCS) or Facility Registry System; they are documented in manual files maintained in the Regional offices. OECA provided information on this issue in its October 2006 response to OIG's original report, yet OIG did not remove these facilities from its February 2007 draft report. OIG continued to misinterpret data in its analysis and inaccurately concluded that these facilities were majors. In other instances, OIG failed to acknowledge that many SNC violations reviewed were already addressed under existing enforcement orders and compliance schedules (i.e., the facility was in "resolve pending" status in PCS), and thus did not require any additional enforcement. Lastly, facilities were included in the analysis even though a known state-wide data system problem resulted in incorrect SNC determinations. • No Acknowledgement of Ongoing Efforts to Improve Records and Reporting. OIG's draft report fails to consider OECA's ongoing efforts to work with Regions and states, through the State Review Framework (SRF), the Watch List, and ECHO processes, to improve record keeping and reporting associated with compliance monitoring and enforcement activities. OECA believes that its existing efforts are responsive to OIG recommendation 3-1 to implement a quality assurance program addressing the completeness of compliance and enforcement files and recommendation 3-2 to address significant bacteria violations. By not recognizing these recent efforts to establish better management controls and accountability, the draft report leaves the impression that OECA is not actively addressing issues such as timely enforcement and accurate recordkeeping of enforcement files. Despite numerous concerns with the underlying analysis in this draft report, OECA largely accepts OIG's three recommendations and, as noted in the bullet above, is already implementing two out of the three recommendations. Of the remaining recommendation, 2-1, to clarify and enforce EMS guidance, OECA is prepared to implement the part of the recommendation that will clarify the guidance. In response, OECA commits to rescind the 1999 OECA memorandum that conflicts with the NPDES Enforcement Management System (EMS). This will be discussed with the Regional NPDES Enforcement Branch Chiefs at the national NPDES meeting in April, 2007. OECA does not, however, concur with the balance of recommendation 2-1, to enforce the EMS, as the EMS does not impose a set of enforceable standards. This draft report reflects continuing problems with the overall OIG evaluation process, including: poor communication during the analysis leading to the use of incomplete and/or inaccurate information, broad findings that are not supported by the data used, lack of acknowledgment of ongoing efforts by OECA, Regions, and states to address the problem being evaluated, exclusion of relevant information which would directly impact findings and recommendations, and failure to weigh the benefits of OIG's recommendations against the resource burdens and procedural barriers they will entail and the competing needs and demands they will displace. A specific example of poor communication during this evaluation is the fact that even after receiving detailed written comments from OECA and having face-to-face 21 ------- meetings to discuss those comments, OIG staff never contacted EPA staff with any follow-up questions or clarifications in the course of re-writing the draft report. OECA would like to work with the OIG to improve communication and the overall evaluation process. Attached are OECA's detailed comments on the analysis, findings, and recommendations in the February 28, 2007 draft report. If you have any questions, please call Mark Pollins at 202-564-4001. Attachments cc: Walker Smith, OCE Randy Hill, OCE Mark Pollins, OCE Michael Stahl, OC Lisa Lund, OC David Hindin, OC Doug Mundrick, Region 4 Cheryl Newton, Region 5 Mike Michaud, Region 6 22 ------- Appendix C OIG Evaluation of Overall Agency Comments OECA took issue with the overall OIG evaluation process, expressing concerns about the conclusions reached in analysis, poor communications during the evaluation, and the belief that data collected was incomplete or inaccurate. OECA also disagreed with the OIG assessment that its oversight needed improvement. Specifically OECA stated that: • OIG inaccurately interpreted National EPA Enforcement Guidance and Policy. • OIG included facilities that were not classified as major facilities during all or part of the evaluation period. • OIG did not acknowledge OECA's ongoing efforts to improve its records and reporting. Despite OECA's numerous concerns with the underlying analysis, OECA stated it generally accepted OIG's recommendations and was in the process of implementing them. OECA's detailed responses, along with the OIG's evaluation, are provided in Appendix D. The detailed OECA response focuses on denying existence of oversight issues, while agreeing to implement most of the recommended actions. OECA's criticism of the evaluation for poor communication and faulty analysis is rooted in OECA's rationale that written policies can be ignored or interpreted to allow OECA "flexibility" to claim timely and appropriate performance under almost all circumstances. OIG understands OECA's rationale; we simply do not accept that it is valid. OIG's conclusion that OECA has not acted in a timely manner is based on written policy and criteria that OECA prefers to ignore. In fact, one of OECA's corrective actions will be to rescind the policy memorandum clarifying the standard for timely enforcement. The OECA response is illustrative of the control environment that allowed the conditions addressed in the report to exist. Management is responsible for communicating the objectives of internal control and ensuring the organization is committed to sustaining an effective internal control environment. Management's philosophy and operational style set the tone within the organization culture. Culture is defined by management's leadership in setting values of integrity and ethical behavior but is also affected by the relationship between the organization and central oversight agencies. Management's commitment to establishing and maintaining effective internal control should cascade down and permeate an organization's control environment, which aids in the successful implementation of internal control systems. In our opinion, the OECA response is a reflection of weakness in the overall management control environment of the organization. Inaccurate Interpretation of National EPA Enforcement Guidance and Policy This is a disagreement over how to measure EPA's effectiveness ensuring the compliance with the Clean Water Act and how accountable for achieving results it is. EPA's desire for total flexibility undermines the use of internal controls to ensure timely and appropriate enforcement response. OECA interprets the national enforcement guidance and policy, EMS, as granting it 23 ------- total flexibility about when and how to enforce against significant noncompliance. Using this interpretation, OECA concluded that its response was "adequate" 93 percent of the time. We carefully applied OECA's own guidance set out in the EMS to determine how well EPA regions and States are fulfilling their enforcement responsibilities. The EMS represents OECA's own enforcement principles for EPA regions and States. The EMS provides clear and consistent guidance on what constitutes a timely enforcement response. In brief, the EMS states that a formal enforcement response must be taken before the end of the quarter following the quarter in which EPA and/or States designate a facility in SNC. The memorandum OECA wants to rescind clarifies this point. Inaccurate Data Analysis OECA's response actually addresses the inaccurate data in their systems, rather than inaccurate data analysis on the part of OIG. OECA objected to the records of some of the facilities we included in our sample taken from their data system, PCS. We agree with OECA that their systems contain inaccurate data and recommended that they develop a quality assurance program. OECA requested that seven facilities in our review should be removed. We will remove one of the seven. Decherd WWTP in Tennessee was not designated a major discharger until after our study period. We have removed Decherd from our findings. We had included this facility in our study because EPA listed this minor facility as being in SNC during our study period in the PCS database. Also, Windermere and Munford Lagoon were not major facilities during all of our review period although PCS showed them in significant noncompliance during that time. Both facilities were in significant noncompliance for at least three quarters after they became majors. Since the facilities did not receive timely action while they were major facilities, we will keep them in our analysis. We disagree with EPA about the other four facilities. OECA stated that we failed to acknowledge many SNC violations reviewed that were already addressed under existing enforcement orders and compliance schedules. On the contrary, we considered whether a facility was in "resolved pending" as indicated by EPA data systems and data submitted to us by OECA. OECA also stated that we retained several facilities in our review in spite of a known state-wide data system problem that resulted in incorrect SNC determinations. Again, we disagree. We did not evaluate the timeliness of enforcement actions at 25 of the 30 Michigan facilities due to the state-wide data problem. According to Region 5 staff, Michigan's data system problems related only to discharge monitoring reporting and compliance schedule reporting. For that reason, we only included facilities in our analysis that had other types of violations. The five Michigan facilities we analyzed had effluent violations for at least two consecutive quarters. Even though we did not evaluate the timeliness of enforcement actions at the 25 facilities, we did report on the obvious oversight problems that exist from a State having system with unreliable data for several years. No Acknowledgement of Ongoing Efforts to Improve Records and Reporting We included the efforts OECA mentioned in its response in the Noteworthy Achievements section of the report. 24 ------- Appendix D Detailed Agency Comments and OIG Evaluation # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response 1 At a Glance, page 1 OECA's specific responses to each of the findings and recommendations are detailed below. The description of the recommendations should be modified accordingly. See responses below. 2 At a Glance, page 1 OECA questions the accuracy of OIG's estimate of the pounds of excess pollutant released. As discussed below, there is sufficient reason to believe that the OIG's selection of facilities used to calculate a hypothetical estimate of pollutants discharged is flawed due to OIG's inaccurate interpretations of EPA's national enforcement guidance. Further, the OIG erroneously assumes that when an enforcement action is concluded, compliance is immediate. • To illustrate, the OIG presents an estimate of the range of pollutants in excess of permit limits released by using data calculated for 45 facilities that the OIG found to be in SNC for the 3-year period from July 2002 through June 2005. However, many of these facilities were operating under consent orders that provided a schedule for return to compliance. As violating facilities complete the necessary plant upgrades allowing them to meet the original permit limits, pollutant reductions are achieved. Because of the infrastructure changes that are often needed to correct compliance problems, the OIG cannot expect that pollutant reductions will be immediately gained upon completion of an enforcement action. Our selection of facilities used to calculate the estimate of pollutants discharged was not flawed. Our estimates of excess pollutants released represent actual effluent pollutant concentrations as reported by NPDES facilities and recorded in EPA's database, OTIS. We analyzed data from OTIS and PCS to determine the excess pollutants released from facilities in long-term SNC for effluent violations. We calculated the excess pollutant load by determining the pollutant load each facility is permitted to discharge and subtracting that load from the pollutant load each facility actually discharged when it violated its permit limits. We determined the pollutant load actually discharged from facility DMR information by multiplying the actual effluent flow times the actual effluent pollutant concentration times a constant of 8.34. EPA uses this formula to determine mass-based permit limitations and was reviewed by staff from EPA's Office of Wastewater Management. We calculated the permitted pollutant load for each facility using the same formula and each individual facility's pollutant permit limits. We summed the daily excess pollutant loads for each facility over the 3-year time period from the third quarter of 2002 through the second quarter of 2005. As such, our estimates would implicitly consider any reductions being achieved as a result of consent orders or plant upgrades. Our analysis does not expect that pollutant reductions are the immediate result of enforcement actions. We did the analysis to determine the impact on the environment when a facility is not in compliance with its NPDES permit. It may take years for a facility to come back into compliance under 25 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response a consent decree. During that time, a facility with effluent violations is still discharging pollutants into the water above its permitted limits. 3 At a Glance, page 1 There is an error in the shaded background summary - EPA "authorizes" (not "delegates") states to administer the NPDES program. The final report was revised to reflect this comment. 4 Chapter 1, page 1 EPA Oversight of NPDES Compliance and Enforcement Program, p. 2: The placement of the statement, "EPA is expected to take formal enforcement action when states fail to take timely and appropriate action" in the first paragraph of this section, which discusses Clean Water Act (CWA) authority and the NPDES program, implies that the CWA mandates when EPA shall take enforcement action. The CWA does not address this issue. The statement itself is an accurate representation of information presented in the EMS. The statement was moved to the section discussing the EMS criteria on page 3. 5 Chapter 1, pages 1-2 Furthermore; the statement inaccurately represents EPA guidance on this issue (see the August 25, 1986 James Barnes Memorandum to the Regions, "Revised Policy Framework for State/EPA Enforcement Agreements (Barnes Memo)). The Barnes Memo recommends consultation with a state to determine whether the state is moving expeditiously to resolve the violation ~ it does not contemplate automatic formal Federal action once a recommended deadline for state action has passed. The Memo specifies that timeframes are not intended to be rigid deadlines for action. It explains that because authorized states have primary responsibility and EPA clearly does not have the resources to take action on or review in detail any and all violations, EPA will circumscribe its actions to certain cases and areas and will take action based on specific listed factors in the Memo. Further, the Barnes Memo acknowledges that because EPA will not, in most cases, receive real time data, Regions are not expected to be able to take direct enforcement action following the exact same timeframes as those that apply to the administering agency. Finally, the Barnes Memo does not require formal enforcement action in every instance. The Memo establishes that an appropriate response is to be determined by the administering agency based upon a consideration of what is needed, and that the form of the enforcement response must achieve compliance. The redraft suggested below is an accurate interpretation of EPA guidance and does not imply, as does OIG's draft language, that immediate federal The EMS was finalized in 1989, at least 2 years after the Revised Policy Framework for State/EPA Enforcement Agreements (Barnes Memo). The EMS constitutes a system for translating compliance information into timely and appropriate enforcement actions. Also, EMS provides the flexibility for each administering agency to develop management procedures which are best suited to its operations and resources with the goal of most efficiently translating compliance information into timely and appropriate enforcement action. Thus, the flexibility desired by OECA was implicitly considered during development of the EMS and the language regarding timely and appropriate formal enforcement actions to address SNC violations. Any ambiguity regarding timely and appropriate language in the EMS was resolved in an October 6, 1999, OECA memorandum that provided the Water Enforcement Division's official interpretation of the EMS's language regarding the "begin date" for the period allowed for a timely and appropriate enforcement response to a SNC violation. The memo further states that EMS requires that a formal action must be taken to prevent the SNC violation from being reported on a second QNCR and, under QNCR rules, this can be accomplished only if the enforcement action is taken during this "following" quarter. Further, the use of 26 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response action is required when the period for timely state actions has passed, or that such action must be formal. • Delete "EPA is expected to take formal enforcement action when states fail to take timely and appropriate action" from first paragraph on page 2." • Add new paragraph after the two bullets in the second paragraph of this section that includes "EPA's guidance indicates that EPA may consider taking direct enforcement action where: a state requests EPA action, where the state enforcement response is not timely or appropriate, in cases or program areas setting national precedents, or where there is a violation of an EPA order or consent decree. In deciding whether to take enforcement EPA will consider whether: the case is nationally significant, risk or damage to the environment or public health is significant, the violator stands to gain significant economic benefit, or there are repeat patterns of violations and violators." • Add a new paragraph for remaining text beginning, "Compliance with the nation's environmental laws. . ." this "begin date" is consistent with the timeframe for violation status calculations PCS makes for the QNCR and SNC. No change made. 6 Chapter 1, page 2 Types of Noncompliance, p. 2: OIG's description of the QNCR regulations does not distinguish between reportable noncompliance (RNC) and significant noncompliance (SNC). Reporting RNC at NPDES majors is a regulatory requirement. Identifying, addressing, and tracking SNC violations (a subset of RNC) is not a regulatory requirement - it is a management tool used to prioritize violations and focus enforcement resources. In contrast to a regulatory requirement, there is inherent flexibility - by design ~ in the application of a policy/management tool. • "Revise citation as follows" . . . Section 123 .45(a)(2)." The report was revised to include the detailed citation. 7 Chapter 1, pages 2-3 Enforcement Management System Criteria, p. 3: In this chapter, OIG acknowledges the flexibility within the EMS for the permitting authority to exercise discretion in deciding which of a range of possible responses (no response, informal action, formal action) to take in addressing violations, including SNC violations. That acknowledgment does not carry through the end of page 3 or through the According to the EMS, facilities with SNC violations must receive a formal enforcement action from the administering authority that is timely and appropriate, or return to compliance within the quarter following the SNC violation. If formal action is not taken, the State or EPA is expected to produce a written record clearly justifying why an alternative action (other 27 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response rest of the draft report. For example in numerous places on pages 6-8, OIG's findings are based solely on whether "formal" enforcement actions were taken - in contrast to its earlier acknowledgement on p. 3 that a range of responses are available. The EMS includes the expectation that SNC violations receive a formal enforcement action. However, it also acknowledges that there may be circumstances when formal enforcement action is not taken and the administering agency is expected to have a written justification for why the alternative action was more appropriate. • "Replace "formal enforcement action" with "enforcement action" at page 3 and pages 6-8. than a formal action) was more appropriate. If the State, as the NPDES program authority, does not take timely and appropriate formal enforcement action, EPA is expected to take a formal enforcement action. Our file and database review at EPA and States found no written justifications for not taking formal enforcement actions. No change made. 8 Chapter 1, page 3 Compliance and Enforcement Data Systems and Reports, p. 4: The second and fourth sentences in the third paragraph need to be rewritten to distinguish between "noncompliance" and "reportable noncompliance." • Revise second sentence as follows: "The report flags NPDES major facilities that were in reportable noncompliance with permit or enforcement order requirements during the previous six months." • Revise fourth sentence as follows: "reportable effluent violations." While our original statement is accurate, we made these revisions in our final report. 9 Attachment 1, Chapter 1, page 3 Scope and Methodology, p. 5, 6, 8: The total of 132 facilities stated in paragraph 2 on p. 5 as well as in paragraph 4 on p. 15 in Appendix A does not match the total of 128 facilities reflected in Table A-l in Appendix A. Further, at pages 6 and 8 the following sentences are inconsistent: • "At the remaining 35 facilities, none of the actions that we could assess were timely based on the criteria in EPA's EMS" (page 6), • "We were able to assess 45 of the 57 actions at 30 facilities for timeliness and, based on EMS criteria, none of the 45 was timely" (page 8), • We could not determine the timeliness of the remaining 12 formal actions at 10 facilities because available information was not adequate" (page 8). The total numbers of facilities represented in the three bullets above are inconsistent (e.g., 30 facilities, 35 facilities and 40 facilities). • Please clarify, correct and provide Appendix A was revised so that the facilities in Table A-l add to 132 facilities. We evaluated the timeliness of actions at 35 total facilities. Some facilities received more than one enforcement action. Each action was assessed individually, thus one facility may have had an action judged untimely and another action for which we could not assess the timeliness. Thus, 45 actions at 30 of the 35 facilities were untimely. We could not determine the timeliness of the remaining 12 actions that occurred at 10 of the 35 facilities. The apparent inconsistency stems from the overlap that occurred during our analysis because facilities may have had more than one action. Thus, an action was classified as either untimely or not assessed. In contrast, a facility with more than one action could be included as both untimely and not assessed. We added clarifying language in the report. 28 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response accurate facility and enforcement action totals. 10 Chapter 2, pages 3-4 "EPA Guidance Misinterpreted and Inadequate, p. 6 - 7: OECA strongly disagrees with OIG's statements that the EMS guidance that describes timely and appropriate enforcement actions is inadequate. On the contrary, the EMS provides fully adequate guidance on timely and appropriate enforcement responses for various types of violations, including those designated as SNC. OIG acknowledges at page 6 that "Flexibility is needed in any national enforcement program" yet contradicts this statement by continuing to apply the EMS as a set of rigid requirements (rather than flexible guidance). The OIG fails to recognize that a flexible enforcement response is necessary to manage the NPDES enforcement program. The need for flexibility is even greater today than when the guidance was first established because of the significant expansion of the universe of regulated entities covered by the NPDES program without a corresponding increase in EPA resources. On page 6, OIG states that "EMS guidance is unclear regarding appropriateness of enforcement actions ..." OECA strongly disagrees - the EMS is very clear that an appropriate response should,"... reflect the nature and severity of the violations, and, unless there is supportable justification, the response must be a formal enforcement action, or a return to compliance by the permittee . . . In the rare circumstances when formal enforcement action is not taken, the administrating agency is expected to have a written record that clearly justifies why the alternate action (informal enforcement action OECA's comment is directly contradicted by its October 6, 2006, response, which states "that its guidance documents are indefinite in regard to the timeliness standard." Our report states that EPA has indistinct guidance to determine if actions are appropriate because the guidance does not state how expeditiously facilities should come back into compliance after an action is taken. Our report states that we believe the guidance is clear on the definition of "timely." However, we were given inconsistent interpretations of what EPA considers timely. EMS states a timely formal enforcement action must be taken by the end of the quarter following the SNC violation. For instance, if a facility experiences an SNC violation in February of the first quarter, a formal enforcement must be taken by the administering agency by June 30 of the second quarter. The October 6, 1999, memorandum Drovidcs clarification of existins EMS language concerning timely and appropriate enforcement response to SNC. The memo represents the Water Enforcement Division's official interpretation of the EMS language concerning timely and appropriate enforcement. OECA states that the memo has not been rescinded. OECA has chosen not to implement the memo, but the memo is the current guidance. We are concerned that OECA is more concerned about current practice than implementing current guidance. or permit modification) was more appropriate" (see EMS, Chapter 11, Attachment B, p .2 and page 7 of this document). On pages 6-7, the OIG states that "EMS guidance is clear with regard to timeliness" in one paragraph yet relies on a subsequent October 6, 1999 OECA memorandum in another paragraph to define what is timely. The OIG relies on this memorandum and not the EMS guidance that reflects current practice. OECA agreed in its October 2006 meeting with OIG to rescind the 1999 OECA memorandum that conflicts with the EMS. OECA also stressed that 1999 29 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response memo is not being implemented by Headquarters and the Regions. 11 Chapter 2, page 4 OECA disagrees with the OIG's conclusions on page 7 regarding whether the response to East Chicago was appropriate. The implication that there was inadequate enforcement attention at this site between 1988 and 2004 is not valid (see attachment 2 for a PCS list of all formal and informal actions at the facility, including available compliance schedule milestones). It is unclear how the OIG calculated the range of excess pollutant loadings for East Chicago. It is unclear if the OIG used the actual flow data taken from the DMRs during its file review, or used design flow data from PCS. Flow data are not available in OTIS. Since the 1988 consent order, EPA and/or Indiana have issued the following formal actions: a pretreatment administrative order in January 1993, a State-agreed order in November 2003 (not shown in OECA's attachment 2), and a pretreatment enforcement action in 2004. A stipulated court order was completed in October 1991. All other actions included in the attachment are informal actions, such as phone calls and warning letters. Despite these actions, East Chicago had 12 consecutive quarters of effluent violations from July 2002 through June 2005. We evaluated the agreed order and determined it was untimely because it addressed violations from March and April 2000. We could not make a determination on the pretreatment enforcement action taken in 2004 due to a lack of information in PCS and the data files. As stated in our report, we question the appropriateness of an enforcement action taken in 1988 that, according to OECA, "renders all effluent violations resolved pending since that time," while the facility continues to violate permits limits and discharges to an impaired waterbody. For this facility, our estimate of pollutant loadings used the design flow. The report's scope and methodology clearly lays out how we calculated the excess pollutant loadings. 12 Chapter 2, page 5 Insufficient Formal Actions at 22 Facilities, p. 8: OECA is reluctant to accept this finding without further evaluation under a correct interpretation of the guidance. This reluctance is based on OECA's experience reviewing the facilities that the OIG examined in its August 2006 draft report wherein its findings were dramatically different than OIG's findings (e.g., 53 or 93% received an appropriate enforcement response). As explained in OECA's response to the August 2006 draft report, OIG's inaccurate interpretation of the EMS, failure to consider whether a facility was in "resolved pending" status, and the exclusion of relevant information, results in inaccurate findings. The same problems exist in this draft report. See OIG response to comments 7 and 10 above. We found that 22 facilities (21 facilities once Decherd is removed, see response to comment #23) in SNC received insufficient formal actions and 30 facilities did not receive timely actions. We stated that we could not judge the appropriateness of the actions because OECA had not clearly and unambiguously defined how quickly a facility must come into compliance. OECA stated that 53 facilities received an appropriate response, but it failed to state 53 received a timelv response. We considered whether a facility was in 30 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response "resolved pending" as indicated by PCS or OTIS and OECA submitted data. We found that 13 facilities had received no formal action during our review period but had received actions prior to the start of our review period. We considered these actions in our review. However, we found that all 13 facilities had unaddressed SNC violations during our review period. As stated in Chapter 3, OTIS data for Regions 4, 5, and 6 do not show facilities with active enforcement actions as resolved pending even though they are required to do so. Thus, OECA's data systems do not show the correct enforcement status of facilities, as required by regulation. 13 Chapter 2, page 5 OECA bases its findings on follow-up discussions with Regional staff - to OECA's knowledge it does not appear that the OIG based its findings on follow-up discussions with Regional staff. The fact that we disagree with OECA on many of the facts and inferences in this report does not imply that we ignored EPA's views. On the contrary, we carefully reviewed all written comments and verbal comments made at the numerous meetings we attended with OECA and regional staff. Following completion of field work and submission of our discussion draft to OECA on May 26, 2006, we met with OECA and/or regional staff on June 7, June 29, October 24, October 26, and November 8, 2006. Further, OECA submitted data on November 16 (in response to the November 8 meeting with OECA and regional staff) that reflected consultation with regional staff. 14 Chapter 2, page 5 OECA disagrees with the statement that"... facilities with SNC violations must receive a formal enforcement action ... or a return to compliance within the quarter following the SNC violation." OIG's statement, excerpted in part, from the EMS implies that every SNC violation requires a formal enforcement action. In fact, the EMS provides flexibility to Regions and states to address SNC - the EMS allows that an appropriate response may be formal or informal, with justification, or that a facility may return to compliance on its own. As stated previously, the OIG's draft report does acknowledge that informal actions are an option; however, the report focuses solely on whether EPA or the states took formal enforcement action at the 57 facilities. By only evaluating whether a formal enforcement action had been taken, the evaluation ignores the provisions in the EMS which allow an See OIG response to comments 7 and 10 above. Alabama and/or EPA should have issued a formal enforcement action by the end of September 2004. The informal action was issued after the facility had been in SNC for four consecutive quarters. 31 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response informal enforcement response. The EMS recognizes that an informal enforcement response may be the most appropriate response given the circumstances of the case. Informal responses are used successfully by EPA and states to address SNC violations. • For example, Alabama issued a Notice of Violation (NOV) to Centerville to address chlorine total residual. The NOV resulted in the permittee changing its treatment technology for disinfection and submitting a schedule for making such change. Clearly, this informal action induced the permittee to take corrective action. 15 Chapter 2, page 5 The draft report fails to acknowledge the compliance status classification of "resolved pending," i.e., a permittee in compliance with enforcement order and on track to achieve compliance pursuant to a compliance schedule but has not yet achieved full compliance with permit conditions. OECA believes that the OIG misinterpreted SNC violations for several facilities as being "unaddressed by enforcement action" when in fact, they were addressed. Below are just a few examples of facilities that were resolved pending during OIG's 12 quarter review period : We considered whether a facility was in "resolved pending" as indicated by PCS or OTIS and OECA submitted data. We found that 13 facilities had received no formal action during our review period but had received actions prior to the start of our review period. We considered these actions in our review. However, we found that all 13 facilities had unaddressed SNC violations during our review period. As stated in Chapter 3 of our draft report, OTIS data for Regions 4, 5, and 6 do not show facilities with active enforcement actions as resolved pending even though they are required to do so. Thus, OECA's data systems do not show the correct enforcement status of facilities, as required by regulation. To overcome this issue, OIG utilized OECA's own extensive data collection efforts provided on November 16, 2006. See OIG response to comments 16 to 19 below. 16 Chapter 2, page 6 City of Attalla, AL - Alabama issued an administrative consent order on February 26, 2002, to this facility addressing violations of effluent limits for BOD5 mass and concentration, TSS mass, TSS % removal, BOD5 % removal, pH, and Toxicity, and for Sanitary Sewer Overflows. The consent order required return to compliance within 1095 days upon the effective date of the order. When problems persisted, Alabama filed a complaint in state court on February 10, 2003 resulting in a consent decree which was finalized on May 13, 2005. Alabama did not wait until February 2005 (1095 days) to The consent decree issued on May 13, 2005, is an escalation of an existing action. Thus, it is unclear if this escalation is timely according to the EMS criteria. However, it should be noted that the February 26, 2002, action, which preceded our evaluation time period, would be considered untimely since it addressed 27 sanitary sewer overflows reported from January 1999 through October 2001 and 12 violations of biochemical oxygen demand and total suspended solids from June through November 2001. This facility had 11 quarters of effluent violations from July 32 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response escalate formal enforcement when it believed a stronger approach was necessary to assure the facility will address its problems expeditiously. 2002 through June 2005. 17 Chapter 2, page 6 Lafayette STP, TN- Tennessee issued an administrative order against this facility in April 2002 to address violations of effluent limits for Nitrogen, Ammonia, TSS, fecal coliform, and DO. The order required expansion of the plant (completed in 2004) to achieve compliance. The Lafayette STP was not a major facility until the 3rd Quarter of 2004. The STP continued to have problems with Nitrogen, Ammonia, and Settleable Solids. It is not uncommon for violations to occur during plant expansion and start-up because it takes time for the activated sludge to form, to establish optimum activated sludge wasting rate, mix liquor concentration, and to reach a steady state of operation. When violations persisted after the start-up of the new plant and allowing for a period of time to reach steady state of operation, Tennessee issued a second administrative order in May 2006. The Facility Registry System, which relies on PCS data, lists Lafayette as a major since August 2002. (See Comment #23) The April 2002 administrative order did not address settleable solids. During our review period, the facility was in SNC for seven quarters for settleable solids that were not addressed by formal enforcement actions. Moreover, this facility had three consecutive quarters of SNC effluent violations for settleable solids after third quarter 2004, when OECA claims this facility became a major. Thus, even under OECA's facility classification, this facility should have received a formal enforcement action before the end of second quarter 2005. 18 Chapter 2, page 6 Austin, IN- Per the PCS comment field for this facility, the state's 2002 order specifically addressed the NPDES schedule violations. The only issue was that the order was not properly linked in PCS to show that the schedule violations were resolved pending. Austin Municipal Wastewater Treatment Plant had at least four consecutive quarters of compliance schedule violations after the February 2002 Consent Order. These SNC violations were unaddressed by enforcement actions. According to OECA, the state's 2002 Consent Order addressed the NPDES schedule violations that occurred after the order was issued. If issuing an administrative order resolves future compliance schedule problems, we do not understand how OECA has any facilities in SNC for compliance schedules violations and do not understand why it would be a category of SNC. 19 Chapter 2, page 6 Hillsboro, TX- The city of Hillsboro had multiple effluent violations during 2002. Texas asked Region 6 to issue an administrative order addressing these violations while the state developed an enforcement action. An administrative order was issued by EPA Region 6 on March 25, 2003. When the city failed to respond adequately to the order, Region 6 issued another administrative order on June 12, 2003, which also addressed continuing effluent violations. The city submitted an acceptable compliance plan on July 17, 2003, which was monitored closely while Texas continued to OIG's analysis indicates that both of these enforcement actions were untimely. The March 2003 Administrative Order addressed violations from January 2002 to October 2002. The June 2003 Administrative Order addressed violations from November 2002 to March 2003. 33 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response develop its enforcement action. The permittee has been in resolved pending status from April 2004 through the present except for the quarter of April through June 2005 when they had additional effluent violations. These violations were addressed by a state administrative order issued on July 14, 2006. This action required the facility to come into final compliance with effluent limits no later than October 12, 2006. The facility complied with this requirement and has had no further violations to date. 20 Chapter 2, page 7 OIG's draft report still includes Michigan facilities (e.g., Flint Ink and Great Lakes Tissue) even though a known state-wide data system problem, discussed at length with the OIG in response to its August 2006 draft, resulted in incorrect SNC determinations and subsequent management decisions not to respond to some effluent violations. DMR data input issues in Michigan existed from October 2003 through September 2005. As such, reports of violations occurring during this timeframe, including non-reporting violations, are questionable. OECA believes EPA Region 5 appropriately exercised its enforcement discretion to make the decision not to respond. OECA's earlier comments only referred to State-wide data system problems involving DMRs and compliance schedule reporting. As a result of those comments, we eliminated 25 of the 30 Michigan facilities from our timeliness assessment. We retained five Michigan facilities that had effluent violations in our timeliness review. If PCS coding issues extend to effluent violations, as OECA now suggests, we are even more concerned about the integrity of EPA's data systems. However, without additional evidence of those newly reported data problems, we are retaining these five cases in the final report. 21 Chapter 2, page 7 • For example, the 12 consecutive quarters of effluent violations cited in OIG's draft report at Flint Ink were actually toxicity violations that occurred for one quarter (February and March 2000). As explained to the OIG on several occasions, these violations continued to appear due to improper coding in PCS. Further, Michigan worked with Flint Ink to find the source of the toxicity and discovered an issue with a water additive. Flint Ink's permit was subsequently modified to withdraw permission to use the additive and the company ceased use. A multi- media inspection conducted by Region 5 in 2006 confirmed cessation of the additive and that the company was moving a majority of its process out of state thus eliminating all discharges to surface waters. The toxicity violations were addressed and resolved by state action (compliance assistance and permit modification), and the general non- reporting violations were a direct result of Michigan's data system issues. OECA reported that the known State-wide data system problems involved only reporting issues such as discharge monitoring reporting and compliance schedule reporting. Flint had 12 quarters of effluent violations during our review period. Flint was still in SNC for effluents violations in September 2006 according to EPA's data system, ECHO. OECA's data provided on November 16, 2006, did not state that the 12 quarters of effluent violations were toxicity violations from February and March 2000. OECA's response is contradictory. OECA states that the violation is due to improper coding. Then it states there was a toxicity problem, and that it was addressed and resolved through State action. From OECA's response and our analysis, we believe Flint was in SNC and did not receive timely action. 34 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response 22 Chapter 2, page 7 • The two consecutive quarters of mercury violations at Great Lake's Tissue cited in OIG's draft report were accurately portrayed. However, since they occurred at the very end of OIG's review period, OECA does not understand how the conclusion was reached that no action was taken. PCS indicates, and the Region has confirmed, that Michigan referred the case to the state Attorney General in July of 2005. A timely enforcement action should have been taken by Michigan to address the mercury effluent violations at Great Lakes Tissue by June 2005, which is prior to the end of our evaluation period. 23 Chapter 2, pages 7-9 OIG completely disregarded OECA's October 2006 comments to the August 2006 draft report relating to a number of minor facilities in Regions 4 and 6 that were inaccurately included in his analysis as majors- these facilities were not classified as majors during the analysis and therefore not subject to SNC screening during all or part of the OIG's evaluation period. Region 4 has reconfirmed that 3 facilities in Tennessee (Lafayette STP, Decherd City STP and Munford Lagoon) were not major facilities during all or part of the OIG's 12 quarter review period. Lafayette did not become a major facility until the 3rd quarter of 2004 when the plant expanded. Two other Region 4 facilities in TN (Decherd City STP and Munford Lagoon) were not classified as major facilities until the 4th Quarter of 2004 and the 4th Quarter of 2005, respectively. Neither of these facilities had SNC violations reported in the QNCR during the OIG evaluation period and therefore should not have been subject to SNC evaluation by the OIG (see attachment 2 for letters from Region 4 files documenting status change dates). Region 6 has reconfirmed that 4 facilities in Texas (New Boston, SWS Holdings, Texas Dept. of Criminal Justice, and Windermere Utility) were not major facilities during the OIG's 12 quarter review period. Although PCS is the legal database of record for the NPDES program, dates of status change are not tracked in PCS - the date associated with a change in facility status from a minor to a major is manually tracked in files at the Regional office. For three of the four Texas facilities (New Boston, SWS Holdings, Texas Dept. of Criminal Justice), the OIG's conclusions were based on dates it found in the Facility Registry System (FRS) - an integrated, comprehensive, multi-media We removed Decherd (TN0020508) from our sample. We included Decherd, as well as the other facilities OECA questions, in our sample because EPA databases showed that facility was a major facility in SNC throughout our review period. However, our findings regarding the other six facilities are still valid. We obtained data on the status of the six facilities from the Facility Registry System. According to Office of Environmental Information staff and data documentation, the Facility Registry System pulls information from PCS. The Facility Registry System would only show that a facility is a major if PCS was updated and also considered it a major. The Facility Registry System shows the status of facility (major or minor) and the date in which the last change was made in the Facility Registry System. Office of Environmental Information staff verified that the date the information was last updated would only change if the data in PCS changed. The Facility Registry System confirms that Lafayette STP (TN0020877) has been a major since August 2002. Even using OECA's data interpretation, Lafayette had three consecutive quarters of SNC effluent violations for settleable solids after becoming a major. This facility did not receive a timely formal enforcement action to address those SNC violations. According to the Facility Registry System, the Texas Dept. of Criminal Justice (TX0031577) has been a major since November 2001; New Boston (TX0026018) has been a major since May 2002; SWS Holdings (TX0070955) has been a major since August 2002; Windermere 35 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response data system maintained by EPA's Office of Environmental Information (OEI). OEI staff confirmed that FRS does not track dates associated with a change in facility status - it tracks dates of permit application receipt and issuance. Thus, it is not possible to determine when a facility became a major by looking in the FRS. For example, the August 30, 2002 date that the IG based its conclusion on for SWS Holdings was the date the permit application was received, not the date the facility became a major. FRS is not the legal database of record for NPDES program. It should be noted that the new ICIS-NPDES data system will improve on the current functionality available in PCS by providing a "begin date" and "end date" to track the conversion. In addition, as a result of the State Review Framework project, the Integrated Data for Enforcement Analysis System (IDEA) now contains quarterly "snapshots" of the major/minor and active/inactive facility status of the NPDES permitted universe. This feature is an additional source of historical information that OECA now can access as needed. For one of the four Texas facilities (Windermere Utility), the OIG based its conclusion that the facility was a major on the fact that the facility has a design flow of 2 MGD. EPA's definition of a major municipal facility is a facility with a design flow > 1 MGD, however, this facility is not a municipality. It has an SIC code of 6552 and is therefore subject to EPA's definition of major (which is a complex formula considering a number of factors) for an industrial facility (see attachment 2 for letters from Region 6's files documenting status change dates). (TX0074853) has been a major since March 2004; and Munford Lagoon (TN0062499) has been a major since September 2004. Windermere had three consecutive quarters of SNC violations after March 2004 that were not addressed with formal enforcement actions so it did not receive timely action. Similarly, Munford Lagoon had three consecutive quarters of SNC violations after September 2004 that were not addressed with formal enforcement actions. As additional evidence, we spoke with the consulting engineer for the City of New Boston, who stated that New Boston has been a major for at least 10 years. 24 Chapter 2, page 9 Untimely Enforcement at 30 Facilities, p. 8 - 9: OECA is reluctant to accept this finding because of OIG's misinterpretation of the EMS timeliness standard. See comments 25 to 27 below. 25 Chapter 2, page 9 As noted above, the OIG relies on a 1999 memorandum in spite of OECA's explanation that the timely standard in the memorandum is not being implemented. OECA stated in October 2006 that the timely standard needs to be clarified and we will rescind the 1999 memo. We relied on the 1999 memorandum because it provides the Water Enforcement Division's official interpretation of the EMS language concerning timely and appropriate enforcement. OIG does not believe rescinding the memo will clarify the timeliness standard. On the contrary, rescinding the memo will make the timeliness standard open to interpretation. 36 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response 26 Chapter 2, page 9 The OIG ignores flexibility in the EMS language regarding timeliness (see Chapter II, Attachment B). The purpose of guidance, such as the EMS, is to provide recommendations and set program expectations - not establish regulatory requirements (see Appalachian Power Co. v. EPA, 208 F. 3d -1015 (D.C. Cir. 2000) and Barrick Goldstrike Mines Inc . v. Browner, 215 F.3d 45 (D C . Cir. 2000), finding that EPA guidance cannot have the binding effect of a rulemaking). The OIG continues, however, to treat the recommendations in the EMS as rigid requirements on the Agency. Our evaluation does not advocate using the EMS as regulatory requirements. It simply uses OECA's own program principles to assess how well the Agency is issuing timely and appropriate formal enforcement actions for long-term SNC violations. In our view, OECA needs to have criteria in place so that it can provide adequate oversight to region and State programs. 27 Chapter 2, pages 9-10 The OIG fails to take into account the time needed to work with a state, gather more evidence to support a case, allocate resources to an action and negotiate complex corrective measures to resolve noncompliance. The OIG does not acknowledge that state enforcement processes and timelines must be considered in determining reasonable timeliness. To illustrate, many states have the authority to issue monetary penalties within a compliance order. Issuance of such an order may require longer time frames than in the EMS due to public comment and participation requirements, or because a hearing is scheduled, or because of complex negotiation schedules. Thus, judgments on what is a reasonable time table for action must ultimately be case-specific in consideration of complex compliance problems that may require long term studies and complex injunctive relief (see Barnes Memo, Criteria for Direct Federal Enforcement in Delegated States, at pages 21 - 25). See OIG response to comments 7 and 10 above. The OIG's evaluation applies the programmatic principles set forth in the EMS to determine how well EPA regions and States are fulfilling their enforcement responsibilities. The EMS represents OECA's own enforcement principles for EPA regions and States. The EMS also provides the flexibility for each administering agency to develop management procedures which are best suited to its operations and resources with the goal of most efficiently translating compliance information into timely and appropriate enforcement action. Thus, the flexibility desired by OECA, to consider many of the issues mentioned, is implicitly considered in the EMS. We appreciate the need for flexibility; however, we believe that OECA needs to have criteria in place so that it can provide appropriate oversight to region and State programs. 28 Chapter 2, page 10 • For example, it is incorrect for OIG to conclude on page 8 of the draft report that EPA's Administrative Order against ConocoPhillips was untimely. There are complex selenium water quality standard and permitting issues associated with this facility. EPA was in extensive negotiations with state partners, as well as Fish and Wildlife Service and the facility to resolve the complex permitting issues. The flexibility inherent in EPA's enforcement guidance is critical to being able to account for such complex circumstances. EPA issued an Administrative Order against ConocoPhillips Company on August 14, 2003, to address effluent violations for selenium beginning July 2002. The action was untimely because it was not taken by December 2002 (the end of the next quarter), as required by EMS. According to PCS, the facility continued to exceed selenium permit limits through June 2005. 37 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response 29 Chapter 2, page 10 • For example, the OIG states on page 8 of the draft report that it could not determine timeliness of 12 formal actions involving 10 facilities because the available information was not adequate. Furthermore, the OIG states on page 9 of the draft report that for Columbiana WWTP, AL, timeliness could not be assessed because the action was not linked to the violation. OECA does not understand this statement given that Alabama provided compliance and enforcement files containing a copy of a Consent Order, Complaint and Consent Decree for this facility to the OIG during its onsite file review on February 6-7, 2006 and the OIG indicates that Alabama's files were 100% complete in Table 3-1 on page 11 of the draft report. As stated in the report, we could not determine the timeliness of the remaining 12 formal actions at 10 facilities because available information was not sufficient. Specifically, the enforcement actions were not directly linked to the SNC violation addressed or when the SNC violations occurred. Without this information, we could not determine whether the action was timely. Alabama did have files on all eight facilities from our sample. The Consent Decree was in the file. However, the Consent Decree simply accelerates an existing compliance schedule for a Consent Order issued on March 21, 2002. The language in the document does not directly link the action to an SNC violation. Therefore, we could not address timeliness. 30 Chapter 2, page 10 It is not clear why the IG did not take advantage of the Watch List for its analysis. The Watch list and associated data provides much better insight related to case-specific circumstances and OECA's tracking of enforcement timeliness. The OIG was provided access to the Watch List, but it is unclear whether it was used. The OIG evaluated whether EPA conducted (1) timely enforcement on NPDES major facilities in SNC, and (2) proper oversight over regions and States. The Watch List is derived from EPA's data systems. To avoid errors, we utilized the original data in EPA's data systems. Our methodology is presented in Appendix A. 31 Chapter 2, page 10 Timely Enforcement Helps Minimize Excess Pollutant Discharges, p. 9: See OECA response at page 3, At A Glance, concerning the accuracy of the pollutant reduction estimate presented here. See OIG response to comment 2 above. 32 Chapter 2, pageslO-11 Recommendation 2-1, p. 9: Clarify and enforce EMS guidance to ensure that timely and appropriate formal enforcement actions are taken against NPDES facilities in SNC. OECA Response: OECA concurs with the recommendation to clarify its NPDES timeliness guidance. As discussed with OIG, OECA will rescind the 1999 OECA memorandum in conflict with the EMS and will discuss this with the Regional NPDES Enforcement Branch Chiefs at the national NPDES meeting in April, 2007. OECA disagrees, however, with the OIG's commendation to "enforce EMS guidance." The EMS is a management tool which facilitates evaluation and performance review. It is not a set of rigid regulatory requirements which EPA imposes on the states or Regions as the use of the term "enforce" suggests. Rescinding the guidance without a satisfactory replacement will not "ensure that timely and appropriate formal enforcement actions are taken against NPDES facilities in SNC." If OECA rescinds the 1999 memorandum, it will need to explain how it will unambiguously define timely and appropriate enforcement. We understand OECA's concerns about the use of the word "enforce" in the recommendation. We will change the word "enforce" to "implement" in the final report. 38 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response 33 Chapter 3, page 11 Chapter 3: Inadequate Records and Reporting Inhibit EPA Oversight of NPDES Major Facilities, p. 10 - 13. Compliance and Enforcement Files Are Incomplete, p. 10: The authorized states are responsible for keeping accurate records. As such, EPA does not require that all information be stored in duplicate in the Regional offices. The OIG appears to assume that EPA should have hard copies of all state files. This redundancy is neither practical nor efficient. EPA has the ability to request information at any time and has the ability to audit state files. The State Review Framework (SRF) established a file review protocol that will be implemented in all 50 states by the end of FY 2007. When a state is not keeping accurate records (e.g., inspection reports or enforcement actions), the SRF process will identify this and formal recommendations and milestones will be made to ensure better file management. To the extent that violations or problems noted in files are not in the databases, the SRF process will also address this. An accurate history of the compliance and enforcement activities at a facility is important for oversight and making future enforcement decisions. The lack of accurate information inhibits EPA's ability to provide effective oversight to NPDES major facilities and thus protect human health and the environment from excess levels of toxic or harmful pollutants. We recognize that the State Review Framework may help address file deficiencies at States noted in this report. However, only the active oversight and management of the State Review Framework process will result in improvements. It is not clear what steps EPA plans to take to improve its oversight and management in this area. 34 Chapter 3, page 11 EPA's Enforcement Data Systems Are Inaccurate, p. 11 - 12: OECA does not agree with the statement that EPA's data systems are inaccurate. The OIG provides no context supporting this broad statement but rather appears to make this statement based upon a specific, known problem in one state (Michigan). The OIG should indicate that EPA requires states to enter at least 95% of discharge monitoring report (DMR) and effluent limit data (which generate SNC status). Historically there are only two states that do not meet or come very close to this goal - Michigan, and Oregon. As EPA provides CWA data to hundreds of thousands of users via the ECHO web site, it is unfortunate that the OIG report classifies the data systems as inaccurate because there are data translation issues in two of fifty states. The ECHO web site clearly caveats these problems. These caveats should have first alerted OIG staff that Michigan data should be avoided and could not be used to draw accurate conclusions about facility SNC status. In addition, OECA, and the Regional office pointed out to the OIG staff on numerous occasions that the Michigan data should not be used in the OIG study; however, the OIG decided to include the flawed data anyway. In regard to the SNC status in other Our findings about NPDES program enforcement data systems being inaccurate go well beyond our findings in Michigan. Our report contains several examples of EPA's data system problems, including several that EPA has acknowledged elsewhere. EPA's data systems do not accurately reflect the SNC status of facilities. Once a formal enforcement action is issued for an SNC violation, a facility should be designated as "resolved pending" for that violation, according to Title 40, Code of Federal Regulations, Section 123.45. However, OECA states that Regions 4, 5, and 6 (the three regions reviewed) do not designate facilities in "resolved pending." Thus, facilities in these regions may appear to remain in SNC status in EPA's data systems for a long time despite operating with an existing enforcement action. EPA staff also stated that once a facility receives a violation for submitting DMRs late, that violation will continually show as a violation in EPA's data systems until regions or States manually correct the problem. 39 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response states evaluated, the OIG audit points out that many facilities that are flagged in the data systems as having SNC violations are actually covered under enforcement orders (thus are technically "resolved pending"). EPA is aware that many Regions and states prefer to keep the SNC flag on to keep pressure on the facility to complete enforcement order milestones. OECA believes that states should have the flexibility to manage SNC data as the states determine is appropriate. Keeping the SNC flag on is more informative to the public (e.g., it puts more pressure on the owner/operator of the plant to come back into compliance). • Revise the title of this section to read "SNC and violation data are inaccurate in Michigan" - this revised language is consistent with what OECA has already posted on its web site. The State of Michigan has had data problems. We could not evaluate the timeliness of enforcement actions at 25 of the 30 Michigan facilities due to the State- wide data problem. Even though we did not evaluate timeliness of enforcement actions at the 25 facilities, we did report on the obvious oversight problems that exist from a State having system problems for years. The ECHO Website notes the following data accuracy issues. 1. A number of EPA-authorized NPDES facilities in Region 10 do not have up-to-date permit limits entered in PCS. 2. DMR non-receipt is not being tracked for California Clean Water Act NPDES permits. 3. In Indiana, many non-receipt violations are erroneous and are currently being investigated with resolution pending. 4. In Delaware, Virginia, and West Virginia, some pH violations are in error. No revision is necessary. 35 Chapter 3, page 12 Many Violations Found During Inspections not Reported in PCS, p. 12: This is an issue that has concerned OECA, and we have been working with the Regions and states to improve such reporting since 2003. As noted in OECA's October 2006 response, the number of reported violations rose from 220 in 2003 to 4,441 in 2005. The OIG Report fails to acknowledge substantial progress made by OECA in this area. EPA has also been responsive to state requests for clarification on reporting standards by finalizing more clear guidelines in regard to how such information should be reported to PCS. To ensure EPA has a full record of violations found, EPA has also proposed to the states that violation reporting be expanded to non-major facilities. The steps taken by OECA are encouraging; however, OECA must implement a quality assurance program to ensure that violations found during inspections are reported in PCS. 36 Chapter 3, page 12 Bacteria not Reported as SNC, p. 12: The permit parameters that are "eligible" to be automatically tracked are specifically noted in the RNC regulation, so EPA would be unable to simply begin tracking these as RNC or SNC without a regulatory change. Despite this obstacle, OECA has piloted additional See OIG response to comment 38 below. 40 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response management controls to assist in tracking bacteria violations at direct dischargers. For direct discharge facilities, the Watch List process provides Regions and states with the capability to flag facilities that have continuing exceedances - which includes bacteria and other pollutants that are not evaluated for SNC. While only some Regions and states are required to report under this pilot, other Regions and states use the Watch List as way to monitor these other parameters that are not included in the regulation. OECA is considering whether this pilot should be extended. OECA is also developing new reporting capabilities that will enable users to automatically access these violations. In OECA's October 2006 comments to the OIG, EPA's existing NPDES SNC policy was described which includes the discretion to flag "any other violation of concern" (such as bacteria) as SNC. OECA is working with an EPA-state advisory group to supplement the existing NPDES SNC policy to address significant violations from wet weather point sources (Combined Sewer Overflows, Sanitary Sewer Overflows, Concentrated Animal Feeding Operations and stormwater) which include the most significant sources of pollutants (including bacteria) not covered by the existing SNC Policy . 37 Chapter 3, page 12 Recommendations 3-1, page 12: Implement a quality assurance program addressing the completeness of compliance and enforcement files, accuracy of EPA data systems, and reporting in PCS the violations found during inspections. • OECA Response: OECA concurs and believes that its existing processes - which include SRF, the Watch List and ECHO - provide a reasonable amount of quality assurance. These substantial efforts were not discussed in the OIG report. In regard to inspection-related violations, EPA continues to negotiate with states to obtain this reporting, and is just beginning a new Regional pilot to ensure that EPA-determined violations are accurately tracked. OECA's existing activities (State Review Framework, the Watch List, and ECHO) do not represent a quality assurance program. The Watch List and ECHO do not appear sufficient to address completeness of files, data accuracy, and reporting inspection violations. Moreover, the State Review Framework may help States address these issues, but not necessarily EPA regions. OIG is not convinced that these existing efforts will sufficiently address our recommendation. OECA needs to implement a quality assurance program addressing the completeness of compliance and enforcement files, accuracy of EPA data systems, and reporting in PCS the violations found during inspections. 38 Chapter 3, pages 12-13 Recommendation 3-2, page 12: Establish controls allowing EPA leadership to identify significant noncompliance by bacteria-only violators for enforcement action. • OECA Response : OECA concurs and is pursuing this in three other ways that may If OECA believes these actions will establish controls so EPA leadership can identify SNC by bacteria-only violators for enforcement action, we will accept their response. We will need to obtain an action 41 ------- # OECA Response, Section/Page OECA Comments (Attachment 1 of its response) OIG Response have the same intended effect of addressing bacteria without the significant resource expenditure it would take to re-open the table of parameters found in the QNCR regulation at 40 CFR Part 123 .45 : 1) use of Watch List pilot criteria which includes bacteria, 2) develop better analytical tools that can assist the Regions and states in easily flagging these problems and tying the problems to watershed quality, and 3) continue to provide discretion for Regions and states to elevate serious bacteria violations to SNC status via the existing NPDES SNC policy . In addition, to improve public access to information about such violations, OECA will be adding effluent report charting (including bacteria violations) to its ECHO database in 2007. In addition, the most significant sources of bacteria (combined sewer overflows, sanitary sewer overflows, concentrated animal feeding operations, and storm water) are included in an effort underway by OECA's Office of Civil Enforcement to develop a wet weather SNC policy. plan and milestone dates for when these actions will occur. 42 ------- Appendix E Distribution Office of the Administrator Assistant Administrator for Enforcement and Compliance Assurance Regional Administrator, Region 4 Regional Administrator, Region 5 Regional Administrator, Region 6 Agency Followup Official (the CFO) Agency Followup Coordinator General Counsel Principal Deputy Administrator for Enforcement and Compliance Assurance Director, Office of Regional Operations Director, Water Enforcement Division, Office of Enforcement and Compliance Assurance Office of Enforcement and Compliance Assurance Followup Coordinator Regional Audit Followup Coordinator, Region 4 Regional Audit Followup Coordinator, Region 5 Regional Audit Followup Coordinator, Region 6 Associate Administrator for Congressional and Intergovernmental Relations Associate Administrator for Public Affairs Acting Inspector General 43 ------- |