53 FR 37082-37247 Friday, Sept. 23, 1988 40 CFR Parts 280 and 281, Underground Storage Tanks; Technical Requirements and State Program Approval; Final Rules--Preamble Section III. Today's Final Rule
This section provides a summary of EPA's final rule. It also identifies and describes several major points of departure from the proposed rule, several alternative strategies, public comments on them, and the Agency's rationale for the direction of the final rule in several other key areas. More detailed summaries of all the public comments and the Agency's responses to them can be found in the "Comment and Response Summaries Background Document" that has been placed into the public docket in support of today's final rule.
EPA is today promulgating regulations for underground tanks storing either petroleum or hazardous substances other than hazardous wastes regulated under Subtitle C of RCRA. These requirements establish measures for both new and existing UST systems to prevent, detect, and clean up releases from these systems. These final requirements of Part 280 fulfill the mandates of RCRA sections 9003(a), (c), and (e), and sections 9009 (a) and (b). The major elements of today's final rule are noted below.
o New UST systems must be designed and constructed to retain their structural integrity for their operating life, in accordance with national consensus codes of practice (see Table 1 provided earlier in this preamble). All tanks and attached piping used to deliver the stored product must be protected from external corrosion. Cathodic protection must be monitored and maintained to ensure that UST systems remain free of corrosion.
o Nationally recognized industry installation standards must be followed in placing new UST systems in service (see Table 1). Owners and operators of new USTs must certify that proper installation procedures were followed and identify how the installation was accomplished.
o Owners and operators of both new and existing UST systems must follow proper tank filling practices to prevent releases due to spills and overfills. In addition, owners and operators of either new or upgraded UST systems must use devices that prevent overfills and control or contain spills.
o Tanks must be repaired in accordance with nationally recognized industry codes (see Table 1). These national codes include several tests that must be conducted to ensure quality repairs.
o To close UST systems, industry-recommended practices must be followed: the UST system can be removed from the ground or left in place after removing all regulated substances and cleaning the tank, filling it with an inert substance, and closing it to all future outside access ( see Table 1). In addition, owners and operators must perform an assessment at the time of UST closure to ensure that a release has not occurred at the site. If a release has occurred, then corrective action must be taken.
o Release detection must be instituted at all UST systems. For petroleum UST systems, several methods will be allowed, although tank owners and operators must adhere to requirements concerning their use. In addition, owners and operators must follow special requirements for pressurized delivery lines. Petroleum UST systems are not required to have secondary containment with interstitial monitoring. All new or upgraded UST systems storing hazardous substances, however, are required to have secondary containment with interstitial monitoring, unless an alternate release detection method is approved by the implementing agency. The owners and operators must demonstrate to the implementing agency that a release detection method will detect releases of the stored substance in a manner no less stringent than the release detection methods allowed for petroleum USTs and that a method of corrective action is available to clean up a release of the hazardous substance should one occur.
o Generally, release detection at existing UST systems must be phased in over a 5-year period based on the age of the tank. The oldest UST systems (usually unprotected from corrosion) are required to phase in release detection within 1 year, and the newest tank systems (usually protected from corrosion) by the end of the 5-year period. Release detection for all pressurized delivery lines must be retrofitted within 2 years.
o Periodic tank tightness testing (every 5 years) combined with monthly inventory control is allowed at new or upgraded UST systems for 10 years after new tank installation or existing tank upgrade. After 10 years, monthly release detection is required.
o Either monthly release detection or a combination of annual tank tightness testing with monthly inventory control is required of substandard existing USTs until they are upgraded. Existing UST systems must be upgraded or closed within 10 years of the effective date of the final rule, or within 1 to 5 years if a release detection method is not available that can be applied during the required phase-in period for release detection. Upgrading of petroleum UST systems includes retrofitting of corrosion protection and both spill and overfill controls at all tanks. Upgrading of hazardous substance UST systems also includes secondary containment and interstitial monitoring or an alternate release detection method approved by the implementing agency.
o Tank owners and operators must report suspected releases. Indications of a release must be reported to the implementing agency, including positive results from release detection methods, unless the initial cause of the alarm has been immediately investigated and the alarm is found to be false. After reporting suspected releases, owners and operators must perform release investigation and confirmation tests and, where a release is confirmed must begin corrective action.
o Owners and operators of leaking UST systems must follow measures for corrective action. Immediate corrective action measures include mitigation of safety and fire hazards; removal of saturated soils and floating free product; and an assessment of the extent of further corrective action needed. A corrective action plan would be required for long-term cleanups addressing ground-water contamination, although these cleanups could begin upon notification of the implementing agency by the owner and operator. Cleanup levels would be established on a site-by-site basis as approved by the implementing agency.
Today's final rule includes four release detection requirements that represent significant changes to the proposed rule:
(1) More frequent tank tightness testing (annual) of unprotected tanks during the initial 10-year upgrading period;
(2) Less frequent monitoring of new and upgraded tanks until age 10;
(3) Phase-in of release detection over 5 years at existing tanks based on age; and
(4) More stringent release detection for all pressurized piping systems.
The final requirements in each of these areas, and EPA's rationale for revising the proposal, are summarized below and shown graphically in Figure 3. The shaded areas indicate the major changes from the proposed approach. More detailed discussions of these major points of departure are also provided in later sections of today's preamble.
In the proposed rule (§§ 280.40 and 280.41), EPA provided owners and operators of existing UST systems without corrosion protection with two options for release detection, but only during the proposed 10-year upgrade period. The two options were: 1) to perform monthly release detection monitoring or, 2) to conduct monthly inventory control combined with a tank tightness test every 3 years. The option of tank tightness testing (in combination with inventory control) was proposed because tightness testing is already widely practiced. The proposed frequency of testing every 3 years was based on the Agency's belief that the industry could not accomplish testing of such a large universe of existing tanks on a more frequent basis because the supply of testing services could not respond to such a large demand.
Several commenters suggested that the proposed tank tightness testing alternative be allowed in the final rule but only if the testing were required more frequently. EPA now shares the concerns of those commenters. The Agency is particularly concerned about unprotected existing tanks, which have the greatest probability of leaking. New information on causes of release has prompted the Agency to believe that tightness testing conducted once every 3 years leaves too much time between tests during which an unprotected steel tank may develop corrosion holes and release product. Therefore, in the final rule the Agency requires that tank testing be performed every year during the 10-year upgrade period, instead of every third year as was proposed. Because unprotected steel tank systems pose the greatest threat to the environment and public health, EPA believes that increasing the frequency of tightness testing will provide better environmental protection and is therefore warranted.
With regard to EPA's initial concern that such a frequency would be impractical or difficult to achieve, numerous public comments were received from the release detection industry indicating their belief that more frequent tank tightness testing is feasible. Commenters also verified that tank tightness testing is already a commonly used release detection method, and its technology and resources are currently widely available.
EPA's proposed approach for new USTs required either frequent-to-continuous (at least monthly) release detection monitoring or semiannual tank tightness testing in combination with monthly inventory control. The Agency proposed frequent-to-continuous monitoring at new tanks because, the more frequently release detection is applied, the more likely an actual release will be detected when it occurs, as explained in detail in the proposal (52 FR 12720), and thus it was believed to be needed to ensure adequate protection of the environment and public health.
Several commenters questioned the proposal to require more frequent release detection at new protected tanks than at existing (older) unprotected tanks. New information on causes of release (see section II.F. of the preamble) confirms that new tanks are the least likely to leak. The probability of new or upgraded tanks leaking, particularly early in their operational lifetime, is extremely low. Therefore, the Agency agrees with commenters who suggested that new tanks should be allowed to be monitored less often than proposed.
Today's final rule replaces the proposed semiannual tank testing with a requirement to conduct tank testing at least every fifth year during the first 10 years of the UST's operational life at a new or upgraded UST. As proposed, however, tank testing must be used in combination with monthly inventory control to be considered an approved release detection method. Because long-term experience with tank corrosion protection technology is not yet extensive, and because existing bare steel tanks rarely fail before 10 years of age, today's final rule requires monthly release detection at protected and upgraded tanks only after they pass their tenth year of operation.
EPA proposed to phase in release detection at existing UST systems over a 3-year period for USTs unprotected from corrosion and a 5-year period for USTs protected from corrosion. This approach was proposed because it was believed to be the fastest schedule reasonably implementable and would require unprotected tanks (which have the greatest potential to leak) to have release detection in place sooner than protected tanks. In the preamble to the proposal (52 FR 12677), the Agency identified three approaches towards phasing in of release detection: (1) a rapid phase-in (over 1 to 2 years), (2) a gradual phase-in (over 3 to 5 years), and (3) a class approach requiring a rapid phase-in at the most sensitive UST locations and more gradual phase-in elsewhere. After considering public comments received on this issue, the Agency has decided in today's final rule to retain the gradual phase-in approach but to base the phase-in on the age of the UST system, with the oldest tanks being required to have release detection first.
The preamble to the proposal solicited public comment on the appropriateness of considering tank age, vulnerability of hydrogeology, distance to drinking water wells, distance to vital ecological systems, and any other factors in the selection of a phase-in approach for release detection. Most commenters supported the gradual phase-in to take place over time periods ranging from 3 to 10 years, but recommended that it be based on age. A few recommended the rapid phase-in approach, and most of those recommended that the phase-in be based on age.
Today's final rule requires the oldest tanks, with or without corrosion protection, to apply release detection within the first year and the newest tanks no later than 5 years after the effective date of today's rule. This approach is believed to offer several advantages: (1) EPA has concluded that the phasing in of release detection at all existing UST systems in the large universe is implementable within 5 years; (2) the age-based sequencing schedule ensures that implementation of the requirements will take place in an orderly fashion that will spread out the demand for release detection services and avoid bottlenecks; (3) it gives higher priority to older, unprotected tanks, which have been found to have the greatest potential to leak; and (4) it still allows the newer, protected tanks, which are the least likely to be leaking, to wait until the fifth year, as did the proposed rule. In choosing the gradual approach, the Agency agrees with those commenters who cited the rapid approach as being impossible to implement with such a large universe of UST systems. In addition, the gradual approach is supported by numerous existing industry, state, and local management programs which utilize the gradual approach as the most feasible phase-in approach. The reasons the Agency is taking this gradual phase-in approach are discussed in more detail later in this preamble (section IV.D.2.). The class approach was deemed unimplementable for other reasons discussed in more detail later in this section of the preamble.
The proposed rule contained release detection requirements for pressurized piping in addition to those requirements for new tank systems. Specifically, automatic release detection and shutoff devices were proposed for the pressurized piping at new tank systems unless continuous release detection or interstitial monitoring was installed. Suction piping systems were exempt from the proposed requirements if other minimum construction requirements were met (see 52 FR 12774). Information obtained by the Agency at the time of the proposal indicated that 20 to 30 percent of all reported releases are due to piping failures. EPA also suspected that pressurized piping systems, which are reportedly the most commonly used withdrawal system at new retail motor fuel installations, were responsible for the larger releases.
Information obtained by EPA after publication of the proposal (see section II.F. of today's preamble) indicates that piping system failures are responsible for a much greater percentage (80 to 85 percent) of release incidents than previously thought. Public comments received by EPA confirmed the Agency's belief that failure of pressurized piping systems frequently results in large releases.
In today's rule, EPA has set forth more stringent release detection requirements for pressurized piping systems. Systems must be: (1) Equipped with automatic shutoff devices or automatic flow restrictors and use either annual line testing or monthly release detection monitoring; or (2) have continuous interstitial or vapor monitoring combined with either shutoff devices or alarms. These requirements will ensure that product releases from the most vulnerable portion of the UST systems are minimized. (Today's preamble section IV.D.3.b. discusses these requirements more fully.)
In the preamble to the proposed rule (52 FR 12673-12675), EPA described three basic regulatory options that were considered for new petroleum tanks: (1) Protected single-walled tanks with release detection; (2) secondary containment with interstitial monitoring; and (3) a class approach under which more protective requirements would apply to UST systems located where a release could pose a particularly high risk. The Agency solicited comments on each of these general approaches to regulate new petroleum tanks, and announced that, based on comments received and on any additional data gathered prior to promulgation, it would give further consideration to each of them before finalizing today's rules. Numerous comments were received concerning these options, and EPA developed additional information after the proposal that is relevant to this issue. (EPA solicited comment on this new information in the December 23, 1987, Supplemental Notice and in the March 31, 1988, Notice of Availability.)
Based on EPA's information and on public comment, the Agency has decided to proceed with the proposed approach of protected single-walled tanks with release detection for new petroleum UST systems. The Agency's basic rationale for this choice as explained in the preamble of the proposal remains unchanged. The following discussion briefly highlights some of the concerns raised by commenters and the influences that the new information had on this final decision.
a. Option 1: Protected Single-Walled Tanks With Release Detection. Several commenters supported EPA's proposed approach of allowing protected single-walled tanks with release detection to be used at all new petroleum UST systems. They agreed that this approach is protective of human health and the environment, confirmed that it is already widely in use by industry and in numerous state programs, and that it is available as a viable and effective alternative to interstitial monitoring.
Some commenters, however, suggested that the approach of protected single-walled UST system with leak detection approach is not as implementable as secondary containment. They stated that secondary containment is a more straightforward technical approach because, in the long run, it simplifies implementation by allowing relatively easy determination of the UST's compliance status. Another disagreement with this option is the belief expressed by some commenters that the probability of releases to the environment occurring at single-walled tanks with release detection is much greater compared to UST systems with secondary containment and interstitial monitoring.
EPA agrees with those commenters who suggested that there will probably be more releases to the environment from protected single-walled UST systems with release detection than from systems equipped with interstitial monitoring. One of the advantages of secondary containment is the potential for detecting a leak before a release to the environment actually occurs. Secondary containment thus provides a second barrier against release.
Although there are several reasons why EPA has not mandated secondary containment with interstitial monitoring at new petroleum UST systems, the most important reason is that it is not believed to be necessary to protect human health and the environment at such systems (in contrast to hazardous substance UST systems, discussed in section III.C.3. below). The new causes of release information (discussed previously in today's preamble) shows that the use of new, protected, single-walled USTs has, so far, resulted in the virtual disappearance of failures, because these new USTs have preventive controls for the main cause of past tank failures: corrosion of unprotected, bare steel. This new information also revealed that protected UST piping that is carefully installed and immediately tested is expected to significantly reduce the occurrence of release incidents from this other major component of the UST system (a reduction of at least two-thirds according to experienced installer estimates). Piping failures will not, however, be prevented to the same degree as tank failures.
Thus, the Agency has concluded that petroleum releases will be dramatically reduced using protected UST systems. This fact must also be considered with the knowledge that numerous effective methods of release detection are available and are being used at petroleum UST systems nationwide. These methods of detection will enable actions to be taken that will minimize the extent of the few releases that do occur. Also, the nature of petroleum products and the widely available technologies for their clean up provides the means to ensure that adverse impacts from such releases (when they occur) can be managed and remediated.
In consideration of all of the above factors, the Agency has concluded that single-walled protected UST systems combined with the release detection required today will adequately protect human health and the environment.
NOTE: As discussed earlier in this section of the preamble, today's final approach to new petroleum UST systems has been further tailored to allow infrequent tightness testing of protected or upgraded tanks (combined with inventory control) for the first 10 years of their operating life, with more release monitoring required of the UST system piping--including continuous monitoring of pressurized lines.
b. Option 2: Secondary Containment With Interstitial Monitoring. Several commenters supported the use of secondary containment with interstitial monitoring for various reasons, including: product releases would be completely contained and prevented from adversely impacting the environment and public health; it is a more rapid and reliable form of release detection; the cost is comparable to single-walled UST systems with release detection; and the need for conducting site assessments and corrective action would be avoided. Others, however, opposed its required use with new petroleum tanks for reasons such as: the greater capital and installation costs do not justify the environmental gains that would be achieved (in comparison to the single-walled approach); and this approach is not compatible with current trends in industry that are well underway in upgrading existing UST systems.
As previously stated, EPA agrees that secondary containment with interstitial monitoring would most likely result in fewer releases to the environment compared to protected single-walled UST systems with release detection. UST systems having secondary containment and interstitial monitoring are not perfect, however, and failures of these systems will also occur and result in some releases into the environment that will have to be remediated. Although protected single-walled systems would result in more releases, the Agency has concluded that this increase is not a significant added threat to human health and the environment given that release detection will minimize the extent of these additional releases, and the availability of petroleum cleanup technologies is widespread and capable of alleviating any resultant adverse impacts. In addition, the Agency is concerned that many owners and operators would delay upgrading their existing petroleum UST systems to the extent allowed by law or even to the point of noncompliance, because of the perception on the part of many commenters of the significance of higher capital and installation costs that would result from requiring secondary containment and interstitial monitoring (compared to protected single-walled USTs with release monitoring).
c. Option 3: Classification Approach. EPA recognized from the outset that releases from petroleum UST systems located in certain sensitive areas pose a greater risk of harming human health and the environment than others. As a consequence, one of the regulatory options the Agency considered extensively in developing the final rule was a federal classification approach based upon the potential impact of a release. Under this approach, a class or classes of UST systems located in high-risk areas would be subject to more protective requirements than UST systems located in less sensitive areas.
Although the proposed baseline standards for prevention and detection of releases made no differentiation based on class, the Agency requested comment in the April 17 proposal on the general desirability and feasibility of a classification approach to regulating UST systems. EPA also sought comment on a specific, two-tiered classification scheme. Under this scheme, owners and operators of UST systems located in high-risk areas, defined as the area within a specified distance of a public drinking-water well, would be required to use secondary containment. The baseline standard of protected, single-walled tanks with release detection would be allowed in low-risk areas.
EPA received several comments that favored or opposed inclusion of a federal classification approach in the final rule. Several commenters in favor of a class approach suggested, and the Agency has considered since proposal, alternative regulatory schemes with respect to how requirements should differ among UST systems in different classes. The schemes examined included accelerating the schedule for upgrading of existing UST systems to new tank standards or for compliance of existing UST systems with release detection requirements in high-risk areas; imposing more stringent design requirements (e.g., secondary containment) in high-risk areas; and imposing more stringent design requirements in all areas except those designated as low risk by the implementing agency. The Agency also explored several potential criteria proposed by commenters for differentiating among classes. Hydrogeologic criteria, such as proximity to ground water used for drinking water, were considered most extensively. The criterion EPA selected for detailed analysis was distance to a public water well. (For discussion of the results of the analysis see the Regulatory Impact Analysis for the Technical Regulations.)
As pointed out by several commenters, the concept of a classification approach to regulating UST systems is appealing for many reasons. Because the potential impact of a release is greater in more sensitive areas, a classification approach tailors the level of protection to the risk posed in a particular area. EPA's analyses of various classification schemes indicated that, without considering the costs of implementation, the benefits of a classification approach, primarily in terms of corrective action costs avoided, could be significant. Cleanup of contamination is especially difficult and expensive in sensitive areas, such as where ground water used for drinking water is affected. Classification could also provide a priority-setting scheme for enforcement and corrective action.
Despite the advantages, EPA has not included a classification approach in the final rule. Commenters' arguments against a federal classification approach influenced the Agency's decision. EPA agrees with commenters that the baseline requirements set by the final rule will adequately protect the environment in all areas while also encouraging timely voluntary compliance by avoiding unnecessary additional complexity and providing reasonable flexibility for UST system owners and operators. Although EPA supports the concept of differential protection based on the potential impact of a release, the Agency believes that, for this program, classification at the federal level is neither feasible nor practical. EPA is particularly concerned about the potential hindrance to state program approval and the difficulties of implementing a classification approach at the federal level.
Due to the size and nature of the community to be regulated, the success of this program depends largely upon implementation at the state and local levels. Most states, however, have not developed classification systems. Development of appropriate and workable classification schemes could take significant time and resources given the number of environmental and other factors that must be considered. The Agency is concerned that the steps necessary to define criteria and then identify high-risk and low-risk areas in states that have not yet done so could delay implementation of the program and divert scarce resources from efforts to achieve the improvements of the baseline UST regulatory requirements, which will provide most of the benefits. This additional complication may discourage states from seeking approval altogether. The Agency has concluded that the potential reluctance of states to implement this program as a consequence of requiring a classification approach could result in less successful protection of the environment and human health.
In addition, even without implementation delays, a federal classification scheme is not likely to provide significant real additional protection. Several commenters stated (and the Agency has concluded) that, at the federal level, any criteria used to classify would have to be simple (such as specified distance to a water well) because implementation of a classification scheme based on more complex criteria would not be feasible. Yet, given the diverse hydrogeologic conditions that exist in the United States, no single, simple classification criterion would be appropriate everywhere. For example, a distance to drinking-water well criterion, as presented in the preamble to the proposal, is meaningless in locations such as Phoenix, Arizona, where the depth to ground water is so great that releases from UST systems will rarely contaminate drinking-water wells; or in places like Dade County, Florida, where ground water may move so fast that a release could travel the standard distance within a short period of time. Because defining and identifying high-risk areas is so highly dependent on local (site-specific) hydrogeologic and other factors, EPA believes that state and local implementing agencies can make the most meaningful determinations of classes.
Although the Agency has concluded that a classification approach at the federal level is neither feasible nor desirable, EPA believes that a classification approach to regulating UST systems at the state or local levels, where local environmental conditions are better known, may be both feasible and appropriate. Under today's approach, the Agency allows but does not require states to use a classification approach if it is appropriate for the conditions in the state. For example, states that have already developed a classification scheme may decide to use it to regulate USTs. For such states, the potential difficulties associated with implementing a classification approach to regulating UST systems may be significantly reduced.
a. Mandatory Upgrading or Replacement of Substandard UST Systems. In the preamble to the proposal, the Agency identified four regulatory approaches to scheduling the upgrade or replacement of substandard existing UST systems: (1) Rapid upgrade or replacement (within 3 to 5 years of final rule promulgation), (2) gradual upgrade or replacement (within 6 to 12 years of final rule promulgation), (3) no required upgrade or replacement, and (4) scheduling of upgrade/replacement based on a class approach (upgrade or replace tanks located in environmentally vulnerable areas first). EPA proposed to implement scheduling of mandatory upgrade/replacement using the gradual approach (within 10 years after the effective date of the final rule) and has retained this requirement in today's final rule.
Many commenters supported the proposed 10-year compliance period for the reasons identified in the preamble to the proposal (52 FR 12676). However, some recommended a shorter time period, stating their belief that industry resources could meet the demand and that a faster upgrading schedule would prevent a significant number of future product releases. In contrast, other commenters suggested a longer compliance period than proposed and argued that more time was needed to become familiar with the UST regulations, that owners of multiple tank systems would not be able to finance upgrades/replacements in 10 years or less, and that the service industry would need more time to implement upgrades for this large universe of existing tank systems.
As stated in the proposal, the Agency agrees that if upgrading could be completed in the shorter compliance period (3 to 5 years) it would decrease the number of product releases and thereby provide greater protection of human health and the environment. However, the Agency also agrees with those commenters who believe that the large number of existing UST systems essentially precludes industry from implementing upgrade requirements faster than the proposed 10-year period. Furthermore, EPA is in agreement with many of these same industry commenters who indicated that a 10-year compliance period is a reasonable time period and would not be overly burdensome. EPA is aware of numerous industry upgrading programs that have already started and will be completed within this time frame.
EPA continues to believe that the alternative of not requiring upgrade/replacement of substandard systems is simply unacceptable because the Agency has concluded that UST owners probably would not upgrade if it is not required. The universe of 1.4 to 2 million UST systems largely consists of unprotected, bare steel tanks. The new causes of release information summarized previously in this preamble (and derived from the "Causes of Release from UST Systems" report) confirms that the unprotected segment of the UST universe is very likely to leak due to corrosion and piping failures and, therefore, presents a significant threat to the public health and environment. Historical tank replacement rates, established trends in the closure of existing retail motor fuel businesses over time, and the imposition of this new regulatory program are expected together to result in rapid decrease in these substandard UST systems nationwide. However, a totally voluntary program of UST system upgrading is not expected by EPA to result in the timely upgrade or replacement of UST systems on a nationwide basis, particularly given the large number and heterogeneous nature of this universe of existing UST systems. In order to ensure human health and the environment are protected, the Agency has established a clear national goal of upgrading all substandard UST systems within 10 years that rejects a purely voluntary approach as inadequate. This goal is intended to prompt all UST owners and operators to plan for and undertake the upgrading steps that are needed to protect human health and the environment.
Another option considered by the Agency in development of the proposal was to further require upgrade and replacement based on the use of a class approach that scheduled more rapid UST system upgrades/replacements in the most vulnerable areas (e.g., ecologically sensitive sites or sites in close proximity to drinking water sources) while allowing other UST systems to be phased in on a more gradual basis. Many commenters recommended staggering the implementation schedule for upgrade/replacement based on other factors (besides class) such as tank age, tank size and type, proximity to human populations, and the measured corrosivity of a site. The common concern among these commenters was that a staggered schedule would prevent all owners from being allowed to wait until the last minute to bring their existing substandard tank systems into compliance and would, thereby, ensure a more even, implementable, and serviceable demand for tank upgrade/replacement over the 10-year period.
The Agency has decided not to require the phase-in of upgrading based on the class approach strategy, or on the basis of any other factor, for several reasons. Although many factors on which to base tank upgrade/replacement have been identified, because of the serious ramifications to the owner and operator associated with the technical requirements for upgrading, EPA believes no single risk-based factor is appropriate or applicable to mandate nationwide for all existing UST systems. Today's general 10-year requirement in the final rule also allows implementing agencies in approved states, or under State law, the opportunity to decide the staggered schedule on which upgrades should be conducted and what factor(s) are most applicable to their tank population. Regardless of what type of state action is taken, upgrading is already rapidly taking place through numerous industry programs. Today's final approach allows them to continue to use the numerous different phase-in approaches that are already successfully underway. EPA believes this flexibility should also encourage others to upgrade because they can set their own schedules to meet the 10-year deadline. Thus, EPA has concluded that much of this regulated community will have completed these voluntary programs of closure and upgrading within the mandated 10-year period, and, therefore, the number of existing USTs that otherwise will have to be upgraded by the end of this upgrading period is expected to be relatively small. The implementation of today's final requirements for release detection (and the financial responsibility requirements to be provided later) are also expected to prompt rapid upgrading and should work to ensure that the implementation difficulties at the end of the 10-year period that were cited by some commenters are not encountered due to a backlog of upgrading demand.
In summary, EPA believes that the 10-year upgrade requirement will not present an undue burden on industry and is implementable, but is needed to prompt a significant portion of this regulated community into making the upgrades and improvements necessary to protect human health and the environment. It will also provide industry and the states with some flexibility that will allow (and thereby encourage) them to determine their own approaches and schedules for ensuring tank upgrade/replacement programs are implemented expeditiously within the nationally mandated timeframe.
b. Methods of Release Detection. The majority of existing UST systems are not currently being monitored for releases to the environment. Therefore, EPA continues to believe that a fundamental goal of today's regulatory approach must include the establishment of reliable release detection at all UST systems. In the proposal (52 FR 12676), the Agency required all existing substandard UST systems to comply with one of the six release detection methods proposed for new USTs (52 FR 12713). The dual objective of this proposed approach was to allow for the continued development of release detection technologies and also to ensure that only sound and reliable release detection methods were used. Particularly for petroleum UST systems, the Agency believed many releases would be detected if an appropriate method was selected and properly used.
In the preamble to the proposal (52 FR 12676), EPA identified that a wide variety of release detection methods (applied either internally or externally to the tank) were currently on the market but with limited data available concerning their performance. This lack of performance data and field experience was corroborated by many commenters. Although numerous comments were received regarding the release detection methods proposed (see section IV.D. of today's preamble for a summary and discussion of these comments), the Agency did not receive any comments that opposed the general approach of mandating the use of at least one of the allowed methods at existing UST systems. Other commenters noted their successful use of different types of methods.
Performance data obtained from commenters and from EPA-sponsored research since proposal support the Agency's decision to retain the proposed approach in the final rule. (EPA solicited public comment concerning this new information on March 31, 1988; 52 FR 10403.) The Agency is convinced that the proposed release detection methods, when properly applied and performed as specified in the final rule, will successfully detect the large number of leaks believed to already have occurred. For example, over the past year in California and Florida, several thousand UST system releases have been identified using these methods of release detection, and there appear to be few instances where releases have been later discovered through other means (e.g., off-site impacts) when release detection was being properly conducted at a site.
UST systems containing hazardous substances were proposed to have secondary containment with interstitial monitoring unless the owner or operator obtained a variance by demonstrating that another release detection method would be effective in detecting releases of that substance from a single-walled UST (see 52 FR 12677, April 17, 1987). The reasons for selecting this proposed approach were: (1) information is not available on the applicability and demonstrated reliability of release detection methods at most hazardous substance USTs, and (2) few industry and state and local agency UST programs yet apply to hazardous substance USTs and, thus, there is little collective experience on management of such USTs. The Agency recognized that some release detection methods for petroleum UST systems might also perform well at some hazardous substance USTs. Hazardous substance USTs, however, store an array of individual chemical compounds that vary widely in their physical and chemical characteristics and that may not be readily or reliably detected using release detection methods developed for single-walled UST systems storing petroleum products. They also may not be able to be as readily addressed by corrective action technologies that are already widely available for petroleum releases. Therefore, the proposed approach required secondary containment at all new hazardous substance UST systems, unless owners and operators could demonstrate that an alternative release detection method could be reliably applied and operated.
Several commenters were opposed to the proposed secondary containment approach at new hazardous substances UST systems because: (1) they do not believe that chemicals designated as hazardous substances are any more hazardous than petroleum products (and, therefore, should not have to meet more stringent requirements), and (2) EPA has not provided a technically valid explanation of why single-walled USTs are appropriate for petroleum products but not for hazardous substances. It was also suggested that, instead of having a lengthy and costly variance procedure on a case-by-case basis, the Agency should incorporate a variance in the final rule for certain classes of hazardous substances.
The Agency has decided to retain secondary containment for all new hazardous substance USTs in today's final rule because EPA has not received any additional information or data since proposal to convince the Agency that reliable and appropriate release detection methods are generally available for hazardous substance USTs. The Agency continues to believe that under certain conditions, hazardous substances will pose a greater risk to human health or the environment than petroleum products. Furthermore, there are proven, reliable and available release detection methods and cleanup technologies for petroleum, whereas the applications of these same release detection methods and cleanup technologies are neither proven nor available for hazardous substances.
Also retained in the final rule is the proposed exception for those owners and operators who, by means of a variance, can demonstrate that a release detection method (other than interstitial monitoring) can be successfully installed and operated at a specific hazardous substance UST system to detect releases of the stored substance. An additional requirement in the final rule is that an owner and operator must also demonstrate that a corrective action method is available to clean up the release. The Agency decided not to structure the variance procedure by classes of hazardous substances because physical properties of chemicals within a class can be quite variable (for example, wide ranges in solubility and volatility) and require different corrective action approaches, and one release detection method would not reliably detect all of the chemicals within that class. (A more detailed discussion on the final rule regarding hazardous substance USTs is presented in section IV.D. of today's preamble.)
A detailed discussion of the methods of release detection allowed for existing petroleum UST systems is also presented in section IV.D. of today's preamble.
An important facet of today's final strategy for regulating underground storage tanks is to ensure that public and private drinking water supplies are protected, and that necessary steps are taken to abate other health and safety threats (such as fires and explosions) due to present and future releases from UST systems. As described previously in this preamble, tens of thousands of UST systems are believed to have already leaked substantial quantities of regulated substances into the environment and numerous public and private wells have already been threatened or destroyed. As discussed in the preamble to the proposal (52 FR 12678), the development of a regulatory program for corrective action started with three assumptions: (1) The need to implement the requirements through the extensive participation of state and local UST programs; (2) the approach taken must be able to work as the nature of the UST corrective action problem changes over time; and (3) a process must be set up that ensures owners and operators get quickly to the task of cleaning up all releases, although the completion of corrective action may take a long period of time at any one site.
The proposal provided for separate but very similar corrective action processes for UST systems that contain petroleum and those that contain hazardous substances. In the proposal, the Agency solicited comment on whether the proposed requirements should be integrated into one subpart. The proposal also divided the regulatory requirements for corrective action into two stages: (1) Immediate abatement actions that all owners and operators must take in response to a release; and (2) long-term investigation and remedial action measures that may be taken on a site-specific basis to protect human health and the environment. The Agency solicited public comment on the approach taken to these immediate abatement (Stage I), and long-term remedial action (Stage II) issues in the proposal. These and numerous other corrective action issues are discussed in more detail in the following section of this preamble that addresses the section-by-section analysis of the final corrective action rules (see section IV.F.). The following discussion briefly highlights the major decisions and influences (including public comment) that went into the development of the Agency's final approach to corrective action for UST systems storing regulated substances.
a. Corrective Action for Regulated Substances. The proposal provided separate corrective action processes for UST systems storing petroleum and those storing hazardous substances (proposed Subparts F and G). The stated purposes of this approach were: (1) To avoid confusion over which procedures must be followed at a release site; and (2) to construct an approach towards hazardous substances corrective action that was very similar to the one for petroleum, except that it was intended to move the owners and operators more quickly to collect, array, and assess the information needed to determine the nature, extent, and hazard of the corrective action release (52 FR 12682). The dividing line between Stage I and Stage II of the process was intended to occur earlier in the response to a hazardous substance release, thus following more closely the RCRA Subtitle C hazardous waste tank approach to corrective action. This proposed approach was an attempt to provide a response process for hazardous substances that recognized the relatively greater hazards that could be posed by hazardous substances releases, and that ensured that there was consistency with the corrective action approach required for hazardous waste tanks. EPA solicited comments on whether the petroleum and hazardous substance requirements for corrective action should be merged in the final rule.
In response to public comments received, and several other revisions made in the final corrective action rules, the Agency has decided to consolidate the requirements for UST corrective action into Subpart F for all regulated substances (both petroleum and hazardous substances) for the reasons discussed below. All of the comments received by EPA concerning the merging of the two proposed subparts were in favor of it, although they made different suggestions on how to accomplish this consolidation (discussed in more detail later in the section-by-section analysis of this preamble).
Although the preamble to the proposal described the Agency's intention to foster quicker investigatory and cleanup actions at hazardous substances release sites, as was pointed out by some commenters, the few differences in the actual regulatory language between the two separate subparts was not likely to attain this end. In fact, most of the requirements within the two proposed subparts were the same. The most significant differences were that a free product investigation was not a required step for hazardous substance responses, and a more rapid (within 30 days) and detailed reporting of certain information was required by owners and operators with hazardous substance releases (see proposed § 280.74(b)) than for those with petroleum releases. Prompted by these public comments, additional review of these proposed differences has led the Agency to agree that they do not necessarily lead to faster corrective actions at hazardous substance release sites. Furthermore, even if this end was achieved somewhat for hazardous substance releases, the Agency does not believe it would protect human health and the environment to move less quickly towards the remedial actions required with petroleum releases. Thus, in the merger of the two subparts in today's final rules, the Agency has developed a process that ensures the most rapid reporting, investigation, and corrective action at all UST release sites that is believed to be attainable (the details of today's consolidation are discussed in more detail later in section IV.F. of today's preamble).
Because the requirements of the two proposed subparts were largely the same, EPA also believes that today's merger of them into one consolidated Subpart F for all regulated substances avoids a significant source of potential confusion within the regulated community. Despite the slight procedural differences between the two subparts, the Agency is now convinced that the work involved in investigating petroleum and hazardous substance releases will be very similar. Also, given that the similarities in the characteristics of different petroleum products are expected to lead to more rapid and widespread understanding of how to respond routinely to petroleum releases (mostly gasoline), the actual investigation, initial abatement of hazards, and the availability of corrective action technology for petroleum releases may result in more rapid and effective release responses at sites with petroleum releases than those with a hazardous substance release.
The Agency continues to believe that under certain conditions, hazardous substance releases can pose a greater hazard to human health and the environment than petroleum releases. However, the consolidated requirements in today's final rules have been designed to allow the implementing agency to cause the owner and operator to move as rapidly as is necessary to identify and control any such additional potential threats. For example, under the final rules, the identification and recovery of free product now must be considered at both petroleum and hazardous substance release sites. This required consideration of free product recovery, however, does not hinder the progress of corrective action at sites with releases of hazardous substances because the presence and recovery of free product is already something that must be commonly considered at all release sites (whether petroleum or hazardous substances). The final rule allows enough flexibility to ensure that this type of action is tailored, under the direction of the implementing agency, to site conditions and the type of substance released.
b. Stage I: Investigation of Releases and Immediate Corrective Action. EPA proposed requirements for the first stage of the corrective action process (Stage I) by establishing immediate steps that must be taken to abate imminent health and safety hazards whenever a release is confirmed. The basic approach assured that the following activities were undertaken by the owner and operator at all sites with a confirmed release: (1) Immediate notification of the implementing agency; (2) actions needed to stop further releases; (3) mitigation of safety hazards due to fire and explosion; (4) removal of contaminated soils; (5) investigation of the existence and extent of floating free product; and (6) initiation of the removal of free product and submittal of free product recovery plan unless directed to do otherwise by the implementing agency.
Numerous public comments on issues related to these proposed requirements are discussed in more detail later in this preamble. In general, many commenters believed the proposal was unclear as to what exactly had to be done, and by whom, to comply with the requirements. EPA, in response to these commenters, has revised the language in the final rule to clarify which elements of the initial response actions are mandatory and which are discretionary. The mandatory requirements are intended to ensure three goals are accomplished:
o To bring leaking UST sites under control with respect to immediate health and safety hazards;
o To stabilize the site so that contamination will not worsen as investigation and potentially long-term remedial actions are considered; and
o To be self-implementing in that these measures emphasize the responsibility of the owner and operator to take immediate action without awaiting approval of the implementing agency.
All sites with releases of regulated substances must investigate the area around the UST system to be able to characterize the size and nature of the release (e.g., a small spill or a large, continuing slow release). The findings of this initial investigation must be reported to the implementing agency, which has the discretion to require a more extensive site characterization based on these initial findings. The proposal has also been revised to make this more extensive examination of ground-water and soil contamination mandatory when the initial investigation reveals that a significant release has occurred (e.g., free product floating on the ground water or saturated soils in the subsurface), even in the absence of direction from the implementing agency. Thus, today's final rule requires all releases that seriously threaten or impact ground water to be automatically investigated by the owner and operator to characterize the extent of ground-water contamination and any soil contamination remaining at the site.
c. Stage II: Long-Term Corrective Action Options. EPA also proposed requirements for the second stage of the corrective action process (Stage II) addressing long-term remediation of contaminated soils and ground water. The Agency solicited public comment on three regulatory options for establishing long-term cleanup requirements: (1) national cleanup standards, with a variance provision; (2) site-specific standards to match the risk presented; and (3) a predetermined class approach. The proposal emphasized the second approach, with its site-specific cleanup targets based on (1) the data from the detailed site investigation of soil and ground-water contamination performed by the owner and operator and (2) the site-specific risk present as determined by the implementing agency using exposure and risk assessment techniques.
EPA received public comments for and against each of the three options for establishing cleanup levels. The majority of commenters preferred the proposed site-specific approach as the best way to accommodate the diversity of UST releases nationwide. Those in favor of establishing national cleanup standards in the regulations expressed the opinion that this would expedite cleanups and provide greater national consistency in cleanup results. Supporters of the class approach cited the wisdom of tailoring cleanup efforts to a predetermined assessment of risk based on prospective classifications tied to ground-water vulnerability. EPA has decided to retain the site-specific approach in the final rules.
For the same reasons that were cited in the preamble to the proposal (52 FR 12680-12682) and supported by most commenters, EPA continues to believe that the site-specific approach is the most appropriate because it allows implementing agencies the necessary flexibility to address corrective action based on the unique circumstances of the site. It also enables state and local governments to build upon their own experiences when assessing the need for and extent of corrective action. EPA is not convinced that a national standards approach will result in more rapid and consistent levels of corrective action. No data were provided that demonstrate this position. In fact, the information available to the Agency about UST cleanup technologies indicates that the level of actual cleanup achieved is much more dependent on the limitations associated with current corrective action equipment and the problems posed by individual site conditions, than it is on the cleanup levels established in a corrective action plan (or by regulation). In selecting the site-specific approach, the Agency notes that it does not preclude the use of the other two approaches by implementing agencies. Some states already are using a statewide standards approach.
The Agency continues to believe that the site-specific risk assessment process can be streamlined to ensure that the corrective action needed at a site can be identified rapidly and initiated quickly. EPA maintains the position expressed in the proposal's preamble (52 FR 12681) that site-specific cleanup levels will have to account for potential exposure of the public to contamination at the site. If a drinking water supply, public or private, is affected or threatened by the release, then the cleanup levels in the corrective action plan should be established using health-based levels as the target for cleanup of the release, unless an alternative source of drinking water can be provided to the potentially affected public. The Agency is developing further guidance concerning the use of risk assessment and exposure assessment techniques for releases at UST sites. These efforts are expected to further streamline the site-specific approach to regulation promulgated today. (Further discussions of the site-specific approach to standard setting and corrective action plans are provided later in the section-by-section analysis in section IV.F. of today's preamble.)