Jump to main content.

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 II. Background



II. Background

A. Subtitle I of RCRA

B. Operating Principles

C. Summary of April 17 Proposed Rule

D. Public Comment on the Proposal

E. Summary of the Supplemental Notice and the Notice of Availability of New Information

F. Influences on the Final Rule

1. Scope of the Problem

2. New Cause-of-Release Information

3. Industry Codes and Practices

4. Industry Trends

5. UST System Technology Development

6. Leaking USTs Present A Unique Regulatory Challenge

7. Emerging State and Local UST Programs and EPA's Approach to Regulation

G. Conclusions Since Proposal



A. Subtitle I of RCRA

The Hazardous and Solid Waste Amendments of 1984 extended and strengthened the provisions of the Solid Waste Disposal Act as amended by the Resource Conservation and Recovery Act (RCRA) of 1976. One major portion of RCRA as amended, Subtitle I, provides for the development and implementation of a comprehensive regulatory program for "underground storage tanks" containing "regulated substances" and releases of these substances to the environment.

Subtitle I defines "underground storage tank" as a tank system, including its piping, that has at least 10 percent of its volume underground. Throughout this preamble and final rule, the terms "underground storage tanks," "USTs," and "UST systems" include both the underground storage tank vessel and the underground piping connected to it.

"Regulated substances" are defined as substances defined as hazardous under the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA), except hazardous wastes regulated under Subtitle C of RCRA, and petroleum.

Subtitle I excludes the following from the definition of USTs:

o Farm or residential tanks of 1,100 gallons or less capacity storing motor fuel for noncommercial purposes;

o Tanks storing heating oil for consumptive use on the premises where stored;

o Septic tanks;

o Pipeline facilities (including gathering lines) regulated under the Natural Gas Pipeline Safety Act of 1968, the Hazardous Liquid Pipeline Act of 1979, or state laws comparable to these Acts;

o Surface impoundments, pits, ponds, and lagoons;

o Storm-water or wastewater collection systems;

o Flow-through process tanks;

o Liquid traps or associated gathering lines directly related to oil or gas production and gathering operations; and

o Storage tanks situated on or above the floor of underground areas, such as basements and cellars.

Subtitle I contains several major provisions for the regulation of UST systems. Section 9002 requires UST system owners to notify states of the existence of their UST systems. These notification requirements were addressed in a final rule published by EPA on November 8, 1985 (50 FR 46602). Section 9002, as amended by the Superfund Amendments and Reauthorization Act of 1986 (SARA), also requires that states use the notifications they receive to compile tank inventories. Under federal grant agreements, states are providing EPA with aggregated data from these notifications.

Under section 9003, EPA must promulgate regulations applicable to all owners and operators of UST systems as necessary to protect human health and the environment. In promulgating these regulations, section 9003(b) authorizes the Administrator to distinguish between types, classes, and ages of underground storage tanks. Section 9003 requires EPA to issue design, construction, installation, and compatibility standards for new tanks, as well as requirements applicable to all tank owners and operators concerning leak detection, recordkeeping, reporting, closure, corrective action, and financial responsibility.

Section 9003(h), as amended by SARA, gives EPA--and states under cooperative agreements with EPA--authority to clean up petroleum releases from UST systems or to require their owners and operators to do so. It also establishes a trust fund to finance these activities.

Section 9004 permits EPA to authorize states to implement their own UST programs in place of the federal requirements if the state's requirements are "no less stringent" than EPA's and provide for adequate enforcement. Programs which are less stringent in certain areas may receive approval for an interim period.

Other provisions of Subtitle I pertain to definitions (section 9001); entry, inspection, and information gathering (section 9005); enforcement (section 9006); federal facilities (section 9007); state authorities (section 9008); and studies and reports to Congress required of EPA (section 9009).

This preamble and final rule pertain to the requirements mandated by sections 9003(a), (c), (e) and (g); they also meet the study requirements of sections 9009(a) and (b). Final rules for state program approval requirements, under section 9004, are found elsewhere in today's Federal Register. Final rules for financial responsibility requirements for petroleum UST systems, under sections 9003(a) and (d), will be promulgated by EPA at a later date.

Section 9003(c) requires EPA to establish the following minimum technical requirements for all UST systems: to maintain a leak detection system or comparable system designed to identify releases to protect human health and the environment; to maintain records of any such release detection system; to report releases and corrective action taken; to take corrective action in response to a release; and to close tanks to prevent future releases. Under section 9003(e), EPA must also establish performance standards for new UST systems. At a minimum, these standards must include design, construction, installation, release detection, and compatibility standards.

Until the promulgation of today's final rule, section 9003(g) established an "Interim Prohibition" that allowed installation of UST systems after May 8, 1985, only if the UST system is protected from corrosion, prevents releases due to corrosion or structural failure for the operational life of the tank, and is constructed of material compatible with the substance to be stored. The law allowed an exemption from the requirement of corrosion protection if an UST system is located at a site having a soil resistivity measured at 12,000 ohms/cm or greater. An interpretive rule concerning the Interim Prohibition was published on June 4, 1986 (51 FR 20418). These Interim Prohibition requirements are replaced by today's final rule for new tank standards, except in those few cases where the Agency has decided to defer regulatory action on some types of UST systems. For these deferred UST systems, the Interim Prohibition will continue to apply until EPA takes action in the future either to regulate or not regulate them (see §§ 280.10 and 280.11 of the final rule).

B. Operating Principles

Faced with the mandate of Subtitle I, EPA recognized several unusual aspects of the regulated universe that have created special problems in developing an effective regulatory approach. First, the regulated universe is immense, including over 2 million UST systems estimated to be located at over 700,000 facilities nationwide. Second, over 75 percent of the existing systems are made of unprotected steel, a type of tank system proven to be the most likely to leak and thus create the greatest potential for health and environmental damage. Third, most of the facilities to be regulated are owned and operated by very small businesses, essentially "Mom and Pop" enterprises not accustomed to dealing with complex regulatory requirements. Fourth, numerous technological innovations and changes are now underway in various sectors of the UST system service community.

In response to the unique aspects of this regulated community, and the clear need for comprehensive management of USTs during their operating life, EPA has identified and followed several key operating principles, described briefly below, in developing the final regulations for USTs.

o The UST program must be based on sound national standards that protect human health and the environment.

o The UST regulatory program must be designed to be implemented at the state and local levels. State and local governments have been and continue to be the authorities most capable of effective oversight of UST systems and response to releases.

o The regulations must be kept simple, understandable, and easily implemented by the owner and operator in order to facilitate voluntary compliance. Section 9003(b) specifically indicates that technical capability can be considered in developing the Subtitle I rules.

o The regulations must not inhibit new UST technological developments.

o The regulations must be designed to retain the flexibility necessary to accommodate, where possible, the special needs of the UST regulated community, which is largely composed of small businesses with limited resources available for capital improvements.

o In order to encourage the utmost voluntary compliance, the regulations should build upon current industry trends and tie into and utilize ongoing industry initiatives toward more sound UST management. Towards this end, Section 9003(b) specifically authorizes EPA to consider industry practices and consensus codes in developing appropriate UST regulations. The Agency expects the nationwide use of these new management practices to yield direct environmental benefits.

By reflecting these operating principles in the final UST regulations, the Agency believes it has taken the most effective approach toward protecting human health and the environment.

C. Summary of April 17 Proposed Rule

On April 17, 1987, EPA proposed regulations for USTs storing either petroleum or hazardous substances (other than hazardous wastes regulated under Subtitle C of RCRA) (52 FR 12662). These proposed regulatory measures set requirements for both new and existing UST systems to control the major causes of releases from these tank systems and included, among other things, corrosion controls, proper installation requirements, and spill and overfill prevention measures.

Figure 1 illustrates several key aspects of the regulatory program proposed in April 1987; requirements for corrosion protection and monthly release detection at all new UST systems; the phase-in of either monthly release detection or periodic tightness testing combined with inventory control at all existing USTs, within 3 years if unprotected from corrosion and within 5 years if protected; and the upgrading of all existing UST systems to the new tank standards within 10 years. In addition, the proposed new and upgraded tank standards for hazardous substance USTs required secondary containment with interstitial monitoring.


Figure 1 does not illustrate several other requirements and standards proposed for both new and existing UST systems: the design and installation of new UST systems; spill and overfill prevention; UST system repair; system closure; release detection methods and performance; and release confirmation, reporting, and response (including corrective action requirements). These proposed requirements and the manner in which they may have changed in the final rule are described in detail later in today's preamble. Several major points of departure from the proposal are identified and discussed in section III. below.

D. Public Comment on the Proposal

EPA received over 5,000 individual comments from over 1,100 commenters on the April 17 proposal, including verbal and written comments from the three public hearings. In general, these public comments supported the Agency's overall approach to the proposed regulations and the substantive requirements for new and existing UST systems.

Many comments addressed specific parts of the proposed rule, suggesting changes or calling attention to potential problems. These specific comments are discussed below in today's preamble. In summary, many comments tended to center on three areas: concern for the impact the UST rule would have on small businesses, calls for for EPA to adopt more stringent requirements for certain sensitive areas, and suggestions on the best way to phase in release detection. (See the corresponding sections in today's preamble for a full discussion of these comments.) The financial responsibility requirements generated more public comment than any other single area. Comments on this issue will be addressed in the financial responsibility final rule that will appear in a later Federal Register.

E. Summary of the Supplemental Notice and the Notice of Availability of New Information

After the April 17 proposal appeared, EPA realized that some aspects of the technical standards needed to be clarified and that more public comment on these matters was needed. Consequently, EPA published a Supplemental Notice on December 23, 1987 (52 FR 48638). This Supplemental Notice dealt with four areas pertaining to the proposed technical requirements:

(1) Use of "static inventory control" to monitor used oil UST systems;

(2) A listing of substances subject to petroleum UST standards;

(3) Alternatives to release monitoring for piping and tanks protected from external corrosion; and

(4) An alternative definition of "flow-through process tank."

Public response concerning these issues is discussed in later sections of this preamble.

On March 31, 1988, EPA published a Notice of Availability of additional information for public comment. It announced the availability of information pertaining to several technical areas of the proposed rule including general operating requirements, release detection and tank closure. This new information was submitted by commenters, gathered in meetings or conferences and produced by Agency research programs. Few public comments were provided concerning these documents.

F. Influences on the Final Rule

In the preamble to the April 17 proposal, the Agency discussed the scope and nature of the problem posed by UST systems and several important influences on the development of the proposal (52 FR 12665-12671). Today's final rule builds on that earlier information and has benefited from numerous comments provided by the public on the issues highlighted in the proposal. The following section briefly discusses several areas that have received further consideration from EPA in the development of today's final rules.

1. Scope of the Problem

The preamble to the proposed rule (52 FR 12665) presented estimates of the number of leaking UST systems based on EPA studies, local government experiences, and industry estimates. Among the statistics cited were the percentage of systems failing tightness testing, the percentage of systems actually leaking, the correlation of tank age to failure, and the extent and impact of soil and ground-water contamination from USTs. After the proposed rule was issued, EPA completed an additional study of the causes of release from UST systems. This new study was placed into the public docket and announced as available for comment in the December 23, 1987, Supplemental Notice. This study, "Causes of Release from UST Systems," and the public comments on it were important in developing today's preamble and final rule.

a. Current Estimates of "Non-Tight" UST Systems. In the preamble to the proposal (52 FR 12665), EPA cited evidence that numerous UST systems are non-tight and may be leaking. This evidence was based largely on three studies: EPA's "Underground Motor Fuel Tanks: A National Survey" reported tank tightness testing results and found that 35 percent of over 450 tank systems surveyed nationwide failed tightness testing; Suffolk County's UST program data revealed that 26 percent of over 6,000 tank systems tested in this New York county failed; and a Chevron-sponsored testing program found that nearly 10 percent of over 3,000 of their UST systems failed.

Commenters responding to the proposal who had experience with tightness testing provided various claims that between 11 and 48 percent of existing UST systems failed under test conditions. In an EPA-sponsored meeting, a group of experienced, independent installation contractors expressed their expert judgment that increased awareness of the UST problem, use of better tanks, and use of better installation and maintenance procedures have decreased the probability of present-day systems testing non-tight to about 20 percent, in contrast to the 50 percent of UST installations they believed would have tested non-tight several years ago.

After publication of the proposal, EPA studied several additional pieces of information concerning causes of release from UST systems. For example, EPA further reviewed the records of over 10,000 tightness test results from local UST programs (in Suffolk County, New York; Austin, Texas; and San Diego, California). EPA also analyzed an extensive and detailed historical set of records from a Texas tank testing company (the Service Station Testing Company of San Antonio, Texas). The EPA-sponsored report, "Causes of Release from UST Systems," is based on all these data and concludes that approximately 25 percent of existing UST systems are found to be non-tight when tested using current methods and that loose tank fittings or faulty piping causes 84 percent of these tightness test failures. Figure 2 summarizes the Agency's findings concerning the causes-of-release profile as derived from tank testing results and documented follow-up at over 10,000 UST systems conducted nationwide.


b. Estimated UST Systems with Releases. Current indications concerning the number of UST systems nationwide that have had releases in the past or are now leaking are less precise than the tank tightness data, but the Agency believes the information that is available is significant nonetheless. As discussed in the preamble to the proposed rule, in many places in the nation that are still without state or local UST regulatory programs today, release detection only takes place when someone sees or smells the release (52 FR 12665). These historical data suggest that only about 10 percent of release incidents in these areas are discovered by inventory controls or mechanical release detection methods. Although the large number of incidents that are known to threaten or have contaminated ground-water wells is significant, it is not an accurate prediction of leaking UST systems because most UST systems are not located near wells.

As the result of aggressive UST monitoring programs in two states, over 5,000 UST sites in California and 3,000 sites in Florida have been identified as having had releases during the past three years. These recent discoveries already exceed 10 percent of the number of UST sites in each of these states, and the number of releases identified in just these two states may soon exceed the cumulative total of releases reported to all the states up until 1985 (see 52 FR 12665). At a more local level, UST system programs in Dade County, Florida, and San Jose, California, have also identified (through required release detection and system closure procedures) that well over 10 percent of their UST facilities have had some noticeable or significant releases into the surrounding soil and groundwater. Thus, the initial findings by state and local UST regulatory programs that are particularly aggressive appear to corroborate an industry-sponsored study (than previously cited Chevron investigation) that found approximately 10 percent of their 1,000 UST facilities located throughout the South and Southwest of the United States have had an adverse impact on nearby ground water in the form of released product floating on top of the ground-water table (see 52 FR 12666).

Public comments received in response to the proposal concerning this subject are not conclusive. Some industry sources provided very low estimates, claiming that from 0 to 3 percent of UST systems have had releases. Others claimed the actual number of leaking tank systems could be as high as 50 percent in some areas. Many estimated that the actual range is somewhere between 8 to 20 percent of UST sites, and the average of all estimates reported by commenters falls into this range. As shown in the previous figure (Figure 2), approximately 25 percent of all systems are now testing as non-tight. About 15 percent of the sites whose systems were tested actually proved to have a leak under normal operating conditions (tank and delivery piping leaks), and this proportion falls within the above-estimated range provided by the commenters.

Some commenters attempted to provide additional insight into the relationship between tank age and failure by indicating that tanks begin to fail tightness tests (and leak) at a much greater frequency after 12 years. In addition, the recent EPA causes-of-release study includes one study that indicated 10 to 13 percent of the tanks that are 12 to 13 years old were non-tight. This is more than double the proportion of non-tight tanks tested in other age groups. In another study, of the tanks actually found to be leaking, 42 percent of the leakers were 15 to 20 years old, and 30 percent of the leakers were 10 to 15 years old. All of the tanks that were leaking were made of bare steel. This demonstrates that the critical age in a typical unprotected steel tank's life is the period between 10 to 20 years of age when breakthrough from corrosion is most likely to begin.

Some commenters disputed the severity of the threat posed to the nation by leaking UST systems, which was discussed in the preamble to the proposed rule. For example, one suggestion was that the data presented in the preamble indicate less than 0.008 percent of the total area of the United States is affected by contamination due to leaking UST systems. In general, EPA finds this line of reasoning unpersuasive. In particular, this argument ignores, (1) that population density in the nation is not uniform (with most areas being sparsely populated or unpopulated); (2) that tank systems are generally located near populated areas to provide the fuel for these centers of human activity and; (3) that there are numerous documented cases of drinking water wells that have been threatened or already destroyed by leaking UST systems nationwide. The dispersal of leaked contaminants within ground-water aquifers can also affect an area many times larger than the soil-contaminated area. Further information gathered over the coming years of UST program implementation will ascertain the full magnitude of the impacts that leaking UST systems pose in terms of contamination to the nation's environment, but EPA concludes that the evidence collected to date, including the information provided by commenters on the proposal, clearly supports the need for today's final rules.

2. New Cause-of-Release Information

EPA's new information concerning releases from UST systems comes primarily from public comment and an EPA-sponsored study ("Causes of Release from UST Systems") that was made available for public comment in the December 23, 1987, Supplemental Notice. Review of this information has resulted in the following findings concerning the major causes of releases today:

o Most releases do not come from the tank portion of UST systems, because piping releases occur twice as often as tank releases;

o Spills and overfills are the most common causes of releases;

o Various nonoperational UST components at the top of USTs are loose and leak in the event of overfills;

o Although the older bare steel tanks fail primarily by corrosion, the "new generation" USTs (i.e., coated and cathodically protected steel, fiberglass-clad steel, and fiberglass tanks) have nearly eliminated failure induced by external corrosion;

o Corrosion, poor installation techniques and workmanship, accidents, and natural events (e.g., frost heaves) are the four major causes of failure for piping; and

o When piping fails, pressurized systems pose a significant added threat of sudden, large releases.

Thus, the major causes of releases from UST systems are due to failures of unprotected tanks, leaks in delivery piping, leaks from vent pipes and fittings on top of the tank, and spill and overfill errors. Comments received on the original proposal (52 FR 12665-12668) and the Supplemental Notice concerning causes of release generally tend to corroborate the above findings. The following information summarizes some of the most relevant findings that are important in guiding today's standard-setting.

a. Tanks. Most existing tanks are made of bare steel. Numerous tank failure histories indicate that when bare steel tanks fail they almost always do so because of external corrosion. Of all of the current causes of release, corrosion of bare steel (tanks and pipes) is by far the most important.

Tank manufacturers have responded to this problem with a "new generation" of tanks. Innovative tanks began to appear about 20 years ago in the United States in three basic forms: fiberglass-reinforced plastic (FRP); steel with a corrosion-resistant coating and cathodic protection; and steel-FRP composite. A dramatic acceleration in the use of new generation tanks occurred with the introduction of the federal law's "Interim Prohibition" three years ago. These protected tanks now are estimated to account for about 20 to 25 percent of existing USTs. Although "new" in terms of protective designs, some of each of the new types of tank systems have been in the ground for over 20 years. Reported failures observed in the field due to corrosion (or other reasons) are very rare.

Failures (leaks) at all existing FRP tanks appear to have occurred at less than a rate of 0.05 percent per year of the total FRP tanks installed nationwide. Many commenters and other sources support the field estimates collected by EPA that less than 0.5 percent of the total number of existing FRP tanks have ever leaked. Although some installation-related failures have occurred in the past, heightened installer awareness of proper practices and techniques appropriate to FRP technology, manufacturer-sponsored contractor education programs, and production quality assurance appear to be responsible for a consistently decreasing failure rate of FRP tanks. The most important reported failure mode for these tanks is improper installation practices.

One new tank type, the STI-P3, is a favorite of corrosion engineers. These steel tanks have an external noncorrodible coating and a factory-applied metal anode that sacrifices itself to protect any bare spots on the tank, and the tank vessel is electrically isolated from any attached piping. Very few failures have ever been reported, and those failures are due to installation damage or improper maintenance, not design. In Ontario, Canada, where STI-P3 tanks have been widely used, the number of tank releases due to corrosion is reported to be declining as old tanks are replaced with STI-P3 tanks.

The steel-FRP composite tanks have not been used as widely as either the FRP or coated and cathodically protected tanks described above. Approximately 65,000 have been installed in this country. No corrosion-related failures have been reported. Many commenters suggested that this type of tank has several advantages over both FRP and coated and cathodically protected steel tanks, such as durability, no need for maintenance, and an added barrier between the tank and the environment should the steel tank be breached by internal corrosion.

As the threat of external corrosion is reduced by new tank designs, internal corrosion may eventually become the primary cause of failure for steel tanks. Internal corrosion, however, occurs far less frequently and takes longer to manifest itself than external corrosion. Many commenters have reported problems with internal corrosion under the drop tube (i.e., fill pipe located within the tank) of steel tanks. Data submitted from the tank lining industry confirm these reports. The tank manufacturing industry, however, began to respond to this problem several years ago by including "striker plates" under all openings of their new tanks.

Lining tank interiors is another way to prevent releases due to internal and external corrosion. Tank interior lining has been employed by major corporations and small businesses both as a short-term solution for potentially leaking tanks and as a preventive measure for temporarily giving structurally sound, non-leaking existing tanks the same protection from corrosion-induced releases that "new generation" tanks have. Data indicate this to be a successful procedure for extending an existing tank's operational life. Even when employed in the absence of external cathodic protection, failure rates are reported to be very low, apparently because current industry consensus codes only recommend the use of lining when the tank shell is assessed to be able to withstand the expected rate of corrosion at the site (determined by assessing the tank's existing condition).

b. Piping. Most commenters rated delivery piping the most significant source of releases and reported releases occurring twice as frequently from piping as from bare steel tank releases. Two types of piping systems are commonly used: suction piping, which is used in low-volume applications where only a few dispensers are needed; and pressurized piping, which is used in high-volume applications where many dispensers are fed from one tank. Each piping system has unique advantages and disadvantages, discussed below.

Suction piping is considered by commenters to be safer than pressurized piping because it operates at less than atmospheric pressure. If the pipe develops a leak, air or ground water is usually drawn into the pipe instead of product leaking out. Commenters suggested, however, that suction piping systems do not operate efficiently in a number of settings, such as at high altitudes, in hot climates, or in high-volume delivery situations.

Pressurized piping systems reportedly are used at about 95 percent of new retail motor fuel system installations. If the delivery line is breached, free product is released until the pressure in the pipe equals the pressure outside the pipe. Without add-on instrumentation or devices, large volumes of product can be pushed out of breaches in the piping when product is delivered to the pump. Pressurized piping simply pushes more volume to meet this increase in demand, releasing large amounts of product quickly into the environment.

Comments received by EPA indicate that the releases from pressurized piping systems can be catastrophic in the absence of monitoring and automated pump flow restriction devices. Incidents involving releases of thousands of gallons have been reported to EPA by experienced installers. It is estimated that at least 70 percent of the volume of product lost through pressurized pipe releases could be avoided by retrofitting each line with a simple, inexpensive, continuous in-line pressure monitor that automatically restricts product flow in the presence of a significant line leak.

Both suction and pressurized piping are often damaged by external corrosion. Cathodic protection of steel piping would significantly reduce corrosion failures. Presently, most steel piping is protected by galvanizing and coating, or coating and wrapping. The threaded portions at joints are the most common failure points because the protection is removed from them while threading and is never replaced. In these cases, cathodic protection would reduce joint failures. Other joint failures result from untightened joints, cross-threaded joints, or improperly made joints. Improving the installer's education and skills in the complex task of pipe installation would reduce these piping failures.

Also, installers and others have estimated that piping is damaged 10 percent of the time at new installations between the installation of equipment and completion of paving. They strongly recommend that a test of new equipment before start-up is essential as a sound practice, particularly with pressurized piping.

Natural forces and accidents also cause piping failures. The piping is near the surface of the ground and, thus, subject to frost heaves and overloading. In addition, the starting and stopping of product delivery causes the piping to move and shift. This eventually causes joint failure in many piping systems. "New generation" piping systems comparable to the "new generation" of tanks are under development but not widely used.

c. Nonoperational Components. Nonoperational components consist of tank bung holes, tank manholes, vent and fill lines, vapor recovery lines, and manifold piping (the piping used in connecting tanks together). These components, all located above the top of the tank, are called nonoperational because releases from these sources do not occur under normal operating conditions. Releases from them are usually unseen because they are underground. These releases are episodic and usually of small volume, because they only occur when the tank is overfilled or when manifolded tanks are filled through the piping connecting the tanks together. Generally, when an overfill occurs, the volume of product contained in the fill tube above the loose nonoperational component will be forced out into the environment until the product level in the UST drops below the leaking component. These leaking, nonoperational components are reported to be most often caused by improper installation practices, such as loose bung hole plugs not being tightened at installation or vent lines being handtightened on top of the tank.

Two solutions are available to stop this type of release: either ensure proper installation of these different types of fittings or eliminate overfills. Elimination of overfilling of the tank is the surest remedy and is probably the easiest to accomplish with overfill shutoff devices now widely available. Most releases associated with nonoperational components would be prevented if overfills were successfully eliminated.

d. Spills and Overfills. In addition to episodic releases from nonoperational components, there is an even more prevalent source of release that takes place at the tank fill port during tank filling. Although usually small in volume, spill and overfill releases are probably the most common causes of release from UST systems. These releases usually occur at the surface of the ground around the top of the fill pipe when the delivery truck's hose is disconnected from the fill pipe. Most of these releases go unreported due to the typically small volume of product lost (generally, less than the volume of the delivery truck's hose). Most excavated bare steel tanks, however, show evidence of spilled material, such as dissolved asphalt coating near the fill pipe. Regulatory officials in Dade County (Florida) cite spills and overfills as the primary cause of release--45 percent of reported releases. These surface releases are at least twice as numerous as tank or piping releases.

Spills most often occur at the fill pipe opening when the delivery truck's hose is disconnected, usually releasing only a few gallons. Overfills occur far less frequently but usually release much larger volumes. Overfills generally result in a release from loose, nonoperational components located above the tanks (as discussed in the previous section ), or from the top of the tank's vent pipe as product is forced out during overfilling of the system. Experienced installation contractors emphasize to EPA that the control of spills and control of overfills are two different problems and equipment that controls one may not control the other.

3. Industry Codes and Practices

In the preamble to the proposed rule (52 FR 12670), EPA identified numerous industry consensus codes and recommended practices that influenced the development of the proposed regulatory program. A table was provided listing several codes and practices concerning the proper management of UST systems that have been developed, mostly in the past decade, by industry associations, nationally recognized professional organizations, and independent testing laboratories. Since the proposal of the federal rule over a year ago, these consensus code-making groups and industry standard-setting activities have continued at an increased rate. (Refer to section IV.H.1. for a more detailed discussion clarifying the use of codes developed by nationally recognized organizations or independent testing laboratories.)

Table 1 reflects a sampling of the current status of this national consensus code-making network. The codes and standards marked with an asterisk have been reviewed, updated, or revised over the past year. For example, last summer, the American Petroleum Institute reviewed several of its recommended practices (e.g., API 1631 and 1615) and improved the guidance provided in these documents. In addition, several new codes are now under development or have been recently added. For example, the National Leak Prevention Association was formed and developed an industry consensus code for the interior lining of tanks (NLPA 631).


Table 1 provides a snapshot of the depth and scope of the collective wisdom that has developed over the past several years in the United States concerning the proper standards for UST systems. The recent updates and additions to this list of industry consensus codes and recommended practices reflect what appears to be a resurgence of interest in several areas of sound UST management practices. Numerous commenters on the proposal cited specific developments in these consensus codes and provided copies of some of the most recent codes that were added or updated and revised. In general, they provided a reminder to EPA that this improving body of knowledge should be understood and considered during the development of today's final rule.

4. Industry Trends

The nature and extent of the public response to the proposal has generally confirmed that a significant level of voluntary industry upgrading and replacement programs is already underway. The closure and replacement trends briefly discussed in the preamble to the proposal (52 FR 12671) were confirmed by some commenters. Numerous major oil companies, independent marketers of retail motor fuel, transportation fleet operators (government and private), and various manufacturers with USTs have clearly embarked on their own UST system management programs before promulgation of today's final rule.

One of the most encouraging trends has been the increasing use of protected UST systems nationwide. EPA estimates that there are currently over 450,000 UST systems in use today that are protected from external corrosion. This expanding use of an important and necessary approach to the prevention of releases has increased rapidly over the past three years, particularly after the Interim Prohibition went into effect in May 1985 (section 9003(g) of RCRA), making illegal the installation of bare steel UST systems. EPA estimates that, during the past three years over 50,000 new protected tanks were installed annually. To date, over 210,000 FRP tanks, 120,000 coated and cathodically protected tanks, and 65,000 steel-FRP composite tanks have been installed. In addition, an estimated 70,000 tanks have had their interiors lined to prevent releases, and another 50,000 UST systems are estimated to have been provided with field-installed corrosion protection systems. According to tank manufacturers, the number of new bare steel tanks installed over the past three years has dropped precipitously.

Early findings from several local UST programs visited by EPA staff since the proposal have confirmed the beginnings of an accelerated rate of closure of old, substandard UST systems (for example, in Suffolk County, New York; Austin, Texas; Dade County, Florida; and Sunnyvale, California). A rapid rate of closure appears to manifest itself when each of these, and several other, local UST programs initiate their release detection requirements. Thus, EPA's earlier projections that as much as three-fourths of the existing UST universe will be closed, upgraded, or replaced to the requirements of the Interim Prohibition within 10 years appears to be realistic (52 FR 12671). Today's final rule will further ensure that these positive changes are accomplished and increased.

EPA believes these are important developments because the successful implementation of this program depends a great deal on the regulated community's voluntary compliance. The Agency is convinced that the most wide-spread compliance will be facilitated by technically sound standards that are capable of easy implementation by a highly varied regulated community. Thus, as much as possible, the federal technical requirements must rely on familiar industry codes and build on recognized and effective trends occurring in the field of UST management that are consistent with protection of human health and the environment.

5. UST System Technology Development

As discussed in the preamble to the proposal, the array of technical control options available to address causes of releases, as well as response to releases that occur, also appears to be growing. Numerous commenters provided technical information to EPA about newly available equipment being marketed for use in the prevention, detection, and correction of releases from UST systems. New equipment for use either inside or outside of the UST system to detect releases from the underground tank or attached piping is being introduced for sale at many trade shows and industry fairs nationwide, including less expensive methods for providing secondary containment with interstitial monitoring for both existing and new UST systems.

Several new types of equipment capable of preventing spills and overfills appear to be less costly and more easily retrofitted to existing tanks than earlier models. For example, several new models of line leak detectors have recently been introduced to the UST market that are more sophisticated and sensitive than older types of equipment. In addition, numerous companies have contacted EPA about their development efforts in the area of retrofitting preventive devices onto existing tanks or refurbishing old tanks. Makers of nylon tanks are in the process of soliciting the approval of Underwriters Laboratories of Canada in that country. Finally, vapor gas vacuum extraction techniques for use in the subsurface cleanup of volatile substances are also under accelerated development and investigative use in several places nationwide. This cleanup technique has reportedly had widespread use in West Germany for a number of years.

In summary, business in the UST control technology sector appears to be booming, and invention is proceeding at a rapid pace. All this activity is a good indication that in the future, simpler, cheaper, and more dependable equipment will be produced to aid in the prevention, detection, and correction of releases. Experienced persons in UST management in the public and private sectors have told EPA staff that the current level of control technology development by far exceeds any previous efforts within this industry. In order to avoid interfering with this ongoing development of innovative and more environmentally protective new technologies, the Agency has chosen to write regulations that allow room for these new developments.

6. Leaking USTs Present a Unique Regulatory Challenge

EPA believes its approach to setting standards for UST systems on a national scale will have to be different from most national environmental programs because the UST problem is significantly different. This difference is mainly due to three factors: the large number of facilities to be regulated, the comprehensive scope of the regulations, and the nature of the regulated community.

The most significant problem is the sheer size of the regulated community. Nationally, over 700,000 UST facilities account for about 2 million UST systems, an average per state of about 14,000 UST facilities and 40,000 UST systems. Estimates indicate that roughly 75 percent of existing UST systems are unprotected from corrosion. In addition, because a relatively high proportion of UST facilities (10 to 30 percent) already have had a leak, or will soon leak unless measures are taken to upgrade them, the average number of leaking UST systems may range from 1,400 to 4,200 per state in the near future.

The large number of tank owners and tank systems has also led EPA to conclude that the final federal UST standards must include a phase-in period for certain requirements that apply to existing tank systems. Although all federal requirements are in effect immediately for new tanks, owners and operators will have additional time to upgrade existing tank systems to the corrosion protection standard for new UST systems, and to install release detection equipment for existing UST systems. This phased-in approach is needed to establish a reasonable schedule that recognizes the limited capability of 700,000 UST owners and supporting service and manufacturing industries to respond immediately to new national regulations, and provides sufficient flexibility for implementing agencies. The experience of states that have already been operating UST regulatory programs is that it takes several years for most owners or operators of existing UST systems to understand, plan, and arrange for the purchase, scheduling, and installation of necessary services and equipment required by the regulations. The phase-in approach also has the added benefit of allowing time for continued development and improvement of available technologies in the marketplace for prevention and detection of releases from UST systems (as discussed previously in this section of the preamble).

In addition, today's final rule establishes comprehensive requirements for the management of a wide range of UST systems. These final standards for UST systems are designed to reduce the number of releases of petroleum and hazardous substances, increase the ability to quickly detect and minimize the contamination of soil and ground water by such releases, and ensure adequate cleanup of contamination. To do this, the standards in some way must affect every phase of the life cycle of a storage tank system: selection of the tank system; installation, operation and maintenance; closure and disposal; and cleanup of the site in cases of product release. As a result, these standards must be technically adequate to ensure the wide array and needed level of improved performance when implemented. At the same time, these wide ranging requirements must be straightforward enough to be understood and to be carried out successfully hundred of thousands of times nationwide.

A third problem is the nature of the regulated community. A large proportion of USTs are owned by small businesses with $500,000 or less in total assets. For example, 72 percent of all retail motor fuel outlets are owned by small businesses. An important influence in the making of today's technical standards has been EPA's attempt to minimize the regulatory impact on small businesses without compromising the statutory requirements to protect human health and the environment. EPA's efforts to minimize the regulatory impact are discussed in a Regulatory Flexibility Analysis conducted for this rule, as specified by the Regulatory Flexibility Act of 1980, and a summary of that analysis is presented later in this preamble.

Specifically, the Agency is convinced that the national UST standards must be kept simple and implementable by state and local officials because many UST facilities are owned and operated as small local businesses, such as "Mom and Pop" gasoline service stations and convenience stores. These small entrepreneurs, who are used to operating their business with minimal regulation, will be significantly affected by environmental regulations for UST systems. The experience of state and local agencies with UST programs is that large national businesses that own tanks are generally willing and have already begun to comply with UST requirements. Owners of small businesses, however, generally need constant reminders and technical assistance to bring them into compliance. Given the nature of this regulated community, a regulatory program often will be most effectively carried out by the level of government closest to the problem, and thus able to respond quickly and to create a visible presence.

7. Emerging State and Local UST Programs and EPA's Approach to Regulation

Many states and localities have adopted requirements applicable to UST systems. Although these state and local requirements are diverse and vary in stringency, EPA believes that the formulation of federal standards should build upon the many effective state and local programs now in operation or about to begin operating in order to utilize this reservoir of accumulated UST experience in a way that can rapidly develop into a strong federal-state partnership for addressing this national concern. Section 9004 also indicates Congressional intent that states with effective programs are to play a major role in implementing the program.

At least 18 states and hundreds of local programs are currently addressing the ground-water contamination and cleanup problems posed by leaking UST systems through established regulatory programs. Several states, such as California, New York, and Florida, and local UST programs such as those in Suffolk County (New York), Dade County (Florida), and Austin (Texas), have established specific UST system regulations that include standards for design, construction, and installation of new UST systems; closure, retrofitting, and repair of existing UST systems; and release detection and corrective action requirements for all UST systems. EPA believes this type of state and local UST program activity nationwide will increase significantly with today's promulgation of EPA's technical standards. Similar to the experiences in the three lead states identified above, other states will begin to wrestle for the first time with the reality of how to implement their UST programs. As the dangers posed by existing UST systems become more widely known, local UST programs and involvement should increase significantly over current levels.

Given the large number of UST facilities, tank systems, and potential cleanups needed, EPA is convinced that many aspects of this regulatory program will be most effectively carried out at the state level of government. Local government involvement in this regulatory program will be important. For example, a small city with about 700 facilities and 2,000 tank systems within its jurisdiction should be able to implement a manageable regulatory program. If each of those 700 facilities installs one new tank during the next five years, that would be about three installations per week. If that small city requires a city inspector to be present at each installation, an inspector must be in the field three times a week for this task alone. In addition, the inspector could be required to be present for periodic tank testings, closures, upgrading or retrofit, and cleanups.

Confronted with the above implementation realities, EPA has developed a more decentralized approach for addressing the realities of the national UST regulatory program. This approach is based on several critical factors. First, as more and more state and local governments become involved, the work of the UST program must be routinely carried out in thousands of jurisdictions nationwide. Several operating state and local UST programs already report that they are very busy "running the store," expressing surprise at the size of the regulated community, and that fairly simple tasks must be routinely repeated numerous times for the implementing agency to be successful in bringing UST systems into, and maintaining, compliance.

Second, visits to several state and local UST programs have shown that they have often developed their own unique requirements and methods of implementation, adapted to the types of tanks, physical environment, and regulated community with which they are concerned while they are, at the same time, all geared towards solving similar technical problems. They need the flexibility to continue and to improve upon approaches which address the specific environmental needs of their communities. They have common implementation problems, however, and have expressed the need for better technical aids, such as data management tools.

Third, many state and local governments that already implement UST programs report a significant level of visible on-site monitoring, requiring a constant "regulatory presence" to effectively ensure this regulated community's compliance with UST requirements. A significant environmental gain is achieved through implementation at the local level by these individual UST programs. Thus, improving their performance will produce maximum environmental benefits and best ensure the success of the UST program nationwide. As the head of the "distribution system" of UST-related technical information and implementation tools, EPA believes that its implementation efforts should be focused on serving the network of state and local programs through listening to their concerns and helping them solve implementation problems with tools that improve the effectiveness of their programs.

Finally, EPA believes that a more decentralized approach to the federal implementation of the UST program is needed to ensure real gains in protection of human health and the environment. Because there are so many UST sites nationwide, it would be very difficult to establish a credible federal implementation presence through compliance monitoring and enforcement at the federal level. A more realistic and effective approach is for EPA to provide support tools and guidance to state and local regulators that can be used to improve their programs' compliance performance.

In adopting this role, the Agency has recognized that it must not only establish sound national standards but, more importantly, must focus on improving the performance of the state and local implementing agencies. Approval of state programs to operate "in lieu of" the federal program takes on a new meaning under this approach because it becomes a basic soundness test to ensure that the work associated with implementation of these state or local requirements will, in fact, cause the needed level of improvement in UST system management when carried out by the regulated community. The requisite state enforcement authority and technical standards must be ensured and will be the focus for approval by EPA. Thus, overall successful performance and implementation of this new national program is less focused on implementing detailed, national technical standards than it is on establishing the national UST program in a way that ensures effective, environmentally protective programs at the grassroots level and improving the performance of these programs over time. EPA's final requirements for state program approval are presented in detail elsewhere in today's Federal Register.

Thus, in recognition of its approach to UST implementation, EPA has attempted to establish final technical standards that are protective of health and the environment but, at the same time, are simple, understandable and implementable by state and local officials. EPA also recognizes that there is often more than one proper way to address specific technical problems that are the focus of the final regulations. Therefore, the Agency has attempted to identify and offer as many effective alternative technical approaches as possible particularly where this flexibility can be applied in the future by the implementing agencies. In this way, the final technical requirements remain focused on the key environmental problems which the implementing agencies face. Promoting the network of state and local implementation is the best way to ensure that significant protection of human health and the environment will be achieved by today's final requirements.

G. Conclusions Since Proposal

EPA has drawn several conclusions from the influences discussed above and in the background sections of the preamble to the proposed rule (52 FR 12663-12671). Some of these conclusions support the direction and emphasis set forth in the proposal, and others indicate a need for change in the final rule.

Given the large size of the existing regulated universe and the proportion of these UST systems that have leaked or are presently leaking, there is a need to finalize today's rules as an important step resources. The number of sites needing significant cleanup due to a number of poor past UST management practices is expected to be in the tens of thousands nationwide.

Cause-of-release information related to unprotected tanks supports EPA's proposed approach for upgrading of unprotected tanks. The new information, however, indicates a need for more frequent monitoring of unprotected tanks than was proposed. By contrast, protected tanks appear to need less frequent monitoring than proposed. Also, pressurized piping systems need more stringent monitoring than was proposed.

Increased activity in the review and improvement of national consensus codes supports EPA's proposed reliance on these codes as providing the most up-to-date consensus practices and expertise concerning what constitutes proper UST system management. The nationwide increase in the use of protected systems, the recent number of tank closures, and the development of new prevention, detection, and corrective action technologies are encouraging. The final rules must be designed to foster and take advantage of these trends. They must be simple and easily implementable by the regulated community to ensure the maximum level of voluntary participation by tank system owners and operators. The Agency continues to believe that the size and nature of this regulated universe presents several unique regulatory challenges that necessitate the phase-in of some of the requirements for existing UST systems to ensure that genuine implementation is accomplished.

The continuing and rapid emergence of numerous state and local UST programs is expected and will be encouraged by EPA because this is where the "real work" of this new national program must actually take place. The Agency's approach to UST program implementation must start with a technically sound set of national standards. These requirements, however, must be kept simple and implementable because most improvements in actual UST performance (and protection of human health and the environment) are expected to be achieved by working closely with state and local governments over time to increase the level of the effectiveness of their UST programs.

[ Back to Section I. ][ Ahead to Section III. ]

Top of page

Local Navigation

Jump to main content.